Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach [1 ed.] 1119603617, 9781119603610

Citation preview

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach Edited by

Ali Gholamrezanezhad, MD

Associate Professor of Clinical Radiology, Keck School of Medicine Universityof Southern California Los Angeles, CA, USA

Majid Assadi, MD, MSc

Professor, Department of Radiology, School of Medicine Director, Nuclear Medicine and Molecular Imaging Research Center Bushehr University of Medical Sciences Bushehr, Iran

Hossein Jadvar, MD, PhD, MPH, MBA

Professor of Radiology, Urology, and Biomedical Engineering Keck School of Medicine and Viterbi School of Engineering University of Southern California Los Angeles, CA, USA

This edition first published 2023 © 2023 John Wiley & Sons Ltd All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions. The right of Ali Gholamrezanezhad, Majid Assadi, and Hossein Jadvar to be identified as the authors of the editorial material in this work has been asserted in accordance with law. Registered Offices John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com. Wiley also publishes its books in a variety of electronic formats and by print-on-demand. Some content that appears in standard print versions of this book may not be available in other formats. Trademarks: Wiley and the Wiley logo are trademarks or registered trademarks of John Wiley & Sons, Inc. and/or its affiliates in the United States and other countries and may not be used without written permission. All other trademarks are the property of their respective owners. John Wiley & Sons, Inc. is not associated with any product or vendor mentioned in this book. Limit of Liability/Disclaimer of Warranty The contents of this work are intended to further general scientific research, understanding, and discussion only and are not intended and should not be relied upon as recommending or promoting scientific method, diagnosis, or treatment by physicians for any particular patient. In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of medicines, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each medicine, equipment, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. Library of Congress Cataloging-in-Publication Data applied for ISBN: 9781119603610 (hardback) Cover Design: Wiley Cover Images: © semakokal/iStock/Getty Images, wenht/iStock/Getty Images Set in 9.5/12.5pt STIXTwoText by Straive, Pondicherry, India

Dedicated to Mojgan, Donya, and Delara, with love. . . Hossein Jadvar To my family especially my mother, Maryam, my wife, Moloud, my sons, Arian and Aiden. For their endless sacrifices they have made to make my life most rewarding Majid Assadi To those contributed to my education and excellence, especially my mother (the best teacher I have ever had), Fatemeh, my brother, Hadi, my wife, Farzaneh, and my son, Adrian and To my patients, those who I served with the deepest gratitude and appreciation. Ali Gholamrezanezhad

vii

Contents List of Contributors Preface xvii

x

 1

Introduction to Correlative Imaging: What Radiologists and Nuclear Medicine Physicians Should Know on Hybrid Imaging 1 Prathamesh V. Joshi, Alok Pawaskar, and Sandip Basu

 2

Basic Principles of Hybrid Imaging 30 Leda Lorenzon, M. Bonelli, A. Fracchetti, and P. Ferrari

 3

Cross-sectional Correlate for Integrative Imaging (Anatomical Radiology) 52 Antonio Jesús Láinez Ramos-Bossini, Ángela Salmerón-Ruiz, José Pablo Martínez Barbero, José Pablo Martín Molina, José Luis Martín Rodríguez, Genaro López Milena, and Fernando Ruiz Santiago

 4

Radiopharmaceuticals 133 Ferdinando Calabria, Mario Leporace, Rosanna Tavolaro, and Antonio Bagnato

 5

Diseases of the Central Nervous System 163 Hiroshi Matsuda, Eku Shimosegawa, Yoko Shigemoto, Noriko Sato, Hiroyuki Fujii, Fumio Suzuki, Yukio Kimura, and Atsuhiko Sugiyama

 6

PET Imaging in Gliomas: Clinical Principles and Synergies with MRI 194 Riccardo Laudicella, C. Mantarro, B. Catalfamo, P. Alongi, M. Gaeta, F. Minutoli, S. Baldari, and Sotirios Bisdas

 7

Diseases of the Head and Neck 219 Florian Dammann and Jan Wartenberg

 8

The Role of Noninvasive Cardiac Imaging in the Management of Diseases of the Cardiovascular System 257 Ahmed Aljizeeri and Mouaz H. Al-Mallah

 9

Vascular System 285 Ahmad Shariftabrizi, Khalid Balawi, and Janet H. Pollard

10

Diseases of the Pulmonary System 308 Murat Fani Bozkurt and Bilge Volkan-Salanci

11

Thoracic Malignancies 333 Sanaz Katal, Thomas G. Clifford, Kanhaiyalal Agrawal, and Ali Gholamrezanezhad

viii

Contents

12

A Correlative Approach to Breast Imaging 351 Shabnam Mortazavi, Sonya Khan, Kathleen Ruchalski, Cory Daignault, and Jerry W. Froelich

13

Correlative Imaging of Benign Gastrointestinal Disorders 383 Mariano Grosso, Michela Gabelloni, Emanuele Neri, and Giuliano Mariani

14

Gastrointestinal Malignancies 407 Janet H. Pollard, Paul A. DiCamillo, Ayca Dundar, Sarah L. Averill, and Yashant Aswani

15

Hepatobiliary Imaging Janet H. Pollard

16

Correlative Imaging in Endocrine Diseases 485 Sana Salehi, Farshad Moradi, Doina Piciu, Hojjat Ahmadzadehfar, and Ali Gholamrezanezhad

17

Correlative Imaging in Neuroendocrine Tumors 512 Ameya Puranik, Sonal Prasad, Indraja D. Devi, and Vikas Prasad

18

Nephro-urinary Tract Pathologies: A Correlative Imaging Approach 521 Salar Tofighi, Thomas G. Clifford, Saum Ghodoussipour, Peter Henry Joyce, Meisam Hoseinyazdi, Maryam Abdinejad, Saeideh Najafi, Fahad Marafi, and Russell H. Morgan

19

Correlative Approach to Prostate Imaging Soheil Kooraki and Hossein Jadvar

20

Correlative Imaging of the Female Reproductive System 554 Sanaz Katal, Akram Al-Ibraheem, Fawzi Abuhijla, Ahmad Abdlkadir, Liesl Eibschutz, and Ali Gholamrezanezhad

21

Musculoskeletal Imaging 577 George R. Matcuk, Jr., Jordan S. Gross, Dakshesh B. Patel, Brandon K. K. Fields, Dorian M. Lapalma, and Daniel Stahl

22

Spine Disorders: Correlative Imaging Approach 625 Azadeh Eslambolchi, Amit Gupta, Jay Acharya, Christopher Lee, and Kaustav Bera

23

Osteoporosis: Diagnostic Imaging and Value of Multimodality Approach in Differentiating Benign Versus Pathologic Compression Fractures 659 Daniela Garcia, Shambo Guha Roy, and Reza Hayeri

24

Emergency Radiology 671 Sean K. Johnston, Russell Flato, Peter Hu, Peter Henry Joyce, and Andrew Chong

25

Correlative Imaging of Pediatric Diseases 693 Seth J. Crapp, Rachel Pevsner Crum, Nolan Altman, Jyotsna Kochiyil, Eshani Sheth, and Caldon J. Esdaille

26

Infection/Inflammation Imaging 717 Christopher J. Palestro and Charito Love

27

Imaging the Lymphatic System 747 Girolamo Tartaglione, Marco Pagan, Francesco Pio Ieria, Giuseppe Visconti, and Tommaso Tartaglione

28

Lymphoma and Myeloma Correlative Imaging 772 Pavel Gelezhe, Sergey Morozov, Anton Kondakov, and Mikhail Beregov

456

533

Contents

29

Clinical Application of PET/MRI 788 Laura Evangelista, Paolo Artoli, Paola Bartoletti, Antonio Bignotto, Federica Menegatti, Marco Frigo, Stefania Antonia Sperti, Laura Vendramin, and Diego Cecchin

30

68

31

Artificial Intelligence in Diagnostic Imaging 826 Martina Sollini, Daniele Loiacono, Daria Volpe, Alessandro Giaj Levra, Elettra Lomeo, Edoardo Giacomello, Margarita Kirienko, Arturo Chiti, and Pierluca Lanzi

32

Radionuclide Therapies and Correlative Imaging Ashwin Singh Parihar and Erik Mittra

Ga-FAPI, a Twin Tracer for 18F-FDG in the Era of Evolving PET Imaging 814 Reyhaneh Manafi-Farid, GhasemAli Divband, HamidReza Amini, Thomas G. Clifford, Ali Gholamrezanezhad, Mykol Larvie, and Majid Assadi

Index 871

838

ix

x

List of Contributors Maryam Abdinejad Department of Radiology, Namazi Hospital, Shiraz, Iran Department of Nuclear Medicine, Namazi Hospital, Shiraz, Iran Ahmad Abdlkadir Department of Nuclear Medicine, King Hussein Cancer Center, Amman, Jordan Fawzi Abuhijla Department of Radiation Oncology, King Hussein Cancer Center, Amman, Jordan Jay Acharya Radiology, Keck School of Medicine of USC, HCCII Lower Level Radiology, Los Angeles, CA, USA

P. Alongi Unit of Nuclear Medicine, Fondazione Istituto G. Giglio, Cefalù, Italy Nolan Altman Nicklaus Children’s Hospital, Miami, FL, USA HamidReza Amini Khatam PET-CT Center, Khatam Hospital, Tehran, Iran Paolo Artoli Nuclear Medicine Unit, Department of Medicine, University of Padua, Padua, Italy

Kanhaiyalal Agrawal Department of Nuclear Medicine, All India Institute of Medical Sciences, Bhubaneswar, India

Majid Assadi Department of Radiology, School of Medicine, Nuclear Medicine and Molecular Imaging Research Center Bushehr University of Medical Sciences Bushehr, Iran

Hojjat Ahmadzadehfar Department of Nuclear Medicine, Klinikum Westfalen, Dortmund, Germany

Yashant Aswani University of Iowa, Carver College of Medicine, Iowa City, IA, USA

Akram Al-Ibraheem Department of Nuclear Medicine, King Hussein Cancer Center, Amman, Jordan Ahmed Aljizeeri King Abdulaziz Cardiac Center, Riyadh, Saudi Arabia King Saud bin Abdulaziz University for Health Sciences, Riyadh, Saudi Arabia King Abdullah International Medical Research Center, Riyadh, Saudi Arabia Mouaz H. Al-Mallah Houston Methodist DeBakey Heart & Vascular Center, Houston Methodist Hospital, Houston, TX, USA

Sarah L. Averill University of Iowa, Carver College of Medicine, Iowa City, IA, USA Iowa City Veterans Administration Healthcare System, Iowa City, IA, USA Antonio Bagnato Department of Nuclear Medicine and Theranostics, “Mariano Santo” Hospital, Cosenza, Italy Khalid Balawi University of Iowa Carver College of Medicine, Iowa City, IA, USA

List of Contributors

S. Baldari Department of Biomedical Sciences and Morphological and Functional Imaging, Nuclear Medicine Unit, University of Messina, Messina, Italy

B. Catalfamo Department of Biomedical Sciences and Morphological and Functional Imaging, Nuclear Medicine Unit, University of Messina, Messina, Italy

José Pablo Martínez Barbero Department of Radiology, Virgen de las Nieves University Hospital, University of Granada, Granada, Spain

Diego Cecchin Nuclear Medicine Unit, Department of Medicine, University of Padua, Padua, Italy

Paola Bartoletti Nuclear Medicine Unit, Department of Medicine, University of Padua, Padua, Italy

Arturo Chiti Department of Biomedical Sciences, Humanitas University, Milan, Italy IRCCS Humanitas Research Hospital, Milan, Italy

Sandip Basu Radiation Medicine Centre, Bhabha Atomic Research Centre, Tata Memorial Centre Annexe, Parel, Mumbai, Maharashtra, India Homi Bhabha National Institute, Mumbai, Maharashtra, India Kaustav Bera Case Western Reserve University School of Medicine, University Hospital Cleveland Medical Center, Cleveland, OH, USA Mikhail Beregov Federal Center for Cerebrovascular Pathology and Stroke, Department of Radiology and Functional Diagnostics, Moscow, Russia Antonio Bignotto Nuclear Medicine Unit, Department of Medicine, University of Padua, Padua, Italy Sotirios Bisdas Department of Brain Repair and Rehabilitation, UCL Queen Square Institute of Neurology, University College London, London, UK Lysholm Department of Neuroradiology, The National Hospital for Neurology and Neurosurgery, UCLH NHS Foundation Trust, London, UK

Andrew Chong Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA Thomas G. Clifford Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA Seth J. Crapp Pediatric Teleradiology Partners, Miami, FL, USA Rachel Pevsner Crum Nicklaus Children’s Hospital, Miami, FL, USA Cory Daignault Minneapolis VA Medical Center, Minneapolis, MN, USA Florian Dammann Department of Diagnostic, Interventional and Pediatric Radiology, Inselspital, University Hospital Bern, Switzerland Indraja D. Devi Department of Nuclear Medicine, Tata Memorial Hospital, Homi Bhabha National Institute (HBNI), Mumbai, Maharashtra, India

M. Bonelli Department of Medical Physics, Central Hospital of Bolzano, Bolzano, Italy

Paul A. DiCamillo University of Iowa, Carver College of Medicine, Iowa City, IA, USA

Ferdinando Calabria Department of Nuclear Medicine and Theranostics, “Mariano Santo” Hospital, Cosenza, Italy

GhasemAli Divband Nuclear Medicine Center, Jam Hospital, Tehran, Iran Khatam PET-CT Center, Khatam Hospital, Tehran, Iran

xi

xii

List of Contributors

Ayca Dundar University of Iowa, Carver College of Medicine, Iowa City, IA, USA Liesl Eibschutz Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA Caldon J. Esdaille Howard University College of Medicine, Washington, DC, USA Azadeh Eslambolchi Pediatric Radiology Section, Mallinckrodt Institute of Radiology, Washington University in St Louis, School of Medicine, St. Louis, MO, USA Laura Evangelista Nuclear Medicine Unit, Department of Medicine, University of Padua, Padua, Italy Murat Fani Bozkurt Department of Nuclear Medicine, Hacettepe University Faculty of Medicine, Ankara, Turkey P. Ferrari Department of Medical Physics, Central Hospital of Bolzano, Bolzano, Italy Brandon K. K. Fields Keck School of Medicine, University of Southern California, Los Angeles, CA, USA Russell Flato Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA A. Fracchetti Department of Medical Physics, Central Hospital of Bolzano, Bolzano, Italy Marco Frigo Nuclear Medicine Unit, Department of Medicine, University of Padua, Padua, Italy Jerry W. Froelich Radiology, University of Minnesota, Minneapolis, MN, USA Hiroyuki Fujii Department of Radiology, National Center of Neurology and Psychiatry, Kodaira, Japan

Michela Gabelloni Diagnostic and Interventional Radiology, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy M. Gaeta Section of Radiological Sciences, Department of Biomedical Sciences and Morphological and Functional Imaging, University of Messina, Messina, Italy Daniela Garcia Department of Radiology, Mercy Catholic Medical Center, Darby, PA, USA Pavel Gelezhe Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department, Moscow, Russia European Medical Center, Radiology Department, Moscow, Russia Saum Ghodoussipour Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ, USA Section of Urologic Oncology, Rutgers Cancer Institute of New Jersey, New Brunswick, NJ, USA Ali Gholamrezanezhad Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA Edoardo Giacomello Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milano, Italy Jordan S. Gross Department of Radiology, University of California, Los Angeles, Los Angeles, CA, USA Mariano Grosso Regional Center of Nuclear Medicine, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy Amit Gupta Radiology, Medicine and Biomedical Engineering, Case Western Reserve University School of Medicine, Cleveland, OH, USA Cancer Imaging Program, Case Comprehensive Cancer Center, Cleveland, OH, USA Diagnostic Radiography, University Hospital Cleveland Medical Center, Cleveland, OH, USA

List of Contributors

Reza Hayeri Department of Radiology, Mercy Catholic Medical Center, Darby, PA, USA Meisam Hoseinyazdi Shiraz University of Medical Sciences, Shiraz, Iran Department of Radiology, Namazi Hospital, Shiraz, Iran Peter Hu Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA Hossein Jadvar Professor of Radiology, Urology, and Biomedical Engineering, Keck School of Medicine and Viterbi School of Engineering, University of Southern California, Los Angeles, CA, USA Sean K. Johnston Department of Radiology, Division of Emergency Radiology, Keck School of Medicine of USC, LAC+USC Medical Center, Los Angeles, CA, USA Prathamesh V. Joshi Department of Nuclear Medicine & PET-CT, Kamalnayan Bajaj Hospital, Aurangabad, Maharashtra, India Radiation Medicine Centre, Bhabha Atomic Research Centre, Tata Memorial Centre Annexe, Parel, Mumbai, Maharashtra, India Peter Henry Joyce Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA Sanaz Katal Nuclear Medicine Fellow, Medical Imaging Department, St Vincent’s Hospital Melbourne, Australia Sonya Khan Los Angeles and Veterans Administration, Greater Los Angeles Healthcare Systems, University of California, Los Angeles, CA, USA

Jyotsna Kochiyil Mount Sinai Medical Center, Miami Beach, FL, USA Anton Kondakov Central Clinical Hospital of the Russian Academy of Sciences, Nuclear Medicine Department, Moscow, Russia Pirogov Russian National Research Medical University, Department of Radiology and Radiation Therapy, Moscow, Russia Soheil Kooraki Department of Molecular and Medical Pharmacology, David Geffen School of Medicine at UCLA, University of California, Los Angeles, CA, USA Pierluca Lanzi Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milano, Italy Dorian M. Lapalma Department of Radiology, University of Southern California, Los Angeles, CA, USA Keck School of Medicine, University of Southern California, Los Angeles, CA, USA Mykol Larvie Department of Radiology, Cleveland Clinic, Cleveland, OH, USA Riccardo Laudicella Department of Biomedical Sciences and Morphological and Functional Imaging, Nuclear Medicine Unit, University of Messina, Messina, Italy Christopher Lee Keck School of Medicine of USC, HCCII Lower Level Radiology, Los Angeles, CA, USA Mario Leporace Department of Nuclear Medicine and Theranostics, “Mariano Santo” Hospital, Cosenza, Italy Alessandro Giaj Levra IRCCS Humanitas Research Hospital, Milan, Italy

Yukio Kimura Department of Radiology, National Center of Neurology and Psychiatry, Kodaira, Japan

Daniele Loiacono Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milano, Italy

Margarita Kirienko Department of Nuclear Medicine, Istituto Nazionale per lo Studio e la Cura dei Tumori, Milano, Italy

Elettra Lomeo IRCCS Humanitas Research Hospital, Milan, Italy

xiii

xiv

List of Contributors

Leda Lorenzon Department of Medical Physics, Central Hospital of Bolzano, Bolzano, Italy Charito Love Radiology, Albert Einstein College of Medicine, Bronx, NY, USA Reyhaneh Manafi-Farid Research Center for Nuclear Medicine, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran C. Mantarro Department of Biomedical Sciences and Morphological and Functional Imaging, Nuclear Medicine Unit, University of Messina, Messina, Italy Fahad Marafi Jaber Al-Ahmad Center for Molecular Imaging, Kuwait City, Kuwait Giuliano Mariani Regional Center of Nuclear Medicine, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy José Pablo Martín Molina Department of Radiology, San Cecilio University Hospital, University of Granada, Granada, Spain George R. Matcuk, Jr. Department of Imaging, Cedars-Sinai Medical Center, Los Angeles, CA, USA Hiroshi Matsuda Integrative Brain Imaging Center, National Center of Neurology and Psychiatry, Kodaira, Japan Federica Menegatti Nuclear Medicine Unit, Department of Medicine, University of Padua, Padua, Italy Genaro López Milena Department of Radiology, Virgen de las Nieves University Hospital, University of Granada, Granada, Spain F. Minutoli Department of Biomedical Sciences and Morphological and Functional Imaging, Nuclear Medicine Unit, University of Messina, Messina, Italy

Erik Mittra Department of Diagnostic Radiology, Division of Nuclear Medicine & Molecular Imaging, Oregon Health & Science University, Portland, OR, USA Farshad Moradi Department of Radiology, Division of Nuclear Medicine, Stanford, CA, USA Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins Medical Institution, Baltimore, MD, USA Sergey Morozov Chief innovation officer, Osimis S.A., Belgium Shabnam Mortazavi Radiology, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA Saeideh Najafi Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA Emanuele Neri Diagnostic and Interventional Radiology, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy Marco Pagan Nuclear Medicine, Cristo Re Hospital, Rome, Italy Christopher J. Palestro Radiology, Donald & Barbara Zucker School of Medicine at Hofstra/Northwell, Hempstead, NY, USA Nuclear Medicine & Molecular Imaging, Northwell Health, New Hyde Park, NY, USA Ashwin Singh Parihar Department of Nuclear Medicine, Postgraduate Institute of Medical Education and Research, Chandigarh, India Mallinckrodt Institute of Radiology, Washington University School of Medicine, St Louis, MO, USA Dakshesh B. Patel Department of Radiology, University of Southern California, Los Angeles, CA, USA Keck School of Medicine, University of Southern California, Los Angeles, CA, USA

List of Contributors

Alok Pawaskar Radiation Medicine Centre, Bhabha Atomic Research Centre, Tata Memorial Centre Annexe, Parel, Mumbai, Maharashtra, India Department of Nuclear Medicine & PET-CT, Shri Siddhivinayak Ganapati Cancer Hospital, Miraj, Maharashtra, India Doina Piciu Department of Endocrine Tumors and Nuclear Medicine, Institute of Oncology Ion Chiricuta and University of Medicine Iuliu Hatieganu, Cluj-Napoca, Romania Francesco Pio Ieria Nuclear Medicine, Cristo Re Hospital, Rome, Italy Janet H. Pollard University of Iowa Carver College of Medicine, Iowa City, IA, USA Sonal Prasad Berlin Experimental Radionuclide Imaging Center, Berlin, Germany Department of Nuclear Medicine, CharitéUniversitaetsmedizin, Berlin, Germany Vikas Prasad Department of Nuclear Medicine, University Hospital, Ulm, Germany Ameya Puranik Department of Nuclear Medicine, Tata Memorial Hospital, Homi Bhabha National Institute (HBNI), Mumbai, Maharashtra, India Antonio Jesús Láinez Ramos-Bossini Department of Radiology, Virgen de las Nieves University Hospital, University of Granada, Granada, Spain José Luis Martín Rodríguez Department of Radiology, San Cecilio University Hospital, University of Granada, Granada, Spain Shambo Guha Roy Department of Radiology, Mercy Catholic Medical Center, Darby, PA, USA Kathleen Ruchalski Radiology, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA Sana Salehi Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA

Ángela Salmerón-Ruiz Department of Radiology, Virgen de las Nieves University Hospital, University of Granada, Granada, Spain Fernando Ruiz Santiago Department of Radiology, Virgen de las Nieves University Hospital, University of Granada, Granada, Spain Neuro-traumatology Hospital, Virgen de las Nieves University Hospital, School of Medicine, University of Granada, Granada, Spain Noriko Sato Department of Radiology, National Center of Neurology and Psychiatry, Kodaira, Japan Ahmad Shariftabrizi University of Iowa Carver College of Medicine, Iowa City, IA, USA Veterans Affair Medical Center, Iowa City, IA, USA Eshani Sheth Mount Sinai Medical Center, Miami Beach, FL, USA Yoko Shigemoto Department of Radiology, National Center of Neurology and Psychiatry, Kodaira, Japan Eku Shimosegawa Department of Molecular Imaging in Medicine, Osaka University Graduate School of Medicine, Suita, Japan Martina Sollini Department of Biomedical Sciences, Humanitas University, Milan, Italy IRCCS Humanitas Research Hospital, Milan, Italy Stefania Antonia Sperti Nuclear Medicine Unit, Department of Medicine, University of Padua, Padua, Italy Daniel Stahl Keck School of Medicine, University of Southern California, Los Angeles, CA, USA Atsuhiko Sugiyama Department of Neurology, Graduate School of Medicine, Chiba University, Chiba, Japan Fumio Suzuki Department of Radiology, National Center of Neurology and Psychiatry, Kodaira, Japan Girolamo Tartaglione Nuclear Medicine, Cristo Re Hospital, Rome, Italy

xv

xvi

List of Contributors

Tommaso Tartaglione Radiology, IDI-IRCCS, Rome, Italy Rosanna Tavolaro Department of Nuclear Medicine and Theranostics, “Mariano Santo” Hospital, Cosenza, Italy Salar Tofighi Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA Laura Vendramin Nuclear Medicine Unit, Department of Medicine, University of Padua, Padua, Italy Giuseppe Visconti Plastic Surgery, Lymphedema Center, A. Gemelli Hospital, Sacro Cuore Catholic University, Rome, Italy

Bilge Volkan-Salanci Department of Nuclear Medicine, Hacettepe University Faculty of Medicine, Ankara, Turkey Daria Volpe Department of Biomedical Sciences, Humanitas University, Milan, Italy IRCCS Humanitas Research Hospital, Milan, Italy Jan Wartenberg Department of Nuclear Medicine, Inselspital, University Hospital Bern, Switzerland

xvii

Preface Medical imaging has come a long way since the discovery of X-rays by Wilhelm Rontgen, for which he received the Nobel Prize in 1901. For over a century, medical imaging has evolved remarkably with discoveries and the development of innovative technologies which in combination with major strides in understanding the biology of health and disease have contributed significantly to the concept of precision health and precision medicine. These milestones include, but are not limited to, the discovery of radioactivity and positron and technical developments of the radiotracer concept, cyclotron, computed tomography (CT), ultrasonography (US), magnetic resonance imaging (MRI), single photon computed tomography (SPECT), and positron emission tomography (PET). Advances in computer technology have also provided opportunities for sophisticated incorporation of radiomics, artificial intelligence, and deep learning (AI/DL) algorithms in medical imaging. Over the past decade, it has become clear that hybrid imaging (e.g. PET/CT, PET/MRI, SPECT/CT) provides a broader view of disease that was unavailable previously. For example, it is now recognized that a small lymph node may harbor a tumor while a large lymph node may be benign. Another example is visualization of tumor infiltration in marrow space without concordant structural abnormalities. Such comprehensive information provides opportunities for enhanced imaging assessment of the patient, which has been demonstrated to impact clinical management and improve patient outcome. The editors of this book have assembled an international team of expert imaging specialists to compile comprehensive coverage of correlative imaging in the domains of

diagnostic radiology and nuclear medicine, addressing all major organ systems and major diseases (cardiovascular, neurologic, oncologic, infection, and inflammation, in both adults and children). In all chapters, there is emphasis on correlative imaging and how one imaging modality complements another in a synergistic way. As appropriate, the reader is introduced to the relevant anatomy and physiology. Modern topics of radiomics, AI/DL, and theranostics are discussed. This image-rich book will appeal to physicians, allied healthcare professionals, and trainees (medical students, residents, fellows). The editors regret any potential errors and omissions and commit to remedy any shortcomings in any future editions. We dedicate this book to the memory of Sanjiv “Sam” Gambhir, MD, PhD, Chair of Radiology at Stanford University. Sam was our mentor, friend, and colleague. He was larger than life with deep intellect, contagious generosity, and remarkable humility. The entire scientific community and indeed humanity itself lost a glorious soul from his untimely passing in July 2020. Ali Gholamrezanezhad Clinical Radiology, University of Southern California, Los Angeles, CA, USA Majid Assadi Nuclear Medicine, Bushehr University of Medical Sciences, Bushehr, Iran Hossein Jadvar Radiology, Urology, and Biomedical Engineering, University of Southern California, Los Angeles, CA, USA

1

1 Introduction to Correlative Imaging What Radiologists and Nuclear Medicine Physicians Should Know on Hybrid Imaging Prathamesh V. Joshi1,2, Alok Pawaskar 2,3, and Sandip Basu2,4 1

Department of Nuclear Medicine & PET-CT, Kamalnayan Bajaj Hospital, Aurangabad, Maharashtra, India Radiation Medicine Centre, Bhabha Atomic Research Centre, Tata Memorial Centre Annexe, Parel, Mumbai, Maharashtra, India 3 Department of Nuclear Medicine & PET-CT, Shri Siddhivinayak Ganapati Cancer Hospital, Miraj, Maharashtra, India 4 Homi Bhabha National Institute, Mumbai, Maharashtra, India 2

Introduction Correlation is defined as a connection or relationship between two or more things that are not caused by chance [1]. Medical research is naturally based on finding the relationship between the known and the unknown [2]. Correlation has been an integral part of medicine. A clinician correlates signs and symptoms with the results of medical imaging, pathology or laboratory investigations. A  nuclear medicine physician or radiologist correlates findings of medical imaging with another imaging modality or laboratory investigation such as tumor marker levels, hormone levels etc. Correlative imaging comprises combining complimentary information provided by different imaging techniques for better interpretation of pathology. In this chapter, our aim is to familiarize readers with the basics of correlative imaging, the strengths and shortcomings of various imaging modalities, and how the correlation among them leads to better understanding of pathologies. The main emphasis of this chapter will be on “fusion imaging”, which has proved to be the best available form of correlative imaging at present.

Correlative Imaging Medical imaging has come a long way since Roentgen first discovered the X-ray in 1895 [3]. Today X-ray, fluoroscopy, computed tomography (CT), ultrasonograpy, single-photon emission tomography (SPECT), positron emission tomography (PET), magnetic resonance imaging (MRI), PET-CT, SPECT-CT, and PET-MRI form the gamut of medical

imaging. Table 1.1 provides a brief review of the different tomographic imaging modalities which form the crux of correlative imaging. Each imaging modality has its own strengths and shortcomings. The utilization of an individual modality depends on multiple factors: 1) Patient-related factors: age of patient, organ of interest, claustrophobia, contrast allergy, pregnancy etc. 2) Modality-related factors: availability, radiation exposure, resolution, need of morphological versus functional information 3) Physician-related factors: expertise of radiologist/ nuclear medicine physician or preference of referring physician 4) Miscellaneous: financial burden of examination, insurance coverage etc. Depending on these multiple factors, an imaging modality is utilized as the investigation of choice during workup of a particular patient. However, it is not uncommon that imaging findings are nonspecific and rather than leading to a definitive diagnosis they lead to a spectrum of differential diagnoses. Through “fusion imaging” or “hybrid imaging” radiologists/nuclear physicians frequently utilize correlative imaging with the intent to narrow down the differentials and/or pinpoint the diagnosis. Correlative imaging can be defined as “imaging the same sample (field of view [FOV] or subject) sequentially or simultaneously with different imaging modalities to obtain complimentary/additive information.”

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach, First Edition. Edited by Ali Gholamrezanezhad, Majid Assadi, and Hossein Jadvar. © 2023 John Wiley & Sons Ltd. Published 2023 by John Wiley & Sons Ltd.

2

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Table 1.1

Overview of the salient attributes of important tomographic imaging modalities. PET

SPECT

CT

MRI

Principle

Three-dimensional distribution of positron-emitting labeled radiotracers

Computer-generated image of local radioactive tracer distribution in tissues produced through the detection of single-photon emissions from radionuclides introduced into the body in the form of SPECT radiotracers

Combined X-ray transmission source and detector system rotating around the subject to generate tomographic images

Strong magnetic field and radio waves to create detailed images of the organs and tissues within the body

The tracer/ contrast used

Positron-emitting radio-pharmaceuticals

Gamma-ray-emitting radio-pharmaceuticals

Iodine-containing contrast medium

Gadolinium-based contrast agents

Resolution

++

+

+++

+++

Functional assessment

+++

++

+

++

Radiation exposure

++

+

+++

None

Allergy/acute side effects

No

No

Yes

Yes

Measurable parameter/ quantification unit

PET tracer uptake/standardized uptake value

−

Attenuation value/ Hounsfield unit

Apparent diffusion coefficient/mm2/s

CT, computed tomography; MRI, magnetic resonance imaging; PET, positron emission tomography; SPECT, single-photon emission tomography.

Correlative imaging has been in practice in the form of comparative imaging for many years, a typical example of this being the “hot spot” observed in bone scans interpreted as metastatic or degenerative based on comparing it with MRI or CT of the same bone. However, now fusion imaging techniques such as SPECT-CT, PET-CT, and PET-MRI have emerged as most widely accepted form of correlative imaging. Conventionally, SPECT and PET had been domains of nuclear physicians while CT and MR had been the radiologist’s forte. With the advent and rapid success of fusion imaging there is a need for combined knowledge of both nuclear medicine and radiology for accurate interpretation of fusion imaging findings. In the next section, we aim to familiarize nuclear physicians and radiologists with the basic principles of tomographic techniques utilized in correlative/fusion imaging.

Positron Emission Tomography–Computed Tomography PET-CT: What a Radiologist Should Know about PET Basics of PET-CT Positron Emission Tomography

PET is a tomographic technique that measures the threedimensional distribution of positron-emitter labeled radiotracers. PET allows noninvasive quantitative assessment of biochemical and functional processes. The most commonly

used tracer at present is the 18F-labeled glucose analogue fluorodeoxyglucose (FDG), and it is the workhorse of PET-CT imaging at present. Though most commonly utilized in oncological imaging, FDG-PET has many other nononcological applications now: dementia, myocardial viability, and infection imaging to name a few. Hence FDG is utilized here as example to demonstrate PET tracer characteristics to radiologists. Principle of FDG PET imaging Enhanced glucose metabolism of cancer cells (primarily dependent on anaerobic glycolysis or the Warburg effect) forms the fundamental basis of FDG PET/CT imaging of malignancies. The increased glucose utilization by the malignant cells is characterized by high expression of glucose transporters (GLUTs, namely GLUT1 and GLUT3) and upregulation of hexokinase activity [4]. Glucose is taken up by tumor cells by facilitated transport (via GLUT) and then undergoes glycolysis with the formation of pyruvate under aerobic conditions. However, under hypoxic conditions (such as in a necrotic tumor), glucose is metabolized under anaerobic conditions with resultant increased tumor lactate levels. FDG is a radiopharmaceutical (RP) analog of glucose that is taken up by metabolically active tumor cells using facilitated transport similar to that used by glucose (Figure  1.1). Despite the chemical differences, cellular uptake of FDG is similar to that for glucose. FDG passes the cellular membrane through facilitated transport mediated by the GLUTs, of which more than 14

●

Introduction to Correlative Imaging CELL CYTOPLASM

hexokinase GLUCOSE

GLUCOSE

GLUCOSE-6-PO4

Glusose-6phosphatase

GLUT receptors

glycolysis

CELL CYTOPLASM

hexokinase F-18 FLUORODEOXYGLUCOSE (FDG) GLUT receptors

FDG

FDG-6-PO4

Glusose-6phosphatase

glycolysis

Figure 1.1 Mechanism of FDG uptake and metabolic trapping inside the cell.

different isoforms have been identified in humans, differing in their tissue distribution and affinity for glucose. GLUT1 is the most common glucose transporter in humans and is, together with GLUT3, overexpressed in many tumors [5–7]. Like glucose, it undergoes phosphorylation to form FDG-6phosphate; however, unlike glucose, it does not undergo further metabolism. At the same time, expression of the enzyme glucose-6-phosphatase is usually significantly decreased in the malignant cells, and FDG-6-phosphate thus undergoes only minimal dephosphorylation, hence becoming “metabolically trapped” in cancer cells [8]. The distribution of FDG in normal organs and pathological lesions is detected by PET scanners. ●

●

Preparation for FDG-PET and scan acquisition Patients are advised to fast and not consume beverages, except for water, for at least 4–6 hours before the administration of FDG to decrease physiologic glucose levels and to reduce serum insulin levels to near basal levels. Oral hydration with water is encouraged. Intravenous fluids containing dextrose or parenteral feedings also should be withheld for 4–6 hours  [9]. FDG is injected intravenously and the PET scan is typically acquired 50–90 minutes after FDG injection. Normal biodistribution and physiological variants Physiological FDG uptake is seen in the brain, myocardium, liver, spleen, stomach, intestines, kidneys and

urine, lymphoid tissue, bone marrow, salivary glands, and testes (Figure 1.2). Breast, uterus, ovary, and thymus can show variable FDG uptake. Causes of Physiological FDG Uptake and Normal Variants Mimicking Pathology

As increased FDG uptake is not limited to malignant tissues alone, for the appropriate interpretation of FDG PET-CT imaging the interpreting radiologist needs to be aware of the physiological causes of FDG uptake as well as commonly encountered physiological variants [10–15]. Table  1.2 summarizes and enumerates the different physiological causes and sites of FDG uptake that can mimic disease and the suggested interventions to reduce them. Quantification of FDG uptake and SUV While interpreting a PET-CT scan, it is the relative tissue uptake of FDG (or any other PET RP) that is of interest to the reporting physician. Visual analysis is sufficient in most cases, but the standardized uptake value (SUV) is a commonly used measure of FDG uptake and it is routinely mentioned in PET-CT reports. The basic expression for SUV is [16] ●

SUV

r a w

3

4

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

Brain

Heart Spleen

Liver

Kidney Bone marrow

Urinary bladder Testes

Figure 1.2 A typical example of the physiological distribution of FDG uptake in a conventional vertex-to-mid thigh whole-body PET study. (a) Maximum intensity projection (MIP) image of a PET scan. (b) Three columns depicting (left to right) trans-axial PET only, trans-axial CT only, and trans-axial fused PET-CT images of physiological distribution.

where r is the radioactivity activity concentration (kBq/ml) measured by the PET scanner within a region of interest (ROI), a is the decay-corrected amount of injected radiolabeled FDG (kBq), and w is the weight of the patient (g), which is used as a surrogate for distribution volume of tracer. If all the injected FDG is retained and uniformly distributed throughout the body, the SUV everywhere will be 1 g/ml regardless of the amount of FDG injected or patient size  [17, 18]. Commonly SUVmax of lesions (maximum SUV) is provided in reports, which is the SUV of most avid voxel in ROI. The reproducibility of SUV measurements depends on the reproducibility of clinical protocols, for example dose infiltration, time of imaging after 18F-FDG administration, type of reconstruction algorithms, type of attenuation maps, size of the ROI, and changes in uptake by organs other than the tumor [9]. SUV or SUVmax values are often

utilized as a marker of change in the metabolic activity of pathology and hence it is important to reproduce the scan conditions during the follow-up PET-CT scan performed for response evaluation.

What Nuclear Medicine Physicians Need to Know about CT PET alone is limited by poor anatomic detail, and correlation with some other form of imaging, such as CT, is desirable for differentiating normal from abnormal radiotracer uptake  [8]. Hence PET-CT morpho-metabolic imaging emerged as an ideal single investigation for oncology practice. However, this also mandates the nuclear physician to have adequate knowledge of the CT component of imaging as well as the various interventions employed in CT acquisition.

Introduction to Correlative Imaging

Table 1.2

Characteristics and causes of physiological uptake of FDG and methods to circumvent them.

Causes/sites of FDG uptake

Physiology behind FDG uptake

PET-CT appearance

Interventions to reduce uptake

Brown adipose tissue (BAT)

Nonshivering thermogenesis requires glucose for glycolysis as a source of adenosine triphosphate, which in turn is utilized in fatty acid oxidation BAT is innervated by the sympathetic nervous system and expresses beta-adrenergic receptors, which are stimulated by cold

FDG uptake in fat density (−150 to −50 HU) in neck, shoulder, and paraspinal regions (Figure 1.3) Less common in perirenal, perigastric regions FDG uptake in BAT is more common in younger patients, females > males

Making patients wear warm clothing and providing a blanket in the waiting suite to avoid cold-induced BAT activation. Premedication with beta-blockers or diazepam

Vocal cords

Phonation-related laryngeal muscle contraction

Symmetrically increased FDG uptake in both vocal cords (Figure 1.4)

If the region of interest is the larynx, the patient should be instructed to avoid talking after FDG injection

Myocardium

Glucose as substrate for energy (GLUT1 and insulin-sensitive GLUT4)

Variable, focal or diffuse without corresponding morphologic abnormality on CT

Fasting before FDG PET-CT (4–12 hours) High-fat, low-carbohydrate diet before scan Premedication with unfractionated heparin before FDG injection

Thymus

Physiological uptake in pediatric patients (especially in postchemotherapy setting, known as “thymic rebound”)

Inverted V-shaped/butterfly pattern of anterior mediastinal uptake on the transaxial view and absence of lesion on corresponding CT (Figure 1.5)

The uptake has a diffuse characteristic pattern: no specific intervention

Lactating breasts

Due to secretory hyperplasia and the increased expression of GLUT-1

Bilateral breast reveal diffuse FDG uptake, but if infant is suckling unilateral breast only that side can show diffuse FDG uptake (Figure 1.6)

The uptake has a diffuse characteristic pattern: no specific intervention

Urinary system

FDG excretion in urine

Usually does not affect scan interpretation Focal retention in kidneys/ureter/ urinary bladder can mimic pathology

Dual point/delayed postvoid imaging with or without diuretic intervention

Ovary

FDG uptake in corpus luteal cyst

Ovoid FDG uptake with smooth margins or a rim of FDG uptake with a photopenic center (Figure 1.7)

Correlation with menstrual history

Endometrium

FDG in menstrual flow

FDG uptake in endometrium in a diffuse uniform pattern (Figure 1.8)

If being evaluated for gynecological pathology, PET-CT scan should be scheduled in the postmenstrual phase

Colon

Related to bowel motility The uptake in cecum and right colon could be result of higher lymphocytes in these regions

Typically heterogeneous and can vary in distribution from mild focal to diffuse uptake Often, there is higher uptake within the cecum and right colon

Uptake pattern: no interventions

(Continued)

5

6

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Table 1.2

(Continued)

Causes/sites of FDG uptake

Physiology behind FDG uptake

PET-CT appearance

Spinal cord

Inadequate clearance of FDG from the artery of Adamkiewicz, which originates on the left side of the aorta between the T9 and T11 vertebral segments Increased cross-sectional area of the spinal cord

The physiological FDG uptake is visualized in the cervical spinal cord peaking at C4 level, and in the lower thoracic spinal cord peaking at the T11–T12 segments (Figure 1.9)

Skeletal muscles

Exercise induces glucose uptake in skeletal muscles Labored breathing can increase FDG uptake in intercostal muscles and diaphragm In postmeal state, insulin increases GLUT (GLUT-4) mediated skeletal muscle glucose uptake

If related to exercise, usually symmetrical FDG uptake in muscles with no abnormal enhancement or lesion on CT If related to meal/insulin diffuse FDG uptake in skeletal muscles (usually also accompanied with cardiac FDG uptake) If related to labored breathing, intercostal muscles and diaphragm reveal symmetrical increased FDG uptake (Figure 1.10)

Interventions to reduce uptake

Patients should avoid strenuous exercise for 48–72 hours before scheduled scan Fasting status should be confirmed before FDG injection

BAT, brown adipose tissue; FDG, fluorodeoxyglucose; GLUT1, glucose transporter 1; GLUT4, glucose transporter 4; PET-CT, positron emission tomography/computed tomography.

(a)

(b)

(c)

(d)

Figure 1.3 (a) FDG uptake in brown adipose tissue in bilateral cervical (red arrows), paraspinal, and perirenal regions as shown in MIP image, (b) transaxial PET-only image of the neck region, (c) transaxial CT-only image, and (d) fused PET-CT image.

Introduction to Correlative Imaging

(a)

(b)

Figure 1.4 (a) Transaxial CT-only image of vocal cords. (b) The fused PET-CT image of the same region shows symmetrical increased FDG uptake in bilateral vocal cords (arrows). This patient was groaning due to painful skeletal secondaries, resulting in hypermetabolism in the vocal cords.

(a)

(b)

as the most important invention in radiological diagnosis since the discovery of X-rays [19, 20].

Principle of CT

Figure 1.5 (a) MIP image of PET-CT of a 10-year-old boy showing physiological FDG uptake in the thymus (black arrows). (b) Hypermetabolism in the soft tissue neoplasm in the occipital region (red arrow).

Computed Tomography Although the potential applications of X-rays in medical imaging diagnosis were clear from the beginning, the implementation of the first X-ray CT system was made in 1972 by Godfrey Newbold Hounsfield (Nobel prize winner in 1979 for Physiology and Medicine), who constructed the prototype of the first medical CT scanner and is considered the father of CT. After this, CT was immediately welcomed by the medical community and has often been referred to

The CT scanner creates cross-sectional images by projecting a beam of X-rays through one plane of an object (patient) from defined angle positions performing one revolution. These X-rays are generated by a rotating X-ray tube (Figure 1.11). As the X-rays pass through the patient‚ some of them are absorbed, while some are scattered and others are transmitted. The process of X-ray attenuation refers to the intensity reduction involving those X-rays which are scattered or absorbed. X-rays which are attenuated due to the interactions with the object do not reach the X-ray detector. Photons transmitted through the object at each angle are collected on the detector and visualized by computer, creating a complete reconstruction of the patient. The three-dimensional (3D) gray value data structure gained in this way represents the electron density distribution in the area of interest [19]. The ability of matter to attenuate X-rays is measured in Hounsfield units (HU). By definition, water is assigned a density value of 0 HU and air a value of −1000 HU. Attenuation values for most soft tissues fall within 30–100 HU. Notable exceptions are lungs, with attenuation values approaching −1000 HU (due to high air content), and mineralized tissues such as bone, with attenuation values of approximately 1000 HU [21].

Intravenous and Oral Contrast in CT Scanning Intravenous Contrast

Differences in the CT attenuation of healthy tissue and  pathology can improve the quality of the images

7

8

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

(d)

Figure 1.6 (a) MIP image of PET-CT showing FDG uptake in bilateral breasts of a nursing mother (red arrows), (b) transaxial FDG PET of breast region, (c) CT of breast region and (d) fused PET-CT of breast region.

(a)

(b)

Figure 1.7 (a) CT image of pelvic region and (b) fused PET-CT of same region showing increased FDG uptake in a corpus luteal cyst in the left adnexal region (arrow).

Introduction to Correlative Imaging

(a)

(b)

Figure 1.8 FDG PET-CT of a 27-year-old female. (a) Transaxial CT of pelvic region and (b) fused PET-CT image of pelvic region revealing FDG uptake in fluid in the endometrial cavity (arrow) corresponding to menstruation.

(a)

(b)

(c)

(d)

(e)

(f)

Figure 1.9 (a) Sagital CT, (b) PET, and (c) fused PET-CT images revealing physiological FDG uptake in the cervical spinal cord (arrows). (d) Transaxial CT, (e) PET, and (f) fused PET-CT images showing focal FDG uptake at the T11-T12 level in the spinal cord (arrows).

9

10

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

Figure 1.10 (a) In a carcinoma larynx patient, the MIP image of FDG PET-CT reveals a hypermetabolic lesion in the neck corresponding to the site of primary malignancy (black arrow). (b) Fused PET-CT image shows increased FDG uptake in the intercostal muscles and diaphragmatic crura (white arrows). (c) Transaxial CT of the same region. The augmented FDG uptake in these muscles of respiration was the result of labored breathing due to narrowing of the airway caused by the laryngeal malignancy.

Rotating X-ray tube

X ray beam

Stationary detector ring

for CT imaging is in the molar concentration range. Since use of intravenous contrast is known to be associated with adverse effects in susceptible population and allergies, caution needs to be exercised during their use. When diagnostic contrast-enhanced CT with intravenous contrast media is to be performed (after the PET/CT examination), indications, contraindications, and restrictions have to be assessed by a qualified physician/radiologist. Medication that interacts with intravenous contrast (e.g. metformin for the treatment of diabetes) and relevant medical history, especially compromised renal function, have to be taken into consideration [23]. Gastrointestinal Contrast Agent

Figure 1.11 Basic principles of a CT scan.

(i.e. greater signal-to-noise and contrast to noise ratios) and hence facilitate detection of abnormality. Hence, contrast imaging agents are often used for better visualization of the tissue of interest by CT [21, 22]. Today, a wide range of ionic and nonionic contrast agents is available and effective diagnostic dose of a contrast agent

Depending on the ROI, gastrointestinal luminal contrast agent may be administered to improve the visualization of the gastrointestinal tract in CT (unless it is not necessary for the clinical indication or it is medically contraindicated). This is more commonly done via oral administration and less commonly by the rectal enema route for evaluation of colonic pathologies. It should be noted that the contrast agents alter the attenuation caused by tissues and hence result in overestimation of SUV values used in PET quantification (more so with IV contrast as compared to gastrointestinal) [24].

Introduction to Correlative Imaging

CT Protocols in PET-CT After the advent of PET-CT in the 1990s, the initial PET-CT acquisition protocols utilized CT as a fast transmission source for attenuation correction, with little additional information for anatomic localization. However, these CT protocols could not generate diagnostic quality CT images. These protocols can be largely considered as low-dose CT scans. The effective dose due to CT procedures in such lowdose CT scans is typically 3–6 mSv [25]. However, after realizing the logistic advantages of a single examination for functional (PET) and morphological (CT) information, CT is now being utilized as a fast transmission source as well as a state-of-the-art diagnostic tool to maximize image quality. This protocol involves optimal acquisition parameters together with oral and intravenous contrast agents. These protocols can be largely considered as diagnostic CT scans. The effective dose due to CT procedures in such diagnostic CT scans is typically 11–20 mSv. There are numerous variations in CT protocols and they are discussed in detail in FDG PET-CT guidelines [23, 26]. The representative two approaches are shown in Figure 1.12.

Display of Fused PET-CT Images In PET-CT scanners (prototype shown in Figure 1.13), the patient lies still on a bed which is then translated through fixed mechanically aligned coaxial CT and PET gantries so that the data acquired are precisely co-registered in

Figure 1.13 Prototype of a PET-CT scanner available in clinical practice, the GE Discovery IQ Gen2 PET-CT scanner.

space  [27]. The PET acquisition typically occurs immediately after the CT acquisition to minimize the effects of patient motion. After reconstruction, the high-resolution anatomical images (from CT) are overlayed with the functional images (from PET) to provide the precise localization of hypermetabolic regions. The images consist of PET only, CT only, and fused PET-CT, which are viewed in the transaxial, coronal, and sagittal planes. Additionally, a cine maximum intensity projection (MIP) image provides a specific type of rendering in which the brightest voxel (the voxel with maximum FDG uptake) is projected into the 3D image. This MIP image enables a “gestalt” impression of the study [28]. An example of a typical display is shown in Figure 1.14.

Protocol1: When CT is used for attenuation correction and localization only (not intended as a clinically diagnostic CT scan)

CT topogram

Low dose CT scan

PET acquisition

Protocol2: When CT is intended to be a diagnostic CT scan

CT topogram

Deep inspiration thoracic CT, with 20 seconds delay from beginning of IV contrast infusion

A wholebody diagnostic CT (with shallow breathing), with 45 seconds delay after thoracic CT (in equilibrium or venous phase of contrast)

PET acquisition

Figure 1.12 Schematic representation of representative PET-CT acquisition protocols.

11

12

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

(j)

Figure 1.14 Staging FDG PET-CT of a 53-year-old female diagnosed with locally advanced carcinoma of the left breast (blue arrow) with metastatic lesion in body of D7 vertebra (red arrow). The following components of PET-CT are seen: (a) maximum intensity projection (MIP) image, (b) trans-axial PET only image, (c) coronal PET only image, (d) sagittal PET only image, (e) trans-axial CT only image, (f) coronal CT only image, (g) sagittal CT only image, (h) trans-axial PET-CT fusion image, (i) coronal PET-CT fusion image, (j) sagittal PET-CT fusion image

Artifacts in PET-CT Fusion Recent PET-CT scanners allow excellent fusion of the PET and CT images and thus improve lesion localization and interpretation accuracy. Moreover, the employment of the CT data for attenuation correction has led to high patient throughput  [29]. Although PET-CT imaging offers many advantages, this dual-modality imaging also poses some technical challenges due to a few artifacts. The reader interpreting PET-CT scans needs to be aware of these limitations. The artifacts can be broadly divided into following categories: 1) Motion artifacts (respiratory or patient related): Although the CT and PET acquisitions are performed without changing the patient position, voluntary or involuntary movements of patient can result in misregistration of PET and CT images. Most commonly such misregistration artifacts are observed in lesions of the lungs and liver. An example is shown in Figure 1.15.

Efforts have been made to minimize such image degradation by the generation of a respiratory motion corrected or four-dimensional PET-CT during which the PET data are acquired in synchronization with respiratory motion [30]. 2) Attenuation correction artifacts: The presence of highdensity material in the patient’s body either in the form of high-density material like bone cement or venous pooling of intravenous contrast/barium from previous studies in bowel loops can result in artifactual FDG uptake due to exaggerated attenuation correction at these sites. A clinical example is shown in Figure 1.16. 3) Beam hardening artifact: This artifact appears as multiple linear bands of abnormal attenuation traversing a body part adjacent to high-attenuation objects, such as metal prosthesis, dental fillings, chemo ports, and pacemakers. Patients need to be instructed to remove metallic objects before scan acquisition and a note should be made of fixed/in situ metallic prosthesis/implants. An example is shown Figure 1.17. Some implants/prosthesis can result

(a)

(b)

(d)

(e)

(c)

(f)

Figure 1.15 50-year-old male, known smoker, referred for characterization of a solitary pulmonary nodule in the basal region of the lower lobe of the right lung. (a) A focus of increased FDG uptake is noted (red arrow), which does not correspond to any morphological abnormality in fused PET-CT (b) and CT only (c). The acquisitions were repeated with shallow breathing to minimize the lung motion and the second set of images (d)–(f) reveal focal FDG uptake in a 14 × 14 mm sized nodule in the basal region of the lower lobe of the right lung (blue arrow), suspicious of neoplastic pathology.

(a)

(b)

(c)

(d)

Figure 1.16 MIP image of FDG PET-CT of a 36-year-old female for staging lymphoma. The focal uptake observed in the right axillary region (black arrows in (a) and (b), and white arrow in (c)) was artifactual due to pooling of intravenous contrast material in the right subclavian vein. The high density of contrast (red arrow in (d)) resulted in high attenuation correction and resultant artifactual FDG uptake in the PET image.

14

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

in false-positive PET findings due to changes in attenuation correction factors. A similar artifact can be seen when the PET-CT scan is acquired with the hands on the sides of the trunk and hence the easiest way to prevent this artifact is to perform the scan with arms up or down, depending on clinical indication. 4) Truncation artifact: Truncation artifacts in PET-CT are due to the difference in size of the FOV between the CT (50 cm) and PET (70 cm) tomographs [31]. These artifacts are frequently seen in large patients or patients scanned (a)

(b)

(c)

(d)

with arms down, such as in the case of head and neck malignancy. When a patient extends beyond the CT FOV, the extended part of the anatomy is truncated and consequently is not represented in the reconstructed CT image. Truncation also produces streaking artifacts at the edge of the CT image, resulting in an overestimation of the attenuation coefficients used to correct the PET data. This increase in attenuation coefficients creates a rim of high activity at the truncation edge (see the example in Figure 1.18), potentially resulting in misinterpretation of the PET scan  [32]. Therefore, in PET-CT imaging, it is

Figure 1.17 Beam hardening artifact caused by a metallic implant in the right femur, seen as linear bands of abnormal attenuation (arrows in (a)–(c)). No artifact is noted in the PET-only image (d).

(a)

(b)

(c)

Figure 1.18 Truncation artifact in a large patient resulting in the rim of FDG uptake in PET image (a, arrow), loss of information in CT only image (b), and FDG uptake without morphological data in fused PET-CT (c, arrow).

Introduction to Correlative Imaging

crucial that technologists carefully position the patient at the center of the FOV and with arms above the head to reduce truncation artifacts [29]. 5) Radiopharmaceutical related: Although relatively rare, focal FDG uptake (or other PET radiotracer uptake) without any CT demonstrable lesion needs to be interpreted with caution, especially in lung paren(a)

chyma (see the example in Figure  1.19). Such uptake can be the result of iatrogenic FDG micro-embolus at the time of injection [33] and when such a finding can affect management of a patient, a follow-up scan can be performed to avoid false-positive interpretation. In the next section, few clinical case examples present (Figures 1.20–1.22).

(b)

(c)

Figure 1.19 Focal FDG uptake seen in the PET image (a, arrow) and fused PET-CT (b) does not correspond to any nodule/lesion in the corresponding CT trans-axial slice of the right lung (c). Such a pattern can be the result of an iatrogenic micro-embolus of FDG caused during injection.

(a)

(b)

(c)

(d)

(e)

Figure 1.20 A 28-year-old male recently diagnosed with non-Hodgkin’s lymphoma for staging FDG PET-CT evaluation. The supradiaphragmatic and infra-diaphragmatic lymphadenopathy was apparent on CT and was suggestive of stage III NHL. However, the hypermetabolism in spleen (a and c, arrow) and left iliac bone (e, arrow) could be appreciated in PET and fused PET-CT images and hence indicated splenic as well as bone marrow involvement, upstaging disease to stage IV. Note that in the CT-only images (b) and (d) the spleen and left iliac bone appear unremarkable.

15

16

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

FDG PET-CT Clinical Examples

PET-CT Tracers Beyond FDG

The following section consists of the pictorial demonstration of clinical case examples (Figures 1.20–1.22) where fusion imaging of FDG PET and CT resulted in accurate diagnosis of pathology and its extent that would have been otherwise difficult to reach.

Although 18-F-FDG is the most widely used tracer in fusion imaging in the form of PET-CT, many other RPs have made a significant impact in patient management and have become part of routine patient management. The list of such RPs is exhaustive and beyond the scope of this chapter; a few common ones are listed in Table 1.3 [34–38] with case examples shown in Figures 1.23–1.25.

Table 1.3

PET beyond FDG: the important contemporary tracers and their clinical applications.

PET radiopharmaceutical

Mechanism of uptake

Clinical use

Gallium-68 and F-18 PSMA labeled PSMA targeted ligands (small-molecule PSMA inhibitors)

Binding to PSMA

Prostate carcinoma: biochemical recurrence, staging high-risk cases, and treatment planning for peptide receptor radioligand therapy

Gallium-68 DOTANOC/ DOTATOC/DOTATATE (DOTAconjugated peptides)

Binding with somatostatin receptors expressed in neuroendocrine tumor cells

Neuroendocrine tumor imaging and treatment planning for peptide receptor radionuclide therapy

Fluorine-18 fluoro-dopamine

Analog of l-DOPA, to trace the dopaminergic pathway and to evaluate striatal dopaminergic presynaptic function

Evaluation of movement disorders Evaluation of congenital hyperinsulinemia

Fluorine-18 sodium fluoride

18

Diagnosis of skeletal metastases

F is substituted for hydroxyl groups in hydroxyapatite and covalently bonds to the surface of new bone

DOPA, dihydroxyphenylalanine [2-amino-3-(3,4-dihydroxyphenyl) propanoic acid; DOTA, 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid; DOTANOC, DOTA-Nal3-octreotide; DOTATOC, DOTA-Tyr3-octreotide; DOTATATE, DOTA-Tyr3-octreotate; FDG, fluorodeoxyglucose; PET, positron emission tomography; PSMA, prostate specific membrane antigen.

(a)

(b)

Pancreatic duct Common bile duct

Figure 1.21 A 56-year-old male complained of chronic, intermittent abdominal pain and recent onset jaundice, nausea. Ultrasonography revealed an overdistended gall bladder without any calculus. FDG PET-CT was performed with suspicion of pancreatico-biliary neoplastic pathology. (a) The contrast-enhanced CT revealed a dilated pancreatic duct and common bile duct, the “double duct sign” (white arrow), indicating pathology in ampullary region. However, on contrast-enhanced CT alone no obvious morphological lesion could be identified in the ampullary/duodenal region. However, in the FDG PET images, focal hypermetabolism is seen in the right lumbar region of the abdomen. When the PET and CT images are fused (b), the hypermetabolism corresponds to the ampullary region of the duodenum (green arrow) and indicates an ampullary lesion obstructing the pancreatic and common bile ducts. On endoscopy, an ulcerated lesion was found in the second part of the duodenum, which revealed ampullary carcinoma on histopathology.

Introduction to Correlative Imaging

(a)

(b)

(c)

(d)

(e)

(f)

Figure 1.22 Pre- and postchemotherapy FDG PET-CTs of a metastatic carcinoma rectum. The pretherapy PET-CT MIP (a), fused PET-CT (c), and CT-only (d) images reveal FDG avid hepatic metastasis. Posttreatment images (b), (e), and (f) reveal complete metabolic response, as seen by resolution of FDG uptake and partial morphological regression, as seen by the reduction in the size of the lesion.

(a)

(b)

(c)

Figure 1.23 MIP image of 68Ga-PSMA-11 PET-CT for evaluation of biochemical recurrence of prostate carcinoma in a 61-year-old male. Increased PSMA expression seen in pelvic region (a, black arrow). (b) CT and fused PET-CT reveal increased PSMA expression in perirectal lymph nodes (white arrows). (d) Scan pattern suggests metastatic lymphadenopathy as a cause of rising PSA levels.

SPECT–CT Imaging Introduction Among the fusion or correlative imaging modalities, PET-CT is the often discussed modality, mainly due to its widespread clinical and research applications, although we must emphasize the potential and increasing applications of SPECT–CT, which is often underestimated and may be

another powerful correlative imaging tool in the future. Gamma camera (Figure  1.26) has been in use for getting functional information on the physiological, biochemical, and metabolic processes in the various organs in the body. The tracer used in gamma camera imaging is usually specifically targeted to obtain information from a particular organ system. Hence, once administered into the patient’s body, the tracer accumulates in the target organ system. The more specific the tracer, the more information from

17

(a)

(b)

(c)

(d)

(e)

Figure 1.24 MIP image of 68Ga DOTA-Nal3-octreotide PET-CT images of a 51-year-old male patient with clinical suspicion of neuroendocrine tumor. He complained of recurrent vomiting and abdominal pain, and was found to have substantially elevated serum chromogranin A levels. (b) Transaxial CT image and (c) fused PET-CT revealed increased somatostatin receptor expression in a small nodular lesion in the second part of the duodenum (white arrow). (d) Transaxial CT revealed enlarged perilesional lymph node. (e) Increased somatostatin expression was seen in the enlarged perilesional lymph node (blue arrow). Biopsy of the duodenal lesion revealed grade 1 neuroendocrine tumor.

(a)

(b)

(c)

(d)

Figure 1.25 (a) and (b) MIP images of 18F-fluoro-DOPA PET-CT of a 1-month-old baby with recurrent severe hypoglycemia due to congenital hyperinsulinism. This rare and grave condition is the result of islet cell hyperplasia, which can be either focal or diffuse. (c) and (d) Transaxial fused PET-CT images reveal diffuse radioactive dopamine uptake was noted in the pancreas, marked with arrows, suggestive of diffuse islet cell hyperplasia. In the focal type only partial pancreatectomy of the hyperfunctioning focus is performed, while in the diffuse type a near-total pancreatectomy may be required.

Introduction to Correlative Imaging

Figure 1.26 Schematic representation of gamma camera. Processing computer

Image on display monitor

Photomultiplier tubes (PMTs) {Conversion of light to electrical signal} Detector crystal [Nal(TI)] {Conversion of gamma rays to light} Collimator Source of radioactivity (Patient)

the target organ may be collected about a particular pathophysiological process. However, at the same time, whatever little morphological information is obtained by background tracer activity diminishes significantly. Hence nuclear medicine techniques often lack anatomical landmarks. Also, there is definite loss of data in the planar imaging due to the attenuation of gamma rays coming from organs deep inside the body. SPECT entails 3D reconstruction of tracer distribution within the patient body with the help of data collected by rotating detectors around the patient body. This helps to achieve better anatomical information, for example in the case of bone scan or myocardial perfusion imaging. CT scanning, on the other hand, is a 3D reconstruction of X-ray attenuation value maps providing morphological details like size, shape, and location. Use of contrast agent in CT primarily provides information about perfusion and the changes in perfusion pattern occurring in various disease processes. However, CT often does not provide any information on the functional or metabolic status of organs in the body. Many disease processes show pathophysiological changes much before morphological changes are manifested. Also, in presence of anatomical distortions secondary to various treatments, anatomical imaging interpretations are difficult and often uncertain because of changes in symmetry and perfusion pattern. Hence it is of vital importance to understand that nuclear medicine (SPECT) and anatomical imaging (CT) are not competitive to each other, but in fact complimentary in nature. The fundamental advantages of this are as follows:

i) Combined SPECT-CT images have the best of both worlds. They have all the anatomical information lacking in SPECT images and functional information lacking in CT images. CT also helps in proper localization of tracer uptake to ultimately help in correct diagnosis and treatment. ii) Not only this, CT attenuation maps are used for attenuation correction and this improves the quality of SPECT images. There are many applications of SPECT-CT that are well established clinically. As new advanced systems are becoming widely available, further improving the accuracy of image fusion and shortening acquisition times, the newer applications are becoming more evident. Apart from applications in oncology, interesting uses of SPECT-CT are seen in the areas of minimally invasive surgery and cardiology. We shall start our treatise with some technical information about these systems before going into the clinical applications.

SPECT–CT System Information Combining SPECT and CT images acquired from different standalone machines has often been challenging. This is because usually the studies are acquired on different dates, on different machines by different operators using different protocols. This creates differences in the position of the patient body, extremities as well as spinal curvatures, as table positions may differ with different systems. Furthermore, it is not possible to match respiratory, cardiac motion, and position of stomach, intestines,

19

20

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

and urinary system at different time frames. Software fusion techniques have been developed which can register and fuse images from multiple sources  [39, 40], but they are best suited for correlating images from rigid structures such as the brain [41] and skeleton [42, 43]. However, for thorax and abdomen, due to inherent movement of internal organs, software fusion has remained challenging. Data sets from two fundamentally different imaging modalities with different spatial resolution and with few common landmarks make it further complicated. A system offering same geometry for acquisition of SPECT and CT images almost simultaneously, such as SPECT-CT scanning, offers much better data sets for fusion. Here the patient remains on the same table and in the same position while undergoing SPECT and CT acquisitions separated by a few minutes. The practical advantages of SPECT-CT fusion imaging are multifold: i) These systems are able to superimpose functional information from nuclear medicine data sets onto anatomical information from CT scans, greatly improving the confidence of the reporting diagnostician. ii) These systems are able to facilitate attenuation correction of SPECT data with patient-specific attenuation maps acquired from CT [44, 45]. Because of this, there is improvement in the spatial resolution, contrast, and signal-to-noise ratio of the image. iii) There is improvement in the functional data quality aided by CT, which shows great promise in quantification of RP uptake  [46, 47]. This is very useful for (i) better radiation dosimetry [48, 49] and (ii) monitoring response to therapy. The concept of combining structural and functional information was conceived and implemented in prototype

form at the very beginning of emission and transmission CT, most notably the work by Kuhl, Hale, and Eaton, who obtained the first trans-axial transmission CT scan of a patient’s thorax using their Mark II brain SPECT scanner in the mid-1960s [50]. However, use of transmission imaging with an external radionuclide transmission source was introduced for attenuation correction in SPECT  [51, 52] and PET [53, 54] only in the 1980s. In this system, external transmission scanning was used with SPECT to perform both attenuation correction and anatomical localization. However, a transmission scan provides poor quality anatomical details and contrast resolution, hence these systems never grew into routine applications. Over the last decade or so, combined SPECT-CT scanners have become commercially available which acquire data from SPECT and CT on the same gantry. The patient remains in same position and on the same table, which is then sequentially moved from one modality to another. The final data acquired is then transferred to a single computer which does data correction, image reconstruction, integration, and display and allows analysis for better diagnosis. The early SPECT-CT systems tried simultaneous acquisition of SPECT and CT data  [55, 56], but the problem with a simultaneous SPECT-CT acquisition system was in designing a common detector with sufficient temporal and energy resolution to discriminate the primary radionuclide photons from both the X-ray signal and the scatter of the radionuclide photons. This problem remains unsolved. The “modern” SPECT-CT system was originally developed by Hasegawa et  al. at the University of California, San Francisco in the mid-1990s  [45]. These systems have SPECT and CT gantries in tandem (in-line) which can acquire patient data sequentially and send it to the same computer for further fusion and processing (Figure 1.27).

SPECT-CT gantry

Patient table

SPECT detectors

Figure 1.27 Schematic diagram of the modern SPECT-CT system.

CT

Introduction to Correlative Imaging

General SPECT–CT Protocols The first commercial SPECT-CT combination that was designed as a single unit was the GE Millenium™ hybrid SPECT/PET/CT camera equipped with a HawkEye™ single-slice CT (GE Healthcare, Haifa, Israel). The CT produced images with a slice thickness of 1 cm and a 256 × 256  matrix size with a spatial resolution of about 3.5 mm. Because of its coarse resolution, the CT was not regarded as a diagnostic CT. Reconstruction was performed using filtered back projection. The system was also offered with a 1-in. crystal and a coincidence unit for PET imaging. This system was the commercial implementation of the very successful research by the late Hasegawa et al. [57]. Almost all current SPECT-CT systems offer highresolution diagnostic CT units as part of their SPECT-CT systems. One of the manufacturers (Mediso) even offers a complete SPECT/CT/PET with high-resolution lutetium-yttrium oxyorthosilicate (LYSO) detector technology as part of their AnyScan family of systems. The CT scan in these devices is a diagnostic CT scan which is used for attenuation correction based on individual patient-based tissue density data. The SPECT-CT workstations allow image reconstruction, and three-plane (transaxial, coronal, sagittal) and 3D display, including MIP and surface volume rendering. SPECT, CT, and fused images are shown on the same screen, and an interconnected pointer is available to exactly colocalize the morphological and functional areas of interest identified in either one of the two study components. The cost of SPECT-CT systems is considerably higher than that of a conventional gamma camera, especially for devices including a full diagnostic capability CT. Higher cost has constrained the availability of this technology to places with limited financial resources. New SPECT devices have recently been developed using CdTe/CdZnTe semiconductors instead of the classic NaI (Tl) scintillation crystals. Such newer systems are smaller, and have higher sensitivity and intrinsic resolution than conventional cameras.

between 3/8 and 3/4 in. (9.5 and 19 mm) depending on the intended usage for radionuclides emitting lower energy photons (e.g. 99mTc) and/or higher energy photons (e.g. 131I). Thicker crystals improve the photo-peak efficiency but degrade the intrinsic spatial resolution. Acquisition on SPECT-CT systems is performed in a sequential mode. The SPECT-CT systems using a diagnostic CT component have higher spatial resolution and faster scanning time. However, diagnostic CT delivers higher radiation doses. The SPECT component is acquired by a rotating, dual-head, variable angle sodium-iodide scintillation camera. SPECT acquisition currently requires a routine scanning time of approximately 20–30 minutes, depending on the radiotracer and the axial length of the body area scanned. CT is usually acquired in matrices of 512 × 512 with the newest CT scanners or 256 × 256 in older scanners, and has to be resized into slices with the same pixel format and slice width as SPECT. SPECT is reconstructed using iterative methods incorporating photon attenuation correction based on the X-ray transmission map and scatter correction. Since X-ray and radionuclide data are not acquired simultaneously, the SPECT images are not contaminated by scatter radiation generated during the X-ray image acquisition. Also, since the patient is not removed from the table, both imaging components are acquired with a consistent and identical patient position, allowing accurate image registration if we assume that the patient has not moved during the entire duration of the SPECT-CT study. The current SPECT systems are equipped with detection devices or sensors to automatically calculate noncircular orbits: the detection methods can be based on optical detection using light-emitting diodes. Aside from the safety aspect, noncircular orbits ensure closer distances to the patient for each of the SPECT projections as well as easy setup procedures, which together reduce the overall time that the patient spends on the imaging table. Companies now offer advanced and integrated diagnostic SPECT-CT solutions with 2-, 6-, or 16-slice diagnostic CTs. Basically, the SPECT-CT is a hybrid or a combination of each of the two systems that have the same characteristics as the respective standalone devices.

SPECT–CT Acquisition

Image Reconstruction

Most clinical SPECT systems still rely on the Anger camera principle discussed earlier, where the location of a photon interaction (i.e. scintillation) site in the body is calculated as the center of gravity of the position-dependent energy signals from a two-dimensional (2D) array of photomultiplier tubes (PMTs) attached to the back of the scintillation crystal  [58]. The available NaI (Tl) crystal thicknesses vary

Most of the newer SPECT systems use statistical iterative reconstruction for creating images  [59, 60]. An iterative procedure includes some kind of model of how the images are formed by the acquisition system. For a SPECT system, an image is formed by the collimation of emitted photons such that only those photons parallel to the collimator hole will interact with the crystal and produce scintillation

21

22

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

events. Hence, these so-called projections are created as 2D representations of all photons detected along the projection lines and this gives rise to a distance-dependent resolution. The greater the distance of the camera head from the radioactive source, the poorer the resolution. There is no information about source depth in the 2D projections. The purpose of the reconstruction algorithm is therefore to create an estimate of the source distribution in 3D. From the CT in SPECT-CT systems, attenuation maps are generated and incorporated in the image formation model. In an iterative reconstruction algorithm, one starts with an initial estimate of the internal and unknown radionuclide distribution. This estimate is usually an image set with equal and constant voxel values. SPECT projections are then calculated from the initial estimate using a computer model of the imaging system. The underlying assumption is that if the calculated projections match the measured projections, then the internal unknown activity distribution in the patient matches the estimate. Initially, these are most likely nonagreeing, so the initial estimate needs to be modified. Therefore, updates are made in the estimate based on a comparison between calculated projections of the estimate using the physics of the imaging process and the measured projections. By back projection, the deviations between these two projection sets finally form an error image of weighting factors that is used to update the initial image estimate. This procedure is placed in an iterative loop in which the calculated and measured projections are compared until the deviation is smaller than a selected criterion (convergence has been reached). Hence, when this happens, the reconstruction loop is stopped. However, in actual practice, as projection data is noisy, the iterative reconstruction procedures are stopped after only a small number of iterations because the noise in the final image goes on increasing for a large number of iterations. The most popular iterative reconstruction algorithm implemented is ordered subset expectation maximization (OSEM)  [61] due to the speed with which it reaches a good estimate of the activity distribution. However, there are many other factors, such as nonhomogeneous photon attenuation, contribution from scattered photons, and blurring due to the collimator response, which need to be accounted for during reconstruction of the image. The compensation for these factors is relatively easy in iterative reconstruction, but there are still some issues when combining these two systems. It should be remembered that a SPECT acquisition is averaged over many respiratory cycles, which blurs the image, while rapidly acquired CT images are instantaneous images and may not match with the averaged SPECT image. The spatial resolution of the SPECT system is also far lower than that of CT. Hence, if high-resolution CT images are used for

attenuation correction, there is a possibility that artifacts may appear at boundaries between different attenuating tissues because of the spill-out of events. Often this effect is reduced by smoothing the CT images with a Gaussian kernel that results in a spatial resolution of the CT images comparable with the spatial resolution of the SPECT images. Also, depending on the acquisition parameters (voltage and mAs), the CT image may not be optimal for attenuation correction and the scaling from the X-ray bremsstrahlung spectra to the specific photon energy used in SPECT may not be optimal. Hence, when using CT only for attenuation correction, so-called low-dose protocols can be used which are optimized, such as a longer scanning time to average the respiratory movements and matched spatial resolutions [62]. The problem of scatter correction has been taken care of by various methods. Siemens and GE have implemented some form of the dual-energy window method based on the estimation of scatter from additional energy windows for subtraction from projection data or used within iterative reconstruction methods as an additive term [63]. However, it is generally known that the distribution of scatter in additional lower scatter windows does not reproduce the scatter distribution in the main photo-peak energy window. Alternatively, model-based scatter compensation has been used that does not rely on additional energy window data collection. Instead, it models the scatter in the main photopeak energy window by using pre-calculated scatter kernels or in real-time calculates the scatter based on theoretical cross-sections of first-order scattering. Philips has implemented a version of the effective scatter source estimator method, originally developed by Frey and Tsui [64]. There is also a need for some compensation for the limited spatial resolution due to the collimator design. This is generally implemented in the iterative reconstruction algorithm, modeling the depth-dependent blurring caused by photons that reach the detectors. The effect of partial volume is an apparent reduction of the radioactive count density/X-ray density that occurs when an organ/tumor/ defect only partially resides within the “sensitive volume of the imaging instrument (in space or time)” [65–68]. From the review of Erlandsson et al. [68] it is clear that an extensive body of work exists in an effort to minimize the partial volume effect. The main aim of partial volume correction (as well as resolution correction) is to improve the quantitative accuracy of the structure under investigation. CT can be used for correction of partial volume in SPECT images owing to its high sampling frequency. Patient body movement can occur during both the SPECT and CT portions of the study. The acquisition times for conventional SPECT–CT systems can vary from 10 minutes to more than 30 minutes depending on area scanned and

Introduction to Correlative Imaging

count rate. During acquisition, SPECT typically takes much more time than CT. All manufacturers offer registration tools to manually or automatically register SPECT and CT in case movement occurs between those two scans. The motion during the CT portion of the acquisition is usually minimized owing to short scan times as well as breadthholding by the patient for diagnostic CT scan. All manufacturers have some form of motion correction for SPECT itself using the consistency of the projections. These algorithms usually work best for small and simple motions. However, for excessive and complex motions, the SPECT-CT acquisition needs to be repeated.

New SPECT–CT Scanners with Solid-state Detectors The most common scintillation material used in gamma camera is NaI(Tl). It is readily available and cost-effective. If we review the principal characteristic of gamma cameras, then the absorbed photon energy is converted to visible light in proportion to the deposited energy and this light is then detected by PMTs. PMTs convert this light to electrons at the cathode surface of the PMTs and the signal is amplified using a sequence of dynodes to create an electrical signal. This signal is then processed to get information about the energy and position of the original gamma rays. Because there are many steps involved, the uncertainty in the measured energy signal is quite large, resulting in an energy resolution in the order of 10% full width at half maximum at 140 keV. Furthermore, a large number of PMTs are required to determine the location of the interaction. That is why the size of the scintillation camera head is quite large and requires sophisticated tuning methods so that all PMTs provide similar amplitude signals for the same imparted energy. Recently, commercial SPECT systems based on cadmium–zinc–telluride (CdZnTe or simply CZT) have been introduced. CZT is a solid-state detector material that generates signals from the collection of induced charge created by the ionizations from photoelectric interactions or Compton scattering. Here each photon is directly converted into an electrical signal. The major advantages of these new CZT modules are their small size and the absence of PMTs, which allows for a compact camera. It also increases the overall system sensitivity significantly. The improvement in sensitivity can either be used to reduce the acquisition time of a given administered activity or reduce the radioactivity administered for the same acquisition time. The major contrasting points between a conventional gamma camera and the CZT-based SPECT-CT system are given in Table 1.4. The GE Healthcare full-size

Table 1.4 The major contrasting points between conventional (NaI-based) gamma cameras and a cadmium–zinc–telluride (CZT)-based SPECT-CT system.

Attributes

Conventional gamma camera (NaI)

CZT-based gamma camera

Type of crystal

Scintillation

Semiconductor

Interaction of gamma rays

Produces light

Produces electron–hole pair

Photomultiplier tube

Necessary to convert light signal to electrical signal and amplify it

Not required

Detector assembly

Bulky

Sleek

Sensitivity

+

+++

Spatial resolution

+

+++

Energy resolution

+

+++

Image contrast

+

+++

Image quality

+

+++

Acquisition time

+++

+

Radiation exposure

+++

+

Available field of view

++

+++

Cost

+

+++

CZT based SPECT–CT system (Discovery NM/CT 870 CZT) provides up to 75% reduction in injected dose or scan time, improved system spatial resolution, from 4.3 to 2.8 mm, and exceptional energy resolution, 6.3% compared to 9.5%, according to data provided by them. They also claim greater than 40% improvement in SPECT contrast-tonoise ratio, 67% reduction in detector frame size, from 7.5 to 2.5 cm, and 25% greater optimal FOV than Nal  [69]. There are newer CZT systems, such as the VERITON CZT camera (Spectrum Dynamics, Caesarea, Israel), which are similar to PET-CT cameras in architecture with detectors arranged in a ring configuration around the patient [70].

Clinical Examples of SPECT–CT As there are chapters dedicated to the clinical applications of SPECT-CT system later in the book, we restrict ourselves to a few clinical examples. The selected clinical examples of SPECT-CT acquisitions shown in Figures 1.28–1.32 are such that, if not for combined imaging, the diagnosis might have been missed or delayed.

23

(a)

(b)

(c)

ANTERIOR

POSTERIOR

Figure 1.28 A 51-year-old female with a presenting complaint of left hip joint pain. (a) The bone scan showed increased tracer uptake in the upper part of the left hip joint. (b) and (c) The SPECT-CT scan of the pelvis showed subchondral cysts in the head of the left femur, suggestive of osteoarthritis. In this case the bone scan helped to find the active osteoblastic reaction in the left hip joint region as cause of pain and the CT component helped to localize abnormal tracer uptake. The morphological findings of osteoarthritis on CT added specificity to bone SPECT findings.

(a)

(b)

SPECT

CT

SPECT-CT fused image ANTERIOR

POSTERIOR

Figure 1.29 A 70-year-old male with diagnosed carcinoma of prostate. His serum PSA level was 34 ng/ml. (a) The whole-body bone scan shows focal increased tracer uptake in the upper dorsal vertebra. (b) The SPECT-CT scan of the thorax shows increased tracer uptake corresponding to sclerotic lesion involving the D2 vertebral body consistent with solitary skeletal metastasis. In this case, the bone scan detected a solitary osteoblastic lesion in the upper dorsal vertebra and CT localized it to the D2 vertebra showing sclerosis, hence confirming metastatic disease.

(a)

Early 20 min image

Delayed 2 hours image

(b)

Figure 1.30 A 35-year-old male with recurrent pancreatitis. He was found to have elevated serum parathyroid hormone levels during evaluation and was referred for a 99mTc-MIBI parathyroid scan. (a) Dual phase parathyroid scintigraphy in anterior static images showing focal abnormal tracer uptake near the lower pole of the left lobe of the thyroid gland (blue arrow) with focal tracer retention seen in the delayed image. (b) SPECT-CT of the neck showing abnormal tracer uptake corresponding to a 14 × 14 mm nodule in the left upper paratracheal region inferior to the lower pole of the left lobe of the thyroid gland suggestive of parathyroid adenoma. The SPECT-CT scan helped in the exact localization of the parathyroid lesion and aided with better surgical planning.

(a)

(b)

(c)

(e)

(d)

(f)

Figure 1.31 A 70-year-old male with chronic kidney disease and suspected pulmonary thromboembolism. The 99mTcmacroaggregated albumin lung perfusion scan (a, b) showed a perfusion defect in the right middle lobe. The SPECT-CT scan (e, f) showed a large perfusion defect in the middle lobe of the right lung with no morphological abnormality, such as a space-occupying lesion, consolidation or fluid accumulation on the CT scan. The CT scan (c, d) showed well-aerated lung parenchyma in the region of the perfusion defect, potentially eliminating the need for a ventilation scan to diagnose pulmonary thromboembolism.

26

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

SPECT

CT

SPECT-CT

ANTERIOR

POSTERIOR

Figure 1.32 A 30-year-old female who delivered a baby 2 years previously. She presented with complaints of pubic region pain. Her whole-body bone scan (a) showed mildly increased tracer uptake in the pubic bones (right > left). (b) The SPECT–CT scan of the pelvic region showed foci of increased tracer uptake in bilateral pubic bones (Rt > Lt) along with widened pubic symphysis joint space and subtle erosions of the right pubic bone suggestive of osteitis pubis. In this example, the whole-body bone scan and SPECT scan were able to pinpoint the possible site of pathology, while the CT scan confirmed the diagnosis of osteitis pubis with characteristic morphological features.

Conclusion In summary, in this treatise we have elaborated on the fundamental principles, imaging protocol, and clinical applications of the hybrid imaging modalities PET-CT and SPECT-CT to provide the readers with a comprehensive working knowledge on the subject. A plethora of illustrative clinical case examples have been provided deliberating on (i) the normal physiological variants and benign pathologies

mimicking malignancy, (ii) the enumeration and depiction of artifacts in PET-CT fusion imaging, and (iii) the value of combined imaging, where the PET-CT and SPECT-CT fusion imaging resulted in accurate diagnosis of pathology. With a number clinical and research applications continuing to evolve, it is expected the correlative anatomicalfunctional approach would play an important role in clinical decision-making for both oncological and nononcological applications.

References 1 Correlation, Definitions and Synonyms (2022). https:// www.macmillandictionary.com/dictionary/british/ correlation. 2 Altman, D.G. and Altman, E. (1999). Practical Statistics for Medical Research. Chapman & Hall/CRC.

3 Bradley, W.G. (2008). History of medical imaging. Proc. Am. Philos. Soc. 152 (3): 349–361. 4 Smith, T.A. (1998). FDG uptake, tumor characteristics and response to therapy: a review. Nucl. Med. Commun. 19: 97–105.

Introduction to Correlative Imaging

5 Basu, S., Hess, S., Nielsen Braad, P.E. et al. (2014). The basic principles of FDG-PET/CT imaging. PET Clin. 9 (4): 355–370. 6 Macheda, M.L., Rogers, S., and Best, J.D. (2005). Molecular and cellular regulation of glucose transporter (GLUT)proteins in cancer. J. Cell. Physiol. 202: 654–662. 7 Ganapathy, V., Thangaraju, M., and Prasad, P.D. (2009). Nutrient transporters in cancer: relevance to Warburg hypothesis and beyond. Pharmacol. Ther. 121: 29–40. 8 Kapoor, V., McCook, B.M., and Torok, F.S. (2004). An introduction to PET-CT imaging. Radiographics 24: 523–543. 9 Delbeke, D., Coleman, R.E., Guiberteau, M.J. et al. (2006). Procedure guidelines for tumor imaging with FDGPET/ CT 1.0. J. Nucl. Med. 47: 885–894. 10 Basu, S., Kwee, T.C., Surti, S. et al. (2011). Fundamentals of PET and PET/CT imaging. Ann. N. Y. Acad. Sci. 1228: 1–18. 11 Aukema, T.S., Vogel, W.V., Hoefnagel, C.A., and Valdes Olmos, R.A. (2010). Prevention of brown adipose tissue activation in 18F-FDG PET/CT of breast cancer patients receiving neoadjuvant systemic therapy. J. Nucl. Med. Technol. 2010; 38: 24–27. 12 Lakhani, A., Khan, S.R., Bharwani, N. et al. (2017). FDG PET/CT pitfalls in gynecologic and genitourinary oncologic imaging. Radiographics 37: 577–594. 13 Bhatt, G., Li, X.F., Jain, A. et al. (2013). The normal variant 18F FDG uptake in the lower thoracic spinal cord segments in cancer patients without CNS malignancy. Am. J. Nucl. Med. Mol. Imaging 10: 317–325. 14 Prabhakar, H.B., Sahani, D.V., Fischman, A.J. et al. (2007). Bowel hot spots at PET-CT. Radiographics 27: 145–159. 15 Karunanithi, S., Soundararajan, R., Sharma, P. et al. (2015). Spectrum of physiologic and pathologic skeletal muscle (18) F-FDG uptake on PET/CT. AJR 205: W141–W149. 16 Weber, W.A., Grosu, A.L., and Czernin, J. (2008). Technology insight: advances in molecular imaging and an appraisal of PET/CT scanning. Nat. Clin. Pract. Oncol. 5: 160–170. 17 Thie, J.A. (2004). Understanding the standardized uptake value, its methods, and implications for usage. J. Nucl. Med. 45: 1431–1434. 18 Kinahan, P.E. and Fletcher, J.W. (2010). Positron emission tomography-computed tomography standardized uptake values in clinical practice and assessing response to therapy. Semin. Ultrasound CT MR 31 (6): 496–505. 19 Cantatore, A. and Müller, P. (2011). Introduction to computed tomography. DTU Mechanical Engineering. Kgs. Lyngby, 76. 20 Kalender, W.A. (2006). X-ray computed tomography. Phys. Med. Biol. 51: 29–43.

21 Lusic, H. and Grinstaff, M.W. (2013). X-ray-computed tomography contrast agents. Chem. Rev. 113: 1641–1666. 22 Mattrey, R.F. and Aguirre, D.A. (2003). Advances in contrast media research. Acad. Radiol. 10: 1450. 23 Boellaard, R., Delgado-Bolton, R., Oyen, W.J. et al. (2015). FDG PET/CT: EANM procedure guidelines for tumour imaging: version 2.0. Eur. J. Nucl. Med. Mol. Imaging 42: 328–354. 24 Mawlawi, O., Erasmus, J.J., Munden, R.F. et al. (2006). Quantifying the effect of IV contrast media on integrated PET/CT: clinical evaluation. AJR 186: 308–319. 25 Beyer, T., Antoch, G., Muller, S. et al. (2004). Acquisition protocol considerations for combined PET/CT imaging. J. Nucl. Med. 45 (Suppl. 1): 25–35S. 26 Alessio, A.M. and Kinahan, P.E (2012). CT Protocol Selection in PET-CT Imaging. https://www.imagewisely. org/~/media/ImageWisely-Files/NucMed/CT-ProtocolSelection-in-PETCT-Imaging.pdf. 27 Kennedy, J.A., Israel, O., Frenkel, A. et al. (2007). Improved image fusion in PET/CT using hybrid image reconstruction and super-resolution. Int. J. Biomed. Imaging 46846. 28 Hofman, M.S. and Hicks, R.J. (2016). How we read oncologic FDG PET/CT. Cancer Imaging 16: 35. https:// doi.org/10.1186/s40644-016-0091-3. 29 Sureshbabu, W. and Mawlawi, O. (2005). PET/CT imaging artifacts. J. Nucl. Med. Technol. 33 (3): 156–161. quiz 163–164. 30 Nehmeh, S.A., Erdi, Y.E., and Ling, C.C. (2002). Effect of respiratory gating on reducing lung motion artifacts in PET imaging of lung cancer. Med. Phys. 29: 366–371. https://doi.org/10.1118/1.1448824. 31 Carney, J., Townsend, D.W., Kinahan, P.E. et al. (2001). CT-based attenuation correction: the effects of imaging with the arms in the field of view [abstract]. J. Nucl. Med. 42: 56P–57P. 32 Mawlawi, O., Pan, T., Cody, D.D. et al. (2004). Evaluation of a new CT truncation correction algorithm for accurate quantification of PET/CT images (abstract). J. Nucl. Med. 45: 413P. 33 Tokmak, H. (2013). Focal FDG uptake in lung parenchyma without structural alterations on CT. World J. Nucl. Med. 12: 38–40. 34 Fendler, W.P., Eiber, M., Beheshti, M. et al. (2017). 68 Ga-PSMA PET/CT: joint EANM and SNMMI procedure guideline for prostate cancer imaging: version 1.0. Eur. J. Nucl. Med. Mol. Imaging 44 (6): 1014–1024. 35 Virgolini, I., Ambrosini, V., Bomanji, J.B. et al. (2010). Procedure guidelines for PET/CT tumour imaging with 68Ga-DOTA-conjugated peptides: 68Ga-DOTA-TOC, 68Ga-DOTA-NOC, 68Ga-DOTA-TATE. Eur. J. Nucl. Med. Mol. Imaging 37 (10): 2004–2010. https://doi.org/10.1007/ s00259-010-1512-3.

27

28

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

36 Barthlen, W., Blankenstein, O., Mau, H. et al. (2008). Evaluation of [18F]DOPA PET-CT for surgery in focal congenital hyperinsulinism. J. Clin. Endocrinol. Metab. 93: 869–875. 37 Ibrahim, N., Kusmirek, J., Struck, A.F. et al. (2016). The sensitivity and specificity of F-DOPA PET in a movement disorder clinic. Am. J. Nucl. Med. Mol. Imaging 6 (1): 102–109. 38 Beheshti, M., Mottaghy, F.M., Paycha, F. et al. (2015). (18) F-NaF PET/CT: EANM procedure guidelines for bone imaging. Eur. J. Nucl. Med. Mol. Imaging 42 (11): 1767–1777. 39 Hill, D.L., Batchelor, P.G., Holden, M., and Hawkes, D.J. (2001). Medical image registration. Phys. Med. Biol. 46: R1–R45. [PubMed: 11277237]. 40 Hutton, B.F., Braun, M., Thurfjell, L., and Lau, D.Y.H. (2002). Image registration: an essential tool for nuclear medicine. Eur. J. Nucl. Med. Mol. Imaging 29: 559–577. [PubMed: 11914898]. 41 Gholipour, A., Kehtarnavaz, N., Briggs, R. et al. (2007). Brain functional localization: a survey of image registration techniques. IEEE Trans. Med. Imaging 26: 427–451. 42 Barratt, D.C., Penney, G.P., Chan, C.S. et al. (2006). Selfcalibrating3D-ultrasound-based bone registration for minimally invasive orthopedic surgery. IEEE Trans. Med. Imaging 25: 312–323. [PubMed: 16524087]. 43 Ma, B. and Ellis, R.E. (2003). Robust registration for computer-integrated orthopedic surgery: laboratory validation and clinical experience. Med. Image Anal. 7: 237–250. [PubMed: 12946466]. 44 LaCroix, K.J., Tsui, B.M.W., Hasegawa, B.H., and Brown, J.K. (1994). Investigation of the use of x-ray CT images for attenuation correction in SPECT. IEEE Trans. Nucl. Sci. 41: 2793–2799. 45 Blankespoor, S.C., Wu, X., Kalki, K. et al. (1996). Attenuation correction of SPECT using x-ray CT on an emission-transmission CT system: myocardial perfusion assessment. IEEE Trans. Nucl. Sci. 43: 2263–2274. 46 Liu, A., Williams, L.E., and Raubitschek, A.A. (1996). A CT assisted method for absolute quantitation of internal radioactivity. Med. Phys. 23: 1919–1228. [PubMed: 8947907]. 47 Da Silva, A.J., Tang, H.R., Wong, K.H. et al. (2001). Absolute quantitation of regional myocardial uptake of 99mTcsestamibi with SPECT: experimental validation in a porcine model. J. Nucl. Med. 42: 772–779. [PubMed: 11337575]. 48 Koral, K.F., Zasadny, K.R., Kessler, M.L. et al. (1994). CT-SPECT fusion plus conjugate views for determining dosimetry in iodine-131-monoclonalantibody therapy of lymphoma patients. J. Nucl. Med. 35: 1714–1720. [PubMed: 7931676]. 49 Koral, K.F., Dewaraja, Y., Li, J. et al. (2000). Initial results for hybrid SPECT-conjugate-view tumor dosimetry in 131 I-anti-B1 antibody therapy of previously untreated

50

51

52

53

54

55

56

57 58 59 60

61

62

63

64

patients with lymphoma. J. Nucl. Med. 41: 1579–1586. [PubMed: 10994741]. Kuhl, D.E., Hale, J., and Eaton, W.L. (1966). Transmission scanning: A useful adjunct to conventional emission scanning for accurately keying isotope deposition to radiographic anatomy. Radiology 87: 278–284. [PubMed: 5915433]. Bailey, D.L., Hutton, B.F., and Walker, P.J. (1987). Improved SPECT using simultaneous emission and transmission tomography. J. Nucl. Med. 28: 844–851. [PubMed: 3494829]. Tsui, B.M.W., Gullberg, G.T., Edgerton, E.R. et al. (1989). Correction of nonuniform attenuation in cardiac SPECT imaging. J. Nucl. Med. 30: 497–507. [PubMed:2786944]. Huang, S.C., Hoffman, E.J., Phelps, M.E., and Kuhl, D.E. (1979). Quantitation in positron emission computed tomography: 2. Effects of inaccurate attenuation correction. J. Comput. Assist. Tomogr. 3: 804–814. [PubMed: 315970]. Carson, R.E., Daube-Witherspoon, M.E., and Green, M.V. (1988). A method for post injection PET transmission measurements with a rotating source. J. Nucl. Med. 29: 1558–1567. [PubMed: 3261786]. Mirshanov, D.M. (1987). Transmission-Emission Computer Tomograph. Tashkent Branch, All-Union Research Surgery Center, USSR Academy of Medical Science, USSR. Kaplan, C.H. (1989). Transmission/emission registered image (TERI) computed tomography scanners. International patent application PCT/US90/03722. Madsen, M.T. (2007). Recent advances in SPECT imaging. J. Nucl. Med. 48: 661–673. Anger, H.O. (1958). Scintillation camera. Rev. Sci. Instrum. 29: 27–33. Bruyant, P.P. (2002). Analytic and iterative reconstruction algorithms in SPECT. J. Nucl. Med. 43: 1343–1358. Knoll, P., Kotalova, D., Kochle, G. et al. (2012). Comparison of advanced iterative reconstruction methods for SPECT/CT. Z. Med. Phys. 22: 58–69. Hudson, H.M. and Larkin, R.S. (1994). Accelerated image reconstruction using ordered subsets of projection data. IEEE Trans. Med. Imaging 13: 601–609. Ljungberg, M. and Pretorius, H. SPECT/CT: an update on technological developments and clinical applications. Br. J. Radiol. 91: 20160402. Ogawa, K., Harata, Y., Ichihara, T. et al. (1991). A practical method for position-dependent Comptonscatter correction in single photon emission CT. IEEE Trans. Med. Imaging 10: 408–412. Frey, E.C. and Tsui, B.M.W. (1996). A new method for modelling the spatially-variant, object dependent scatter response function in SPECT. IEEE Med. Imaging Conf. 2: 1082–1086.

Introduction to Correlative Imaging

65 Hoffman, E.J., Huang, S.C., and Phelps, M.E. (1979). Quantitation in positron emission computed tomography: 1. Effect of object size. J. Comput. Assist. Tomogr. 3: 299–308. 66 Pretorius, P.H., King, M.A., Pan, T.S. et al. (1998). Reducing the influence of the partial volume effect on SPECT activity quantitation with 3D modelling of spatial resolution in iterative reconstruction. Phys. Med. Biol. 43: 407–420. 67 Hutton, B.F. and Osiecki, A. (1998). Correction of partial volume effects in myocardial SPECT. J. Nucl. Cardiol. 5: 402–413.

68 Erlandsson, K., Buvat, I., Pretorius, P.H. et al. (2012). A review of partial volume correction techniques for emission tomography and their applications in neurology, cardiology and oncology. Phys. Med. Biol. 57: R119–R159. 69 GE Healthcare. (2021). https://www.gehealthcare.in/ products/molecular-imaging/nuclear-medicine/ nm-ct-870-czt. 70 Desmonts, C., Bouthiba, M.A., and Enilorac, B. (2020). Evaluation of a new multipurpose whole-body CzT-based camera: comparison with a dual-head Anger camera and first clinical images. EJNMMI Phys. 7:18. https://doi. org/10.1186/s40658-020-0284-5.

29

30

2 Basic Principles of Hybrid Imaging Leda Lorenzon, M. Bonelli, A. Fracchetti, and P. Ferrari Department of Medical Physics, Central Hospital of Bolzano, Bolzano, Italy

General Introduction Since the discovery of X-rays in 1895, noninvasive medical imaging has become established as an essential tool for evaluating and managing patients. Traditional imaging techniques, including X-ray, ultrasound, computed tomography (CT), and magnetic resonance imaging (MRI), can be used to depict morphological changes of patients’ anatomy. Conversely, functional imaging modalities, such as planar scintigraphy, single photon emission computed tomography (SPECT), positron emission tomography (PET), and magnetic resonance spectroscopy (MRS), are able to depict metabolic or functional changes, which are precursors of the changes accompanying pathological conditions. Each single imaging modality has its own strengths and drawbacks. Radionuclide imaging, such as PET and SPECT, offers very high molecular sensitivity (in the pico-nanomolar range) but provides relatively poor anatomic resolution. On the other hand, anatomical imaging modalities, such as CT and MRI, have high spatial resolution, but relatively poor sensitivity (in the millimolar range). Despite the clinical usefulness of stand-alone anatomical and functional imaging modality, the increased information that comes from integration of multiple modalities breaks new ground. To obtain multimodality imaging, it is essential to have an accurate spatial correlation between anatomical and functional information. Software-based image registration can fuse images from two or more different image datasets after they are acquired separately on different scanners, at different times. Software tools allow the display of transverse, coronal, and sagittal sections of the two image volumes, either side by side or as fused images, through an alpha-blending fusion algorithm in which the

two imaging modalities are superimposed. However, image registration can be challenging in areas of the body that can bend and flex (such as thorax and abdomen), and the quality of the resulting fused image could be affected. Hybrid imaging was developed with the aim of increasing the accuracy of the correlation between different types of image data, [1–3]. The description ‘hybrid’ denotes the use of one imaging unit that physically combines complementary imaging modalities to produce different image datasets, intrinsically co-registered. Examples of hybrid imaging devices include PET/CT, SPECT/CT, and PET/ magnetic resonance (MR). As we will explain in this chapter, hybrid imaging offers not only a way to simplify the image registration and fusion processes, but also several other advantages over standalone tomographic imaging techniques. We will briefly describe some basic physics concepts of single and combined imaging techniques: PET, SPECT, ultrasound, CT, and MRI.

Computed Tomography Introduction CT became computationally and technically feasible for the first time in the 1960s, when computer technology became available, albeit still in its infancy. Some original ideas underlying the tomographic theory had already been developed by the Bohemian mathematician J.H. Radon in his famous 1917 publication [4]. Based on this theory, it is generally possible to univocally derive the values of a function within a given spatial region knowing the values of line integrals through the region itself and using transform methods.

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach, First Edition. Edited by Ali Gholamrezanezhad, Majid Assadi, and Hossein Jadvar. © 2023 John Wiley & Sons Ltd. Published 2023 by John Wiley & Sons Ltd.

Basic Principles of Hybrid Imaging

The formulation of this theory was very general and abstract, but it can also be applied to regions of the human body, such as transversal layers. The line integrals are represented by X-ray projections through these layers and along different angular directions. The function to be determined is the distribution of the linear attenuation coefficient (LAC) μ(x, y). The greater the number of measured projections, the better the determination of the distribution. The great innovation and power of the CT technique lies in the possibility of representing in the images whole volumes of the human body in a three-dimensional (3D) way through a division into layers (tomograms), which in turn are divided into small parallelepipeds called voxels, instead of superpositioning images of entire anatomical sections, typical of conventional radiographic technique. This advantage is also shared with other imaging modalities such as magnetic resonance tomography (MRT), PET, SPECT, and ultrasound. The first practical realization in the medical field of a CT scanner was made by the engineer G.N. Hounsfield in 1972. The first CT scanners were limited in size and could only scan the head. Due to the limited power of radiation sources and the limited number of detectors, acquisitions with these single-layer scanners lasted many minutes. The acquisition mode with the first scanners was sequential (step and shot). In 1989 the first spiral scanner was introduced [5], where the acquisition and movement of the table are continuous. In 1998 the first multislice CT (four slices) was introduced, and in 2000 the first combined PET/ CT system appeared. Modern CTs have up to 256 detector rows and up to 512 layers, and can acquire a chest or abdomen image in less than a second.

Physical Principles of the CT Technique The physical principle on which the CT technique is based is the measurement of the overall attenuation of the X-rays through a layer of the human body, along different directions (for simplicity, let us initially examine acquiring a single layer with the patient table at rest). In practice, many transverse “radiographic projections” are acquired for each slice by rotating the X-ray tube and the detector around the patient’s body. For each trajectory traveled by the X-rays in the tissue, the intensity of the transmitted radiation I(si, θj) is measured as a function of the discrete position si, along the row of detector elements, and of the discrete angle θj, subtended by the acquisition system. The ratio between the intensity of the incident radiation and the transmitted radiation, I0/I(si, θj), represents the attenuis called the ation. The function pi, j0 log I 0 / I si , j0

Rotation X Ray tube

Fan beam

Detector array

Figure 2.1 Simplified diagram of CT acquisition system. At the bottom is the detector array (normally consisting of 64, 128 etc. detector rows and each row has around 800–900 detector elements) at the top the X-ray tube. The detector array and X-ray tube are solidly fixed together and go around the patient.

projection. The set of all projections constitutes the so-called sinogram, and the mathematical transformation that simulates the CT acquisition is called the Radon transform. A simplified diagram of the measurement setup is shown in Figure 2.1. The acquisition system revolves around the patient. In modern CT scanners, during one rotation up to about 2500 projections can be acquired. Each projection is composed of a number of values equal to the number of detector elements in the row. The set of all projections constitutes the raw data of a slice. To speed up the acquisition and therefore also reduce any movement artifacts, a large number of detector rows (up to 256) and a “pyramidal” beam, widened along the patient’s longitudinal direction, are used in modern CT scanners. A CT examination can be carried out in sequential or spiral (helical) mode. In the first case, the acquisition of a slice alternates the table feed and the feed is equal to the slice collimation (slice thickness). In the second case, patient movement and acquisition are continuous. In helical mode, the table speed can be set through the so-called pitch, p, defined as the ratio between table feed per rotation and the slice collimation. If p is less than 1, the acquired slices are partially overlapping (oversampling); for p greater than 1 the slices are

31

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

slightly spaced and it is necessary to interpolate the raw data. The pitch is chosen according to the diagnostic question and the anatomical region. For the thorax region, for example, where there are moving organs, p greater than 1 is normally chosen to reduce movement artifacts; for the abdomen, p less than 1 is chosen to improve image quality. In the past, it seemed that the helical mode was more often used in CT examinations. With the enormous technological developments of recent years, which have led CT manufacturers to enormously increase the number of detector rows, it has been necessary to return, for acquisitions with more than 128 rows, to the sequential mode. In fact, for detector arrays with more than 128 rows, due to the huge beam aperture in the z direction, and therefore also the huge table feed in helical mode, it is no longer possible to reconstruct images with enough quality.

Representation of CT Images The acquired CT raw data contain the patient’s anatomical information but cannot be visually interpreted. To make this information visible, a “reconstruction” operation (Radon antitransform) must be carried out on the acquired data (see Figure 2.2). As already mentioned, in the reconstruction the projections pi, j0 log I 0 / I si , j0 are transformed into the distribution of LAC μ(xk, yl) = μk, l, where k, l ∈ {1, 2, . . .., 512}. μk, l, however, is not suitable for graphical representation since its value depends on the energy of the photons (i.e. on applied voltage) and also on the unit of measurement chosen. For this reason, Hounsfield has introduced another quantity for the visualization of the images, the so-called

p(si, θj,)

Reconstruction

θj

CT number or CT#, which is calculated by the following affine transformation (affine function): CT # k ,l

xk

k ,l

water water

1000

The unit thus defined is called the Hounsfield unit and is indicated by the symbol HU. In this scale, called the Hounsfield scale, the CT# of water is 0 HU, of air is −1000 HU, of cortical bone is from about 300 to 1000 HU, depending on its density, of fat is from −100 to −80 HU, and of liver is 50–70 HU.

Reconstruction Methods The main reconstruction methods used today are filtered back projection (FBP), the only method used until a few years ago, and the iterative (IR) method, which is gradually replacing FBP in almost all types of examination. The FBP method is based on a mathematical convolution operation on the projections with a specific kernel, which depends on the type of examination, and on the “back projection” of the obtained convolution data on an image matrix, normally of 512×512 size. There are several types of kernels: smoothing kernels are used to improve low contrast resolution, i.e. to visualize soft tissues, sharp kernels to improve spatial resolution, i.e. to view bone structures or contrast medium, and standard kernels are a compromise to have a fairly good low contrast resolution and spatial resolution. Iterative methods normally use FBP reconstruction as a starting point. However, this image is mathematically forward-projected into the raw data space (Radon space). This forward-projected data set is compared with the acquisition

CT#(xk,yl,)

yl

si

32

Figure 2.2 CT acquisition of a Phantom Catphan 600: left, the row data set (sinogram); right, the reconstructed image.

Basic Principles of Hybrid Imaging

data set by creating a correction data set, which is in turn back-projected in the image space, creating a correction image matrix. This cyclic operation is iterated a few times until the correction is negligible. At this point the image is displayed. The most modern iterative reconstruction methods also use a mathematical modeling of the acquisition system (model-based IR) to transform data from the image space to the Radon space to further improve the image quality (reduction of artifacts and improvement of spatial resolution). In addition, statistical models are used to reduce noise. In addition to the possibility of using different kernels depending on the type of tissue to be analyzed (soft tissue, bone etc.), IR methods also allow the possibility, through a parameter called strength, which normally goes from 1 to 5, of setting the “degree of iterative action” of IR, going from an FBP-like image to a very “plastic” image.

Double-layer detector technology, developed in Philips Healthcare, does not use two different X spectra incident on the patient, but does a spectral analysis during the acquisition phase precisely through a double-layer detector. This detector is able to acquire “high” energy photons separately from “low” energy ones. From a conceptual point of view, the dual-energy technique is the forerunner of hybrid techniques, even if chronologically it came after some of them. In fact, in this case two identical techniques were combined with different acquisition parameters to improve, integrate, and add further diagnostic information with respect to the two individual acquisitions.

Nuclear Medicine Tomographic Imaging Single-photon Emission Tomography

Dual-energy CT CT sometimes has problems discerning bone structures from adjacent contrast medium enhanced structures, as these materials can give similar attenuations. To overcome these problems, dual-energy scanners have been developed that allow the attenuation coefficient to be studied also as a function of kV: μ = μk, l, m(kV). Since μ, particularly at low kV, depends not only on material density but also on its composition, through dual-energy acquisitions it is possible not only to distinguish bone from vessel structures with contrast medium, as mentioned above, but also to carry out material composition analysis. This technique also allows materials to be analyzed and distinguished, such as uric or nonuric kidney stones, etc. [6]. The dual-energy CT technique (DECT) consists of two simultaneous acquisitions of the same anatomical region but with two different voltages (two different X spectra), one normally of 80 or 100 kV and one of 140 kV. There are several technologies that allow for dual-energy acquisition: dual-source, fast kV switching, and double layer detector. Dual-source technology (DSCT), developed by Siemens Healthcare, is based on two X-ray tube-detector acquisition systems offset by approximately 90° and mounted inside the same gantry. In this case the two X-ray tubes can be supplied with two different voltages. A DSCT scanner can also be used with two identical voltages and high pitches to make fast acquisitions (flash acquisition mode, used, for example, in cardiac examinations) or with identical voltages and standard pitches to increase the effective mAs and so improve image quality in robust patients. The fast kV-switching technology, developed in GE Healthcare, uses a special high-voltage generator capable of switching the kV very quickly between low and high values.

Introduction

Single-photon emission computed tomography (SPECT) is an imaging technique that uses a conventional gamma camera to produce three-dimensional images or maps of functional processes within the body. As for other diagnostic nuclear medicine procedures (i.e. PET), it involves the administration to a patient of a radiolabeled compound that is then selectively absorbed by specific organs, tissues or cells within the human body. Most radiopharmaceuticals used in SPECT are labeled with radionuclides that emit gamma (γ) radiation. This radiation is emitted by nuclei that are in an excited state. These nuclei will then reach their ground state, either instantaneously or with some time delay (case of metastable nuclei) through the emission of one or more gamma A ) or through the process of internal photons (AZ X* ZX conversion. The most commonly used radionuclide is technetium-99m (99mTc), a metastable nuclear isomer of technetium that emits 140 keV gamma rays. Physical Principles of SPECT

Typically, a scintillation camera system (also known as an Anger camera or gamma camera) is used for imaging gamma rays. Designed by Hal O. Anger about 60 years ago [7], the gamma camera is based on the use of a single large area (e.g. 50 cm × 40 cm of thallium-activated sodium iodide, NaI(Tl)), scintillation crystal, equipped with a lead collimator (typically with parallel-holes design), that allows only photons traveling in a given direction to reach the detector (Figure  2.3). The collimator design is the most important factor in determining the sensitivity and spatial resolution of the imaging system. Behind the crystal, a light guide is optically coupled to an array of light-sensitive photomultiplier

33

34

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach Reconstruction

Figure 2.3 Schematic description of the basic components of an Anger scintillation camera, showing the scintillation crystal, the PMTs and the position logic circuit. The detail of a PMT is shown on the right.

tubes (PMTs) that proportionately convert the distribution of scintillation light into electronic signals. A PMT consists of a photocathode and a series of dynodes in an evacuated glass enclosure. The dynode chain provides internal amplification, resulting in an electrical output with a good signalto-noise ratio. The PMT outputs are then processed using the Anger logic electronics, generating output signals that represent the spatial position at which a photon interacts with the crystal. The total energy deposited by the absorbed gamma ray is computed by adding up the signals from all PMTs. A multichannel analyzer is used to measure the energy spectrum of the detected photons. Using the multichannel analyzer, an energy window is set on the photopeak(s) of the radioisotope. To deal with the limited energy resolution of the scintillation detector, the upper and lower level discriminators of the energy window are typically set to ±10% from the primary gamma energy. A computer system then displays two-dimensional (2D) projection images of the radioactivity distribution in the patient. SPECT images are obtained from 2D projection images, acquired at different views around the patient, using appropriate reconstruction algorithms (see section Image Reconstruction). SPECT configurations typically involve the use of one or more scintillation cameras (up to four) in conjunction with a gantry that allows rotation of the detection system around the patient (Figure 2.3).

Image Degrading Factors

Quantitative accuracy and quality of SPECT images are affected by several physical and clinical degrading factors. Physical factors include noise, attenuation, scatter, deadtime, and partial volume effects (PVEs) (due to the limited spatial resolution). There also other factors related to the patient, such as voluntary and involuntary motions. Noise is related to the statistical fluctuations in the number of detected counts. The counting statistics in nuclear medicine studies depend on camera sensitivity, the amount of injected activity, and the acquisition time. Noise varies also with the reconstruction algorithm and the implemented corrections. Attenuation from photoelectric absorption and Compton scattering is one of the major degrading factors of image quality. The amount of attenuation depends on the length of the path that the photon must travel in the tissue before reaching the detector and on the LAC of the tissue at that photon energy. Photons that have been scattered before reaching the radiation detector carry misleading information regarding the decay location. The fraction of scattered photons is high, about 20–30% in brain studies and about 30–40% in cardiac and body SPECT studies for 140 keV photons. Images obtained with a gamma camera are also degraded by the spatially varying collimator-detector response

Basic Principles of Hybrid Imaging

(CDR), whose shape is the primary factor determining the image spatial resolution in SPECT [8].

traditional NaI(Tl) but one of the following scintillators is chosen [9]: ●

Positron Emission Tomography Introduction

● ●

+

PET is based on β decay, in which nuclei with an excess of protons (p) achieve stability through the transformation of a proton into a neutron (n) and the emission of a positron (β+) and a neutrino (ν): p

n

The positron, inside the matter, follows a tortuous path of a few millimeters, progressively reducing its kinetic energy through Coulomb interactions with atomic electrons. To the attainment of thermal energy, the interaction of positron with electron takes place through the annihilation process (Figure  2.4), in which positron and electron convert their mass into energy, creating two 511 keV photons emitted simultaneously in opposite directions in the center of mass reference frame. Physical Principles of PET

As in gamma cameras, in almost all PET systems the detection of annihilation photons is performed with scintillator crystals coupled to PMTs. However, differently from SPECT, due to the need to detect high-energy photons (511 keV) and the requirement for good timing resolution, the scintillator crystal is not constituted by the

γ 511 keV β+ = positron

°

0 18

β+

e− = electron γ = photon

e− Annihilation reaction

●

bismuth germanium oxide, BGO lutetium oxyorthosilicate doped with cerium, LOS lutetium yttrium oxyorthosilicate doped with cerium, LYSO gadolinium orthosilicate, GSO.

The most commonly used arrangement of scintillation crystals and PMTs is the block detector (Figure  2.5), in which a group of crystal elements shares a smaller number of PMTs. A light guide, interposed between the matrix of crystal and the PMTs, allows scintillation light to be shared between PMTs. The interaction position of the annihilation photon inside a block is determined from the output signals of the PMTs using a weighted centroid algorithm, similar to the Anger logic of a gamma camera. Experimentally determined look-up tables are then used to perform a linearity correction. In a conventional block arrangement, each block operates independently of other surrounding blocks: the light is not transferred between blocks and PMTs of one block are unaffected by signals from an adjacent block. PMTs could also be located at the corners of an adjacent block, resulting in an alternative design with a better spatial resolution (the sharing detector in Figure 2.5). A complete scanner system is constructed as a cylindrical assembly of block detectors in a ring structure that surrounds the patient. To acquire multiple transverse slices simultaneously, several rings of detectors are used. In 2D acquisition mode, an array of septa is inserted between the detector rings, providing a collimation. When data acquisition is performed without any interplane septa, the scanner operates in 3D mode and coincidences from all axial angles in the field of view will be accepted, thus increasing system sensitivity by a factor of 4–6 as compared with 2D mode. All newer models of PET operate in 3D mode only. Unlike rotating camera SPECT systems, full-ring PET systems simultaneously acquire all projections required for tomographic image formation. Data Acquisition

γ 511 keV Effective positron range

Positron path β+

Figure 2.4 Positron–electron annihilation reaction.

Data acquisition in PET exploits the specificities of the annihilation process and is based on coincidence detection. Since annihilation photons are generated collinearly and simultaneously, only the photons that are revealed by two opposite detectors within a time difference of less than about 10 ns (coincidence window) are considered; all the other revealed photons (not in temporal coincidence) are rejected. Coincidence detection defines a line of response (LOR) along which the annihilation event has occurred and

35

36

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Segmented scintillator crystal PMT

PMT

PMT

PMT (a)

Standard block detector

PMT

(b)

Sharing detector

Figure 2.5 PET block detector arrangement.

obviates the need for a collimator to restrict the possible photon direction (as in SPECT). Each point on the LOR has an equal probability of having generated the annihilation event. In PET systems with time-of-flight (TOF) technology, the difference in the arrival times of the two photons is used to determine the most likely location of the annihilation event along the LOR. The spatial uncertainty ∆x, in the localization of the annihilation, through the measurement of TOF, is influenced by timing resolution ∆t through the following relation: x

c t 2

The incorporation of the TOF information improves the signal-to-noise ratio, significantly reducing the accidental random events and allowing high count rates to be handled [10]. Besides true coincidences, which occur when the two annihilation photons reach the detectors without interactions with the surrounding medium, there are also false coincidences. In fact, PET scanners also detect scattered, random, and multiple coincidences (Figure 2.6). In a scattered coincidence, at least one of the detected photons has undergone a Compton scattering event, which has changed its energy and trajectory, causing the incorrect identifica-

tion of the LOR. When two photons from two independent annihilation events are revealed in coincidence, a random coincidence has occurred. In multiple coincidences, more than two photons are detected in different detectors within the coincidence time window, with no possibility of defining a LOR uniquely. A figure of merit that is used to describe the performance of a scanner system is the noise equivalent count rate (NECR): NECR

T

T2 R 2S

where T, S, and R are the true, scatter, and random coincidence count rates, respectively. Another important parameter for the assessment of PET performance is the spatial resolution, expressed by the full-width at half maximum (FHWM) of the point spread function and parameterized with the following relation: FWHM

K

d/2

2

b2

0.0022 D

2

r2

p2

where d is the crystal size, b is related to the characteristics of the interaction between the gamma ray and the scintillator, the factor 0.0022D is the error associated with the nonperfect collinearity of annihilation photons (D is the

Basic Principles of Hybrid Imaging Gamma ray Assigned LOR Annihilation event

lost Scatter

lost True

Scatter

Random

Multiple

Figure 2.6 Different types of coincidences detected in a PET system.

detector ring diameter), and r is a parameter related to the nonzero positron range before the annihilation. The factor p is called the parallax error, whose origin is in the lack of information on the position of the depth of interaction point. K is related to the reconstruction algorithm and the choice of reconstruction parameters. The limited spatial resolution and the discrete image sampling of continuous 3D activity distribution produce the so-called PVE, which affects nuclear medicine images both qualitatively and quantitatively. Due to PVE, a small tumor will appear larger but less “aggressive” (less metabolically active) than it actually is.

Wl

Wp

Wl

Wp

Wu

Quantitative Data Corrections in SPECT and PET To achieve accurate quantitative information in PET and SPECT images, corrections for dead time, randoms (only for PET), scatter, partial-volume effects, and attenuation need to be performed. For more details on the quantitative correction methods for SPECT we suggest Bailey and Willowson [11], and Willowson et al. [12], while for PET we recommend Zaidi [13]. A brief description follows. Dead time losses are first modeled as a combination of paralyzable and nonparalyzable components and then the parameters of the model are determined experimentally through repeated measurements of a decaying source. In PET, random coincidences are usually subtracted using a delayed coincidence window, while scatter subtraction is performed via energy discrimination combined with models that simulate the scattering process (an example is the single scatter simulation method [14, 15]). Scatter events in SPECT are generally estimated using additional energy windows adjacent to the photopeak window. In the dual-energy window (DEW) method, only one energy window, located in the Compton edge before the photopeak window, is used. The triple-energy window (TEW) technique, reported by Ogawa et al. [16], uses two

110 120 130 140 150 160 Energy (keV)

120 130 140 150 160 Energy (keV)

Figure 2.7 Energy window settings used in DEW and TEW scatter correction. Wp is the photopeak window, which in case of 99m Tc is centered at 140 keV. Wl and Wu are the lower and upper energy windows, respectively.

narrow-scatter energy windows, one above and one below the photopeak window (Figure  2.7). More accurate and complex solutions, based on Monte Carlo estimations of the scattered events, have been also developed [17, 18]. A large variety of partial volume correction (PVC) methods have been developed and can be grouped in two categories: post-reconstruction and during-reconstruction approach. For additional information refer to Hoffman et al. [19], Rousset et al. [20], Müller-Gärtner et al. [21], and Teo et al. [22]. The most important correction for quantitative PET and SPET imaging is attenuation. Emitted gamma or annihilation photons, before being detected, must travel through the patient’s body. During this travel, they are subject to absorption and scatter, which leads to photon attenuation

37

38

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

PET image without attenuation correction

PET image with attenuation correction

Figure 2.8 Transaxial PET images from a NEMA IEC Body Phantom with (right) and without (left) attenuation correction.

(Figure 2.8). If the tissue is homogeneous, the attenuation is expressed mathematically by: I0 I

e

L

where I  and I0 are the attenuated and nonattenuated radiation intensities, respectively, μ is the LAC and L is the thickness of tissue. If the tissue is not homogeneous, the attenuation is expressed by a slightly more complicated formula: I0 I

x,y,z dl T

where T represents the trajectory of the photon in the body, μ(x, y, z) is the 3D LAC distribution (see section  Reconstruction Methods), and dl is the path element along the trajectory. For attenuation correction, the spatial distribution of LACs (attenuation map) in the patient’s body is therefore required. With the advent of dual modality (SPECT/CT and PET/CT) systems, the attenuation map can be obtained from CT image [23]. This first requires transformation of the CT data, which are expressed in Hounsfield units, into a map of LACs at the energy of the radionuclide gamma emission and then the obtained image must be down-sampled to the SPECT or PET format. The most widely used method to obtain the attenuation map from the acquired CT images is known as the bilinear segmentationscaling method [24, 25]. A threshold is first used to separate the bone component from the soft tissues of the CT image, and then separate scaling factors are used for the bone and nonbone components. In standalone PET systems, attenuation maps can be estimated from a transmission scan, where an external positron source is rotated around the patient and the attenuation of the transmitted photons is determined. In clinical centers than only have a SPECT scanner, not a hybrid SPECT/CT scanner, Chang’s method has also been widely

used. This method assumes that the attenuation coefficient is uniform over the body region. This assumption, however, leads to imperfect attenuation correction.

Image Reconstruction The final stage to obtain the in vivo 3D radiotracer distribution from multiple 2D projections is image reconstruction. Tomographic reconstruction techniques can be divided into two groups: analytical and iterative. Analytical methods include the well-established FBP method (see section Reconstruction Methods). The iterative approach allows more accurate results to be obtained since some physical degrading factors such as attenuation, scatter, dead time, and system-response function can be compensated for by modeling their effects in the projection and back-projection steps of the reconstruction algorithm. The most common iterative algorithm is the maximum likelihood expectation maximization (MLEM) [26], which accounts for the Poisson statistical nature of the emission data. However, due to its slow convergence rate, in commercial system an accelerated version of this iterative algorithm, known as the ordered subsets expectation maximization (OSEM) algorithm, is employed. In OSEM, projection data are broken up into subsets and the MLEM algorithm is applied only on a subset of projections. In general, OSEM speeds up MLEM by a factor that is approximately equal to the number of subsets used. More details on tomographic image reconstruction can be found in Cherry et al. [27], Valk et al. [28], and Wernick and Aarsvold [29].

Recent Developments Recent developments in the context of PET detector technology have involved the introduction of digital detectors (or more precisely solid-state read-out systems). With this

Basic Principles of Hybrid Imaging

term, we refer to the use of solid-state detectors, such as silicon photomultipliers (SiPMs), instead of the traditional PMTs. SiPM is a multipixel array of semiconductor avalanche photodiodes (APDs) biased to be used above avalanche breakdown in the so-called Geiger mode. Each SiPM can operate as a single-photon counter. Commercially, two designs of SiPM are available (Figure 2.9): the so-called analog SiPM (aSiPM) and the digital SiPM (dSiPM). In aSiPM, multiple SiPM pixels are connected and the signal from each pixel is summed by means of dedicated integrated circuits. Conversely, in dSiPM electronics are incorporated into each photodiode to provide position and energy signals as a direct digital output, avoiding the need for independent spatial decoding  [30]. dSiPMs allow 1 : 1 detector coupling, resulting in higher spatial and timing resolutions. SiPMs have several inherent advantages over PMTs, such as high quantum efficiency, compact and flexible shape, insensitivity to magnetic fields up to 9.4 Tesla and low-cost production [31, 32]. They also have good intrinsic timing resolution, allowing an improvement of the TOF performance. Physical performance of digital PET/CT systems has been evaluated  [33, 34] in comparison with traditional PMT-based systems, showing higher sensitivity, better performance under high count-rate conditions, improved spatial resolution, gain in sensitivity, improved accuracy, and better image quality, thus enabling reduced acquisition time or lower radiotracer dose. Other innovative solutions concern the introduction of continuous bed motion, which allows variable motion speed and optimal collection of PET photon counts for each body’s region. Recent advances in SPECT technology have involved the introduction of solid-state detectors, which have replaced

not only the PMTs (as in digital PET) but also the scintillator, thus providing a direct conversion of gamma photon energy to an electronic signal. The two most commonly used detector materials are cadmium zinc telluride (CdZnTe or CZT) or, less commonly, cadmium telluride (CdTe). Detectors are designed in small elements (approximately 2.5 mm × 2.5 mm square) and spatial localization of photon interaction consists of identifying the elements that have produced the signal. The advantages of solid-state detectors over conventional cameras are an improvement of the spatial resolution (in applications where the influence of collimator resolution is reduced, that is, when the camera can be placed close to the radioactive source), the full use of the detection area, since no Anger logic is performed for spatial localization, and superior energy resolution (around 5.5% compared to 9–10% of a conventional gamma camera). However, the sensitivity is reduced, and at 140 keV it is of the order of 70% for a 5 mm-thick CZT detector compared to almost 100% for the 9.5 mm-thick sodium iodide crystals coupled to PMTs. The even lower sensitivity for 511 keV annihilation photons is the reason why this technology cannot be implemented in PET detection. CZT technology for a conventional sized gamma camera was introduced in 2016 (GE Healthcare Discovery NM/CT 670 CZT). In 2018, a completely new design of SPECT system was introduced. The system is composed of 12 CZT independent detector arms which provide 360° coverage and allow each detector to be positioned as close as possible to the patient. In the field of software innovations, a new penalized likelihood iterative reconstruction algorithm that suppresses the noise in a large number of iterations has become available [35, 36].

Analog SiPM

Digital SiPM

Analog Silicon Photomultiplier Detector Vbias Readout ASIC

SiPM

Digital Silicon Photomultiplier Detector Vbias

Discriminator TDC

Time

Shaper

Energy

∫

ADC

Vbias

Cell Electronics Recharge

Detector + Readout

Cell Electronics Trigger Network

TDC

Time

Photon counter

Energy

Figure 2.9 Comparison between an analog SiPM and a digital SiPM. Source: Image courtesy of Philips Healthcare.

39

40

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Magnetic Resonance Imaging Introduction Felix Bloch at Stanford and Edward Purcell at Harvard in 1946 observed the nuclear magnetic resonance (NMR) phenomenon  [37, 38]. An MR scanner consists of three main hardware components: a main magnet, a magnetic field gradient system, and an RF system. The function of the magnet is to generate a strong uniform static magnetic field for polarization of nuclear spins in an object. To provide good image quality the magnet must have moderate homogeneity over a large volume. The magnetic field gradient system consists of three orthogonal gradient coils designed to produce time-varying magnetic fields. Gradients are a crucial component of an MRI scanner for signal localization and gradient strength, and are normally measured in units of millitesla per meter (mT/m). The RF system consists of a transmitter coil that can generate a rotating magnetic field, for excitation of a spin system, and a receiver coil that converts processing magnetization into an electrical signal. A single coil can be used as both a transmitter and receiver coil. MR images are extremely rich in information content. The image pixel value is in general dependent on intrinsic parameters: the nuclear spin density p, the spin lattice relaxation time T1, the spin–spin relaxation time T2, and molecular motions. The imaging effects of these parameters can be suppressed or enhanced by a set of operator-selectable parameters: repetition time (TR), echo time (TE), and flip angle (α). An MR image obtained from the same anatomical site can look drastically different with different data acquisition protocols.

Figure 2.10

Rotation of a proton about an axis.

Figure 2.11

Precession of a proton around B0.

Physical Principles of MRI Atomic nuclei are made up of protons with positive charge and neutron with zero charge. The basis of the NMR phenomenon is a property called “spin”, which can be represented as a motion of rotation of the particles around the axis (Figure 2.10). Under the effect of a static magnetic field (B0), nuclei are characterized by a precession of the rotation axis around the direction of B0 (Figure 2.11). The precession frequency (ω) is determined by the intensity of B0 and a constant (characteristic of each nucleus), called the gyromagnetic ratio (γ), according to the Larmor equation: ω = γB0. In the case of hydrogen nuclei, the gyromagnetic relationship is approximately 42.6 MHz/T. The magnetic contribution of each individual proton can be described with a vector. The spin is quantized, thus it cannot take continuous values, only discrete ones: +1/2 and −1/2.

MRI Signal Formation MRI signal formation depends on the following: Static magnetic field (B0) of high intensity (usually 0.5– 1.5 T in clinical diagnostic activity). This magnetic field is usually aligned parallel to the major axis of the patient (defined by convention as the z or longitudinal axis). The plan perpendicular to it, i.e. the axial plane of the patient, is called the xy or transverse plane. ● Atomic nuclei of the biological tissues. ● Radio frequency (RF) coils to transmit and receive electromagnetic waves. ●

Basic Principles of Hybrid Imaging

In the absence of B0, the protons are randomly oriented in all directions in space, without any preferential direction or resulting magnetization. The insertion of protons in a high-intensity B0 determines the following phenomena: ●

●

protons behave like small magnets, aligning themselves in the direction of B0 protons can assume a down-state (magnetic dipole moment parallel to B0) and an up-state (magnetic dipole moment anti-parallel to B0).

Protons in the up-state have a higher content energy compared to those in the down-state. With B0 equal to 1 T, there is a minimal difference in the number of protons in the two states. Such a small difference is measured in parts per million (ppm): for every 1 000 000 spins with magnetic dipole moment antiparallel to B0 there are 1 000 007 spins with parallel magnetic dipole moment. Although this difference of population states is of small entity, it is the basis of the MR signal formation [39]. The magnetization vector of each proton is not therefore perfectly aligned to the B0 but describes an ideal rotation cone around the z axis. In conditions of equilibrium, the sum of the small magnetic vectors can be represented by a single vector called macroscopic magnetization (MM), which has a direction toward that of B0 and an intensity proportional to the difference of population between the down- and up-states. Since the direction of this MM lies on the z axis and is parallel to B0, it is usually called MM longitudinal (MML) or Mz. As the intensity of B0 increases, the intensity of vector MML also increases, with a consequent increase in the MRI signal. To measure this MM and therefore get the MR signal, we need to flip it from the z axis, where it is dominated by the B0, to the perpendicular plane, xy. To overturn the MM on the transverse plane xy and therefore obtain a signal, we

Figure 2.12 Vector MM flipping under the effect of an RF pulse from the z axis to the xy plane. z

send a polarized electromagnetic wave (Figure 2.12) in the orthogonal plane at B0 (circular polarization) and with a frequency equal to that of the Larmor (resonance) frequency. In this condition, the system resonates and produces a passage of energy from the electromagnetic wave to the proton system, with consequent excitement. The radiofrequency is called the 90° RF pulse. The MML vector can, however, be deflected by an arbitrary angle (α), different from this borderline case, by changing the duration and intensity of the RF pulse.

Free Induction Decay and Spatial Encoding On cessation of the RF pulse, the MM is reversed on the plane transverse and on this plane it precedes with a frequency defined by Larmor equation. If we have a radio frequency coil placed transversely with respect to MML, the variation of the magnetic field generated by the precession of the MMT inside the coil itself generates an electrical signal measurable according to Faraday’s law. This MRI signal is called free induction decay (FID) and is the expression of the return energy of protons at initial conditions (Figure 2.14). To produce a 3D image, the FID resonance signal must be encoded for each dimension. The encoding in the axial direction (B0 direction) is accomplished by adding a gradient magnetic field to B0 (Gz) (Figure 2.13). This gradient causes the Larmor frequency to change linearly in the axial direction. Gy is then applied, causing the resonant frequencies of the nuclei to vary according to their position in the y direction. Gy is removed and Gx is applied perpendicular to Gy: the resonant frequencies of the nuclei vary in the x direction due to Gx and have phase variation in the y direction due to Gy. 2D Fourier transform is then used to transform the encoded image to the spatial domain.

Impulso RF

z

y y

X

X

41

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

e Slic

tion

loca

Frequency

Selective Excitation

Magnetic field

42

Mz

time

Low

High Frequency

Figure 2.13 Slice selection.

Longitudinal Magnetization Recovery: Relaxation Time T1 The equilibrium recovery of the relationships between protons with spins aligned with the field (parallel) and protons with spins against the field (anti-parallel), and the consequent reconstitution of MML occurs in a process called longitudinal relaxation or spin-lattice relaxation. Considering an MRI sequence, the TR is the time between one RF excitation and the next. To get a signal difference dependent on the different T1  values of two tissues (the T1-weighted image) it is therefore necessary to use a sufficiently short TR because the tissues do not have time to recover all MML and the greater the percentage of recovered MML, the shorter the T1 of the tissue (Figure 2.14).

Transverse Magnetization Decay: Relaxation Time T2 At the end of the RF stimulation (90° RF pulse), the proton system is in phase coherence. After the removal of the 90° pulse, this order progressively decreases, resulting in the loss of MMT. This process is called spin-spin relaxation. Unlike T1 relaxation, in which there is a “right” speed for maximum efficiency in the spin-lattice relaxation process, for T2 relaxation there is a linear relationship with the individual proton’s speed. The TE is a technical parameter of acquisition, which indicates how long from the end of RF stimulation the signal is sampled. With the same quantity of magnetization on the transverse plane, to a specific TE, tissues with higher T2 have lost few MMT quantities and therefore produce a higher signal. As a consequence, to get a signal difference depending on the different T2 values of the tissues (image weighed in T2), it is necessary to use a long TE (Figure 2.14).

Mxy

Figure 2.14 3D representation of FID. The projection on the z axis shows T1 relaxation Mz(t). The projection on the xy plane shows the T2 relaxation Mxy(t).

Ultrasound Imaging Ultrasound imaging (sonography) is a diagnostic medical procedure that uses high-frequency sound waves to generate dynamic images of organs, tissues or blood flow inside the body.

Physical Principles In physics, sound is a mechanical wave, a pressure disturbance that results from vibration. The two conditions that are required for the generation of a sound wave are a vibratory disturbance and an elastic medium, the most familiar of which is air. Sound waves propagate longitudinally in all materials. Sound waves are often simplified as sinusoidal plane waves, characterized by the following properties: ●

●

●

●

Amplitude: maximum displacement of the medium particles from their equilibrium position. Frequency (f): the number of cycles or pressure changes that occur in 1 s. It is measured in Hertz (Hz). Frequency is determined by the sound source only and not by the medium in which the sound is traveling. Wavelength (λ): the distance traveled in one cycle. It is inversely proportional to the frequency. Higher frequency results in higher resolution images but less penetration in tissues. Propagation speed: the distance traveled per unit time. It depends upon the type of medium and its state. In air, the propagation speed of sound waves is of the order of 340 m/s, in water it is 1450 m/s. The average speed in

Basic Principles of Hybrid Imaging

Infrasound

Acoustic

Ultrasound 20 kHz

2 MHz

20 Hz 200 MHz

Figure 2.15 Frequencies of sound waves.

tissues, used in sonography to estimate the distances traveled by ultrasound in the body, is set at 1540 m/s. There is a relation between the wavelength of a sound wave, its frequency, and the speed of the wave (V), such that: V

f

Humans are generally capable of hearing sounds between 20 and 20 kHz. Sounds below 20 Hz are called infrasound, whereas those above 20 kHz are called ultrasound. Ultrasound for medical imaging applications has a range of frequency between 2 and 10 MHz (Figure 2.15).

Interaction of Ultrasound with Tissue As sound waves travel through a patient’s body, there is a gradually reduction in the intensity of the wave in a process known as attenuation (see Table 2.1). This occurs as a result of reflection, refraction, scatter, and absorption. The degree of sound wave attenuation depends on the type of tissue and is expressed by the attenuation coefficient, which is also an increasing function of the frequency: over a given distance, higher frequency sound waves generally have more attenuation than lower frequency waves. Reflection and refraction occur at surfaces that are large compared with the wavelength of ultrasound. When the tissues on either side of the boundaries have differences in acoustic impedance, Z (the product of the density and the propagation speed), a reflection occurs. Reflection of the beam is called an echo and is the basis for ultrasound imaging production. Larger differences in these acoustic impendences result in more reflection (see Table  2.2). Reflection can be specular or diffuse. When the incidence surface is sufficiently regular (as for bone), a specular reflection occurs (Figure  2.16). In diffuse reflection, the sound is back-scattered at many angles. The optimal reflection for image production occurs when the angle of incidence approaches zero and is perpendicular to the tissue of interest. Refraction occurs when the transmitted wave deviates from its original direction due to transition through an

Table 2.1 Attenuation coefficients for different biological tissues.

Materials

Attenuation (dB/cm/MHz)

Blood

0.18

Fat

0.6

Kidney

1

Muscle (perpendicular to the fibers)

3.3

Muscle (along the fibers)

1.2

Brain

0.85

Liver

0.9

Lung

40.0

Skull

20.0

Aqueous humor

0.022

Vitreous humor

0.13

Water

0.0022

Table 2.2 Percentage reflection of ultrasound for normal incidence at various boundaries. Boundary

% reflection

Fat/muscle

0.18

Fat/kidney

0.6

Soft tissue/water

0.2

Bone/fat

49

Soft tissue/air

99

interface between tissues with differing wave speeds (Figure  2.16). The amount of deviation is described by Snell’s law: sin 1 VL1

sin 2 VL2

where VL1 and VL2 are the longitudinal wave velocities in material 1 and 2, respectively.

43

44

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

θ1

θ1

Z1 Z2

θ2

artificially produced piezoelectric ceramics are lead zirconate titanate (PZT), barium titanate, and lead titanate. An ultrasound transducer is an electromechanical device that transforms a potential difference into an ultrasound package and vice versa. It is has five components (Figure 2.18): ●

●

Figure 2.16 Refraction and reflection.

●

●

θ1

●

Z1 θ2

Z2

Backing or damping material that absorbs the backwarddirected energy and stops the vibrating transducer so that only a short, sharp pulse of ultrasound (typically two to three cycles) is produced. A plastic film, which helps protect the active elements and helps transmission of sound into the patient. One or more piezoelectric crystals (generally PZT) placed between two electrodes. Material for the acoustic matching between crystal and tissues (quarter-wavelength matching layer): the thickness of λ/4 allows the transfer of energy from the transducer to the patient to be maximized. Focusing lens.

A short voltage pulse generates vibration in the piezoelectric crystal, causing a sudden change in pressure and therefore the creation of an ultrasound package (Figure 2.19). PZT plate Lens

Figure 2.17 Refraction and artifacts. Electrical leads

Refraction is the main cause of artifacts in clinical ultrasound images (Figure 2.17). When a wave encounters a structure much smaller than its wavelength, energy is scattered in many directions. This typically occurs on interaction with red blood cells. However, the major contribution to attenuation is absorption, where wave energy is lost in heat.

Transducers Ultrasound waves are generated from a device called a transducer. A transducer has the function of transforming a physical quantity (such as temperature, pressure, etc.) into another quantity (e.g. a potential difference) or vice versa. Ultrasound transducers work on the piezoelectric principle. Piezoelectricity is the property of some materials to develop electric charge on their surface when mechanical stress is exerted on them. The electrical response to mechanical stimulation is called the direct piezoelectric effect, and the mechanical response to electrical simulation is called the converse piezoelectric effect. Some crystalline materials, such as quartz and tourmaline, have a natural piezoelectricity. The most commonly

Backing layer

Matching layer

Figure 2.18

Basic components of an ultrasound transducer.

d D(t)

t

ΔV

Figure 2.19 An ultrasound transducer works on the piezoelectric principle.

Basic Principles of Hybrid Imaging

The greatest deformations of the crystal occur when the frequency of the alternating voltage is equal to the resonance frequency of the crystal. The resonance frequency, fresonance, is related to the thickness, d, of the crystal and to a material constant, const, according to: fresonance

const 2 d

●

●

The line along which the echo is located, known from the angular position of the transducer. The depth from which the echo comes, which is determined by measuring the time elapsed between the emission of the impulse and the arrival of the echo (TOF).

Display Modes

Probe Operation The transducer transmits signal pulses about every microsecond, and then waits for the reflected pulses (Figure 2.20). The frequency with which the transducer transmits the ultrasound packets is called pulse repetition frequency (PRF). When the reflected signal reaches the transducer, the piezoelectric crystal undergoes a deformation that creates a small potential difference. This small signal is then amplified with a time gain compensation (TGC), and an increased gain is applied to the deeper signals to compensate for greater attenuation. In this way, the intensity of the echo depends only on the difference in acoustic impedance between the crossed interfaces.

Signal Processing Signal processing consists of recognizing the position of the interface that has produced the reflected wave (the echo) and the intensity of the echo itself. The intensity of the echo derives from the amplitude of the potential difference generated by the piezoelectric crystal, compensated with TGC, and depends on the differences in acoustic impedance of the media. The position of the interface is obtained from two parameters:

Once the information has been gathered, the data can be displayed. There are three main display modes: amplitude (A-mode), brightness (B-mode), and motion (M-mode). In A-mode, each echo appears in the form of a light peak of height directly proportional to the intensity of the signal and positioned at a distance that depends on the depth from which it arrives. Because this mode allows for precise measurements, it is used in ophthalmology for ocular measurements. In B-mode, the varying amplitudes are converted into dots of varying intensity, depending on the echo intensity, used to generate anatomic images with the depth proportional to the TOF of the pulse. Stronger reflections appear as brighter white dots, whereas weaker reflections appear as darker gray dots. M-mode records the position and movement of the echo. The echoes appear in the form of bright points of height directly proportional to the amplitude of the echo: if the interface remains stationary, this is represented by a straight line; in the case of a moving interface, this is represented by a sinusoidal line.

Doppler The Doppler effect (or the Doppler shift) is the change in frequency of a wave in relation to an observer who is moving relative to the wave source.

Pulse Length (PL)

Pulse Repetition Frequency (PRF) – PRF per unit time = 3

Figure 2.20 Transmitted pulses.

45

46

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

f

f0 V0

V0

∆f ≅

V θ

2 f0 v cos θ v0

Figure 2.21 Doppler effect in diagnostic ultrasound. The transducer transmits acoustic waves with a frequency f0. Due to the movement of red blood cells (orange structure), with a speed v with respect to the medium, the acoustic waves are then received by the transducer with frequency variation ∆f.

In diagnostic ultrasound, Doppler shifts are created when transmitted sound waves strike moving red blood cells (Figure 2.21). Considering a single red blood cell, which moves at speed v with respect to the medium and at an angle θ with respect to the direction of ultrasound, which has a frequency  f0 and speed v0, the transducer transmits and receives the acoustic waves with a frequency variation of: f

2 f0 cos 0

Frequency variation generally falls in the range of 20 Hz–20 kHz and therefore in the spectrum of frequencies perceptible to humans. During an examination, it is possible to hear sounds whose frequency is linked to the speed of the red blood cells inside the vessel under test. Not all red blood cells have the same speed, but they have a distribution of velocity that is typically parabolic in the veins, whereas in large arteries it can vary during the cardiac cycle. It is therefore necessary to perform a frequency analysis (using a proper signal processing instrument such as fast Fourier transform) to obtain the frequency distribution and so the distribution of red blood cell velocities. Different types of Doppler ultrasound devices can be found, but the two most common are continuous wave (CW) Doppler and pulsed Doppler. In CW Doppler, there is one crystal that transmits continuously and one that receives continuously. This configuration can accurately record high velocities but cannot determine the exact location of moving red blood cells. In pulsed wave Doppler (PW), the flow is studied at a predetermined distance and the transducer is a single crystal that functions as a transmitter and receiver. In this case,

short pulses are emitted and the transducer itself receives the scattered signal after a certain time interval, which corresponds to the distance between the probe and the vessel being investigated. PW Doppler is inaccurate in the measurement of high-velocity signals. Both types of Doppler ultrasound imaging are represented on a Cartesian plane as graphs of speed (typically in m/s) as a function of time. Doppler information can be presented in two ways: color Doppler and power Doppler. Color Doppler is a panoramic representation of the flow with a chromatic scale: the blood flow that moves away from the probe is colored red, while the blood flow that moves toward the probe is colored blue. Power Doppler shows magnitude of flow: blue represents more intense flow and red less intense flow. Power Doppler is often more sensitive to flow, but it is not able to distinguish the direction of the movement. Power Doppler is mainly used to study the smallest and deepest vessels, or those vessels which, due to too slow flow, cannot be well highlighted with color Doppler.

Hybrid Imaging Hybrid imaging denotes the use of one imaging unit that physically combines complementary imaging modalities to produce different image datasets intrinsically co-registered. Examples of hybrid imaging devices include PET/CT, SPECT/CT, and PET/MR. Hybrid imaging differs from image fusion, which is based on the use of software registration techniques to coregister different image datasets produced on geographically remote imaging devices.

Basic Principles of Hybrid Imaging

SPECT/CT and PET/CT Systems The diagnostic potential offered by the combination of nuclear medicine modality (SPECT or PET) with highresolution anatomical imaging modality (as CT) has driven the development of hybrid systems such as SPECT/CT and PET/CT, introduced clinically in 1999 and 2000, respectively. These systems consist of a single unit that integrates two imaging modalities, allowing data acquisition of each modality in a single patient study. Data are acquired sequentially, with the patient table moving between the two units. Simultaneous data acquisition is not possible due to cross-talk between subsystems. Most SPECT/CT designs involve the use of a dual-head gamma camera with a high-performance multislice CT component to benefit from improved image quality, faster data acquisition, and a broad choice of CT protocols. Traditionally in SPECT/CT systems the functional data act as the primary source of information and the CT data as a secondary source. In 2013, xSPECT technology, a truly revolutionary product in the field of SPECT/CT, was introduced [40]. xSPECT technology represents a new concept, a more complete integration of SPECT and CT, by shifting the imaging viewpoint away from the SPECT frame-ofreference and into the CT frame-of-reference. CT, which is known for its sharper spatial resolution, is the foundation to align SPECT and CT, thus preserving its spatial accuracy. Moreover, xSPECT bone is a dedicated solution of xSPECT for bone SPECT studies. A CT-based tissue segmentation is incorporated into SPECT reconstruction to provide SPECT images with bone anatomy appearance. PET/CT systems are accurately aligned combinations of stand-alone CT and PET scanners with a shared patient bed. They are available commercially with in two designs: a more compact system with a unique gantry or two separated gantries with a gap to allow better patient comfort. The advantage of this hybrid approach is that anatomical and functional images are inherently co-registered. This allows an immediate fusion of transmission images, acquired with CT, with the emission images to provide functional-anatomical maps for precise localization of radiotracer uptake. It has been also shown that, due the synergy of complementary information, the diagnostic information of the hybrid system is superior to that of stand-alone solution. Moreover, CT images can be used for attenuation correction (see section Quantitative Data Corrections in SPECT and PET). Compared with other methods of attenuation correction, such as transmission with external radioactive sources, CT-based attenuation correction (AC) provides advantages in terms of reduction of statistical noise and acquisition time, and a greater accuracy of attenuation maps. However, incor-

rect spatial registration between the two imaging systems, due to patient motion between the two scans or different respiratory dynamics during the two acquisitions, could lead to incorrect correlation between anatomical and functional imaging and to inaccurate attenuation correction. Another source of bias in CT-based AC is the presence of metal implants (e.g. dental implants, hip prosthesis, pacemaker, etc. [41]), which cause artifacts to appear as dark and bright streaks in CT. Metal artifact reduction (MAR) algorithms are currently available for clinical software [42]. Finally, the possibility of performing anatomical and functional procedures in a single session is logistically advantageous for both the patient and the hospital. However, with the inclusion of CT in SPECT and PET components, there is an increase in radiation dose absorbed by patients undergoing nuclear medicine examinations.

PET/MR On the back of the success of PET/CT imaging, in 2010 a new hybrid imaging modality entered the clinical arena. In this year, the first whole-body systems for PET/MR imaging became available on the market. When compared to CT, MR has important diagnostic advantages, namely better soft-tissue contrast and reduced radiation dose. MR can also provide functional and anatomical information at the same time. These aspects make PET/RM an interesting alternative to PET/CT. However, PET/RM development has been hampered by technological problems due to the mutual interference between PET and MRI subsystems. From the PET perspective, PMTs are highly sensitive to magnetic fields and therefore cannot operate inside a MR magnet. From the MR perspective, PET hardware could produce electromagnetic interference with consequent negative effects on MR imaging, such as reduction of SNR, artifacts, inhomogeneity of the static field, and eddy currents [43]. One possible option to address these issues is to spatially separate the PET and MR subunits, in combination with proper shielding. This solution allows the sequential acquisition of PET and MR signals. An alternative approach is to acquire the PET and MR data simultaneously with fully integrated PET/MR systems. The development of fully integrated systems has been possible thanks to technological innovation. Solid-state photodetectors (such as APDs or SiPMs), which are insensitive to the magnetic field, instead of the traditional PMTs were introduced. Nonmagnetic versions of the standard PET components (e.g. resistors, capacitors, connectors) have allowed the main magnetic field homogeneity to be maintained. Scintillator materials with a magnetic susceptibility close to that of human tissue (i.e. lutetium-based

47

48

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

scintillators instead of lutetium gadolinium oxyorthosilicate) have been implemented in combination with proper shielding of PET components to minimize electromagnetic interference in the radiofrequency part of the spectrum. Finally, the introduction of larger MR diameter bore scanners has allowed PET detectors to be positioned between the body radiofrequency coil and the gradient set. Simultaneity is a unique aspect of integrated PET/MR imaging because no other combined imaging system offers this capability. Unlike sequential acquisition, simultaneous acquisition has some important advantages, including temporal correlation of PET and MR imaging data, shorter scan times, smaller footprint, and minimization of misregistration from differences in patient position and interscan motion. Moreover, simultaneous PET-MR imaging offers the ability to correct PET imaging for voluntary and involuntary motion blurring, since real-time 3D MRI can be used to provide temporally varying anatomic information [44]. In PET/RM systems, attenuation correction of PET images is not as simple a task as in PET/CT units. MR images cannot be used directly to derive attenuation maps, as in CT, since the MR signal is not correlated directly with electron density, which is the primary source of photon attenuation. Therefore, it could be that structures (e.g. bone) with a high photon attenuation have a near-zero signal (as air) in conventional MR images. MR-based attenuation correction (MRAC) algorithms can be categorize into three generic groups: segmentationbased approaches, atlas registration, and emission/ transmission-based methods. Each of these algorithms has its advantages and disadvantages. Here, we will only talk about the segmentation-based method, which is the method of choice for all major PET/RM vendors. The segmentation-based algorithm relies on the use of coronal two-point Dixon T1-weighted MR acquisition that allows reconstruction of fat-only, water-only, and fat– water images and results in tissue segmentation of air, fat, muscle, and lungs. Early efforts assigned a predefined LAC (0, 0.022, 0.085, 0.100 cm–1) to each voxel class to generate the attenuation maps. More recently, methods to assign a continuous LAC value have been proposed [45, 46]. Accurate attenuation correction is provided in the soft tissue, but bone tissue is misclassified as soft tissue, leading to important quantification errors (up to 31% [47]) when quantifying osseous lesions. This issue has been partially addressed by the use of modified segmentation methods based on ultrashort echo time sequences (UTE) that are better able to detect bone, a tissue with an inherently short T2  [48] or more efficiently by integrating Dixon-based MR-AC with atlas-based methods [49] to retrospectively add bone information to MR-based attenuation maps.

On an integrated PET/MR scanner, in addition to tissue attenuation, other sources of attenuation should be considered, such as the patient table, patient positioning aids, and radiofrequency coils that are placed within the PET field of view [48]. Hardware component AC is generally performed during the PET reconstruction process using predefined attenuation maps obtained by systems manufacturers from CT transmission scans of the RF coils and the patient table [50, 51]. A source of bias in MRAC is related to truncation artifacts, due to limited MR field of view, which could result in partial or complete truncation of a patient’s arms and in turn generate an undercorrection of emission data  [52]. Truncation correction may be based on MR images  [53, 54] or PET images  [55]. AC from MR imaging is also affected by metal artifacts, which introduce signal voids or local image distortions in the MR-based AC implant volume.

Triple Modality Scanner Commercially, triple modality solutions are available, such as the SPECT/CT/PET system (Mediso AnyScan) or the Trimodality PET/CT and MRI scanners by General Electric. In the latter, two separate systems (PET/CT and MRI) are located in adjacent rooms with a common patient table that is moved from one scanner to another. The patient can undergo PET/CT acquisition followed by MR examination without changing position and whole-body image fusion of PET, CT, and MRI can easily be obtained.

Image Fusion of Real-time Ultrasound Several ultrasound systems for image fusion of real-time ultrasound with CT, MRI, or PET/CT are commercially available. Real-time image fusion requires a tracking system and a co-registration process [56]. Tracking systems allow visualization, in real-time, of the ultrasound probe position in 3D space throughout the procedure. The two most popular tracking methods are: ● ●

electromagnetic tracking (e.g. real-time virtual sonography) mechanical position encoding.

In electromagnetic (EM) tracking, the position of the ultrasound probe is obtained through a small sensor attached to the probe that interacts with a magnetic field generator external to the patient. The sensor acts as a coil in a variable magnetic field: because of Faraday’s Law, a current is generated in the coil when placed in a continuously changing magnetic field. As the ultrasound probe moves, the magnitude of the current in the position sensor changes with respect to the magnetic field. With this information, the position sensor unit, installed in the ultrasound system,

Basic Principles of Hybrid Imaging

calculates the exact location in the (x, y, z) coordinate system, of the probe. Mechanical position encoding uses a robotic arm or a mechanical stepper to track the location of the ultrasound probe by direct attachment of the probe to the device. The position of the probe in 3D space is encoded and automatically sent to the computer through angle sensors located in the robotic arm joints or through position encoders in the arm joints of the mechanical stepper [57]. Most commercial ultrasound systems for real-time ultrasound imaging fusion are based on an EM tracking system. EM tracking only measures the position and orientation of the ultrasound probe in space, and is not directly related to patient’s anatomy. Therefore, external fiducial markers or internal anatomic landmarks should be used to perform co-registration. The previously recorded CT, PET/CT, or MRI dataset is uploaded to the ultrasound machine and co-registration from fiducial markers is performed. Afterward, a data rescaling of CT, MRI, or PET/CT to ultrasound format is performed. Most systems use a rigid transformation matrix, which is simpler and requires fewer co-registration points than a nonrigid matrix. These systems do not compensate for voluntary and involuntary patient motion, which can cause poor alignment. To mitigate this problem, the co-registration should be made in the same respiratory phase as the previously acquired data or more registration points should be added.

Beyond Mathematical Rules: One Plus One is More than Two! Unfortunately, there are no universal diagnostic techniques that can provide all the information for diagnosing diseases. Much more often it is necessary to combine the contributions of multiple techniques to obtain an overall picture. CT, for example, is able to provide 3D information on the anatomy and morphology of the body and organs within the body. In particular, it highlights tissue regions that differ from the surrounding ones by their density, since the density of the material at one given point (x, y, z) is proportional to the LAC ρ(x, y, z) ∝ μ(x, y, z) (in CT discrete space it is ρk, l, m ∝ μk, l, m). For this reason, CT is the technique of excellence for the visualization of high-contrast structures such as bone tissues, the vascular system with contrast

medium, etc., as these structures exhibit a high density compared to neighboring tissues, providing a high contrast to their interface. For high-contrast structures, CT has very high spatial resolution, which in the latest generation of CT scanners can reach 0.2 mm (2.5 lp/mm). MRI can provide detailed anatomical information on tissues that contain hydrogen and therefore in particular on soft tissues, where CT is not so bright. PET, as well as SPECT, by measuring the accumulation of radiopharmaceutical in the tissues, is able to provide functional and metabolic information on the tissues. It is therefore clear that the combination of these complementary techniques in a single device (hybrid imaging) brings significant advantages in terms of time, but also and above all in terms of spatial and temporal accuracy. In hybrid imaging, the information from the different imaging modalities is not simply gathered together but is integrated at a deeper level, achieving more than an additive effect (e.g. PET/CT > PET + CT). In fact, in addition to the potential advantages of having complementary information in the same session and without the need for coregistration, hybrid imaging provides many additional synergistic effects. Anatomical and function information from hybrid systems provides information that cannot be easily inferred from one type of image alone. In addition, these systems offer the possibility of using anatomical data (from CT or MRI) to improve both the visual quality and the quantitative accuracy of the correlated radionuclide imaging. Considering, for example, PET/MR systems. MRderived anatomic information may improve PET quantification, in-motion artifact reduction [thanks to the use of MR sequences acquired repeatedly with high temporal resolution (2–6 s), [58, 59]] or in-attenuation correction. In addition, simultaneous data acquisition (with PET/RM) allows the addition of kinetic, functional, and metabolic information for real-time multiparametric functional imaging. For other examples, see section Dual-Energy CT and section SPECT/CT and PET/CT Systems. Beyond these attractive prospects, hybrid imaging has effectively been shown to benefit clinical practice in many aspects [60, 61]: increasing oncologic staging accuracy [62, 63], changing the therapeutic strategy, assessment of early treatment response [64], to name a few. To conclude, we would like to quote the words of a great philosopher, Aristotele: “The whole is greater than the sum of its parts.”

References 1 Hasegawa, B.H., Iwata, K., and Wong, K.H. (2002). Dual-modality imaging of function and physiology. Acad. Radiol. 9: 1305–1321.

2 Hasegawa, B.H., Tang, H.R., and Da Silva, A.J. (2000). Dual-modality imaging. Nucl. Instr. Meth. Phys. Res. A 471: 140–144.

49

50

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

3 Beyer, T., Townsend, D.W., and Brun, T. (2000). A combined PET/CT scanner for clinical oncology. J. Nucl. Med. 41: 1369–1379. 4 J. Radon, "Über die Bestimmung von Funktionen durch ihre Integralwerte längs gewisser Mannigfaltigkeiten.," Ber. vor Sächs. Akad. Wiss., 69: 262–277, 1917. English translation available [J. Radon}: On determination of functions from their integraI values along certain manifolds. IEEE Trans. Med. Imaging, MI-5(4): 170–176, 1986. 5 Kalender, W., Seiss ler, W., and Vock, P. (1989). Singlebreath-hold spiral volumetric CT by continuous patient translation and scanner rotation. Radiology 173: 414. 6 Lombardi, F., Bonatti, M., and Zamboni, G.A. (2017). Uric acid versus non-uric acid renal stones: in vivo differentiation with spectral CT. Clin. Radiol. 72: 490–496. 7 Anger, H.O. (1958). Scintillation camera. Rev. Sci. Instrum. 29: 27–33. 8 Frey, E.C. and Tsui, B.M.W. (2006). Collimator-detector response compensation in SPECT. In: Quantitative Analysis in Nuclear Medicine Imaging (ed. H. Zaidi), 141–166. New York: Springer. 9 Lecoq, P. (2015). Development of new scintillators for medical applications. Nucl. Instr. Meth. Phys. Res. A 809 (Supplement C): 130–139. 10 Moses, W. Time of flight in PET revisited. IEEE Trans. Nucl. Sci. 50: 1325–1330. 11 Bailey, D.L. and Willowson, K.P. (2013). An evidencebased review of quantitative SPECT imaging and potential clinical applications. J. Nucl. Med. 54: 83–89. 12 Willowson, K., Bailey, D.L., and Baldock, C. (2008). Quantitative SPECT reconstruction using CT-derived corrections. Phys. Med. Biol. 53: 3099–3012. 13 Zaidi, H. (2006). Quantitative Analysis in Nuclear Medicine Imaging. Basel: Springer. 14 Watson, C., Casey, E., Michel, C., and Bendriem, B. (2004). Advances in scatter correction for 3D PET/CT. IEEE Symp. Conf. Rec. Nucl. Sci. 2004 (5): 3–7. 15 Watson, C. (2000). New, faster, image-based scatter correction for 3D PET. IEEE Trans. Nucl. Sci. 47: 1587–1594. 16 Ogawa, K., Harata, Y., Ichihara, T. et al. (1991). A practical method for position-dependent Compton-scatter correction in single photon emission CT. IEEE Trans. Med. Imaging 10: 408–412. 17 Peterson, M., Gustafsson, J., and Ljungberg, M. (2017). Monte Carlo-based quantitative pinhole SPECT reconstruction using a ray-tracing back-projector. EJNMMI Phys. 42: 32. 18 Elschot, M., Lam, M., Van den Bosch, M. et al. (2013). Quantitative Monte Carlo-based 90Y SPECT reconstruction. J. Nucl. Med. 54: 1557–1563. 19 Hoffman, E., Huang, S., and Phelps, M. (1979). Quantitation in positron emission computed tomography:

20

21

22

23

24

25

26

27 28

29 30

31

32

33

1. Effect of object size. J. Comput. Assist. Tomogr. 3: 299–308. Rousset, O., Ma, Y., and Evans, A. (1998). Correction for partial volume effects in PET: principle and validation. J. Nucl. Med. 39: 904–911. Müller-Gärtner, H., Links, J., Prince, J. et al. (1992). Measurement of radiotracer concentration in brain gray matter using positron emission tomography: MRI-based correction for partial volume effects. J. Cereb. Blood Flow Metab. 12: 571–583. Teo, B., Seo, Y., Bacharach, S. et al. (2007). Partial-volume correction in PET: validation of an iterative postreconstruction method with phantom and patient data. J. Nucl. Med. 48: 802–810. Blankespoor, S., Wu, X., Kalki, K. et al. (1995). Attenuation correction of SPECT using X-ray CT on an emission-transmission CT system: myocardial perfusion assessment. IEEE Nucl. Sci. Symp. Conf. Rec. 2: 1126–1130. Kinahan, P., Hasegawa, B., and Beyer, T. (2003). X-raybased attenuation correction for positron emission tomography/computed tomography scanners. Semin. Nucl. Med. 33: 166–179. Kinahan, P., Townsend, D., Beyer, T., and Sashin, D. (1998). Attenuation correction for a combined 3D PET/ CT scanner. Med. Phys. 25: 2046–2053. Shepp, L. and Vardi, Y. (1982). Maximum likelihood reconstruction for emission tomography. IEEE Trans. Med. Imaging 1: 113–122. Cherry, S., Sorensen, J., and Phelps, M. (2003). Physics in Nuclear Medicine. Grune & Stratton, Inc: Orlando, FL. Valk, P., Bailey, D., Townsend, D., and Maisey, M. (2003). Positron Emission Tomography: Basic Science and Clinical Practice. London: Springer-Verlag. Wernick, M. and Aarsvold, J. (2004). Emission Tomography. San Diego, CA: Elsevier Academic Press. Liu, Z., Pizzichemi, M., Auffray, E. et al. (2016). Performance study of Philips digital silicon photomultiplier coupled to scintillating crystals. J. Instrum. 11: 01. Pichler, B., Judenhofer, M., Catana, C. et al. (2006). Performance test of an LSO-APD detector in a 7-T MRI scanner for simultaneous PET/MRI. J. Nucl. Med. 47: 639–647. Spanoudaki, V.C., Mann, A.B., Otte, A.N. et al. (2007). Use of single photon counting detector arrays in combined PET/MR: characterization of LYSO-SiPM detector modules and comparison with a LSO-APD detector. J. Instrum. 2: P12002. Wagatsuma, K., Miwa, K., Sakata, M. et al. (2017). Comparison between new-generation SiPM-based and conventional PMT-based TOF-PET/CT. Phys. Med. 42: 203–210.

Basic Principles of Hybrid Imaging

34 van Sluis, J., de Jong, J., Schaar, J. et al. (2019). Performance characteristics of the digital biograph vision PET/CT system. J. Nucl. Med. 60 (7): 1031–1036. 35 Teoh, E., McGowan, D., Macpherson, R. et al. (2015). Phantom and clinical evaluation of the bayesian penalized likelihood reconstruction algorithm Q.Clear on an LYSO PET/CT system. J. Nucl. Med. 56 (9): 1447–1452. 36 Yang, L., Zhou, J., Ferrero, A. et al. (2014). Regularization design in penalized maximum-likelihood image reconstruction for lesion detection in 3D PET. Phys. Med. Biol. 59 (2): 403–419. 37 Bloch, F. (1946). Nuclear induction. Phys. Rev. 70: 460–474. 38 Bloch, F., Hansen, W., and Packard, M. (1946). The nuclear induction experimen. Phys. Rev. 70: 474–485. 39 Rabi, I. (1937). Space quantization in a gyrating magnetic field. Phys. Rev. 51: 652–654. 40 A. Vija, "Introduction to the xSPECT Technology: Evolving Multi Modal SPECT to Become Context Based and Quantitative.” Molecular Imaging, White Paper. Siemens Medical Solutions USA, Inc., 2013. 41 Nahmias, C., Lemmens, C., Faul, D. et al. (2008). Does reducing CT artifacts from dental implants influence the PET interpretation in PET/CT studies of oral cancer and head and neck cancer? J. Nucl. Med. 49: 1047–1052. 42 Delso, G., Wollenweber, S., Lonn, A. et al. (2013). MR-driven metal artifact reduction in PET/CT. Phys. Med. Biol. 58: 2267–2280. 43 Catana, C., Guimaraes, A., and Rosen, B. (2013). PET and MR imaging: the odd couple or a match made in heaven? J. Nucl. Med. 54: 815–824. 44 Lalush, D. (2017). Magnetic resonance-derived improvements in PET imaging. Magn. Reson. Imaging Clin. N. Am. 25 (2): 257–272. 45 Juttukonda, M., Mersereau, B., Chen, Y. et al. (2015). MR-based attenuation correction for PET/MRI neurological studies with continuous-valued attenuation coefficients for bone through a conversion from R2* to CT-Hounsfield units. NeuroImage 112: 160–168. 46 Ladefoged, C., Benoit, D., Law, I. et al. (2015). PET/MR attentuation correction in brain imaging using a continuous bone signal derived from UTE. EJNMMI Phys. 2 (Suppl 1): A39. 47 Roy, S., Wang, W., Carass, A. et al. (2014). PET attuenation correction using synthetic CT from ultrashort echo-time MR imaging. J. Nucl. Med. 55 (12): 2071–2077. 48 Paulus, D. and Quick, H. (2016). Hybrid positron emission tomography/magnetic resonance imaging: challenges, methods, and state of the art of hardware component attenuation correction. Investig. Radiol. 51 (10): 624–634. 49 Sekine, T., Ter Voert, E., Warnock, G. et al. (2016). Clinical evaluation of ZTE attenuation correction for brain FDG-PET/MR imaging-comparison with atlas attenuation correction. J. Nucl. Med. 57 (12): 1927–1932.

50 Paulus, D., Braun, H., Aklan, B., and Quick, H.H. (2012). Simultaneous PET/MR imaging: MR-based attenuation correction of local radiofrequency surface coils. Med. Phys. 39: 4306–4315. 51 Paulus, D., Tellmann, L., and Quick, H.H. (2013). Towards improved hardware component attenuation correction in PET/MR hybrid imaging. Phys. Med. Biol. 58: 8021–8040. 52 Delso, G., Martinez-Moller, A., Bundschuh, R. et al. (2010). The effect of limited MR field of view in MR/PET attenuation correction. Med. Phys.: 37, 2804–2812. 53 Blumhagen, J., Ladebeck, R., Fenchel, M., and Scheffler, K. (2013). MR-based field-of-view extension in MR/PET: B0 homogenization using gradient enhancement (HUGE). Magn. Reson. Med. 70: 1047–1057. 54 Blumhagen, J., Braun, H., Ladebeck, R. et al. (2014). Field of view extension and truncation correction for MR-based human attenuation correction in simultaneous MR/PET imaging. Med. Phys. 41: 022303. 55 Nuyts, J., Bal, G., Kehren, F. et al. (2013). Completion of a truncated attenuation image from the attenuated PET emission data. IEEE Trans. Med. Imaging 32: 237–246. 56 Ewertsen, C., Săftoiu, A., Gruionu, L. et al. (2013). Real-time image fusion involving diagnostic ultrasound. AJR Am. J. Roentgenol. 200 (3): W249–W255. 57 Kongnyuy, M., George, A., Rastinehad, A., and Pinto, P. (2016). Magnetic resonance imaging-ultrasound fusionguided prostate biopsy: review of technology, techniques, and outcomes. Curr. Urol. Rep. 17 (4): 32. 58 Catana, C., Drzezga, A., Heiss, W., and Rosen, B. (2012). PET/MRI for neurologic applications. J. Nucl. Med. 53: 1916–1925. 59 Baete, K., Nuyts, J., Van Laerea, K. et al. (2004). Evaluation of anatomy based reconstruction for partial volume correction in brain FDG-PET. NeuroImage 23 (1): 305–331. 60 Catalano, O., Masch, W., Catana, C. et al. (2017). An overview of PET/MR, focused on clinical applications. Abdom Radiol. 42 (2): 631–644. 61 de Galiza Barbosa, F., Delso, G., Ter Voert, E. et al. (2016). Multi-technique hybrid imaging in PET/CT and PET/MR: what does the future hold? Clin. Radiol. 71 (7): 660–672. 62 Antoch, G., Stattaus, J., Nemat, A. et al. (2003). Non-small cell lung cancer: dual-modality PET/CT in preoperative staging. Radiology: 229, 526–533. 63 Lardinois, D., Weder, W., Hany, T. et al. (2003). Staging of non-small lung cancer with integrated positron-emission tomography and and computed tomography. N. Engl. J. Med.: 348, 2500–2507. 64 Schwaiger, M., Ziegler, S., and Nekolla, S. (2010). PET/CT challenge for the non invasive diagnosis of coronary artery disease. Eur. J. Radiol. 73: 494–503.

51

52

3 Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology) Antonio Jesús Láinez Ramos-Bossini1, Ángela Salmerón-Ruiz1, José Pablo Martínez Barbero1, José Pablo Martín Molina2, José Luis Martín Rodríguez2, Genaro López Milena1, and Fernando Ruiz Santiago1,3 1

Department of Radiology, Virgen de las Nieves University Hospital, University of Granada, Granada, Spain Department of Radiology, San Cecilio University Hospital, University of Granada, Granada, Spain 3 Neuro-traumatology Hospital, Virgen de las Nieves University Hospital, School of Medicine, University of Granada, Granada, Spain 2

Relevant Imaging Anatomy and Variants This chapter aims to describe the essential aspects of normal anatomy and its variants on cross-sectional imaging from a radiological perspective. The imaging techniques that will be reviewed are computed tomography (CT) and magnetic resonance imaging (MRI). To define the anatomical structures precisely, it is important to remember that information from these two imaging modalities is obtained from different types of sources, namely ionizing radiation in the case of CT and radiofrequency pulses in a magnetic field in MRI. Consequently, the ability of each of these examinations to depict anatomical structures differs in terms of both quality and precision. In general, it can be stated that CT is superior in defining bony structures, while MRI is more advantageous for studying soft tissues, including bone marrow, due to its higher contrast resolution. CT is a cross-sectional imaging method that provides high spatial resolution for anatomical imaging of the human body. The introduction of the helical CT technique in 1989, in which images are acquired during continuous rotation of the X-ray tube with simultaneous movement of the table (and the patient), allowed significant shortening of the scanning time compared to the first sequential acquisition CT devices. The posterior development of multislice CT scanners with an increasing number of detector rows (e.g. from four or 16 rows initially to the latest models with up to 320 rows) made it possible to acquire a large number of slices simultaneously with a tube rotation time of less than 1 second [1, 2].

Although imaging acquisition is performed in the axial plane, submillimeter slices allow the creation of isotropic images with no significant spatial resolution distortion in multiplanar and 3D reconstructions [3]. CT can distinguish different tissues based on their specific X-ray attenuation, providing structural information. Quantitative information can be obtained based on Hounsfield Unit (HU) values, which measure the density of any structure relative to water (0 HU). For instance, the normal density of fat tissue ranges between −190 and −30 HU, and normal soft tissues show attenuation values in the range of 31–100 HU (although increased fat content may reduce this density to as low as 0 or even negative HU values)  [4, 5]. Tendons appear hyperdense relative to muscles, with reported values ranging from 75 to 115 HU. The air contained within the paranasal sinuses, lungs or bowel presents density values up to −1000 HU. Accordingly, so-called “windows” allow optimization of the analysis of certain anatomic regions by setting the value of two parameters: level and width. For example, to study lung parenchyma, a default “lung window” is set at a level of −700 HU with a width of −1000 HU, whereas analysis of brain tissue is conventionally performed with a level of 40 HU and a width of 80 HU. In addition, intravenous administration of iodine contrast can provide relevant information on vascular structures and tissue vascularization. If the acquisition is performed soon after contrast administration (e.g. 15–40 seconds depending on the anatomical region), information on arterial vascularization will be obtained. More delayed acquisitions will provide further information on the venous system and tissue microvascularization.

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach, First Edition. Edited by Ali Gholamrezanezhad, Majid Assadi, and Hossein Jadvar. © 2023 John Wiley & Sons Ltd. Published 2023 by John Wiley & Sons Ltd.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

In current multislice CT scanners, the radiation dose can be reduced by selecting appropriate acquisition protocols (e.g. adjusting voltage, slice thickness, and pitch) as well as by automatic modulation of the tube current. Radiation dose and image noise can also be reduced using iterative reconstruction techniques. Compared to CT, MRI offers better characterization of soft-tissue structures, including the spinal cord. Standard protocols in most anatomical regions should include the basic sequences, namely T1- and T2-weighted images. In addition, fat-suppressed images based on techniques of chemical shift and/or inversion-recovery suppression are usually performed  [6]. As with CT, contrast-enhanced protocols after intravenous gadolinium administration are commonly used to analyze the vascular behavior of different structures. More advanced MRI sequences, such as diffusion or perfusion-weighted images, can help further characterization of lesions. Of note, the reader should know that there is substantial protocol variability according to the anatomical region and type of lesion under study. Because this chapter is particularly focused on anatomy, most figures are based on T1- and T2-weighted images. The chapter is divided into different sections to cover all relevant anatomy “from head to toe.” We believe that this structure will facilitate the understanding of the whole chapter. At the end of the chapter, the reader will have a comprehensive overview of the cross-sectional imaging findings on radiological anatomy that nuclear medicine physicians should be aware of in their daily practice.

­Anatomy of the Brain For an adequate knowledge of the radiological anatomy, it is first convenient to conduct a brief review of the radiological semiology of the brain, i.e. how the different structures that make up the brain are depicted in different techniques or sequences.

Semiology of the Brain in Neuroimaging In general, the density of the gray matter (GM) (cerebral cortex and basal ganglia) is slightly high (about 50 HU) relative to that of the white matter (WM) (about 30 HU) on CT images, probably because of the higher cell density in the former. Comparatively, the cerebrospinal fluid (CSF) is hypodense, with density values close to water (0–5 HU), while parenchymal or dural calcifications and bone structures are hyperdense. To increase the ability for detecting and characterizing lesions, CT examinations can be complemented with the administration of intravenous contrast. Accordingly, vascular CT angiography of

supra-aortic vessels (SAV) and circle of Willis (COW), or cerebral venography studies can be performed. In addition, intravenous contrast enables the assessment of the enhancement pattern of vascularized lesions, such as vascular malformations or tumors with a high rate of neovascularization (e.g. meningiomas and glioblastomas)  [7]. Moreover, it allows the evaluation of the integrity of the blood–brain barrier (BBB), whose disruption determines high-contrast uptake in high-grade glial tumors, metastases or inflammatory lesions [8]. In MRI, each tissue has a specific signal intensity in different sequences. Accordingly, the combination of the different signal intensities allows the differentiation and characterization of normal tissues, which, in turn, enables the detection of abnormalities and the use of diagnostic algorithms  [9]. In general, in T1-weighted sequences, the GM is hypointense relative to the WM, and vice versa in T2-weighted sequences. The CSF is markedly hyperintense in T2-weighted sequences and hypointense in T1-weighted sequences. The use of T2-Flair (Fluid Liquid Attenuation Inversion Recovery) sequences allows the suppression of the free-water signal, and specifically the CSF signal, improving the detection of lesions in the brain parenchyma and the subarachnoid space (SAS). Fat, which is present in the subcutaneous tissue of the skull, facial bones, neck, and back, as well as inside the orbits and deep spaces of the neck (particularly the masticator and parapharyngeal spaces), is hyperintense in both T1- and T2-weighted sequences. Fat-suppression techniques based on different physical approaches (e.g. Dixon, STIR, SPAIR, SPIR or FatSat) are frequently used to improve the detection of lesions. The most commonly used contrast in MRI is based on gadolinium chelates, which shorten T1 relaxation times, increasing signal intensity. Enhancement is mainly secondary to disruption of the BBB and increased vascularization [10]. Of special interest are the T2 gradient echo and magnetic susceptibility weighted imaging (SWI) sequences, which have higher sensitivity to detect ferromagnetic substances. Accordingly, SWI sequences allow the identification of the substantia nigra and the red nuclei in the midbrain. Of note, blood deposits can be recognized in SWI as hypointense lesions, including intraparenchymal hematomas, subarachnoid hemorrhage (SAH), cortical siderosis or microhemorrhage foci from hypertensive angiopathy [9]. Diffusion sequences are based on MRI properties to detect the movement of protons and, consequently, of water molecules. In an ideal free environment, molecules move in all directions in space. However, the human body contains numerous barriers that hinder the free diffusion of water molecules, causing “diffusion restriction,” mainly

53

54

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

in the CNS, the BBB, and cellular barriers (such as myelin). Accordingly, the CNS shows significant restricted diffusion under physiological conditions

The CNS comprises a series of intracranial structures. At the supra-tentorial level, the brain is located in the anterior and middle cranial fossae and contains two hemispheres separated by the falx cerebri. In the posterior fossa, the tentorium cerebelli separates the brain stem (composed of midbrain, pons, and medulla oblongata) from the cerebellum (Figure 3.1).

of the skull and is attached to the inner table; and another one, more internal or meningeal, which makes up the falx cerebri, tentorium, and diaphragma sellae. The dura mater can be distinguished on imaging and is usually less than 2 mm thick. The innermost meningeal layer is the leptomeninx, composed of the arachnoid mater (attached to the inner layer of the inner dura, which is the outer boundary of the SAS) and the pia mater, in direct contact with the brain. The two components of the leptomeninges are not visible on imaging, except in the presence of pathologic processes, in which both layers are usually affected. The fact that these layers are the boundaries of the SAS makes them especially relevant [8, 10].

Anatomy of the Coverings of the CNS

Anatomy of the Brain Parenchyma

The CNS is encased and protected by different bone structures, including the skull, of great anatomical complexity and comprising multiple bones joined to each other and to the facial bones. In addition, the structures of the CNS present an inner lining composed of several membranes known as the meninges [11]. The most external and thickest layer is the dura mater or pachymeninx, which in turn comprises two layers, one that forms the periosteum

One of the cornerstones of the morphological and functional anatomy of the brain is the concept of somatotopy, according to which there is a specific location in the CNS controlling the functions of the rest of the body, especially the motor and sensory functions  [12]. In addition, despite having a similar histological architecture, each structure of the CNS performs different tasks based on its location. The distribution of specific operations in

Compartments and Structures of the CNS

(a)

(b)

(c)

Figure 3.1 Bone anatomy of the cranial fossae. (a) Axial CT image. (b) Coronal CT multiplanar reconstruction (MPR) image. (c) Sagittal MPR CT image. Yellow, anterior cranial fossa; red, middle cranial fossa; blue, posterior cranial fossa.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

different cortical regions highlights the importance of the functional anatomy of the brain [13]. The CNS is mainly made up of two types of tissue: the GM, mostly composed of neuronal bodies, and the WM, arranged around the neuronal axons. The GM is mainly present in two locations, namely the cerebral cortex and the deep GM, with the basal ganglia being its most representative example.

The cortical GM is formed by the cerebral cortex and cerebellum, being responsible for the main functional tasks according to the anatomical location. During evolutionary

development, numerous folds have been formed on the encephalic surface to increase the volume of GM as much as possible. The main consequence of this phylogenetic process is the formation of the so-called sulci and fissures (depressions of the surface), and their corresponding gyri [14]. The brain parenchyma is divided into lobes, which can be defined based on the primary sulci and fissures (Figures 3.2 and 3.3) [11, 14]. Accordingly, the frontal lobes are situated in the anterior pole of the brain, particularly in the anterior fossa and on both sides of the interhemispheric fissure, being separated from the parietal lobes by the central sulcus. The temporal lobes are located in the middle cranial fossa, and the Sylvian fissure separates them from

(a)

(b)

Functional Anatomy of the Cerebral Cortex

Figure 3.2 Sagittal anatomy of the brain in the midline. (a) Sagittal contrast-enhanced T1-3D weighted magnetic resonance image (MRI). (b) Sagittal noncontrast MPR CT image of the skull. 1, corpus callosum; 2, midbrain; 3, pons; 4, medulla oblongata; 5, fourth ventricle; 6, cerebellar vermis; 7, cerebral aqueduct; 8, quadrigeminal plate; 9, sella turcica; 10, frontal lobe; 11, parietal lobe; 12, occipital lobe.

(a)

(b)

Figure 3.3 Lobar and cortical anatomy. (a) Coronal T2-weighted MRI. (b) Axial T1-3D- weighted MRI (plane at the level of the hippocampus). 1, hippocampus; 2, third ventricle; 3, body of the lateral ventricle; 4, temporal lobe; 5, superior temporal gyrus; 6, middle temporal gyrus; 7, inferior temporal gyrus; 8, interhemispheric fissure; 9, occipital lobe; 10, frontal lobe.

55

56

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

the frontal and parietal lobes. The parietal lobes extend along the postero-superior margin of the convexity of the brain, and their posterior boundaries with the occipital lobe are difficult to define since the parieto-occipital fissure is not always present. The occipital lobes are located in the posterior part of the encephalus, in relation to the occipital bones. Moreover, modern functional neuroanatomy usually considers two other lobes, namely the insular cortex, which is located in the depth of the Sylvian fissure and inferior to the frontal operculum, and the limbic lobe, composed of GM structures and located on both sides of the midline, surrounding the corpus callosum. It is important to briefly review the most relevant sulci and fissures of the frontal lobes. In the basal face of the frontal lobe, the supraorbital gyri (medial, lateral, anterior, and posterior) are related to the floor of the anterior cranial fossa, and the gyrus rectus, of parasagittal location on both sides of the midline, is related to the cribiform plate of the

Figure 3.4 Landmarks of cortical anatomy. (a) Sagittal T1-weighted MRI. (b, c) Axial T1-weighted MRI with different angulation. (a) The Sylvian fissure (white asterisk), the superior face of the temporal lobe with the planum temporale (blue line), and the Heschl’s gyrus (red arrow). The opercular region (yellow circle) lies above the Sylvian fissure. (b) How the central sulcus does not normally reach the interhemispheric fissure (green arrow). (c) Typical appearances of the motor area of the hand at the central gyrus, i.e. inverted omega shape on the right side and epsilon shape on the left side.

(a)

(b)

ethmoid bone and the crista galli. On the lateral face of the frontal lobe, the superior, middle, and inferior frontal gyri can be found. In the latter, immediately above Sylvian fissure, there is an area of paramount importance due to its linguistic function, known as the opercular region (Figure  3.4). It is made up of three adjacent gyri which, from anterior to posterior, are called pars orbitalis, pars triangularis, and pars opercularis. The primary motor cortex, located in the precentral gyrus, is of special importance. Following a somatotopic arrangement as defined by the Penfield’s homunculus, the motor areas of the leg and foot (parasagittal), hand and arm (convexity), and mouth and face (lateral aspect) can be found in this gyrus. A number of anatomical landmarks can be used to locate the central sulcus (Figure  3.4). Accordingly, the motor area of the hand usually adopts a characteristic omega (Ω) or epsilon (ε) shape. In addition, it is helpful to know that the central sulcus does not usually reach the

(c)

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

interhemispheric fissure, while the superior frontal sulcus normally ends in the pre-central (rather than the central) sulcus. Furthermore, in MRI, the motor cortex is more hyperintense in T2-weighted images and less well-defined in T1-weighted images, which also helps locate this area. The lateral face of the temporal lobe includes the inferior, middle, and superior temporal gyri. Its superior face contains the so-called “planum temporale,” interrupted in its posterior and deep margin by the Heschl’s gyrus, of great relevance for being responsible for auditory processing based on tonotopic coding (Figure 3.4). On the lateral side (usually in the superior temporal gyrus) lies the Wernicke area, also of major importance. In the medial margin of the temporal lobes lie the hippocampi, of special relevance due to their implication in long-term memory. The hippocampal convolution is formed by a folding of the cortex, contains a high density of GM, and is bounded by the choroidal fissure medially (Figure 3.3). In the parietal lobes, the most relevant part of the cortex is the post-central cortex, responsible for the sensitivity of the body, and with a somatotopic organization similar to the motor cortex. The occipital cortex is located in the posterior and basal pole of the calcarine fissure, being responsible for the visual function.

Anatomy of the Basal Ganglia The other typical location of GM is the basal ganglia (BG or gray nuclei), which are located in the deep part of the cerebral hemispheres and in the cerebellum. The BG are (a)

(b)

connected with each other and with the cortex, being part of numerous functional circuits [15]. In the supratentorial compartment, the claustrum is located between the extreme and external capsules. It has a linear morphology and is hypointense in T1-weighted images, but it is usually only visible in volumetric sequences in 3T MRI equipment. Medially, the lenticular nucleus (composed of the putamen and the globus pallidus in its posterolateral and anteromedial portions, respectively) is located between the external and internal capsules. The putamen is hypointense in T1 and slightly hyperintense in T2, while the globus pallidus is isointense relative to the WM in T1 and hypointense in T2. The caudate nuclei lie lateral to the frontal horns and bodies of the lateral ventricles, being isointense relative to the lenticular nuclei. The thalami, of more posterior and inferior location, are difficult to recognize in T1-weighted images because their signal intensity is similar to that of the WM and slightly hyperintense in T2-weighted images (Figure 3.5). The substantia nigra and red nuclei of the midbrain, as well as the subthalamic nuclei, can be identified as hypointense structures in magnetic susceptibility sequences due to their high content in iron (Figure  3.6)  [16]. In the cerebellum, the dentate nuclei lie on both sides of the fourth ventricle, with a hypointense central area and hyperintense periphery (Figure 3.7). Under normal conditions, the dentate nuclei are isointense to the WM in T1-weighted images. Nevertheless, it has been recently described that, in patients undergoing repeated examinations with intravenous gadolinium, their signal intensity increases due to interstitial (c)

Figure 3.5 Supratentorial axial anatomy, section at the level of the base ganglia. (a) Axial noncontrast CT image. (b) Axial FLAIR T2-weighted MRI. (c) Axial TSE T2-weighted MRI. 1, thalamus; 2, caudate; 3, putamen; 4, globus pallidus; 5, frontal horn of the lateral ventricle; 6, occipital horn of the lateral ventricle; 7, genu of the corpus callosum; 8, splenium of the corpus callosum; 9, frontal lobe; 10, Sylvian fissure; 11, occipital lobe.

57

58

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

Figure 3.6 Anatomy of the basal ganglia in magnetic susceptibility weighted imaging. (a) Coronal image. (b) Axial image at the level of (3) in (a). (c) Axial image at the level of (4) in (a). 1, body of the lateral ventricle; 2, red nucleus; 3, subthalamic nucleus; 4, substantia nigra; 5, cerebral aqueduct; 6, putamen.

(a)

(b)

(c)

Figure 3.7 Infratentorial axial anatomy, sections at the level of the fourth ventricle. (a) Axial noncontrast CT image. (b) Axial contrast-enhanced T1-3D-weighted MRI. (c) Axial TSE-T2-weighted MRI. 1, pons; 2, middle cerebellar peduncle; 3, fourth ventricle; 4, cerebellar vermis; 5, dentate nucleus; 6, temporal lobe; 7, sphenoid sinus.

gadolinium deposits. There are also GM regions around the cerebral aqueduct of Sylvius and in the lateral walls of the third ventricle, although they do not represent actual nuclei [17].

Anatomy of the White Matter Historically, the study of the WM by imaging has been difficult. The similar appearance of WM tracts in the different imaging techniques made their individualization almost

impossible [13]. However, diffusion tensor imaging (DTI) has allowed anatomical maps of WM fibers to be obtained in vivo and noninvasively, leading to a quantum leap in the study of WM. From the classic anatomical perspective, three main groups of WM fibers can be distinguished, namely association, projection, and commisural. The association fibers communicate cortical neurons from one hemisphere with other neurons more or less nearby, usually from the same functional circuit. The projection fibers connect different

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

cortical regions with the deep GM from the BG or run toward the spinal cord. Finally, the commissural fibers are responsible for interconnecting the cerebral hemispheres. The commissural fibers can be easily identified in structural MRI or CT examinations, and should exhibit the same imaging features as the rest of the WM under normal conditions (Figure  3.8). The most relevant anatomical structure within this group is the corpus callosum, formed from anterior to posterior by rostrum, genu (minor forceps), body, and splenium (forceps major). It is located above the lateral ventricles and interconnects the cerebral hemispheres in a massive way. Agenesis of the corpus callosum is usually associated with developmental disorders, although it may be an incidental and asymptomatic finding occasionally. In addition, partial hypoplasia of the corpus callosum can be associated with midline lipomas, of low density in CT and high signal intensity in T1- and T2weighted images in MRI. The anterior commissure connects the deep WM of both temporal lobes and is located anterior to the columns of the fornix. The posterior commissure connects both cerebral hemispheres above the cerebral aqueduct, particularly occipital fibers related to light perception. These commissures are especially relevant since the imaginary plane that joins them in the sagittal projection (intercommisural line) is used as a landmark for the axial plane in most MRI studies. They are easily located in sagittal T1-weighted images but may be (a)

absent as a variant of normality in some cases. The interthalamic adhesion can sometimes be seen in the midline, at the level of the third ventricle, as an anatomical variant. The projection fibers originate in the cortex and direct toward the BG, posterior fossa or spinal cord. The most relevant structure within this group of fibers is the corticospinal tract, which arises from the primary motor cortex of the central gyrus and runs inferiorly through the corona radiata, posterior limb of the internal capsule, and cerebral peduncles, reaching the pons, from which they extend to both sides of the midline. The association fibers are responsible for connecting adjacent cortex areas, which usually belong in the same functional circuit. The “U” or juxtacortical fibers stand out within this group, as well as bundles such as the arcuate fasciculus, which connect the areas related to the language circuit.

Brief Overview of the Cranial Nerves The cranial nerves are a set of 12 nerves that arise from the encephalus and are involved in sensory functions, motor control of facial muscles, and regulation of glands of the neck. We will briefly review the most relevant cranial nerves from the radiological point of view. Both T2 (e.g. T2DRIVE 3D or FIESTA) and T13D volumetric sequences with isotropic voxel and high resolution (less than 1 mm) are required for the study of the cranial nerves (Figure 3.9). (b)

(c)

Figure 3.8 Cerebral commissures. (a) Sagittal T1-weighted MRI. (b, c) Axial T1-weighted MRI in two different patients. The intercommissural line, which is used as a landmark to acquire sequences in the axial plane, can be seen in (a) (dashed red line). (b) The anterior and posterior commissures. (c) The interthalamic adhesion, a variant of normality. 1, anterior commissure; 2, posterior commissure; 3, interthalamic adhesion; 4, third ventricle; 5, corpus callosum. Green line (a) and arrow (c) show the fornix.

59

60

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

The first cranial nerve, or olfactory nerve, is located in the anterior cranial fossa, specifically in the olfactory fossa, on both sides of the midline and the crista galli, and closely related to the gyrus rectus of the basal face of the frontal lobes. Its fibers, hypointense relative to the CSF in T2-weighted images, cross the cisternal space toward the cribiform plate. The second cranial nerve, or optic nerve, arises from the fovea centralis of the ocular globes, and is isointense to the WM in all MRI sequences. In fact, strictly speaking, it is a WM tract and not a peripheral nerve. The optic nerve has an intraconal course within the orbit, a fissural segment by which it exits the orbit, and a cisternal segment prior to crossing over in the optic chiasm, of suprasellar location, and from where the optic tracts and radiations arise. The fifth cranial nerve, or trigeminal nerve, has its nuclei in the brain stem, and emerges on both sides of the

(a)

pons. It presents an anterior course toward the Meckle’s cave, where the trigeminal ganglion is located. At this point, the nerve trifurcates, although its branches are not radiologically evident. It is easily identifiable in its cisternal course in T2-weighted images. Contact areas between the trigeminal nerve and vascular structures are of special relevance for being a potential cause of trigeminal neuralgia. The bundle containing the facial and statoacoustic nerves (VII and VIII cranial nerves, respectively) arises from both sides of the pons, crosses the cistern of the pontocerebellar angle, and enters the internal auditory meatus. In the intra-canalicular segment, we can distinguish the VII cranial nerve (supero-anterior) and the cochlear nerve (infero-anterior). The vestibular branches are of posterior location.

(d)

(b)

(c)

(e)

Figure 3.9 Basic anatomy of the cranial nerves. (a, b, c, e) High-resolution 3D DRIVE T2-weighted MRI. (d, e) Oblique axial T1-3Dweighted MRI. (a) Coronal image showing the olfactory nerve in the olfactory groove (red circle) and the optic nerve in its intraorbital segment (green arrow). (b) Axial image showing the trigeminal nerves (red arrow) reaching the Meckel’s cave. (c) Axial image showing the cisternal segments (cerebellopontine angle) and intracanalicular segments of statoacoustic and facial nerve bundles (blue circle). The arrival of the cochlear nerves (branch of the VIII cranial nerve) at the cochlea can be appreciated (brown arrow). (d) Axial imaging showing the orbital, cisural, and chiasmatic segments, and optic tracts of the II cranial nerve (green line). (e) Coronal image showing the optic chiasm (yellow circle), medial to the internal carotid arteries (green arrow) and above the sella turcica.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

Posterior Fossa In the anterior part of the posterior fossa the brain stem, consisting of the midbrain, pons, and medulla, can be identified (Figure 3.2). This fundamental structure contains WM projection fibers as well as the nuclei of the cranial nerves. The midbrain is connected to the brain through the two cerebral peduncles. The cerebral aqueduct, which connects the third and fourth ventricles, lies in the dorsal aspect of the midbrain, while the quadrigeminal plate is located posteriorly. The cerebellum consists of a central structure, the vermis, and two hemispheres located on both sides of the posterior cranial fossa, divided into numerous gyri known as cerebellar folia (Figure 3.7). In its most inferior portion, close to the foramen magnum, the cerebellar tonsils are located, and they should not surpass this boundary under normal conditions.

Ventricular System The approximate volume of CSF in the CNS under normal conditions ranges from 125 to 150 mL [18]. Approximately 80% of this volume is distributed in the SAS (cortical sulci, basal cisterns, and spinal canal) and the rest occupies the ventricular system. It is constantly produced in the choroidal plexuses by an active metabolic process and drainage of interstitial fluid, reaching 450 mL/24 hours, thus its renewal is constant. Its circulation and flow pattern seem to be influenced by blood pressure peaks, as well as by intracranial pressure, presenting oscillations parallel to those of blood pressure. Its reabsorption mainly occurs via parenchymal capillaries into the glymphatic system, although there is also some reabsorption into the arachnoid granulations, which act as valves, by the choroid plexuses and by the nasal lymphatic system. The spaces containing CSF are the ventricular system and the SAS. The ventricular system consists of four interconnected cavities inside the brain which are covered by a layer of ciliated cells (ependymal cells). At the supratentorial level, both lateral ventricles (consisting of anterior or frontal horns, body, atria, occipital and temporal horns) are connected to the third ventricle through the interventricular foramen (of Monro). The atria are of special relevance, since the choroid plexuses, the key producers of CSF, are normally present at this level. The third ventricle communicates with the fourth ventricle through the cerebral aqueduct (located in the dorsal aspect of the midbrain, closely related to the quadrigeminal plate), which lies in the posterior fossa (Figures 3.2 and 3.7). Its main relations are the posterior aspect of the pons anteriorly, the cerebellar vermis posteriorly, and both cerebellar hemispheres laterally.

Through the two foramina of Luschka (lateral) and the foramen of Magendie (central), the fourth ventricle communicates with the SAS.

­Neck, Paranasal Sinuses, and Orbit The most accepted approach to study the neck distinguishes two anatomical spaces: the supra-hyoid and the infra-hyoid neck. The anatomical landmark for this division is the deep cervical fascia (DCF), which is inserted into the base of the skull and converges into the hyoid bone. There are two main fasciae in the neck: (i) the superficial cervical fascia, which is loose, with fat content and not relevant from the anatomical perspective, and (ii) the DCF, which is composed of several layers (Figure  3.10) and divides the neck into different compartments or functional spaces [19]: ●

●

Superficial or investing layer/fascia (Figure 3.10): envelopes the sternocleiodomastoid (SCD) and trapezium muscles, the parotid gland, the mandible, and the masticatory muscles (masticator space). Deep or prevertebral layer/fascia (Figure  3.10): invests the prevertebral and paravertebral muscles.

Between both the superficial and deep layers there are three important fasciae. The most relevant one is the oropharyngeal fascia or middle layer (Figure  3.10), which defines the pharyngeal mucosal space (containing the nasopharynx and the oropharynx) and includes the pharyngeal mucosa, the constrictor muscles providing muscular support of the pharynx, and lymphatic tissue. There is another fascia that runs anteromedially from the styloid process to the tensor veli palatini muscle and divides the parapharyngeal space into pre-styloid and retro-styloid. Finally, the alar fascia (Figure 3.11), a fold of the prevertebral fascia, is located medial to both carotid arteries and posterior to the buccopharyngeal fascia (between the buccopharyngeal and prevertebral fasciae). The retropharyngeal space, as described below, lies between the buccopharyngeal fascia and the prevertebral fascia (or deep layer). Therefore, the alar fascia separates the retropharyngeal space, more medial, from the carotid space, more lateral (because it has a sagitally-oriented fold). In addition, this fascia divides the retropharyngeal space into two compartments, one located more anteriorly, which is the true retropharyngeal space (extending from the clivus to the first thoracic vertebrae), and another one, more posterior, which is known as the “danger space.” The

61

62

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 3.10 Supra-hyoid neck. (A) Axial T1-weighted MRI at the level of the palate. (B) Axial view at the level of the floor of mouth. Cervical fasciae: I, superficial layer of the deep cervical fascia; II, middle layer of the deep cervical fascia; III, deep layer of the deep cervical fascia. Cervical spaces: a, parapharyngeal space; b, carotid space (retro-styloid parapharyngeal space); c, parotid space; d, masticator space; e, perivertebral space; f, pharyngeal mucosal space; g, retropharyngeal space; h, sublingual space. Anatomical structures: 1, pterygoid muscles; 2, masseter muscle; 3, longus colli muscle; 4, posterior belly of the digastric muscle; 5, sternocleidomastoid muscle; 6, geniohyoid muscle; 7, lingual septum; 8, mylohyoid muscle; 9, submandibular gland; 10, parotid gland (superficial lobe); 11, parotid gland (deep lobe); 12, palatine tonsil; 13, internal carotid artery; 14, buccinator muscle; 15, hard palate; 16, soft palate; 17, semispinalis cervicis muscle; 18, splenius capitis muscle.

(A)

(B)

latter compartment runs more inferiorly to reach the mediastinum, representing a potential communication pathway between the neck and the mediastinum through the retropharyngeal space. This fascial division allows the extension pathways of pathological processes through the cervical spaces to be analyzed in a structured way.

Compartments of the Supra-­hyoid Neck The DCF defines the pharyngeal, parapharyngeal, carotid, parotid, masticator, retropharyngeal and prevertebral mucosal spaces [20]. Pharyngeal Mucosal Space

The pharyngeal mucosal space is divided into the nasopharynx and the oropharynx and is bounded by the middle layer of the DCF, specifically the buccopharyngeal fascia (Figure 3.10). The superior margin of this fascia surrounds the pharyngeal-basilar fascia (Figure  3.12), which is the aponeurosis of the superior pharyngeal constrictor muscle

and is inserted into the base of the skull. The inferior margin of the buccopharyngeal fascia invests the superior and middle pharyngeal constrictor muscles, which provide muscular support to the nasopharynx and oropharynx, respectively. The anatomical nasopharynx are (Figure 3.12):

Nasopharynx

● ● ●

boundaries

of

the

Anterior: anterior nares. Posterior: sphenoid bone. Inferior: soft palate.

The lateral boundaries are more anatomically complex. From anterior to posterior, these include the salpingo-pharyngeus muscle, the cartilaginous part of the Eustachian tube, and the prevertebral muscles (longus colli muscle). These structures form folds on both sides of the nasopharynx, specifically (from anterior to posterior) the recess of the Eustachian tube, the torus tubarius, formed by the cartilaginous end of the tube, and the fossa of Rosenmüller, located between the torus

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

(A)

(B)

(D)

(C)

Figure 3.11 Infra-hyoid neck. (A, B) Axial T1-weighted MRI illustrating the fasciae and spaces of the neck. (C) Axial T1-weighted MRI with numbered anatomical structures. (D). Sagittal noncontrast MPR CT image, detail of the retropharyngeal space. Cervical fasciae: I, superficial layer of the deep cervical fascia; II, middle layer of the deep cervical fascia; III, deep layer of the deep cervical fascia; IV, alar fascia. Cervical spaces: a, posterior cervical space; b, carotid space; c, perivertebral space; d, visceral space; e, retropharyngeal space; f, danger space. Anatomical landmarks: 1, internal carotid artery; 2, jugular vein; 3, thyroid cartilage; 4, cricoid cartilage; 5, laryngeal vestibule; 6, post-cricoid region; 7, retropharyngeal space; 8, scalene muscles; 9, levator scapulae muscle; 10, sternocleidomastoid muscle; 11, trapezius muscle.

(c)

(a) (b)

Figure 3.12 Nasopharynx. Axial T1-weighted MRI at the level highlighted in yellow. 1, levator veli palatini muscle; 2, pharyngobasilar fascia; 3, tensor veli palatine muscle; 4, parapharyngeal space; 5, recess of the Eustachian tube; 6, torus tubarius; 7, fossa of Rossenmüller; 8, pterygoid muscles; 9, mandibular condyle; 10, lumen of the nasopharynx; 11, middle turbinate of the nasal fossa.

63

64

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

tubarius and the longus colli muscle, representing one of the most common early sites of origin of nasopharyngeal tumors. Next to the wall of the nasopharynx, the pharyngobasilar fascia separates the levator veli palatini from the tensor veli palatini muscles (from medial to lateral), the latter of which exhibits a more linear morphology in crosssectional images. The pharyngobasilar is an aponeurosis of hard consistency related to the superior pharyngeal constrictor muscle and is inserted into the base of the skull. In addition, the pharyngobasilar fascia presents an area of muscular deficiency called the sinus of Morgagni, through which the levator veli palatini muscle and the Eustachian tube run toward the base of the skull. Its importance lies in the fact that it provides a potential communication pathway for tumor extension from the nasopharynx to the base of the skull. Oropharynx

The anatomical boundaries of the oropharynx are (Figure 3.13): ● ● ●

● ●

Superior: soft palate. Inferior: superior aspect of the epiglottis. Anterior: the plane formed by the anterior margin of the soft palate, the anterior tonsillar pillar, and the circumvallate papillae, which separates the base and the free edge of the tongue and is bounded by the oral cavity anteriorly. Posterior: first cervical vertebrae. Lateral (from anterior to posterior): anterior tonsillar pillar, formed by the palatoglossus muscle, the palatine tonsil, and the posterior tonsillar pillar, made up of the palatopharyngeus muscle.

Parapharyngeal Space

The parapharyngeal space was classically divided into pre- and post-styloid with respect to the styloid process. Nevertheless, the pre-styloid parapharyngeal space is (a)

(c) (b)

currently referred to as the parapharyngeal space (Figure  3.10), while the former post-styloid space is equivalent to the currently known as carotid space (Figure 3.10). The parapharyngeal (pre-styloid) space is located in the central space of the DCF and predominantly contains fat, in addition to the parapharyngeal venous plexus, other vascular structures such as the ascending pharyngeal artery and, occasionally, minor salivary glands. Its boundaries are established by different layers of the DCF: ●

●

●

medial: middle layer of the DCF, which surrounds the pharyngeal mucosal space. lateral: superficial layer of the DCF, which surrounds the masticator and parotid spaces. posterior: carotid sheath, which surrounds the carotid space and is formed by the three layers of the DCF.

This space extends from the base of the skull to the hyoid bone and communicates with the submandibular space directly. It allows potential dissemination of abnormal processes that extend from nearby spaces. In fact, extension by adjacent lesions is the most frequent cause of pathologic involvement of the parapharyngeal space. The parapharyngeal fat is characteristically displaced in one direction or another based on the space from where the pathologic process arises (Figure 3.14). Carotid Space

As mentioned above, the carotid space corresponds to the classic retro-styloid parapharyngeal space (Figure 3.10). It is located posterior to the styloid process, which extends from the base of the skull to the aortic arch. The three layers of the DCF contribute to form the sheath that surrounds this space. Its supra-hyoid portion contains the internal carotid artery, the jugular vein, lymph nodes, the cranial nerves IX, X and XI (glossopharyngeal, vagus and accessory, respectively), and the cervical sympathetic plexus. Figure 3.13 Oropharynx. Axial T1-weighted MRI at the level highlighted in yellow. 1, palatine tonsil; 2, base of the tongue; 3, anterior tonsillar pillar (palatoglossus muscle); 4, posterior tonsillar pillar (palatopharyngeal muscle); 5, parapharyngeal space.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

(a)

(b)

(c)

(d)

(e)

Figure 3.14 Displacement of the fat in the parapharyngeal space based on the location of lesions in adjacent cervical spaces. (a–e) Axial noncontrast CT images. (a) Lesion in the parotid space: antero-medial displacement. (b) Lesion in the masticator space: posteromedial displacement. (c) Lesion in the perivertebral space: lateral displacement. (d) Lesion in the carotid space: anterior displacement. (e) Lesion in the pharyngeal mucosa: postero-lateral displacement.

Masticator Space

The masticator space is a large space bounded by the superficial layer of the DCF. Its anatomical boundaries are (Figure 3.10): ● ●

Inferior: mandibular angle. Superior: the masticator space extends medially to reach the insertion of the pterygoid muscles in the base of the skull, inside the foramen ovale. Laterally, it extends to the upper insertion of the temporal muscle in the cranial vault. The zygomatic arch divides this space into the

● ●

temporal fossa or supra-zygomatic masticator space and the infra-temporal fossa under the zygomatic arch. content: masticatory muscles: masseter, medial and lateral pterygoid and temporal muscles, which are innervated by the masticatory nerve, the motor portion of the third branch of the V cranial nerve (or trigeminal) which enters the masticator space via the foramen ovale being a potential pathway of perineural invasion from the masticator space to the base of the skull.

65

66

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach ● ● ●

inferior alveolar nerve. Mandible. Parotid duct.

●

●

Buccal Space

The buccal space (Figure 3.15), without true fascial boundaries, is located immediately anterior to the masticator space and contains [21]: ● ● ● ●

Buccal fat pad Facial artery and vein Distal part of the parotid duct Buccinator muscle.

●

Retropharyngeal Space

Parotid Space

The parotid space (Figure 3.10) contains the following anatomical structures: (a)

(b)

(c)

(d)

●

Parotid gland, with intraparotid lymph nodes, surrounded by the superficial layer of the DCF. Facial nerve, which forms a plane that divides the parotid gland into its superficial and deep lobes. Retromandibular vein (Figure 3.15), which lies medial to the course of the facial nerve at the mandibular angle, being the main anatomical landmark to distinguish the superficial from the deep lobe of the parotid gland on cross-sectional CT images, since the facial nerve is usually not visible at this level. External carotid artery.

The retropharyngeal space (Figure 3.11) is a virtual space between the pharyngeal constrictor muscles, surrounded Figure 3.15 Oral cavity (yellow area in the sagittal noncontrast CT image, on the right). (a) Axial T1-weighted MRI at the level of the palate (yellow area in the sagittal noncontrast CT image, on the right). (b) Axial view at the level of the floor of mouth. (c) Coronal T1-weighted MRI. (d) Volume rendering of the floor of mouth with the salivary glands. 1, free edge of the tongue; 2, base of the tongue; 3, sublingual space; 4, submaxillary space; 5, mylohyoid muscle; 6 and 6′, genioglossusgeniohyoid complex, root of the tongue; 7, lingual septum; 8, anterior belly of the digastric muscle; 9, palatine tonsil; 10, anterior tonsillar pillar (palatoglossus muscle); 11, sublingual gland; 12, submandibular glands (encircled by yellow dots); 13, retromolar trigone; 14, buccinator muscle; 15, parotid gland; 16, retromandibular vein.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

by the middle layer of the DCF and the prevertebral muscles, which are in turn surrounded by the deep layer of the DCF. A thin line of fat within this space allows its identification on imaging. The alar fascia separates the retropharyngeal space from the carotid space laterally, and from the danger space posteriorly, representing a pathway for the extension of pathological processes from the retropharyngeal space to the mediastinum (Figure 3.11). Contents: ● ●

Fat. Lymph nodes, only from the level of the nasopharynx to the level of the hyoid bone.

Perivertebral Space

Also called pre-vertebral space, the perivertebral space (Figure 3.10) is bounded by the deep layer of the DCF, which invests the posterior pre- and paravertebral muscles, and is inserted into the transverse processes of the adjacent vertebrae. The perivertebral space contains elements of the brachial plexus, the phrenic nerve, pre- and paravertebral muscles, vertebrae, vertebral arteries and veins, and the spinal cord. Oral Cavity

The oral cavity (Figure 3.15) is located anterior to the oropharynx, from which it is separated by a ring of structures formed by the soft palate, the anterior tonsillar pillars, and the circumvallate papillae. Its boundaries are: ●

● ●

●

Superior: hard palate, upper alveolar ridge and upper dental arch. Lateral: buccae (cheeks). Posterior: circumvallate papillae and anterior tonsillar pillars. Inferior: mylohyoid muscle or floor of mouth, inferior alveolar ridge and inferior dental arch.

The oral cavity contains the submaxillary and sublingual spaces, which are separated by the mylohyoid muscle and communicated by their free edge posteriorly. Infectious processes involving both the submaxillary and sublingual spaces (e.g. Ludwig angina) can extend to the parapharyngeal and retropharyngeal spaces and eventually reach the mediastinum. The sublingual space, located above the mylohyoid muscle and lateral to the root of tongue, is formed by the genioglossus and geniohyoid muscles. It contains the following structures: ● ●

●

lingual nerve IX and XII cranial nerves (glossopharyngeal and hypoglossal, respectively) lingual artery and vein

● ●

sublingual glands deep lobe of the submandibular gland.

The submandibular space, situated inferior and lateral to the mylohyoid muscle, contains the following structures: ● ● ● ● ●

superficial lobe of the submandibular gland anterior belly of the digastric muscle lymph nodes facial artery and vein inferior loop of the hypoglossal nerve.

Compartments of the Infra-­hyoid Neck The infra-hyoid neck is formed by the carotid, perivertebral, retropharyngeal, posterior cervical, and visceral spaces. The visceral space is the only specific compartment of the infra-hyoid neck [22, 23]. The first three compartments have already been explained. Accordingly, this section focuses on describing the posterior cervical and the visceral spaces. Posterior Cervical Space

The posterior cervical space (Figure  3.12) is located between the superficial and deep layers of the DCF, posterior to the sheath of the carotid space and anterolateral to the perivertebral space. It contains fat, the posterior cervical lymph chain, the accessory spinal nerve, and the brachial plexus. Visceral Space

The visceral space (Figure 3.12) corresponds to the pharyngeal mucosal space in the supra-hyoid neck and is bounded by the middle layer of the DCF. It contains the larynx, hypopharynx, thyroid and parathyroid glands, trachea, esophagus, lymph nodes, and laryngeal nerves. Next, the larynx and hypopharynx will be described in more detail. Both structures have close anatomical and functional relations. The larynx (Figure 3.16) consists of a bony skeleton made up of the hyoid bone and a cartilaginous skeleton formed by the thyroid, cricoid, arytenoid, and corniculated cartilages, and the epiglottis; all of them lined with respiratory mucosa. The larynx is divided into three anatomical spaces, namely the supraglottis, glottis, and subglottis. The supraglottis or supraglottic larynx extends from the free edge of the epiglottis to the laryngeal ventricles and its main components include the following:

Larynx

●

Epiglottis, attached to the base of the tongue by the medial and lateral glossoepiglottic folds, which define the valleculae.

67

68

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 3.16 Larynx. (a) Coronal (left) and sagittal views (middle and right) of the larynx in the midline, with detail of the laryngeal ventricle and the pre-epiglottic space in the coronal and sagittal planes, respectively. (b) Axial contrast-enhanced CT images of the supraglottic space. (c) Axial noncontrast CT image at the level of the glottis. (d) Axial contrast-enhanced CT images at the infraglottic level. Hypopharynx: formed by 1, posterior pharyngeal wall; 2, pyriform sinus; 3, post-cricoid region. Supraglottis: 4, epiglottis with glossoepiglotic fold and valleculae located on both sides; 5, free edge of the epiglottis; 6, base of the epiglottis; 7, base of the tongue; 8, aryepiglottic folds; 9, pharyngoepiglottic fold; 10, pre-epiglottic space; 10′, paraglottic space, which also extends to the glottis; 11, false vocal cords or bands. Glottis: 12, vocal cords, the vocal ligament lies more medial; 13, anterior commissure; 14, posterior commissure; 10′, paraglottic space. Other structures: 22, esophagus; 17, thyroid gland; 16, parathyroid; 15, recurrent nerve. Bony and cartilaginous anatomy of the larynx: 18, hyoid bone; 19, thyroid; 20, arytenoid; 21, cricoid; 22, esophagus; 23, trachea; 24, laryngeal ventricle; 25, hypopharynx.

(a)

(b)

(c)

(d)

●

●

●

●

Pre-epiglottic space, located anterior to the epiglottis and posterior to the hyoid bone, thyroid membrane, and thyroid cartilage. Inferiorly, it is continuous with the paraglottic or paralaryngeal space, which lie external to the false vocal cords and, to a lesser extent, the glottis. Both spaces contain fat tissue. Laryngeal vestibule, an air space within the supraglottis. Aryepiglottic folds, which run inferiorly from the epiglottis to the aytenoid cartilages and separate the laryngeal vestibule from the pyriform sinuses (belonging to the hypopharynx). The aryepiglottic folds end in the false vocal cords distally. Laryngeal ventricle, a mucosal fold between the true and false vocal cords. It is usually collapsed.

In the glottis or glottic larynx the following structures can be distinguished:

●

●

●

Vocal cords, formed by the thyroarytenoid muscle and the vocal ligament. Anterior commissure or point of junction of the true vocal cords in the anterior midline. Posterior commissure, a mucosal layer between the arytenoid cartilages. Both commissures must not exceed 1 mm thick in cross-sectional images.

The subglottis or subglottic larynx corresponds to the space between the plane located 1 cm below the vocal cords and the inferior margin of the cricoid cartilage. The hypopharynz is the part of the visceral space of the infra-hyoid neck that extends from the superior edge of the hyoid bone to the cricoid cartilage (Figure 3.16). It lies posterior to the supraglottis, with the exception of the pyriform sinuses, which are situated more laterally. The hypopharynx consists of three parts:

Hypopharynx

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology) ●

●

●

Pyriform sinuses, lateral to the aryepiglottic folds and the laryngeal vestibule. Posterior wall of the hypopharynx, which is continuous with the posterior wall of the oropharynx. Post-cricoid region, which corresponds to the pharyngoesophageal junction and is usually collapsed.

● ●

●

●

Lymph Nodes of the Neck The lymph nodes of the neck are classified into different levels based on cross-sectional anatomy (Figure 3.17). The key anatomical landmarks are the following: (a)

●

Posterior margin of the submandibular gland (which separates level I from level II). Hyoid bone, which separates level II from level III. Cricoid cartilage, which separates level III from level IV (and sublevels VA from VB). Posterior margin of the sternocleidomastoid muscle, which separates levels II, III and IV from level V. Medial margin of the internal carotid artery, which separates levels III and IV from level VI. Levels:

●

(a)

Level I – Level IA (submental). Its boundaries are: ○ Superior: plane of the body of the mandible. ○ Inferior: plane of the hyoid bone. ○ Lateral: medial margins of the digastric muscle. – Level IB (submandibular). Its boundaries are: Medial: digastric muscle. Lateral: body of the mandible. ○ Posterior: posterior margin of the submandibular gland. Level II (upper internal jugular group). Its boundaries are: – Superior: base of the skull. – Inferior: hyoid bone. – Lateral: sternocleidomastoid muscle. – Medial: medial edge of the internal carotid artery. – Level IIA: anterior, medial, lateral or posterior to and inseparable from the jugular vein. – Level IIB: posterior to and separable from the jugular vein. Level III (middle jugular vein group). Its boundaries are: – Superior: hyoid bone. – Inferior: cricoid cartilage. – Posterior: posterior edge of the sternocleidomastoid muscle. – Medial: medial margin of the internal carotid artery. Level IV (lower internal jugular group). Its boundaries are: – Superior: cricoid cartilage. – Inferior: clavicle. – Posterior: posterior margin of the sternocleidomastoid muscle. – Medial: medial margin of the common carotid artery. ○ ○

●

(b)

(b)

●

(c)

(c)

●

Figure 3.17 Lymph node levels. (a–c) Axial noncontrast CT images, with anatomic illustrations on the right. (a) Submental (IA), submaxillary (IB), upper jugular (IIA and IIB), and upper posterior triangle (VA) lymph nodes. (b) Middle jugular (III) and upper posterior triangle (VA) lymph nodes. (c) Low jugular (IV), lower posterior triangle (VB), and central compartment (VI) lymph nodes. 1, posterior border of the submandibular gland; 2, jugular fossa; 3, posterior border of hyoid bone; 4, inferior margin of cricoid; 5, internal carotid artery; 6, manubrium roof; 7, internal jugular vein.

●

Level V (posterior triangle). Its boundaries are: – Anterior: posterior margin of the sternocleidomastoid muscle. – Superior: base of the skull. – Inferior: clavicle. – Posterior: anterior edge of the trapezius muscle.

69

70

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

●

●

– Level VA, from the base of the skull to the cricoid cartilage. – Level VB, from the cricoid cartilage to the clavicle. Level VI (central compartment). From the hyoid bone to the sternal manubrium, medial to both carotid arteries. Level VII (superior mediastinum group). Between the sternal manubrium and the brachiocephalic vein, medial to both carotid arteries.

Sinonasal Region The sinonasal region has a complex anatomy with many variants of normality [24, 25], some of which can predispose to disease by interfering with the sinus drainage pathways in the nasal cavity (Figure 3.18) [26]. The nasal cavity is located in the central region of the face, above the oral cavity, inferior to the anterior cranial fossa and medial to the orbits. Its boundaries are the following:

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

Figure 3.18 Paranasal sinuses. (a–c) Coronal noncontrast MPR CT images (bone window). (d–f) Axial noncontrast CT images (bone window). (g–i) Sagittal noncontrast MPR CT images (bone window). 1, maxillary sinus; 2, frontal sinus; 3, ethmoidal air cells; 4, sphenoid sinus; 5, middle concha; 11, uncinate process; MO, maxillary ostium; MM, middle meatus; HS, hiatus semilunaris. Sphenoethmoidal recess: 12, olfactory fossa; 6, lateral lamella; 7, cribiform plate. Frontal recess: FR (red dots), drainage pathway of the frontal sinus into the middle meatus; 8, nasolacrimal duct; 9, pterygopalatine fossa; 10, lamina papyracea; 13, pterygoid process; 14, clivus; 15, sella turcica.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology) ●

● ● ●

Superior wall: olfactory fossa, which is mainly formed by the cribiform plate. Inferior wall: hard and soft palate. Medial wall: nasal septum. Lateral wall: formed by the three conchae or turbinates (superior, middle, and inferior), which form air passages called meatuses between them: – the nasolacrimal duct drains into the inferior meatus – the frontal sinuses, the anterior ethmoidal air cells and the maxillary sinuses drain into the middle meatus (MM) – the posterior ethmoid cells and the sphenoid sinus drain into the superior meatus.

The paranasal sinuses consist of pneumatized spaces within the facial bones that surround the nasal cavity. In childhood, they are present in a rudimentary form, reaching their final configuration during adolescence: ●

●

●

●

Frontal sinus [3]: This is not visible on X-rays until the age of 6 years. Its posterior wall is very thin and contacts with the meninges and brain. Its base has a lateral part in close relation to the orbital roof, and its medial wall is continuous with the ethmoid bone. Maxillary sinus [1]: This is present at birth. Its superior wall corresponds to the floor of the orbit and is related to the infraorbital nerve. Its inferior wall forms the superior dental arch (root of premolars and first molar). Its posterior wall is the thickest and is related to the pterygopalatine fossa (Figure 3.18). Sphenoid sinus: The pneumatization of the sphenoid sinus is evident after the age of 3 years. It is separated by a thin central septum. Its anterior wall is related to the ethmoid cells (Figure 3.18), its posterior wall is bounded by the clivus (Figure 3.18), its superior wall by the sella turcica, and its inferior wall by the nasopharynx. The cavernous sinuses are located on both sides of the lateral walls. Ethmoid sinuses: These are divided into anterior and posterior air cells, separated by the lateral insertion of the middle concha into the lamina papyracea (called basal lamina). Their lateral wall is defined by the lamina papyracea of the ethmoid bone, which separates them from the orbit. Their superior wall bounds with the frontal sinuses. The anterior nares are situated inferomedially, and the olfactory fossa and the sphenoid sinus lie posteriorly.

●

●

Orbit The bone structure of the orbit has the shape of a pyramid with the base situated anterolaterally and the vertex situated posteromedially, converging at the middle cranial fossa (Figures 3.19 and 3.20). It is formed by the frontal, sphenoid, zygomatic, maxillary, ethmoid, lacrimal, and palatal bones. They form a foramen and two fissures through which vascular and nervous structures communicate the orbit with the middle cranial fossa [28]. Its boundaries are (Figure 3.19): ●

●

●

●

●

●

Key anatomical landmarks in CT of the paranasal sinuses [27]: ●

Ostiomeatal complex: This consists of a group of anatomical structures that allow the drainage of the frontal and maxillary sinuses and the anterior ethmoid

air cells into the MM. It is formed by the maxillary infundibulum (MI), which is the pathway that communicates the maxillary sinus with the MM. Its lateral boundaries are the ethmoid concha [3], which is the most anterior ethmoidal air cells, while it is medially bounded by the uncinate process, related to the lateral wall of the nasal fossa. The MM and the MI converge at the apex of the uncinate process, specifically in the hiatus semilunaris, where the frontal and maxillary sinuses end up draining. Sphenoethmoidal recess: This is the drainage pathway of the sphenoidal sinus and posterior ethmoidal air cells into the superior meatus. Olfactory fossa: This forms the roof of the nasal cavity. Inferiorly, it contains the cribiform plate, through which the crista galli process arises in its middle zone and divides the olfactory fossa into two cavities. On both sides, the lateral lamellae separate the olfactory fossa from the ethmoidal air cells. The lateral lamella is a very thin bony structure, susceptible to laceration by trauma or surgery.

●

Roof: formed by the frontal bone and the lesser wing of sphenoid. Floor: weaker and formed by the zygoma and the maxilla. Medial wall: lamina papyracea of the ethmoid, palatal, maxillary, lacrimal, and lesser wing of sphenoid. Lateral wall: formed by the zygoma and the greater wing of sphenoid. Posteriorly, the orbit communicates with the endocranial cavity via the superior orbital fissure, inferior orbital fissure, and optic canal (Figure 3.20). Superior orbital fissure, between the greater and lesser sphenoid wings, and through which the III, IV, VI, and VI cranial nerves, superior ophthalmic vein, and sympathetic nerve branches pass. It communicates directly with the cavernous sinus. Inferior orbital fissure, between the lateral wall and the floor of the orbit. It communicates with the pterygopalatine fossa, which is an important pathway of

71

72

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(d)

(b)

(e)

(c)

(f)

Figure 3.19 Orbital fissures. (I) Volume rendering of the orbits. (a–d) Axial noncontrast CT images of the orbit (bone window). (e, f) Coronal noncontrast MPR CT images (bone window). 1, inferior orbital fissure; 2, superior orbital fissure; 3, optic canal; 4, pterygopalatine fossa; 5, sphenopalatine foramen; 6, pterygomaxillary fissure. The inferior orbital fissure is in direct communication with the pterygopalatine fossa, located on the posterior part of the facial bones (between the posterior wall of the maxillary sinus and the pterygoid process). The pterygopalatine fossa (4) is an important pathway for the extension of lesions. It communicates with the infratemporal fossa of the masticator space laterally via the pterygomaxillary fissure (6), and with the nasal fossa medially via the sphenopalatine foramen (5). The superior orbital fissure communicates with the cavernous sinus.

extension of adjacent lesions. The second trigeminal branch (V2) and the infraorbital nerve pass through this fissure. The pterygopalatine fossa is located between the posterior wall of the maxillary sinus and the pterygoid processes, communicating with the nasal fossa medially via the sphenopalatine foramen, and with the infratemporal fossa of the masticator space laterally via the pterygomaxillary fissure.

●

Optic canal, located within the lesser sphenoid wing. The optic nerve passes through this canal.

The orbit can be divided into the preseptal space, the ocular globe, and the retro-ocular spaces (Figure 3.20): ●

Preseptal space, anterior to the orbital septum (a): This forms the front edge of the orbit. The orbital septum is inserted into the periosteum of the orbital mar-

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

(a)

●

●

vascular uveal tract, formed by the choroid, ciliary bodies and the iris. The choroid becomes visible on MRI after intravenous contrast administration. retina.

The inner part of the ocular globe is divided into two cavities by the posterior surface of the lens: ●

●

●

(b)

●

●

Figure 3.20 Orbit. (a) Axial T1-weighted MRI. (b) Coronal T2-weighted MRI. 1, inferior rectus muscle; 2, medial rectus muscle; 3, lateral rectus muscle; 4, superior rectus muscle; 5, optic nerve; 6, perioptic subarachnoid space; 7, lacrimal gland; 8, superior oblique muscle; 9, levator palpebrae superioris muscle; 10, sclera; 11, choroid; 12, superior ophthalmic vein; 13, lamina papyracea. Spaces of the orbit: a, preseptal space; b, ocular globe; c, extraconal space; d, intraconal space.

●

●

gin peripherally, and into the levator palpebrae superioris and the tarsal plate centrally. It separates the preseptal soft tissues (eyelids, orbicularis oculi muscle, eyelid fat) from the ocular globe and the retroocular space. Ocular globe (b) [29, 30]: This has a wall formed by three layers: sclera and cornea

the anterior segment contains the aqueous humor and is subdivided into an anterior chamber, which extends from the cornea to the iris, and a posterior chamber, from the iris to the lens the posterior segment, corresponding to the vitreous body, contains the vitreous humor and is located posterior to the lens. Retro-ocular spaces: The space behind the ocular globe is divided into the intraconal and extraconal compartments, based on whether it is inside or outside the musculofascial cone, respectively. The musculofascial cone is formed by the superior, lateral, medial, and inferior rectus muscles. These muscles are inserted posteriorly into the tendinous (or Zinn) ring at the orbital apex. The superior and inferior oblique muscles are not part of the musculofascial cone, although the former is also inserted into the tendinous ring. The levator palpeabrae superioris muscle lies next to the superior rectus muscle, from and they are often difficult to differentiate from each other on imaging. Intraconal compartment: This is inside the musculofascial cone and contains fat, the ophthalmic artery, the superior ophthalmic vein, the optic nerve, the perioptic space, and the III, VI, and VI cranial nerves. Extraconal compartment: This is outside the muscles of the musculofascial cone. It contains the lacrimal gland and ducts, the IV cranial nerve, and fat.

­Thoracic Anatomy: Introduction The thorax is a complex anatomical region containing vital organs with different densities surrounded by a bone structure which exerts a protective effect. Chest CT is the most sensitive and precise diagnostic imaging method both for depiction of anatomical structures and detection of thoracic pathology. The significant reduction in acquisition time allows the study of the thorax during a single breath-hold of few seconds, avoiding respiratory motion artifacts and obtaining submillimeter slices with no spacing between them. This results in high spatial resolution in the axial plane and in multiplanar and 3D reconstructions. In addition, the administration of intravascular iodine contrast considerably increases the contrast resolution of this imaging technique.

73

74

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

CT images should always be visualized in mediastinal window (level 40 HU, window width 350 HU) and lung window (level −600 to −700 HU, width 1000–1600 HU), with a standard reconstruction algorithm for the mediastinum and a high-spatial resolution reconstruction for the lung parenchyma (Figure 3.21) [1]. Chest MRI provides high-contrast resolution for soft tissues, high sensitivity to blood flow, multiplanar capability, and absence of ionizing radiations. In addition, it can be used in patients with allergy to iodine contrast. Traditionally, it has been considered a limited technique for the study of the thorax and lungs due to the low proton density of the normal lung and the magnetic susceptibility artifacts caused by air-tissue interfaces. Technical advances with more powerful gradients and the development of faster sequences have expanded its indications. The examinations are performed with a body coil and include fast sequences (gradient echo, turbo spin echo) weighted on T1 and T2. These fast sequences, along with parallel acquisition techniques and matrix sizing, are strategies to allow accelerated acquisition during a single breath-hold to minimize motion artifacts [31].

Anatomy of the Pulmonary Hilum The pulmonary hilum is a complex anatomical structure containing the lobar and segmental bronchi, pulmonary arteries and veins, bronchial arteries and veins, soft tissue, and lymph nodes. The arrangement of the bronchi,

vessels, and lymph nodes and their constant relationships to certain hilar structures allow their precise recognition. Identification of the specific bronchi is essential in the analysis of the pulmonary hilum. The branching pattern of the bronchial tree is usually constant (Figure 3.22) [32].

Anatomy of the Bronchi and Hilar Vessels Current CT scanners (with 1 mm slice thickness) allow full visibility of the segmental bronchi, although their morphology may vary based on their spatial orientation. The bronchi provide anatomical support to the hilum. The hilar vessels have a constant relationship with the bronchi, thus identification of the hilar bronchi should be the first step to approach the analysis of the hilum. Most of the segmental hilar arteries can be correctly identified based on their close association with the specific segmental bronchi, which can be traced in consecutive images to identify their origin. The assessment of the vascular structures of the hilum is facilitated based on specific anatomic levels with respect to the bronchi [33]. Anatomy of the Right Bronchus

The right main bronchus is shorter than the left one and bifurcates into the right upper lobe (RUL) bronchus and the intermediate bronchus. The carina, right main bronchus, and left upper lobe (LUL) bronchus are often visualized in the same slice (Figure 3.23) [34].

Figure 3.21 Typical Hounsfield unit values of anatomical structures in normal chest CT (left, mediastinal window; right, lung window). Note the right pleural effusion demonstrating water HU values.

Figure 3.22 Normal bronchial tree. LL, lower lobes. Trachea

Apical Right upper lobe Posterior Anterior Bronchus intermedius Middle lobe

Basal segments Lower lobe

(a)

(c)

Left upper lobe

Superior Segs LL

Apical posterior Anterior Lingula

Basal segments lower lobe

(b)

(d)

Figure 3.23 Axial CT image (lung window). (a) Apical segment bronchus of the right upper lobe (arrow). Coronal (b) and axial (c) CT images (lung window). (d) Tracheal bronchus (arrows in b and c) approximately 2 cm above the carina (arrow).

76

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

The RUL bronchus is located at the level or just below the carina. It runs laterally for 1–2 cm before dividing into three segmental branches: apical, anterior, and posterior (Figures  3.23 and  3.24). The posterior wall of the RUL bronchus can be seen as a thin line since it is well delimited from the surrounding lung parenchyma. Its branching pattern is variable, mainly due to differences in the origin of the apical segment bronchus.

and right lower lobe (RLL) bronchi (Figures 3.25–3.27). Its morphology is round or oval in the axial view. Its posterior wall is very well defined, with a maximum thickness of 3 mm, and delimits the apical segment of the RIL bronchus.

Right Upper Lobe Bronchus

●

●

Middle Lobe Bronchus The ML bronchus originates from the anterolateral wall of the intermediate bronchus and marks the origin of the RLL bronchus. It extends 1–2 cm before bifurcating into the medial and lateral segment bronchi. Owing to its orientation, its main segment as well as the bronchi of the medial and lateral segments are often seen 1–2 cm below the origin of the ML bronchus (Figure 3.29).

Apical segment bronchus: This becomes visible above the RUL bronchus usually at the level of the distal trachea (Figure 3.23). It is visualized as a rounded structure of air density in the axial plane. Anterior and posterior segment bronchi: These emerge as a Y-shaped bifurcation (Figure 3.24). Anatomical variations are rare, the most common being the so-called tracheal bronchus.

The pre-bifurcation segment of the RLL bronchus is very short. Near its origin, it gives rise to the bronchus of the apical segment (Figure  3.28), and the basal bronchial trunk continues for a short distance (Figure 3.29) until it splits off into the four basal segment branches of the RLL (medial, anterior, lateral, and posterior) (Figure 3.30) [34].

Right Lower Lobe Bronchi

The term tracheal bronchus consists of a bronchus arising from the trachea or main bronchus which runs toward the upper lobe territory. This abnormal bronchus usually originates in the right side of the tracheal wall, less than 2 cm superior to the carina, and may provide entire support to the upper lobe or its apical segment. If a branch of the upper lobe bronchus is missing, the tracheal bronchus is defined as “displaced”; if the right upper lobe bronchus has a normal trifurcation into apical, anterior, and posterior segment bronchi, the tracheal bronchus is called “supernumerary.”

●

●

Intermediate Bronchus The intermediate bronchus measures between 3 and 4 cm in length from the level of the RUL bronchus and gives rise to the middle lobe (ML) (a)

Apical segment: This emerges at the same level or slightly inferior with respect to the origin of the ML bronchus (Figure 3.28). Basal trunk and segments of the basal pyramid: Below the origin of the apical segment bronchus, the basal bronchial trunk (or basal trunk) extends inferiorly approximately 5 mm (Figure 3.29).

The basal segments have a typical orientation starting medially and anteriorly in a counter-clockwise direction (M-A-L-P) (Figure 3.30). (b)

Figure 3.24 (a) Axial CT image (lung window). Right upper lobe bronchus and apicoposterior segment bronchus of the left upper lobe. Ant, anterior; P, posterior; AP, antero-posterior. (b) Axial CT image (mediastinal window). AT, anterior trunk; SPV, superior pulmonary vein; AS, anterior segment artery; LPA, left pulmonary artery.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

(a)

(b)

Figure 3.25 (a) Axial CT image (lung window). IB, intermediate bronchus; ST, superior trunk. (b) Axial CT image (mediastinal window). RPA, right pulmonary artery; LPA, left pulmonary artery; SPV, superior pulmonary vein.

(a)

(b)

Figure 3.26 (a) Axial CT image (lung window). IB, intermediate bronchus; LM, left main bronchus; ST, superior trunk. (b) Axial CT image (mediastinal window). SPV, superior pulmonary vein; LPA, left pulmonary artery.

(a)

(b)

Figure 3.27 (a) Axial CT image (lung window). IB, intermediate bronchus; Lin, lingular bronchus. (b) Axial CT image (mediastinal window). IPA, interlobar pulmonary artery; SPV, superior pulmonary vein.

77

(a)

(b)

Figure 3.28 (a) Axial CT image (lung window). Middle lobe and apical segment of the right lower lobe bronchi, lingular bronchus, and apical segment of the left lower lobe. ML, middle lobe bronchus; RLL, right lower lobe bronchus; R-Api, apical segment of the right lower lobe bronchus; Lin, lingular bronchus; L-Api, apical segment of the left lower lobe bronchus. (b) Axial CT image (mediastinal window). ML, middle lobe; RLL, LLL, right and left lower lobe arteries, respectively; SPV, superior pulmonary vein.

(a)

(b)

Figure 3.29 Middle lobe and its segments. Trunks of the lower basal lobe. (a) Axial CT image (lung window). ML, middle lobe bronchus; MS, medial segment; LS, lateral segment; BT, basal trunks. (b) Axial CT image (mediastinal window). RLL, LLL, right and left lower lobe pulmonary arteries, respectively; IPV, inferior pulmonary vein.

(a)

(b)

Figure 3.30 Basal segments of the right lower lobe. (a) Axial CT image (lung window). The four basal segments are shown, which in a counter-clockwise direction are the medial basal (M), anterior basal (A), lateral basal (L), and posterior basal (P). (b) Axial CT image (mediastinal window). Seg, segmental arteries; LV, lower pulmonary veins.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

Vascular Anatomy of the Right Hilum

The vascular anatomy is best recognized at specific levels in relation to the bronchi [35]. Next to the level of the carina, one or several branches of the anterior trunk irrigate the apical segment of the RUL and one or several branches of the right superior pulmonary vein drain it. These structures can be seen in cross-sectional images next to the apical segment bronchus (Figure 3.23).

Level of the Carina and Apical Segment Bronchus

Level of  the Right Upper Lobe Bronchus At this level, the anterior trunk has not yet divided and can be easily identified (Figure 3.24). This large vessel is the first main branch of the right main pulmonary artery. A branch of the right superior pulmonary vein, the posterior vein, is located within the angle formed by the bifurcation of the RUL bronchus into its anterior and posterior branches (Figure 3.24). of  the Intermediate Bronchus The interlobar pulmonary artery is located anterior and lateral to the corresponding bronchus (Figures 3.25–3.27). The superior right pulmonary vein is located anterior to the right interlobar pulmonary artery and sometimes two veins can be seen.

Level

Level of the Middle Lobe Bronchus At the level of the origin of the ML bronchus, the pulmonary artery of the RLL is located lateral to the external walls of the ML and RLL bronchi. The pulmonary artery of the ML runs parallel to the ML bronchus (Figures  3.28 and  3.29). The superior right pulmonary vein runs anterior and medial to the ML and RLL bronchi until it enters the upper part of the left atrium. Level of  the Basal Segment Bronchus of  the Right Lower Lobe Inferior to the origin of the ML bronchus, the

pulmonary artery of the RLL usually divides into two shorter trunks, which in turn divide into the four pulmonary arteries of the basal segments (Figures  3.30 and  3.31). The inferior pulmonary veins join to form the inferior pulmonary vein, passing behind the RLL bronchus and its arteries before entering the lower part of left atrium. Anatomy of the Left Bronchus

The distinctive feature of the left main bronchus is its longer length compared to the right one, thus it can be seen during several contiguous slices below the carina [34]. The LUL bronchus is 2–3 cm long (Figure 3.26). In 75% of cases it branches out into an upper trunk and the bronchus of the lingula. The upper Left Upper Lobe Bronchus

trunk gives rise to the bronchi of the anterior and apicoposterior segments (Figures 3.25 and 3.27). In 25% of cases, the LUL bronchus branches into the posterior apical bronchus, the anterior segmental bronchus, and the lingular bronchus. ●

●

Anterior segment bronchus: Visible above the level of the LUL bronchus (Figure 3.24). Oriented anteriorly. Bronchus of the apicoposterior segment: Visible as a rounded air space superior to the origin of the anterior segment bronchus (Figure 3.24).

The lingular bronchus originates from the lower surface of the most distal part of the LUL bronchus and runs inferiorly with an oblique course (Figure 3.27). It subsequently divides into the upper and lower segment bronchi. The upper branch runs more laterally and is closer to the plane of the image in a similar way to the branching pattern of the medial and lateral bronchi of the ML.

Lingular Bronchus

The LLL bronchus displays a branching pattern similar to that of the RLL bronchus, but there are three basal segments (anteromedial, lateral, and posterior) instead of four. Left Lower Lobe Bronchus

●

●

Apical segment bronchus: This originates less than 10 mm distal to the origin of the LLL bronchus, being similar in size and arrangement pattern to the apical segment bronchus on the right side (Figure 3.28). Basal trunk and segments: The basal trunk is the LLL bronchus itself, which gives rise to the basal segments and is visible below the origin of the apical segment (Figure 3.29). As mentioned above, the configuration of the basal segmental bronchi is similar to that of the contralateral basal pyramid bronchi, except for the fact that the medial and anterior basal bronchi typically originate as a common trunk (Figure  3.31). Accordingly, their clockwise arrangement is M-A (anteromedial)-L-P.

Vascular Anatomy of the Left Hilum

On the left side there are more anatomical variants than on the right side, although the constant relationships between vessels and airways remain [35, 36]. Level of  the anterior and  posterior apical segments of  the LUL The apicoposterior segment bronchus and its arteries

and veins are arranged similarly to those on the right side. The artery of the anterior segment of the LUL runs medially to the bronchus of the anterior segment (Figure 3.24).

Level of  the LUL Bronchus The left interlobar pulmonary artery causes a large convexity of the posterior part of the hilum and the upper pulmonary vein causes an anterior

79

80

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

Figure 3.31 Basal segments of the left lower lobe. (a) Axial CT image (lung window). Basal segments in a clockwise direction are the anteromedial basal (AM), lateral basal (L), and posterior basal (P). (b) Axial CT image (mediastinal window). Seg, segmental arteries; IPV, inferior pulmonary veins.

convexity medial to the ascending bronchial trunk (Figure 3.26). Level of the Lingular Bronchus The left interlobar pulmonary artery can be seen in cross-sectional images lateral to the LLL bronchus, posterolateral to the lingular bronchus, and anterolateral to the apical segment bronchus (Figure 3.27). The superior pulmonary vein passes anteriorly and medially to the corresponding bronchus to enter the left atrium.

In this region, the anatomy of the lower part of the left hilum is virtually a mirror image of the right side (Figures  3.30 and  3.31). The branches of the pulmonary artery of the LLL are located laterally and posteriorly to the basal bronchus of the LLL. The inferior left pulmonary vein runs anterolateral to the descending aorta and posterior to the bronchi and arteries to enter the left atrium.

Level of the Basal Pyramid Bronchi of the LLL

Mediastinum The mediastinum is the compartment located between the two lungs, anterior to the spine, posterior to the sternum, and extending from the thoracic inlet to the diaphragm [37]. The aorta and its branches, the large veins, the pulmonary arteries, the trachea, and the main bronchi serve as precise landmarks to anatomically locate other important mediastinal structures [38, 39]. For a better anatomical understanding, the division proposed by Webb [3] will be followed, considering four compartments in the cranio-caudal direction:

1) Superior or supra-aortic mediastinum. 2) Aortic arch and aorto-pulmonary window. 3) Pulmonary arteries, subcarinal space and azygoesophageal recess. 4) Heart and paracardiac mediastinum. Supra-­aortic Mediastinum

The trachea is located in a central position inferior to the thoracic inlet (Figure  3.32). The esophagus is positioned posterior to the trachea and slightly displaced to the left (most often) or to the right. It is usually collapsed and looks like a flattened structure with soft tissue density. There may be small amounts of air or fluid in its lumen. At this level, the supra-aortic vessels, and the subclavian and brachiocephalic veins are also depicted. The latter are the most anterior and lateral visible vessels. They are located immediately behind the clavicular heads. Although they may vary in size, their positions are relatively constant (Figure 3.32). The right brachiocephalic vein has a vertical course along its entire length. The left brachiocephalic vein is longer and runs horizontally across the mediastinum from left to right (Figure  3.32). This vein has a rather variable location in the cranio-caudal axis. The trunks of the brachiocephalic, subclavian, and carotid arteries are located posterior to the veins and adjacent to the anterior and lateral walls of the trachea. Their arrangement is relatively constant. The trunk of the brachiocephalic artery is located in close proximity to the anterior tracheal wall, near the midline or slightly to the right. It is the most variable of all the supra-aortic great vessels.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

(a)

(c)

(b)

(d)

(e)

Figure 3.32 Supra-aortic mediastinum. (a–d) Axial contrast-enhanced CT image (mediastinal window). C, clavicle; T, trachea; M, sternal manubrium; E, esophagus. (e) Axial contrast-enhanced CT image (lung window). Arch of the azygos with posterior azygos vein lying more lateral to its normal course (arrow).

The left common carotid artery is located to the left and slightly posterior to the brachiocephalic trunk. It usually has the smallest diameter of the three supra-aortic vessels. The left subclavian artery is located further posteriorly in most individuals, running along the left side of the trachea or slightly posterior to the midline. Small vascular branches, the internal thoracic arteries and veins, and the vertebral arteries are also visible at this level. There is an anatomical variant known as “aberrant right subclavian artery,” which consists of a right subclavian artery that courses through the mediastinum from left to right between the esophagus and the vertebral body (Figure 3.32d). The thyroid gland is also visible in the superior mediastinum, with its right and left lobes located on both sides of the trachea. The density of the thyroid gland is higher than that of the surrounding soft tissues. There is another congenital anomaly called the azygos lobe, which consists of the arch of the azygos being located in a more superior position than usual, near the point of junction of the brachiocephalic veins. The azygos fissure can be seen defining the pulmonary lobe of the azygos (Figure 3.32e). Aortic Arch and Aorto-­pulmonary Window

At this level several structures can be depicted, including the large mediastinal vessels, the aorta, and the superior

vena cava, as well as the mediastinal spaces and different groups of lymph nodes (Figure 3.33). The aortic arch usually shows a characteristic morphology, although it can be variable. The anterior part of the aortic arch is located anteriorly and slightly to the right of the trachea. It then courses posteriorly, passing to the left side of the trachea. The posterior aortic arch is located anterior and lateral to the spine. The position of the anterior and posterior parts of the aortic arch may vary depending on the tortuosity and the existence of arteriosclerosis. The superior vena cava is visible anteriorly and slightly lateralized to the right side of the trachea, with an elliptical morphology on cross-sectional imaging (Figure 3.33). The appearance of the esophagus is similar to that in upper levels and is located posterior to the trachea, although its course may be variable. It is more frequently located to the left side of the midline (Figure 3.33a–d). The pre-tracheal space is limited by the aortic arch on the left, the superior vena cava and the mediastinal pleura on the right, and the trachea posteriorly. It has a triangular shape and contains mediastinal fat (Figure 3.33a). The pre-vascular space also has a triangular shape and is located anterior to the great vessels (superior vena cava and aorta). The apex of this space is posterior to the sternum,

81

82

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

(d)

(e)

(f)

Figure 3.33 Sub-aortic mediastinum. Azygos arch and aorto-pulmonary window. (a–f). Axial contrast-enhanced CT images (mediastinal window). T, trachea; AA, ascending aorta; DA, descending aorta. Note the anatomical variant consisting of elongated left aortic arch with aberrant right subclavian artery (arrows in e and f).

forming the anterior junction line. This mediastinal compartment contains fat, lymph nodes, and the thymus, which involutes with age. The mediastinal pleural reflections lateral to the pre-vascular space can be concave or convex. At a level below the aortic arch, the ascending and the descending parts of the aorta are visualized as separate structures (Figure 3.33c,d).

In a slightly lower position, the trachea bifurcates into the main bronchi and adopts a triangular morphology prior to the carina. On the right, the arch of the azygos originates from the posterior wall of the superior vena cava, passes over the right main bronchus, and continues along the mediastinum on the right side and anterior to the spine

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

(Figure  3.33b). It delimits the right border of the pretracheal space. Below the level of the arch of the azygos, the azygos vein can be seen [40]. The aorto-pulmonary window is located on the left side of the mediastinum, inferior to the aorta and above the pulmonary trunk (Figure  3.33c). This space contains fat, lymph nodes, the recurrent laryngeal nerve, and the ligamentum arteriosum (although the latter is usually not visible) [41]. The superior pericardial recess contains a small amount of fluid and can be visualized in the pre-tracheal space immediately posterior to the ascending aorta (Figure 3.13d). It should not be mistaken for a lymph node, and can be distinguished by its typical location, contact with the aortic wall, crescent shape, and fluid-like attenuation. The anterior recess of the pericardial space can be visualized anterior to the aortic arch, the ascending aorta, and the pulmonary trunk (Figure 3.33d). Many developmental abnormalities of the aortic arch have been described, such as a double aortic arch, right aortic arch with or without aberrant subclavian artery, and

elongated left aortic arch with aberrant right subclavian artery (Figure 3.33e,f). Pulmonary Arteries, Subcarinal Space, and Azygo-­ esophageal Recess

At the level of the carina and the main bronchi, the pulmonary trunk divides into the right and left pulmonary arteries (Figure 3.34). The left pulmonary artery is slightly above the right one (around 1 cm). The emergence of the right pulmonary artery is almost right-angled relative to the pulmonary trunk. It then crosses the mediastinum from left to right anterior to the carina or the main bronchi (Figure  3.34a). This artery serves as the most inferior bound to the pre-tracheal space and the anterior limit of the subcarinal space. The azygos vein runs parallel to the esophagus on the right side of the mediastinum, contacts laterally with the medial pleural reflection of the RUL, and defines the posterior limit of the azygo-esophageal recess (Figure 3.34). On the left side, the hemiazygos vein (not always visible) runs parallel to the descending aorta (Figure 3.34b). This

(a)

(b)

(c)

(d)

Figure 3.34 (a–d) Axial contrast-enhanced CT images (mediastinal window) at the level of the pulmonary trunk, subcarinal space, and azygo-esophageal recess. AA, ascending aorta; DA, descending aorta; PA, pulmonary artery; RP, right pulmonary artery; LP, left pulmonary artery; RB, right bronchus; LB, left bronchus; E, esophagus. Note the anatomical variant consisting of the hemiazygos vein draining directly into the left brachiocephalic vein.

83

84

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

vein generally drains into the azygos vein through a branch that crosses the midline behind the aorta at the height of the T8 vertebral body. There are several congenital anomalies of the superior vena cava that can be incidentally found as anatomical variants, including the continuation of the superior vena cava with the azygos-hemiazygos system and, more rarely, the dilated hemiazygos vein draining directly into the left brachiocephalic trunk instead of joining the azygos vein (Figure 3.34c,d). The region of the mediastinum located just below the carina is known as the sub-carinal space and is limited laterally by the main bronchi. It is closely related to the esophagus and contains lymph nodes (Figure 3.34a). (a)

The azygo-esophageal recess (Figure  3.34) is located inferior to the tracheal carina and the arch of the azygos. The medial part of the right lung contacts the mediastinum where the azygos vein, the esophagus, and the sub-carinal space are located. The contour of the azygo-esophageal recess is usually concave. Heart and Paracardiac Mediastinum

The pre-vascular mediastinum at the level of the heart becomes thin and practically obliterated because the ventricles come into contact with the anterior chest wall (Figures 3.35 and 3.36). The azygo-esophageal recess can still be visualized posterior to the heart down to the level of the diaphragm. The left paravertebral space is located posterior to the descending (b)

Figure 3.35 Paracardiac mediastinum. (a, b) Axial contrast-enhanced CT images (mediastinal window). AR, aortic root; DA, descending aorta.

(a)

(b)

Figure 3.36 Normal cardiac anatomy. (a, b) Axial contrast-enhanced CT images (mediastinal window). RV, right ventricle; LV, left ventricle; RA, right atrium; LA, left atrium; LVO, left ventricular outflow tract; DA, descending aorta.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

aorta and contains the hemiazygos vein, fat, and small lymph nodes. The heart usually has an oblique position relative to the anatomical axes, in such a way that it is oriented from right to left and from top to bottom. In the upper planes, the outflow tract of the left ventricle and the aortic valve are in a central position within the heart (Figure 3.35a). Next to the aortic root it is possible to visualize the exit of the coronary arteries [42]. The outflow tract of the right ventricle is directed toward the left and is visible anterior or slightly to the left of the outflow tract of the left ventricle. At this level, the entrance of the vena cava into the right atrium, both atrial appendages, and the left atrium are visible (Figure 3.35b). More inferiorly, the four cardiac chambers can be seen, with the atria separated from the ventricles by the atrioventricular sulcus. The interventricular septum is oriented inferiorly and toward the left, being slightly convex anteriorly due to the greater pressure in the left ventricle. The lateral aspect or free wall of the left ventricle is usually three times larger than that of the right ventricle. The left ventricle has an elliptical shape. The right ventricle has a triangular shape and is oriented anteriorly and to the right (Figure 3.36a). In its lower part, the coronary sinus can be seen heading toward the most inferior part of the right atrium (Figure 3.36b). The diameter of the coronary sinus may be increased in the presence of anatomical variants in which a left superior vena cava drains into it. This occurs in two anatomical variants, namely double superior vena cava and persistent superior left vena cava (Figure 3.37).

Thoracic Wall and Pleural Surfaces Several structures arranged in layers surround the lung and wrap around the inner surface of the chest cavity. The layers of the visceral and parietal pleura surrounding the lung and pleural space are less than 0.5 mm thick. (a)

(b)

External to the parietal pleura, the extra-pleural fat separates the parietal pleura from the endothoracic fascia. This fat layer is very thin but may be thicker on the lateral and posterolateral aspects of the ribs. The thoracic cavity is surrounded by the endothoracic fascia, a fibroelastic layer which covers the surface of the intercostal muscles and the corresponding ribs. Anteriorly, this fascia blends with the perichondrium and periosteum of the costal cartilages and the sternum. In its posterior part, the fascia continues with the prevertebral fascia, which covers the vertebral bodies and the intervertebral discs. External to the endothoracic fascia there are three layers of intercostal muscles. The innermost intercostal muscle passes between the inner surfaces of the ribs. It is thin and separated from the intercostal muscles by a layer of fat and the intercostal vessels and nerves. The innermost intercostal muscles are incomplete in the anterior and posterior parts of the thoracic wall. The transverse and subcostal thoracic muscles may occupy the same relative plane. Anteriorly, the transversus thoracis muscle is made up of four or five muscular strips that arise from the xiphoid process or the lower part of the sternum and pass superolaterally from the second to sixth costal cartilages. The internal thoracic vessels lie external to the transversus thoracis muscle. Posteriorly, the subcostal muscles, which are thin and rather variable, extend from the inner surface of the angle of the lower ribs, cross one or two ribs and the intercostal spaces, and reach the inner side of the rib above. On CT, a 1–2 mm thick soft tissue band (the intercostal band) is visible in the anterolateral and posterolateral intercostal spaces at the site in which the lung comes into contact with the thoracic wall (Figure 3.38). This line represents the innermost intercostal muscles, the combination of the visceral and parietal pleural layers, the pleural space containing fluid, the endothoracic fascia and fat layers. (c)

Figure 3.37 Left superior vena cava variants. (a–c) Axial contrast-enhanced CT image (mediastinal window). (a) Double superior vena cava. (b) Persistent left superior vein cava. (c) Dilated coronary sinus.

85

86

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 3.38 Axial contrast-enhanced CT image (mediastinal window). Intercostal band represented by a thin line corresponding to the innermost intercostal muscle (arrows).

In the paravertebral regions, the innermost intercostal muscle is absent.

the azygos and hemiazygous veins, the thoracic duct, the intercostal arteries, and the splanchnic nerves. The esophageal hiatus is located in a more anterior position, in the muscular portion of the diaphragm. The esophagus, the two portions of the vagus nerve, and small vessels pass through this diaphragmatic aperture. The inferior caval foramen crosses the fibrous portion of the central diaphragmatic tendon, anteriorly and to the right of the esophageal hiatus. Of the three diaphragmatic apertures, the aortic hiatus is the one most easily visualized. The esophageal hiatus is visualized as an opening at the junction between the esophagus and the stomach. The inferior vena caval foramen can be located based on the position of the inferior vena cava (IVC).

­Anatomy of the Abdomen Anatomy of the Hepato-­biliary System and Pancreas

Diaphragm The central portion of the diaphragm is not visualized as a single structure, but its location can be inferred based on the position of the lung bases (in an upper plane) and the abdominal organs (in a lower plane)  [43]. As the more peripheral portions of the diaphragm extend inferiorly toward their costal and sternal insertions, the anterior, posterior, and lateral portions of the diaphragm become visible, adjacent to the retroperitoneal fat. The right and left diaphragmatic crura are tendinous structures that originate inferiorly on the anterior surfaces of the lumbar vertebral bodies and the corresponding intervertebral discs and continue with the anterior longitudinal ligament of the spine. The crura ascend anterior to the spine on both sides of the aorta and then cross anteriorly and medially to join the muscular portion of the diaphragm located in front of the aorta to form the aortic hiatus. The right crus is larger than the left one and originates from the first three lumbar vertebrae. The left crus originates from the first two lumbar vertebrae. The diaphragmatic crura have a rounded appearance and it is possible to visualize their progressive fusion with the diaphragm in the upper slices. Diaphragmatic Apertures

The diaphragm is noncontinuous due to various apertures that permit the passage of different anatomical structures from the thorax to the abdomen. The aortic hiatus lies posteriorly and is bounded by the diaphragmatic crura anteriorly and by the vertebral bodies posteriorly. The anatomical structures that pass through this hiatus include the aorta,

Anatomy of the Liver

The liver is located in the right hypochondrium. It is partially peritonealized and covered by a fibrous capsule (Glisson’s capsule). The liver is fixed to the anterior abdominal wall and to the diaphragm by the falciform ligament which contains the ligamentum teres, the coronary ligament, and the two triangular ligaments. In its inferior and medial aspect it is fixed to the stomach and duodenum by the hepatogastric and hepatoduodenal ligaments. The connective tissue of the so-called “bare area” and the IVC, to which the bare area is solidly attached by the hepatic veins (HV), fixes the posterior surface of the liver [44] (Figure 3.39). The liver has external impressions, which can be seen on its surface. The peritoneum covering the surface of the liver folds back toward the underlying parenchyma, with or without an associated ligament and to a variable depth, creating the fissures of the liver fissures. There are four constant fissures, including the ligamentum teres, ligamentum venosum, gallbladder (or major), and transverse fissures. Fissures are clefts visible on the external surface of the liver which communicate with each other and with the intraperitoneal space. The transverse fissure is a peritoneal fold that forms the hepatic hilum. It contains the bifurcation of the portal vein (PV), the first branch of the hepatic artery (HA), and the common hepatic duct [44] (Figure 3.40). Classical anatomy of the liver This is based on external marks which divide the liver into four lobes, namely right, left, caudate (or Spiegel’s), and quadrate lobes. This classification is no longer in use in radiology.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

(a)

(b)

(c)

Falciform ligament Round ligament Ligamentum venosum Gastrohepatic ligament

Hepatic ligaments Figure 3.39 Ligaments of the liver. (a–c) Axial contrast-enhanced CT images (portal venous phase).

Functional anatomy This is based on the intraparenchymal distribution of the afferent blood supply of the liver, with the PV being the main and most easily identifiable landmark. The liver is divided into functional units, each with an independent portal supply, and venous and biliary drainage. Accordingly, removal of one or more units does not compromise the liver function. Following the initial division of the afferent branches of the hepatic hilum, the liver is divided into right and left hemi-livers. After the second division of the hilum branches, each hemi-liver is further divided into two sections: the right hemi-liver into anterior and posterior, and the left hemi-liver into medial and lateral. The third and final division splits each section into superior and inferior segments [44] (Figure 3.41). The most representative system to understand the functional anatomy of the liver is the Couinaud classification (Figure 3.42):

●

●

●

On the transverse plane, the main anatomical landmark is the portal bifurcation through the transverse fissure, which divides the liver into superior and inferior sections. Several vertical planes are then defined for each segment of the liver based on intersections with the transverse plane. The middle vertical plane defines the right and left hemilivers. The anatomical landmark superior to the transverse plane corresponds to the middle hepatic vein (MHV), which separates segment VIII from the superior portion of segment IV. Inferiorly, along the gallbladder and its fissure, this plane separates segment V from the inferior portion of segment IV. The right vertical plane separates the right anterior from the right posterior section. Superior to the portal bifurcation, the anatomical landmark is the right hepatic vein (RHV), which separates segment VII from segment VIII. Inferiorly, the limit is rather arbitrary, with the intersection of the RHV coinciding with the bifurcation of the

87

88

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

Umbilical fissure Ligamentum venosum fissure Transverse fissure Vesicular fissure

Hepatic fissures Figure 3.40 Fissures of the liver in a patient with cirrhosis and moderate free ascites. (a) Axial noncontrast CT image. (b) Coronal contrast-enhanced MPR CT image (portal venous phase). (c) Axial contrast-enhanced CT image (portal venous phase).

●

right portal vein (RPV), separating segment VI from segment V. The left vertical plane separates the medial section (segment IV) from the left lateral section (segments II and III). In its upper third, this plane is represented by the left hepatic vein (LHV), in its middle third by the umbilical portion of the left portal vein (LPV), and in its lower third by the fissure of the ligamentum teres.

The posterior sector is formed by segments I and IX. On the left side, the fissure of the ligamentum venosum separates this section from segment II, while the IVC separates it from segment VIII on the right side. Superiorly, it is bounded by the common trunk formed by the MHV and IVC and, posteriorly, by the horizontal portion of the LPV in the transverse fissure [44].

The classification of the International Hepatic Pancreatic Biliary Association (IHPBA) (Brisbone, 2000) is increasingly being used [45, 46]. It unifies both anatomical and surgical terminology, using Arabic numerals and segment 4 is not divided. In addition, it accepts both “section” and “sector” terms, but they are only equivalent to the Couinaud classification in the right liver (Figures  3.41 and 3.42). The right posterior section (or sector) is made up segments 7 and 6, while the anterior section is formed by segment 5. The left medial section includes segments 4 and 3, and the left lateral sector includes segment 2. The posterior sector described by Couinaud is independent of the rest of the liver parenchyma and receives vascular supply from right and left branches, draining directly into the VCI via the HVs [45, 46].

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

(a)

(b) M hepatic v

R hepatic v 7

L hepatic v 2

8

4a 3

6

5

4b Gall bladder

L portal v bile duct hepatic a

R portal v bile duct hepatic a

(c)

(d)

Figure 3.41 First-order division of the afferent vascular supply of the liver, resulting in two lobes. Segments 2, 3, and 4 belong to the left liver, and the other segments belong to the right liver. (a) Schematic diagram, adapted from Abdel-Misih and Bloomston [44]. (b, d) Axial contrast-enhanced CT image at the level of the hepatic veins (portal venous phase). (c) Coronal contrast-enhanced MPR CT image (portal venous phase). RHL, right hepatic lobe; LHL, left hepatic lobe.

The PV provides 80% of the vascular supply of the liver. The splenic vein and the superior mesenteric vein join to form the PV behind the head of the pancreas. The PV runs toward the hepatic hilum posterior to the bile duct and to the HA at the free edge of the lesser omentum. In the hepatic hilum, the PV bifurcates into left and RPVs. The LPV presents an abrupt angulation in the site of insertion of the venous ligament and becomes vertical (umbilical portion) medial to the ligamentum teres to vascularize segments 2, 3 and 4. The RPV is further divided into right anterior branch for segments 5 and 8, and right posterior branch for segments 6 and 7. The vascularization of the caudate lobe comes from an independent branch of the LPV [47] (Figure 3.43). There are several anatomical variants of the PV  [48] (Table 3.1 and Figure 3.1). The HA provides up to 35% of the vascular supply of the liver. The HA arises from the bifurcation of the celiac trunk and is referred to as common hepatic artery (CHA)

Vascularization of the Liver [4]

at this level. Before reaching the liver, the CHA bifurcates into the gastroduodenal artery and the proper HA. The latter further divides into left and right hepatic arteries at the level of the hepatoduodenal ligament  [47] (Figure 3.44). The HA presents numerous anatomical variants (Table 3.2) [48]. The anatomy of HVs is highly variable  [48]. In most cases (2 cm, consider independent superior and inferior drainage

Table 3.4

Anatomical variants of the middle hepatic vein.

Type

Description

I

Tributary veins for segments 5 and 4 (forming a proximal confluence) and another tributary vein for segment 8

II

Two large segments for segments 8 and 4 (forming a distal trunk), and the other segments are drained by small veins

III

Similar to type I, but with a bifurcation of unequal size for segments 5 and 8, and a small branch for segment 4

the duodenum and allows bile to fill the duct with subsequent retrograde flow into the cystic duct and gallbladder. The volume capacity of the gallbladder is approximately 30 mL, although it can dilate up to 300 mL. It is covered by visceral peritoneum. Anatomically, it is divided into fundus, body, infundibulum (Hartmann’s pouch), and neck. Its blood supply is provided by the cystic artery, which is usually a branch of the right HA despite high anatomical variability exists, including its origin in the left HA, proper HA, or even superior

Type

Description

I

Three veins, two for segment 4 and one for segment 2 Distal confluence The most common variant

II

Common trunk for segments 2 and 3, and receives a tributary for segment 4

III

Similar to type II but there are no tributary veins for segment 4

mesenteric artery. Its venous drainage is via the portal system [51] (Figure 3.46). Pancreas

The pancreas is a lobulated gland that measures between 15 and 20 cm in length and lies in the anterior pararenal space of the retroperitoneum. It is anatomically divided into four parts: head, neck, body, and tail. The pancreatic head is located on the duodenal flexure, to the right of the superior mesenteric vein. The uncinate process is an inferior prolongation of the hea, and is oriented to the left; it has a triangular shape and its antero-medial edge can be straight or concave. The pancreatic neck is the left portion of the head and lies anterior to the superior mesenteric vein. The body and tail of the pancreas are posterior to the lesser sac and stomach (Figures 3.47 and 3.48). The main pancreatic duct (of Wirsung) normally measures 3.5 mm at the level of the pancreatic head, 2.5 mm at the body, and 1.5 mm at the pancreatic tail. Its length ranges from 9.5 to 25 cm. There are approximately 27 different ductal configurations. Usually, the main duct has between 25 and 30 lateral branches which drain into it at right angles. The main pancreatic duct joins the bile duct and drains into the major papilla via the sphincter of Oddi to enter the duodenum. In most cases (80–90%) both ducts join within the sphincter through a muscular wrapping of approximately 10–15 mm in length. The biliopancreatic junction can be long (Y-shape) or short (V-shape). High biliopancreatic sites of junction can favor pancreatic reflux and lead to formation of bile duct cysts [50]. The arterial supply to the pancreatic head is mainly provided by the superior pancreatoduodenal artery, a branch of the gastroduodenal artery (which in turn is a tributary of the CHA), and by the inferior pancreatoduodenal artery, a branch of the superior mesenteric artery. Both of them are located between the pancreas and the duodenum, irrigating the head of the pancreas.

Vascularization of  the Pancreas

93

94

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(b) (a) Fundus

Neck

Body

Cystic duct

Infundibulum

(c)

(d)

Figure 3.46 Gallbladder and bile duct. (a) Diagram, adapted from Mortele et al. [50]. (b) Axial contrast-enhanced CT image (portal venous phase). (c, d) Coronal contrast-enhanced MPR CT image (portal venous phase).

Table 3.6

Anatomical variants of the left intrahepatic bile duct.

Frequency Type (%)

Description

I

55

Drainage of lateral segments (2 and 3) into a single duct next to the umbilical fissure, joining the duct for segment 4

II

30

Drainage of lateral segments, plus segment 4 drained by two ducts, one next to the umbilical fissure and the other one close to the biliary confluence

III

10

Independent drainage of segments 2 and 3, and two bile ducts for segment 4 The bile duct for segment 2 is joined next to the hilum

IV

30

Trifurcation, drainage of segments 2 and 3 together by a short bile duct, and drainage of segment 4 which is joined next to the hilum

Table 3.7 Anatomical variants of the right intrahepatic bile duct. Frequency Type (%) Description

I

56

RAD and RPD) merge to form the RHD, which joins the LHD to form the CHD (most common variant)

II

14

Trifurcation: union of the RAD, RPD, and LHD without RHD

IIIa 5 IIIb 15

RAD joins the LHD RPD joins the LHD

IVa IVb

RAD joins the CHD posterior to the confluence RPD joins the CHD posterior to the confluence

AHD, CHD, common hepatic duct; LHD, left hepatic duct; RAD, right anterior duct; RHD, right hepatic duct; RPD, right posterior duct.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology) Head

Neck

Body

Tail MPD

Santorini

Wirsung

Figure 3.47 Anatomy of the pancreas. Illustration adapted from O’Brien et al. [42].

(a)

(b)

(c)

(d)

Figure 3.48 Pancreas. (a-d) Axial contrast-enhanced CT image (pancreatic phase).

The arterial vascularization of the neck, body, and tail of the pancreas is primarily provided by pancreatic branches of the splenic artery. The longest and most important branch is the great pancreatic artery.

The venous drainage of the pancreatic neck and body is via the splenic vein, while drainage of the pancreatic head is via the superior mesenteric vein and PV.

95

96

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Anatomy of the Gastrointestinal System Stomach

The stomach is composed of five regions, namely the cardia and gastroesophageal junction, fundus, body, antrum, and pylorus. The fundus and body contain acid-secreting glands, while the antrum is lined with an alkaline-secreting epithelium and gastrin-producing endocrine G-cells. The gastroesophageal junction is closely related to the diaphragm. The fundus and body are oriented vertically, with the spleen lying more lateral, the lateral segment of the left hepatic lobe situated medial and anterior, and the abdominal aorta in a posterior location. The antrum and pylorus are oriented horizontally and related to the transverse colon inferiorly and to the left splenic flexure laterally. The fundus of the gallbladder is suspended above the pylorus [53] (Figure 3.49). The stomach is a richly vascularized organ. The arterial supply is provided by five main arterial branches. The left gastric artery, branch of the trifurcation of the celiac trunk, provides vascular support to the superior portion of the lesser curvature. The right gastric artery, branch of the CHA, irrigates the inferior portion of the lesser curvature. The right gastroepiploic artery, branch of the gastroduodenal artery, provides arterial supply to the antrum and inferior part of the body.

Vascularization of  the Stomach

(a)

The left gastroepiploic artery, branch of the splenic artery, irrigates the superior part of the gastric body. Finally, the splenic artery gives off short gastric arteries that supply the superior portion of the gastric body and fundus [53]. Small Intestine

The mucosa of the small intestine (or bowel) is characterized by presenting the valvulae conniventes (between 1.5 and 2 mm thick) which arise from the second portion of the duodenum and decrease in number as the intestine approaches the ileocecal valve. The small intestine is divided into three functional units: duodenum, jejunum, and ileum. The duodenum is a unique segment of the small intestine with both an intra- and an extraperitoneal portion. It is approximately 25 cm long with a transverse diameter of approximately 2.5 cm. Its mucosal folds are 2 mm thick. It has been classically divided into four differentiated parts. The first portion, also known as duodenal bulb, is intraperitoneal and extends from the gastric pylorus to the neck of the gallbladder. The second portion or descending duodenum includes two flexures (superior and inferior) and extends retroperitoneally from the neck of the gallbladder to the lower lumbar spine. The third portion, also retroperitoneal, crosses the midline

Duodenum

(b)

Figure 3.49 Stomach. (a, b) Coronal contrast-enhanced MPR CT image (portal venous phase). Note the presence of a subserosal uterine fibroid with coarse calcifications.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

from right to left. Finally, the fourth portion ascends slightly to reach the ligament of Treitz. The serous surface of the duodenum is closely related to the head of the pancreas and forms the pancreaticoduodenal groove, an anatomical space containing the pancreaticoduodenal arterial arcades, mesenteric veins, and lymphatic structures (Figure 3.50). The main pancreatic duct and the common bile duct usually end in the duodenal papilla to form the ampulla of Vater, which is surrounded by the sphincter of Oddi. The major duodenal papilla is located in the second portion of the duodenum in 75% of cases and in the third portion in the remaining cases [54]. The ligament of Treitz separates the fourth duodenal portion from the first jejunal loops. It lies posterior to the pancreas and anterior to the left renal vein, and is inserted into the duodenum, the esophageal hiatus of the diaphragm, and the celiac trunk. It is not defined by retroperitoneal fat and cannot be identified in CT images [55]. (a)

There are no strictly defined boundaries between the jejunum and ileum, but more than half of the small intestine corresponds to the ileum. An imaginary diagonal line drawn from the right hypochondrium to the left iliac fossa is a conventional landmark used to locate the jejunum in the upper left half and the ileum in the lower right half (Figure 3.51). Jejunum and Ileum

Large Intestine

The colon can be normally distinguished from the small intestine based on its different appearance, diameter, and location. The taeniae coli are three longitudinal bands of approximately 8 mm wide that run along the entire length of the colon and are located on its postero-medial, posterolateral, and anterior walls. The taeniae coli merge at the junction of the vermiform appendix with the cecum and at the recto-sigmoid junction. The haustra are prominent saccules formed in the spaces between the taeniae coli. The size of the haustras is varia-

(b) 1a: Duodenal bulb 2a: Descending duodenum 3a: Portion 4a: Portion Ligament of Tretz Pancreaticoduodenal groove

(d)

(c)

Figure 3.50 Duodenum. (a, c) Axial contrast-enhanced CT image (portal venous phase). (b, d) Coronal contrast-enhanced MPR CT image (portal venous phase).

97

98

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

approximately 9 cm in normal subjects. The transverse colon is about 6 cm in diameter and the descending colon and sigmoid colon are usually about 3 cm in diameter. The wall of the colon is very thin and should measure less than 3 mm thick. The colon usually contains gas, feces in varying degrees of dehydration, and minimal amounts of fluid in its lumen. The colon usually frames the abdomen and is surrounded by homogeneous fat. However, its position is highly variable, although this often lacks clinical significance. Accordingly, the position of the cecum is related to the length of the mesentery and the degree of retroperitonealization. In addition, the colon can be redundant, resulting in the transverse colon descended into the lower abdomen. Moreover, the colon can be located between the liver and the diaphragm (Chilaiditi sign) [56]. Vascularization of the Bowel

Figure 3.51 Small intestine. Coronal contrast-enhanced MPR CT image (portal venous phase). Note the presence of a subserosal uterine fibroid with coarse calcifications.

ble, depending on the contraction of the taeniae coli. The epiploic appendages (appendices epiploicae) are small bundles of fatty tissue that run along the taeniae coli and vary in size in relation to the nutritional status. They usually go unnoticed on CT images, except when affected by inflammatory pathology (epiploic appendagitis) (Figure 3.52). The transverse diameter of the colon is significantly variable. Generally, the largest diameter of the cecum is

Appendices epiploicae

Ascending colon

Transverse colon

The superior mesenteric artery arises from the abdominal aorta at the level of L1, approximately 1.5 cm inferior to the origin of the celiac trunk and superior to the origin of the  renal arteries. It supplies blood to the duodenum via the pancreticoduodenal arcade and irrigates the jejunum, ileum, ascending colon, and usually the transverse colon. The superior mesenteric artery is located to the left of the superior mesenteric vein in its origin. After crossing the third duodenal portion, it enters the mesentery and lies posterior to the mesenteric vein (Figure 3.53). The tributaries that most frequently arise from the superior mesenteric artery are the inferior pancreaticoduodenal artery, the right colic artery, the middle colic artery, the jejunal arteries, the ileocolic artery, and the ileal arteries (Figures 3.54 and 3.55).

Figure 3.52 Colon. Schematic illustration adapted from Horton et al. [56]. Haustra

Descending colon

Terminal ileum

Tenia coli Sigmoid colon

Cecum Appendix

(a)

(b)

(c)

(d)

Figure 3.53 Superior mesenteric artery. (a, c) Axial contrast-enhanced CT image (arterial phase). (b) Sagittal contrast-enhanced MIP CT image (arterial phase). (d) Coronal contrast-enhanced MIP CT image (arterial phase).

(a)

(b)

Figure 3.54 Main branches of the superior mesenteric artery. (a) Axial contrast-enhanced CT image (arterial phase). (b) Coronal contrast-enhanced MIP CT image (arterial phase).

100

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 3.55 Main branches of the superior mesenteric artery. Coronal contrast-enhanced MIP CT image (arterial phase).

The inferior pancreaticoduodenal artery joins the superior pancreaticoduodenal artery (branch of the celiac trunk) to form the pancreatic arcade. The right colic artery, which is absent in up to 80% of patients, supplies blood to the ascending colon, along with the middle colic and ileocolic arteries. The middle colic artery usually arises from the right side of the superior mesenteric artery and courses to the right and inferiorly to reach the right lower quadrant, where it anastomoses with the ileocolic artery. The jejunal arteries (usually four to six branches) arise from the left side of the superior mesenteric artery and supply the jejunum. (a)

The ileocolic artery, which arises from the right side of the superior mesenteric artery, marks the transition between the jejunal and the ileal artery (usually eight to 12 vessels). It supplies the terminal ileum, cecum, and inferior portion of the ascending colon. The inferior mesenteric artery arises from the aorta, approximately 7cm below the origin of the superior mesenteric artery, generally above the level of L3. It gives off several arteries from its left side, including the left colic, colosigmoid, recto-sigmoid, and superior rectal arteries (Figure 3.56). The left colic artery forms an anastomosis with the transverse colon artery, but is absent in up to 12% of individuals. The colosigmoid artery supplies blood to the descending and sigmoid colon. Distal to the origin of the recto-sigmoid artery, the inferior mesenteric artery gives off the superior rectal arteries. The superior mesenteric vein typically consists of a trunk formed by two branches (right and left). It receives blood from numerous veins (ileocolic, gastrocolic, right colic, and middle colic). The superior mesenteric vein joins the splenic vein to form the portal vein. The inferior mesenteric vein has three main tributaries, the left colic, and the sigmoid and superior rectal veins. The inferior mesenteric vein can drain into the splenic vein or into the superior mesenteric vein [57].

Spleen The spleen is an intraperitoneal organ located in the left hypochondrium. It has an inverted comma shape and can measure up to 12 cm long and 7 cm thick under normal conditions. Between 10% and 30% of the population

(b)

Figure 3.56 Inferior mesenteric artery. (a) Axial contrast-enhanced CT image (arterial phase). (b) Coronal contrast-enhanced MIP CT image (arterial phase).

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

presents adjacent nodules corresponding to accessory spleens (splenunculi). The spleen’s anatomical boundaries include the phrenicocolic ligament inferiorly, the diaphragm superiorly and postero-laterally, the stomach medially and antero-laterally, and the left kidney postero-medially. The visceral peritoneum covers the fibrous capsule of the spleen except for the splenic hilum, which is retroperitoneal. The splenic hilum contains the splenic artery and vein, lymphatic vessels, and nerves; it is also the site of insertion of the gastrosplenic and splenorenal ligaments. It is often in direct contact with the pancreatic tail [58].

Mesenteries

The mesenteries are peritoneal reflections which cover different abdominal structures, containing fat, vessels, lymphatic structures, and nerves (Figure 3.57). ●

●

Peritoneum and Mesentery The peritoneum is an extensive, thin serous membrane made up of a layer of simple epithelium (mesothelium) and loose connective tissue. It is the largest and most complex serous membrane in the body and covers both the abdominal cavity and the abdominopelvic organs. The parietal peritoneum covers the anterior, lateral, and posterior abdominal and pelvic walls as well as the anterior surface of the retroperitoneal organs, the inferior surface of the diaphragm, and the superior surface of the pelvis. The visceral peritoneum covers many of the abdominopelvic organs to a variable degree. A number of peritoneal reflections result in the formation of different ligaments, mesentery, and omenta. These reflections contain retroperitoneal areolar tissue, blood, vessels, lymphatic tissue, and nerves. The subperitoneal space represents a communication pathway between the peritoneum and the retroperitoneum. There are intraperitoneal, peritonealized, secondary peritonealized, and partially peritonealized organs [59]. The peritoneal cavity forms a closed pouch in men. In women, it remains open at the orifices of the Fallopian tubes, allowing communication between the intraperitoneal and extraperitoneal pelvis. Its function is twofold. On the one hand, it provides a lubricated surface (by secreting a small amount of sterile fluid that also has defensive properties against local bacteria) that enables the movement of organs without friction. On the other hand, it serves as a transportation system of fluids. Clearance of the peritoneal fluid is performed through continuous circulation in cephalad direction, which is thought to be caused by changes in the abdominal pressure during breathing and by bowel peristalsis [60]. Several peritoneal structures can be distinguished. To recognize them on imaging, it is essential to remember their typical location and their anatomical relations with other organs and vascular structures.

●

Mesentery: A double fan-shaped peritoneal layer that attaches jejunal and ileal loops to the posterior wall of the abdomen. It extends from the ligament of Treitz to the ileocecal valve. It is currently considered an organ itself. Transverse mesocolon: This surrounds the transverse colon and fixes it to the posterior abdominal wall, creating a communication pathway with the retroperitoneum. It divides the peritoneal cavity into two compartments, namely supra and infra-mesocolic. Mesosigma: This fixes the sigmoid colon to the posterior wall of the pelvis.

Ligaments

The ligaments are made up of two peritoneal layers and contribute to the fixation and structural support of the abdominal organs. Under normal conditions, the peritoneal ligaments are named after the anatomical structures which they connect. ●

● ●

●

●

Splenorenal ligament: This attaches the spleen to the anterior pararenal space. The splenic vessels run through this ligament (as mentioned above, the splenic hilum is a retroperitoneal space) (Figure 3.58). Gastrosplenic ligament. Phrenicocolic ligament: This fixes the descending colon to the posterior part of the diaphragm. Suspensory ligaments of the liver: Coronary ligaments, triangular ligaments. Falciform ligament: This fixes the liver to the anterior wall of the abdomen and contains the obliterated umbilical vein,

Omentum or Epiploon ●

●

Lesser omentum: This results from the combination of the hepatogastric and hepatoduodenal ligaments and connects the lesser curvature of the stomach and proximal duodenum with the liver. Additionally, it covers the lesser sac anteriorly. Greater omentum: This is mainly composed of fat and extends anteriorly into the stomach, transverse colon, and small bowel. It is located immediately posterior to the anterior abdominal wall (Figure 3.59).

Peritoneal Compartments

The peritoneal compartments result from the arrangement of the peritoneal structures detailed above, dividing the abdominal cavity into two separate compartments.

101

(a)

(b)

Mesenteries Mesentery Transverse mesocolon Sigmoid mesocolon

(c)

(d)

(e)

Figure 3.57 Mesenteries in patient with ascites due to peritoneal carcinomatosis secondary to ovarian tumor. (a–e) Coronal contrast-enhanced MPR CT image (portal venous phase).

(a)

(b)

(c)

Ligaments Splenorenal ligament Phrenicocolic ligament Falciform ligament Gastrosplenic ligament

Figure 3.58 Peritoneal ligaments in patient with ascites due to peritoneal carcinomatosis secondary to ovarian tumor. (a, b) Coronal contrast-enhanced MPR CT image (portal venous phase). (c) Axial contrast-enhanced CT image (portal venous phase).

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

(a)

(b) Falciform Hepatogastric Liver

Spleen

Gastrosplenic

Stomach Gastrocolic Hepatoduodenal Phrencocollic Ascending mesocolon

Splenic flexure Descending mesocolon Cecum

(c)

Rectum

Sigmoid mesocolon

(d)

Figure 3.59 Peritoneal compartments and omenta in patient with carcinomatous ascites due to ovarian tumor. (a) Diagram, adapted from Wasnik et al. [60]. (b, d) Coronal contrast-enhanced MPR CT image (portal venous phase). (c) Axial contrast-enhanced CT image (portal venous phase).

Supramesocolic Compartment The supramesocolic compartment is the cavity lying superior to the transverse mesocolon. The falcifom ligament divides this space into the left and right compartments (Figure 3.60).

●

Right Supramesocolic Space

The right supramesocolic space is divided into three compartments, namely right subphrenic, right perihepatic, and right subhepatic: ●

●

●

Right subphrenic space: This lies between the right lobe of the liver and the diaphragm. Right perihepatic space: This surrounds the anterior and lateral portions of the diaphragm Right subhepatic space: This is sub-divided into anterior and posterior spaces: The right anterior subhepatic space is in close contact with the PV and communicates with the omental bursa by the omental foramen (of Winslow).

Right posterior subhepatic space is also known as Morison’s pouch or hepatorenal space. The omental bursa (lesser sac) is an enclosed cavity bounded by the hepatogastric ligament, the posterior wall of the stomach, the duodenum, and the caudate lobe anteriorly, by the pancreas and the posterior parietal peritoneum posteriorly, by the diaphragm superiorly, and by the transverse mesocolon inferiorly. The foramen of Winslow communicates the lesser sac with the rest of the abdominal cavity or greater sac.

Left Supramesocolic Space

The left supramesocolic space is divided into anterior and posterior perihepatic spaces, and anterior and posterior subphrenic spaces. ●

Left anterior perihepatic space: This lies anterior to the liver and medial to the falcifom ligament.

103

104

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(c)

(d)

Right subphrenic space Left subphrenic space Falciform ligament Right subhepatic space Right subhepatic anterior Right subhepatic posterior Lesser sac

(b)

Figure 3.60 Supramesocolic compartments in patient with carcinomatous ascites due to ovarian tumor. (a, b) Axial contrastenhanced CT image (portal venous phase). (c, d) Coronal contrast-enhanced MPR CT image (portal venous phase). ●

●

●

Left posterior perihepatic space (gastrohepatic recess): This surrounds the lateral segment of the liver and contacts the stomach. Anterior subphrenic space: This is continuous with the left anterior perihepatic space, separating the gastric fundus from the diaphragm. Left posterior subhepatic space: This surrounds the spleen and is bounded by the phrenicocolic ligament, which separates this space from the left paracolic gutter.

The inframesocolic compartment is divided into right and left spaces. The left inframesocolic space is larger and contains the mesentery. The paracolic gutters are the most lateral parts of these spaces. The right gutter communicates with the right perihepatic and subhepatic spaces superiorly, and the left gutter is bounded by the phrenicocolic ligament superiorly [59–61] (Figure 3.59).

Inframesocolic Compartment

Urinary System Kidneys

The kidneys are retroperitoneal organs with a disposition parallel to the psoas muscles. They measure approximately 10–14 cm long and 3–5 cm wide. The right kidney is around 2 cm lower than the left one and is slightly smaller (approximately 0.5 cm). It is convex laterally and concave medially.

In the craniocaudal orientation, its upper pole lies medially and its lower pole posteriorly. The kidneys are surrounded by a fibrous capsule to which they are firmly attached. This capsule is surrounded by anterior perirenal fat which, in turn, is defined by the Gerota fascia anteriorly and by the Zuckerkandl’s fascia posteriorly. The renal hilum is located medially. It is composed of adipose tissue and a fibrous fascia. It contains the renal vessels, the collecting system, and lymphatic structures. On imaging, the hilum can be identified entering the kidney within the renal sinus (rich in fat). The renal parenchyma is formed by the renal cortex and the medullary pyramids. The renal cortex receives 90% of the renal blood flow and is formed by proximal and distal glomeruli and tubules. The renal medulla contains between eight and 18 pyramids that can be identified on imaging by their typical inverted-cone shape, with the vertex representing the papilla which drains into a minor calyx. The minor calyces are cup-shaped structures that converge into major calyces which drain into infundibula. These, in turn, drain into the renal pelvis. The renal cortex arches between the pyramids forming the columns of Bertin [62]. Anatomical Variants There are frequent anatomical variants of renal development that may lead to a

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

pseudotumor appearance of the kidneys. The most common variants include the following: ●

●

●

●

Column of Bertin hypertrophy: This can simulate a solid focal lesion and is usually located between the upper and middle thirds of the kidney. Dromedary hump: This is a focal interpolar protuberance on the lateral edge of the left kidney, formed by the adaptation of the renal contour to the spleen, with which it is in close contact. Persistent fetal lobulation: During the neonatal period, the fetal kidneys are divided into lobes by grooves, which normally disappear by the end of gestation. However, they can persist into adulthood. Junctional parenchymal defects: These are related to the persistence of a prominent fetal lobulation which includes perirenal fat and leads to fold of the anterior surface of the upper third of the kidneys toward the hilum [63].

The renal arteries arise from the aorta above the L2 level, immediately cranial to the superior mesenteric artery. They reach the renal hilum anterior to the renal pelvis, where they divide into segmental arteries. The first branch is usually posterior and supplies the posterior and inferior renal segments. The anterior branch divides into four arteries: apical, superior, middle, and inferior. The segmental arteries are divided into lobar, interlobar, arcuate, and interlobular arteries.

Renal Arteries

The renal cortex drains into the arcuate and interlobar veins. The lobar veins unite to form the main renal vein. The renal veins usually lie anterior to the renal artery. The left renal vein is three times longer than the right renal vein because it drains into the IVC. Unlike the right renal vein, the left renal vein is joined by the left adrenal vein in its superior portion, the left gonadal vein inferiorly, and a lumbar vein posteriorly before draining into the IVC [64].

Renal Veins

Ureter

The ureter is an extraperitoneal structure surrounded by fat, with an approximate length of 22–30 cm. It is a dynamic organ and not a simple duct through which urine flows. It conveys urine from the renal papillae to the ureteric orifice in the bladder, regardless of gravity. It is divided into three portions: the proximal (superior) ureter, which extends from the pelviureteric junction to the site where the ureter crosses the sacroiliac joint, the middle ureter, which runs through the bony pelvis and the iliac vessels, and the distal (pelvic or inferior) ureter, which extends from the iliac vessels to the bladder. The distal segment is subdivided into

juxta-vesical, intramural, and submucosal portions. Its blood supply comes from the ureteral artery, which runs along the ureter. The superior third of the ureteral artery is irrigated by the aorta and the renal artery, while the middle and inferior ureter receive their arterial supply from the iliac, lumbar, and vesical arteries [65]. Urinary Bladder

When the urinary bladder is distended, it lies almost entirely in the antero-inferior pelvis and enters the major pelvis when it is full. It consists of a muscular bag with the function of collecting, storing, and eliminating urine. It is made up three layers, the internal one corresponding to the urothelium (site of origin of most bladder cancers), muscularis propria (formed by the detrusor muscle of the bladder), and serosa [66]. The urinary bladder has a round or oval shape, with its superior segment being wider, and is divided into fundus, body, apex, and neck. Anteriorly, it contacts the pubic symphysis. In men, it is bounded by the retrovesical space and rectum posteriorly, the prostate, seminal vesicles, obturators, and levator ani muscles inferiorly, and the peritoneum anteriorly. In women, its posterior aspect contacts the vesicouterine space, the uterus, the cervix, and the vagina. The ureteral orifices and the internal orifice of the urethra define an area known as the trigone. The dome of the bladder is covered with peritoneum to a varying degree based on the length of the peritoneal reflections. The rest of the surfaces of the bladder are extraperitoneal. During the neonatal period, the apex of the bladder is connected to the umbilicus through the urachus. This duct can persist in adults as a normal anatomical variant. The neck of the bladder is located inferiorly and contains the internal ureteral orifice, in the lower angle of the trigone. The ureters have a short and oblique intramuscular course before opening at the postero-lateral angles of the trigone [67]. The vascular supply of the bladder is via the external iliac arteries, which give off the superior and inferior vesical arteries. It also receives vascular supply from branches of the obturator, uterine, and vaginal arteries. The bladder’s venous drainage is via a complex plexus which ends in the internal iliac vein [68].

Adrenal Glands The adrenal glands are Y-shaped organs located anterosuperiorly and slightly medial to the superior pole of each kidney in the so-called adrenal space, located in the Gerota fascia and surrounded by fat. The adrenal gland is a bipartite structure comprising the cortex and medulla, each with different embryogenesis, structure, and function. Its

105

106

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

approximate size is 10–12 mm for the body and 6 mm for the limbs [69]. Because they are responsible for hormone production and play a key role in systemic functions, the adrenal glands are richly vascularized. Their arterial supply is via three adrenal arteries (superior, middle, and inferior), which are branches of the superior phrenic, aorta, and renal artery, respectively. Their venous drainage is via a single adrenal vein, the left one draining into the left renal or the inferior phrenic vein and the right one directly into the IVC [70].

Retroperitoneum The retroperitoneal space is an anatomical compartment bounded by the posterior parietal peritoneum and the transversalis fascia. It is divided into five compartments, two lateral, two posterior, and one central (Figures  3.60 and 3.61). Lateral Compartments

The lateral compartments are subdivided into three spaces bounded by fasciae: anterior pararenal, perirenal, and posterior pararenal (Figure 3.62). The anterior pararenal space is bounded by the posterior parietal peritoneum anteriorly, by the anterior renal fascia posteriorly, and by the lateroconal fascia laterally. It contains the posterior face of portions of the ascending and descending colon, the duodenum, and the pancreas (Figure 3.63).

Anterior Pararenal Space

Peritoneal space Retroperitoneal space

The perirenal spaces contain the kidneys and ureters, the adrenal glands, and their corresponding vascular, lymphatic, and nervous structures (Figure 3.64). The perirenal space has an inverted-cone shape, with the tip directed toward the pelvis and the base placed onto the diaphragm. The anterior edge of this space is formed by the anterior (Gerota) and posterior (Zuckerkandl) renal fasciae. The posterior renal fascia merges with the fascia of the quadratus lumborum muscle. Along with the connective tissue that surrounds the great vessels, there is a theoretical duct that connects both perirenal spaces at the level of L3–L5, known as the Kneeland channel, allowing free diffusion between both anatomical locations. The superior part of the anterior renal fascia contacts the bare area (nonperitoneal region of the liver) so that the posterior peritoneum and the anterior renal fascia merge, establishing direct contact between the liver and the kidney. This relation explains the involvement of the perirenal space by different processes affecting the liver. At the inferior end of the perirenal space, the anterior and posterior renal fasciae merge around the ureter. Perirenal Spaces

Posterior Pararenal Space The posterior pararenal space only contains fat and is bounded by the diaphragm superiorly, by the posterior renal fascia anteriorly, and by the transversalis fascia posteriorly (Figure 3.65). The lower part of the anterior and posterior pararenal spaces unite to form the infrarenal retroperitoneal space, which communicates with the prevesical space and the extraperitoneal spaces of the pelvis.

Figure 3.61 Peritoneal and retroperitoneal spaces. Axial contrast-enhanced CT image (portal venous phase).

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

Figure 3.62 Retroperitoneal compartments. Axial contrast-enhanced CT image (portal venous phase).

Lateral space Median vascular space Posterior iliospoas space

Figure 3.63 Anterior pararenal space. Axial contrast-enhanced CT image (portal venous phase).

Posterior parietal peritoneum Laterocanal fascia Anterior renal fascia Ascending colon Descending colon Duodenum Pancreas

Central compartment

Posterior Compartments

The central compartment is a vascular compartment that extends from T12 to L4–L5, located between the two perirenal spaces and posterior to the anterior pararenal space. It contains the abdominal aorta and its branches, the IVC and its afferent vessels, lymphatic chains, and the abdominal sympathetic trunk (Figure 3.62).

The posterior compartments contain the psoas and iliac muscles and are considered to be retroperitoneal despite being located posterior to the transversal fascia because of their frequent involvement in the context of retroperitoneal processes (Figure 3.62).

107

108

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 3.64 Perirenal space. Axial contrastenhanced CT image (portal venous phase).

Anterior renal fascia – Gerota Posterior renla facia – Zuckerkandl

Figure 3.65 Posterior pararenal space. Axial contrast-enhanced CT image (portal venous phase).

Lateroconal fascia Posterior parietal peritoneum Transversalis fascia Anterior renal fascia Posterior renal fascia

Interfascial Plane

The interfascial plane is one of the potential retroperitoneal spaces created between the layers of the renal fasciae. ●

●

Anterior or retromesenteric plane: Thist lies between the anterior pararenal and perirenal spaces and crosses the midline. Retrorenal space: This lies between the perirenal and posterior pararenal spaces. It does not cross the

●

●

midline because it is interrupted by the central compartment. Combined intersfascial space: This results from the fusion of the retromesenteric and retrorenal spaces and extends anterior to the psoas muscle and the retroperitoneum of the pelvis. Lateroconal plane: This lies between the lateroconal fascia and is communicated with the retromesenteric and retrorenal spaces [59, 71, 72].

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

­Anatomy of the Pelvis

Uterus and Cervix

The pelvis is bowl-shaped and presents two openings: one superior, into the abdominal cavity, and one inferior, bounded by the pelvic floor and perineum. Anatomically, the pelvis is usually divided into true and false pelvis by an oblique line extending from the sacral promontory along the anterior face of S1 to the pubic symphysis. The greater or false pelvis is located above this plane, containing the ascending, descending, and sigmoid colon as well as ileal loops and the bifurcation of iliac vessels. The lesser or true pelvis, located below the pelvic brim, contains the reproductive organs, urinary bladder, pelvic ureters, small bowel loops, and rectum [73].

Anatomy of the Female Pelvis The pelvic cavity is separated into an anterior and posterior space by the broad ligament, which is transversely oriented and contains the uterus in its center. This ligament is the main anatomical landmark to understand the female pelvic anatomy. The round ligaments, attached to the uterus, run antero-laterally through the broad ligament to reach the pelvic wall, where they change their course medially, around the inferior epigastric vessels, to exit the pelvis via the internal inguinal ring and inguinal canal [73].

Figure 3.66 Uterus. (a) Axial oblique contrastenhanced MPR CT image (portal venous phase). (b) Sagittal contrast-enhanced MPR CT image (portal venous phase).

The uterus is a smooth, hollow organ which exhibits a pyriform shape with its lower vertex posterior to the urinary bladder and anterior to the rectum. Its position is variable and changes when the bladder is full. It is divided into three segments: fundus, body, and neck (Figure 3.66). The fundus is located in the superior segment of the uterus, with the uterine tubes arising at this level. It is made up of three layers: the endometrium in contact with the cavity, the junctional zone, and the myometrium. The endometrium is the most central part of the uterus, with a variable thickness along the menstrual cycle, being thicker during the secretory phase than in the follicular phase or menstruation. Its thickness also varies depending on the reproductive age; in postmenopausal women, it should be less than 5 mm. The junctional zone is the innermost layer of the myometrium and should be less than 12 mm thick, whereas the myometrium lies external [74] (Figure 3.66). There are two peritoneal folds in contact with the uterus: the peritoneal lining of the uterus reflects anteriorly to cover the posterior wall of the bladder at the level of the cervix, forming the vesicouterine recess, which presents two lateral recesses called paravesical fossae. Posteriorly, the peritoneum is reflected to cover the superior rectum, forming the rectouterine pouch (also known as the cul-de-sac or pouch of Douglas), which is continuous with the para-rectal fossae. This is the most caudal portion of the peritoneal cavity and often presents a mild amount of fluid during ovulation or loops of small intestine [73] (Figure 3.67).

(a)

(b) Uterus Fundus Body Cervix Endometrium Miometrium External cervical orifice Internal cervical orifice

109

110

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(c)

Uterus – peritoneum Vesicouterine pouch Rectovaginal pouch Paravesical fossae Pararectal fossae

(b)

Figure 3.67 Uterus and peritoneum. (a) Axial noncontrast CT image. (b) Axial contrast-enhanced CT image (portal venous phase). (c) Sagittal contrast-enhanced MPR CT image (portal venous phase).

The cervix is separated from the uterine body by the internal cervical orifice [68]. It is cylindrical in shape and protrudes into the vagina, with which it communicates via the external cervical orifice. The anatomical relation between the body and the cervix changes with reproductive age. The body:cervix ratio is 1 : 1 in premenopausal women, and 2 : 1 in reproductive age women. In postmenopausal women, the uterus atrophies and decreases in size to 4–6 cm in length [73].

pelvic infundibulum, which fixes the ovary to the lateral wall of the pelvis. The broad ligament and the mesovarium cannot be identified in CT images unless surrounding ascites is present [74] (Figures 3.68 and 3.69). In the premenopausal woman, the normal ovary presents follicles, which usually lie peripherally and measure

Ovary: Fallopian Tube

The ovaries are ellipsoid in shape and measure between 2.5 and 5 cm in length. They change in location after pregnancy and with the size of the uterus and degree of bladder distension. In nulliparous women, the ovaries lie within the ovarian fossa (of Waldeyer) on the lateral wall of the pelvis, bounded by the obliterated umbilical artery anteriorly, the external iliac vein superiorly, and the ureter and internal iliac artery posteriorly [73, 75]. The ovary is suspended in the pelvis by the mesovarium, which fixes the ovary to the posterior surface of the broad ligament, by the ovarian ligament, which attaches the ovary to the uterus, and by the suspensory ligament or

Ovary Ovary Suspensory ligament of ovary Umbilical artery External iliacvein

Figure 3.68 Ovary. Axial contrast-enhanced CT image (portal venous phase).

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

(a)

(b) Isthmus

Infundibulum

Ampulla

Ovary Fallopian tube

Intramural portion

Uterus Round ligament Ovary

Uterus Broad ligament

Fimbriae

Peritoneum

(c) (d)

Figure 3.69 Ovary and fallopian tubes. (a) Schematic illustration, adapted from Revzin et al. [76]. (b) Axial contrast-enhanced CT image (portal venous phase). (c) Coronal contrast-enhanced MPR CT image (portal venous phase). (d) Coronal contrast-enhanced MPR CT image (portal venous phase).

less than 3 cm. The corpus luteum cyst is an involutional dominant functional cyst that can be recognized because its wall tends to be more irregular [68]. The fallopian tubes arise from the posterior part of the uterus, at the site of junction between the fundus and body. They are approximately 10 cm long and lie in the superior aspect of the broad ligament. Four portions can be distinguished: intramural (within the myometrium), isthmic (medial third), ampullar (middle distal), and infundibular (distal zone). The distal part of the infundibulum is fimbriate and opens into the peritoneal cavity [73, 76] (Figures 3.68 and 3.69).

lower third is located inferior to the base of the bladder, so that the urethra lies anteriorly, the middle third corresponds to the base of the bladder, and the upper third comprises the lateral vaginal fundi (Figure 3.70).

Vagina

Vagina

The vagina is a cylindrical fibromuscular structure measuring approximately 7–9 cm in length (its anterior wall is approximately 6 cm and its posterior wall 9 cm). The vagina extends from the vulva to the cervix superiorly, lying between the urinary bladder and rectum. It forms an angle greater than 90° relative to the uterus. The urogenital diaphragm connects the vagina to the levator ani muscle. It is lined with a stratified squamous epithelium sensitive to strogens. Morphologically, it is divided into thirds: the

Figure 3.70 Vagina. Sagittal contrast-enhanced MPR CT image (portal venous phase).

111

112

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Urethra

The female urethra is a thin-walled tubular muscular canal, measuring approximately 40 mm in length. It runs antero-inferiorly from the internal urethral meatus located in the trigone of the bladder and terminates in an external urethral meatus anterior to the vagina. It has three layers: an internal mucosa, vascular submucosa, and an external muscular layer. The urethropelvic ligaments provide structural support to the urethra [68]. Ligaments of the Pelvis

The broad ligament consists of two layers of peritoneum, which surround the uterus and extend laterally to the external wall of the pelvis. It includes the fallopian tube, ovarian ligament, uterine and ovarian vessels, nerves, lymphatic vessels, and a portion of the ureter, along with loose connective tissue and fat, known as the parametrium (Figure 3.69). On its superior free edge, the broad ligament surrounds the fallopian tube, while the ureter lies in its inferior aspect. The cardinal or transverse cervical ligament arises from the cervix and superior portion of the vagina, and merges with the fascia of the internal obturator muscle. (a)

The ureterosacral ligament extends posteriorly from the lateral face of the vagina to the anterior body of the sacrum at the second or third sacral vertebrae. Finally, it merges with the cardinal ligaments medially. The round ligament is a fibromuscular band that arises from the antero-lateral side of the fundus, runs laterally through the broad ligament along the lateral wall of the pelvis to the internal inguinal ring, and ends in the labia majora [77] (Figure 3.71). The ovary has two supporting ligaments, namely the ovarian ligament and the suspensory ligament or pelvic infundibulum [73]. The suspensory ligament of the ovary is a peritoneal fold containing the ovarian artery and vein. It arises from the ovary and extends antero-laterally over the external iliac vessels to the lateral wall of the pelvis, where it merges with the connective tissue covering the psoas muscle. Vascularization of the Pelvis

The common iliac arteries give off the internal and external iliac arteries approximately at the level of the lower sacroiliac joint. (b)

Ligaments of the pelvis

(c)

Round ligament Suspensory ligament of ovary Uterosacral ligament Cardinal ligament (d)

Figure 3.71 Ligaments of the pelvis. (a, b, d) Axial noncontrast CT images. (c) Axial contrast-enhanced CT image.

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

The paired uterine arteries are branches of the anterior trunk of the internal iliac artery. The uterine artery runs medially above the cardinal ligament through the base of the broad ligament to provide the primary blood supply to the uterus. The artery crosses anterior to the pelvic ureter to reach the cervix and divides into a large uterine branch and a smaller cervico-vaginal branch. Both branches are tortuous and form extensive vascular networks lateral to the uterus and vagina. In the superior segment of the uterus, the uterine artery trifurcates, giving off branches for the fallopian tubes, uterine fundus, and ovary. In addition, the ovaries receive direct blood flow from the ovarian arteries, which arise from the aorta immediately inferior to the renal arteries. The blood supply of the vagina comes from a network of vessels formed by the anastomosis between the vaginal and uterine branches of the internal iliac artery. The lower two thirds of the vagina are supplied by the middle rectal artery and the internal pudendal arteries [68]. Venous Drainage The venous drainage of the uterus, cervix, upper vagina, and ovaries is via an extensive venous plexus within the parametrium. This plexus eventually

(a)

drains into veins that run parallel to the arterial blood supply. Venous drainage is via the gonadal vein, with the right one draining into the IVC and the left one into the left renal vein [68, 73]. Rectum

The rectum is an extraperitoneal organ that extends from the rectosigmoid junction to the anorectal ring. Its proximal limit is approximately 15 cm from the anal verge and can be divided into the lower (5 cm from the anal verge), middle (5–10 cm), and upper (10–15 cm) rectum. It is surrounded by fat and by the mesorectal fascia (MRF), which represents the visceral fascia of the extraperitoneal portion of the rectum. The MRF has a funnel shape opened in its anterior portion, in such a way that the posterior insertion of the peritoneum lies higher than its anterior insertion, which forms the anterior peritoneal reflection (Figure 3.72). The presacral fascia lies posteriorly, in close contact with the sacrum. The MRF and presacral fascia merge at the level of S4 to form the sacrorectal ligament. Anteriorly, the MRF merges with the urogenital septum,

(b) Rectum Rectum Anal verge Presacral fascia Mesorectal fascia Mesorectal fat Levator ani muscle

(c)

Figure 3.72 Rectum. (a) Coronal noncontrast MPR CT image (portal venous phase). (b) Axial contrast-enhanced CT image (portal venous phase). (c) Sagittal contrast-enhanced MPR CT image (portal venous phase).

113

114

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

forming the rectovaginal septum in women, while it merges with the Denonvilliers’ fascia in men  [78] (Figure 3.74). The anal canal is the terminal portion of the large intestine. It is bounded by the rectum at the anorectal junction, which can be recognized because the puborectalis muscle forms a U-shaped sling posteriorly. The inferior boundary is the anal verge. The approximate histological length of the anal canal is 4 cm (Figure 3.72). The wall of the anal canal has a cylindrical shape. The internal muscular layer is made up of smooth muscle and represents the internal anal sphincter. It is the distal prolongation of the circular muscular layer of the rectum. The external layer is a complex external anal sphincter of striated, skeletal muscles, which keeps the anal lumen closed. Between both layers there is fatty tissue called intersphinteric fat [79]. The external anal sphincter has three separate fiber bundles: deep, superficial, and subcutaneous. The deep fibers merge proximally with the puborectalis muscle, similar to a sling (Figure 3.73). Both the external and internal sphincters are subsequently attached to the anococcygeal ligament and to the fibrous perineal body anteriorly. The external sphincter has a series of reinforcements and support structures, which fix the anal canal to pelvic structures, especially the puborectalis muscle (into its superior margin) and the levator ani muscle (Figure 3.73). (a)

The levator ani muscle is the main muscle of the pelvic diaphragm and has three components: ileococcygeal, pubococcygeal, and puborectalis. The vascularization of the rectum is via three rectal arteries. The superior rectal artery is the terminal branch of the inferior mesenteric artery. It supplies the rectosigmoid junction and the upper rectum. It anastomoses with the middle rectal artery (branch of the internal iliac artery) and with the inferior rectal artery (branch of the internal pudendal artery). The middle rectal artery supplies the lower rectum, seminal vesicles, and prostate. In women it can be replaced by the uterine artery. The inferior rectal artery is responsible for the arterial supply of the anal canal.

Anatomy of the Male Pelvis Prostate and Seminal Vesicles

The prostate is an exocrine gland found only in men. It has an inverted-cone shape with the base placed superiorly. It lies in the subperitoneal space, posterior to the pubic symphysis and anterior to the rectum. The base is attached to the bladder neck by the prostatic urethra, an anatomical landmark in the prostate, and the apex rests on the urogenital diaphragm contacting the medial surface of the levator ani muscles. The prostate is made up of a glandular component (70%) and fibromuscular stroma (30%). Its primary function is to secrete an alkaline fluid that makes the semen more fluid. (b)

- Anal canal Anorectal juntion Anal verge External analsphicter Deep part Superficial part Subcutaneus part Internalanal sphincter Interesphinteric space Puborectalis muscle Levator ani muscle

(c)

(d)

Figure 3.73 Anal canal. (a, b) Axial noncontrast CT images. (c) Coronal noncontrast MPR CT image. (d) Coronal contrast-enhanced MPR CT image (portal venous phase).

Cross-­sectional Correlate for Integrative Imaging (Anatomical Radiology)

The ejaculatory ducts that come from the seminal vesicles open into the prostatic urethra in the so-called prostatic utricle, an area of glandular tissue which can be recognized because it protrudes from the posterior urethral wall [80] (Figure 3.74). The seminal vesicles receive sperm from the testicles via the vas deferens and their size changes with ejaculation. There are several classifications of the prostatic anatomy, including traditional divisions into lobes and the more upto-date zonal classification, which distinguishes four zones, namely transitional, central, peripheral, and anterior fibromuscular stroma (Figure 3.75). The volumetric contribution of the transitional, central, and peripheral zones changes with age. Prostatic hyperplasia is characterized by a considerable increase in the transitional zone [68, 80].

Zonal anatomy of the prostate Central zone Transitional zone Anterior fibromuscular stroma Peripheral zone Seminal vesicle

Peritoneal and Extraperitoneal Pelvic Spaces The inferior terminal portion of both paracolic gutters lies at the inlet of the major pelvis, resulting from the anterior peritonealization of the ascending and left colon and its reflection in the abdominal wall. Both gutters communicate with the peritoneal pelvic spaces. (a)

Figure 3.75 Sagittal contrast-enhanced MPR CT image (portal venous phase).

(c) Prostate Apex Medium Base Seminal vesicle Denonvilliers fascia Bladder Rectum Prostate

(b)

Figure 3.74 (a, b) Axial contrast-enhanced CT image (portal venous phase). (c) Sagittal contrast-enhanced MPR CT image (portal venous phase).

115

116

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b) Retzius space

The peritoneum covers the superior surface of the extraperitoneal pelvic organs, forming the peritoneal reflections. Three main reflections can be found in women, namely the rectouterine pouch (of Douglas), the vesicouterine pouch, and the anterior vesical space lying between the anterior parietal peritoneum and the anterior vesical wall. In men, the main peritoneal reflection is the rectovesical pouch. Anterior to the pouch of Douglas, the umbilical folds, which contain the obliterated umbilical artery, divide the pelvic spaces into lateral and medial compartments. On each side, the inferior epigastric artery divides the lateral compartments of the pelvis into lateral and medial inguinal fossae [59]. The prevesical (or Retzius) space is a large extraperitoneal compartment which mainly contains fat. Its anatomical boundaries are the transversalis fascia anteriorly, the umbilicovesical fascia posteriorly, the umbilicus superiorly, and the pubovesical ligament inferiorly. The transversalis fascia is a dense layer of areolar tissue which runs below the diaphragm and covers the posterior aspect of the muscles of the anterior abdominal wall. The umbilicovesical fascia extends in a triangular shape from the umbilicus to the urinary bladder, surrounding the urachus, the medial umbilical ligament (containing the obliterated umbilical arteries) and the urinary bladder, thus separating the prevesical from the perivesical space. Accordingly, the perivesical space contains the urachus, the medial umbilical ligament, and the urinary bladder [81–83] (Figure 3.76).

Spine Anatomy General Anatomy of the Spine The normal spine accounts for 24 pre-sacral vertebrae (seven cervical, 12 thoracic, five lumbar). However, transitional abnormalities are frequent, namely sacralization (L5 is fused to sacrum, described in 200/110 mmHg

1) Patients with ongoing wheezing or a history of significant reactive airway disease testing 2) Second- or third-degree AV block without a functioning pacemaker 3) Sick sinus syndrome or symptomatic bradycardia 4) Severe aortic stenosis 5) Systolic BP < 90 mmHg 6) Uncontrolled hypertension (BP > 200/100 mmHg) 7) Recent (85% of the age-predicted peak heart rate 2) Severe hypotension (systolic BP < 80 mmHg) 3) Severe hypertension (BP > 220/115 mmHg) 4) Arrhythmia 5) Severe chest pain associated with ST depression 2 mm 6) Signs of poor perfusion 7) Patient’s request to stop

1) Severe hypotension (systolic BP < 80 mmHg) 2) Development of persistent high-degree AV block 3) Ventricular tachycardia 4) Severe chest pain associated with ST depression 2 mm 5) Signs of poor perfusion 6) Patient’s request to stop

1) 2) 3) 4) 5) 6) 7) 8) 9) 10) 11) 12)

Moderate to severe angina Marked shortness of breath Signs of poor perfusion Fatigue, dizziness, or near-syncope Patient’s request to terminate the test Excessive ST-segment depression (>2mm from baseline) ST elevation (>1 mm) in leads without Q-waves Sustained supraventricular tachycardia New LBBB Drop in systolic BP of greater than 10 mmHg from baseline Hypertensive response (BP > 230/115 mmHg) Heart rate within 20 beats of therapy indications in patients with implantable cardioverter defibrillators

AV, atrioventricular; BP, blood pressure; LBBB, left bundle branch block; MI, myocardial infarction.

99m

Tc-based radiotracers allow for imaging with a lower radiation dose [21–23]. In addition, 99mTc-based radiotracers are produced from a generator while thallium is cyclotron produced, therefore technetium-based radiotracers are preferred over 201Tl [24]. The most widely used radiotracers for PET MPI in a clinical setting are rubidium-82 and ammonia-13. While rubidium-82 is generator produced,

ammonia-13 is produced from a cyclotron. Both these radiotracers are used for MPI and quantification of myocardial blood flow  [25]. Furthermore, 18-fluorodeoxyglucose (18FDG) is used for metabolic imaging in the evaluation of myocardial viability and cases of inflammation and infections [26, 27]. The characteristics of SPECT and PET radiotracers are summarized in Table 8.3.

259

260

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

One-day rest-stress Tc-99 m 99 m

99 m Tc 8-12 mCi

Tc 24-36 mCi*

30-60 mins

15-60 mins

30-60 mins Rest

Stress Test

Gated Stress

*Dose of second to first tracer injection = 3:1 Time between injection ≥ 2 hours

Two-day stress Tc-99 m Day 2*

Day 1 99 m

99 m

Tc 30 mCi

Tc 30 mCi 15-60 mins Stress Test

30-60 mins Gated Rest

Gated Stress

*For stress only imaging, rest imaging is not necessary if stress images are normal.

Figure 8.1 Table 8.3

Common protocols in SPCET imaging. Characteristics of the commonly used cardiac SPECT and PET tracers.

Cardiac SPECT tracers 201

99m

99m

Photon energy (keV)

67

140

140

Physical half-life (hours)

73

6

6

Uptake mechanism

Active transport

Passive diffusion

Passive diffusion

Preparation

Cyclotron

Generator

Generator

Heart uptake (%)

3

1–1.4

1.2

Elimination

Renal

Hepatic

Hepatic

Redistribution

Significant

Minimal

Minimal

Stress modality

Exercise or pharmacologica

Exercise or pharmacological

Exercise or pharmacological

Tl

Tc-sestamibi

Tc-tetrofosmin

Cardiac PET tracers Ammonia-13

Rubidium-82

Flurpiridaz-18

18-FDG

Positron energy (MeV)

0.49

1.48

0.25

0.25

Physical half-life

10 minutes

76 seconds

110 minutes

110 minutes

Positron range (mm)

2.5

8.6

1

1

Uptake mechanism

Diffusion and metabolic trapping

Active transport

Mitochondrial binding

Glucose transport

Preparation

On-site cyclotron

Generator

Regional cyclotron

Regional cyclotron

Myocardial extraction fraction (%)

80

65

94

3

Indication

Perfusion MPI

Perfusion MPI

Perfusion MPI

Viability

Stress modality

Pharmacological

Pharmacological

Exercise or pharmacological

Clinical status

FDA approved

FDA approved

FDA approved

FDA, US Food and Drug Administration; 18-FDG, 18-fluorodeoxyglucose; MPI, myocardial perfusion imaging.

Phase III clinical trails

The Role of Noninvasive Cardiac Imaging in the Management of Diseases of the Cardiovascular System

­ uclear Cardiology and the N Diagnosis of IHD The initial evaluation of patients with suspected IHD includes confirming that the presenting symptoms are due to IHD, considering the need for further evaluation with invasive coronary angiography and, subsequently the need for revascularization. The fundamental goal of the management of those with known IHD is to guide medical and revascularization therapy, predict future cardiac events such as cardiac death and nonfatal myocardial infarction (MI), and find measures to prevent them. In addition, MPI is one of the most trusted methods of evaluating the physiological significance of known anatomical stenosis. With more than 30 years of experience, nuclear cardiology MPI is arguably the most evidence-based noninvasive modality in the management of patients with stable IHD. MPI has been evaluated across the full spectrum of IHD presentations with rigorous prospective and retrospective studies.

Multiple studies have demonstrated the high diagnostic accuracy of SPECT and PET MPI in the detection of significant coronary artery disease (CAD). In a large meta-analysis of 108 SPECT studies and four PET studies involving 11862 patients, the pooled sensitivities were 92.6% and 88.3% for PET and SPECT, respectively [28], while the pooled specificities were 81.3% and 75.8%, respectively [29]. Another metaanalysis of eight SPECT and 15 PET studies demonstrated sensitivities of 85% and 90% for SPECT and PET, respectively, while the specificities were 85% and 88%, respectively, when compared to invasive coronary angiography  [30]. When compared to invasive fraction flow reserve (FFR), a head-tohead comparison in 208 patients demonstrated sensitivities of 57% and 87% for SPECT and PET, respectively, while the specificities were 94% and 84%, respectively  [31]. A metaanalysis of 37 studies examining the diagnostic accuracy of multiple noninvasive stress tests against invasive FFR showed specificities of 61% and 83%, respectively, and specificities of 84% and 89%, respectively  [32] (Figures  8.2 and 8.3).

SA (Apex→Base)

StrCTAC

RstCTAC

StrCTAC

RstCTAC

HLA (INF→ANT)

StrCTAC

RstCTAC

VLA (SEP→LAT)

StrCTAC

RstCTAC

Figure 8.2 63-year old male with diabetes, hypertension, and hypercholesterolemia, referred for PET MPI for evaluation of chest pain. The perfusion images show severe reversible perfusion defect on the mid and apical segments of the anterior wall and apex (arrows). The coronary angiography revealed severe disease in the mid segment of the anterior descending coronary artery.

261

262

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach SA (Apex→Base) StrCTAC

RstCTAC

StrCTAC

RstCTAC

HLA (INF→ANT)

StrCTAC

RstCTAC

VLA (SEP→LAT) StrCTAC

RstCTAC

Figure 8.3 58-year-old male with diabetes and hypertension referred for evaluation of exertional chest pain. PET MPI demonstrates normal perfusion with high transient ischemic dilatation (1.26) and normal coronary flow reserve (2.3). In view of normal perfusion and normal CFR, she was referred to coronary angiography, which revealed nonobstructive coronary artery disease.

In general, PET has better diagnostic accuracy than SPECT owing to its better spatial resolution and the use of attenuation correction, which results in better image quality and a lower rate of artifact [33] (Figure 8.4). Another advantage of PET over SPECT is the ability of PET to diagnose balanced ischemia that can be easily overlooked in relative perfusion imaging. The ability of PET to measure myocardial blood flow allows for the diagnosis of balanced ischemia due to multivessel disease as well as coronary microcirculatory dysfunction [34]. Coronary flow reserve (CFR), which is calculated as the ratio between stress and rest myocardial blood flow, can also be used to exclude significant left main disease  [20, 28]. A CFR >2 reliably excludes high-risk angiographic stenoses such as left main disease or two-vessel disease, including proximal left anterior descending artery with a negative predictive value (NPV) of 97%  [35] (Figure  8.4). Coronary microvascular dysfunction (CMD) poses a diagnostic challenge and appears to be more common than initially thought. Almost 60% of patients with angina do not have obstructive CAD on invasive coronary

angiography  [36]. The diagnosis of CMD on PET MPI is made when there is a reduced CFR (10%

End-systolic LV volume > 70 mL

Maximum exercise systolic blood pressure

Exercise-induced ventricular arrhythmia

TID > 1.2

LVEF reserve 0.5 for thallium and >0.32 for technetium)

CFR < 1.5

Calcium score >400 CFR, coronary flow reserve; HLR, heart/lung ratio; LV, left ventricular; LVEF, left ventricular ejection fraction; MPI, myocardial perfusion imaging; SPECT, single-photon emission tomography; TID, transient ischemic dilatation.

prognostic capabilities, regardless of the stressor. A large meta-analysis of 14 918 patients who underwent stress MPI demonstrated similar prognostic value between exercise and pharmacological stress tests  [39]. However, it is well known that the inability to exercise portends a poor prognosis, which was demonstrated in the same meta-analysis. The risk of cardiac death and nonfatal MI was 1.2 is associated with increased risk of significant CAD and poor outcome  [46, 47]. However, recent analysis has questioned the diagnostic and prognostic value of TID in patients with normal perfusion  [48]. Increased lung/heart ratio (>0.5 for thallium and >0.32 for technetium) is associated with higher risk of cardiac death  [49, 50]. This is thought to be due to severe CAD leading to diastolic dysfunction. LV dyssynchrony by phase analysis was also found to predict major adverse clinical events including cardiac death and nonfatal MI independent of LVEF [51]. Finally, vasodilator-induced ST-segment change is associated with a worse outcome even in the presence of normal MPI [52, 53]. The unique ability of PET to measure myocardial blood flow and the subsequent calculation of CFR provides an independent and incremental prognostic value in patients referred to PET MPI. Multiple studies have demonstrated the prognostic value of CFR irrespective of the presence and extent of ischemia. PET-derived CFR of 12 hours) following a

high-fat low-carbohydrate diet [71]. This preparation will shift the cardiac myocytes to use fatty acid for energy production. However, it is known that inflamed myocytes will continue to utilize glucose despite these preparations. Hence the metabolic activity of the inflamed cells can be detected by metabolic imaging with 18FDG. The diagnosis of cardiac sarcoidosis is made through the assessment of the inflammation by metabolic imaging in comparison with the presence of a scar on rest PET perfusion imaging [203] (Table 8.6). The sensitivity and specificity of PET in the diagnosis of cardiac sarcoidosis are 89% and 78%, respectively [204]. A small study compared the diagnostic accuracy of CMR and PET in the diagnosis of cardiac sarcoidosis and showed that PET has higher sensitivity while CMR has higher specificity  [205]. Keeping in mind that CMR detects fibrosis while PET detects inflammation, these differences may be related to different stages of the presentation of cardiac sarcoidosis. The presence of perfusion and metabolism defect on 18FDG PET is associated with an almost fourfold higher risk of sudden cardiac

Figure 8.15 Perfusion defects and FDG uptake in a patient with biopsy-proven cardiac sarcoidosis. Perfusion

Inflammation

Table 8.6­­ Interpretation of 18FDG PET imaging for the diagnosis of cardiac sarcoidosis. Annual event rate for sudden cardiac death and sustained arrhythmia (%)

Rest perfusion

18

Disease stage

Normal

No uptake

Normal

Normal or mild defect

Focal uptake (focal mismatch)

Early disease

Moderate defect

Uptake in the area of the defect

Progressive disease

18.4

Severe defect

Low or no uptake

Advanced disease (fibrosis)

31.9 (hazard ratio = 3.9)

FDG imaging

7.3

Source: Data from Blankstein R, Osborne M, Naya M, Waller A, Kim CK, Murthy VL, Kazemian P, Kwong RY, Tokuda M, Skali H, Padera R, Hainer J, Stevenson WG, Dorbala S and Di Carli MF. Cardiac positron emission tomography enhances prognostic assessments of patients with suspected cardiac sarcoidosis. J Am Coll Cardiol. 2014;63:329–336.

273

274

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

death and ventricular arrhythmia [203]. Furthermore, the presence of MDE on CMR and perfusion defect on 18FDG PET is associated with an extremely high risk of sudden cardiac death (32-fold increase) [199]. The answer to the question about which test should be done in the evaluation of cardiac sarcoidosis, PET or CMR, can be answered by understanding the properties of each of these tests and their ability to detect specific stages of the spectrum of the cardiac sarcoidosis. It is likely that these tests have complementary roles and should be utilized according to clinical suspicion. In a study of 91 subjects, adding CMR to PET resulted in reclassification of 45% of the patients suspected to have cardiac sarcoidosis [206]. Cardiac MRI can also help identify other cardiomyopathies, such as LV noncompaction and dilated cardiomyopathy.

­Valvular Heart Disease Most valvular heart diseases present in heart failure regardless of whether the valve pathology is stenosis or regurgitation. Degenerative aortic stenosis is the most common valvular heart disease [207]. Echocardiography is the gold standard in the diagnosis of valvular stenosis [208]. However, recently aortic valve calcification has been used to aid the diagnosis of severe stenosis in questionable cases [209]. CMR has emerged as the gold standard in the evaluation of valvular regurgitations. CMR can calculate the exact regurgitant volumes and fractions by directly measuring the LV stroke volume and forward flow through the aortic valve [210]. In addition, CMR provides accurate evaluation of LV volume and LVEF, which is an important determinant of intervention  [211]. Likewise, CCTA is crucial in the evaluation of patients prior to transcatheter aortic valve replacement. CCTA provides pivotal procedure guidance through the evaluation of vascular access and decisions about the size of the valve  [211]. CCTA also provides information about the complications of infective endocarditis. CCTA is superior to echocardiography in the detection of the extent of infection as well as abscess formation and subsequent development of pseudoaneurysms [211]. The diagnosis of infective endocarditis in prosthetic valves or cardiac implantable electronic devices poses a considerable challenge in clinical practice. Transthoracic and transesophageal echocardiography face major limitations in the diagnosis of prosthetic valve endocarditis [212]. 18FDG PET imaging is increasingly used in the diagnosis of infective endocarditis. It requires similar patient preparation to inflammation imaging [70]. Pooled data for more than 500 patients with suspected infective endocarditis showed sensitivity and specificity of 76.8%

and 77.9%, respectively [213]. The sensitivity and specificity of 18FDG PET in the diagnosis of infection of cardiac devices were 80% and 91%, respectively  [214]. In  addition, 18FDG PET reclassified 90% of cases with suspected endocarditis  [215]. Similarly, SPECT imaging with radiolabeled white blood cells has value in the diagnosis of infective endocarditis (90% sensitivity, 94% PPV, and 100% NPV)  [216]. In addition, positive 18FDG PET is associated with increased risk of adverse cardiac  events, including in-hospital mortality and 1-year mortality [217].

­Diseases of the Aorta The diseases of the aorta include aortic dissection, aortic aneurysm, and aortitis. Acute aortic dissection is a lifethreatening condition that requires prompt diagnosis and urgent surgical management [218]. CCTA is the cornerstone for the diagnosis of aortic dissection, including the location of the dissection and the size of the true and false lumens  [219, 220]. In addition, CCTA provides information about the extension of the dissection to the major branches of the aorta and the presence of intramural hematoma  [221]. The wide availability of CCTA, its fast acquisition time, and superior spatial resolution make CCTA the first choice of imaging modality for acute aortic syndromes. Likewise, CCTA is pivotal in the management of patients with aortopathies and aortic aneurysms [222]. Aortitis could be a manifestation of many diseases, including giant cell arteritis and Takayasu arteritis [223, 224]. Aortitis results from aortic wall infiltration by the lymphocytes and myocytes. The inflammation may extend into the tissue adjacent to the aorta  [225]. The diagnosis of aortitis on 18FDG PET is made when the wall uptake is higher than the liver uptake [226]. 18FDG PET is highly effective in the diagnosis of aortitis (pooled sensitivity and specificity of 80% and 89%, respectively) [227]. 18 FDG can also differentiate between acute and chronic aortitis  [228], and can predict complications such as aneurysm formation  [229]. CMR also provides valuable information about the diagnosis and complications of aortitis [230, 231].

­Conclusions The utility of noninvasive cardiac imaging in cardiovascular diseases is ever-increasing. Nuclear cardiology, CMR, and CCTA have become pivotal in the management many cardiovascular diseases. They have evolved to be

The Role of Noninvasive Cardiac Imaging in the Management of Diseases of the Cardiovascular System

indispensable tools for early detection, accurate diagnosis, guiding therapy, and predicting prognosis. The future holds more promise of advances in technology, software,

and hardware to further consolidate the role of noninvasive imaging to shape the future landscape of cardiovascular diseases.

References 1 Group WCRCW (2019). World Health Organization cardiovascular disease risk charts: revised models to estimate risk in 21 global regions. Lancet Glob. Health 7: e1332–e1345. 2 Ahmed, A.M., Hersi, A., Mashhoud, W. et al. (2017). Cardiovascular risk factors burden in Saudi Arabia: the Africa Middle East cardiovascular epidemiological (ACE) study. J. Saudi Heart Assoc. 29: 235–243. 3 Yeh, R.W., Sidney, S., Chandra, M. et al. (2010). Population trends in the incidence and outcomes of acute myocardial infarction. N. Engl. J. Med. 362: 2155–2165. 4 Rywik, T.M., O’Connor, F.C., Gittings, N.S. et al. (2002). Role of nondiagnostic exercise-induced ST-segment abnormalities in predicting future coronary events in asymptomatic volunteers. Circulation 106: 2787–2792. 5 Hlatky, M.A., Pryor, D.B., Harrell, F.E. Jr. et al. (1984). Factors affecting sensitivity and specificity of exercise electrocardiography. Multivariable analysis. Am. J. Med. 77: 64–71. 6 Detrano, R., Gianrossi, R., and Froelicher, V. (1989). The diagnostic accuracy of the exercise electrocardiogram: a meta-analysis of 22 years of research. Prog. Cardiovasc. Dis. 32: 173–206. 7 Gianrossi, R., Detrano, R., Mulvihill, D. et al. (1989). Exercise-induced ST depression in the diagnosis of coronary artery disease. A meta-analysis. Circulation 80: 87–98. 8 Shaw, L.J., Mieres, J.H., Hendel, R.H. et al. (2011). Comparative effectiveness of exercise electrocardiography with or without myocardial perfusion single photon emission computed tomography in women with suspected coronary artery disease: results from the what is the optimal method for ischemia evaluation in WOMEN (WOMEN) trial. Circulation 124: 1239–1249. 9 Al-Mallah, M.H., Elshawi, R., Ahmed, A.M. et al. (2017). Using machine learning to define the association between cardiorespiratory fitness and all-cause mortality (from the Henry Ford exercise testing project). Am. J. Cardiol. 120: 2078–2084. 10 Al-Mallah, M.H., Juraschek, S.P., Whelton, S. et al. (2016). Sex differences in cardiorespiratory fitness and all-cause mortality: the Henry Ford ExercIse testing (FIT) project. Mayo Clin. Proc. 91: 755–762. 11 Al-Mallah, M.H., Sakr, S., and Al-Qunaibet, A. (2018). Cardiorespiratory fitness and cardiovascular disease prevention: an update. Curr. Atheroscler. Rep. 20: 1.

12 Bruce, R.A., Kusumi, F., and Hosmer, D. (1973). Maximal oxygen intake and nomographic assessment of functional aerobic impairment in cardiovascular disease. Am. Heart J. 85: 546–562. 13 Gibbons, R.J., Balady, G.J., Bricker, J.T. et al. (2002). ACC/AHA 2002 guideline update for exercise testing: summary article. A report of the American College of Cardiology/American Heart Association task force on practice guidelines (committee to update the 1997 exercise testing guidelines). J. Am. Coll. Cardiol. 40: 1531–1540. 14 Gerstenblith, G., Lakatta, E.G., and Weisfeldt, M.L. (1976). Age changes in myocardial function and exercise response. Prog. Cardiovasc. Dis. 19: 1–21. 15 Henzlova, M.J., Duvall, W.L., Einstein, A.J. et al. (2016). ASNC imaging guidelines for SPECT nuclear cardiology procedures: stress, protocols, and tracers. J. Nucl. Cardiol. 23: 606–639. 16 Wilson, R.F., Wyche, K., Christensen, B.V. et al. (1990). Effects of adenosine on human coronary arterial circulation. Circulation 82: 1595–1606. 17 Chan, S.Y., Brunken, R.C., Czernin, J. et al. (1992). Comparison of maximal myocardial blood flow during adenosine infusion with that of intravenous dipyridamole in normal men. J. Am. Coll. Cardiol. 20: 979–985. 18 Ward, R.P., Al-Mallah, M.H., Grossman, G.B. et al. (2007). American Society of Nuclear Cardiology review of the ACCF/ASNC appropriateness criteria for single-photon emission computed tomography myocardial perfusion imaging (SPECT MPI). J. Nucl. Cardiol. 14: e26–e38. 19 Holly, T.A., Abbott, B.G., Al-Mallah, M. et al. (2010). Single photon-emission computed tomography. J. Nucl. Cardiol. 17: 941–973. 20 Al-Mallah, M.H., Sitek, A., Moore, S.C. et al. (2010). Assessment of myocardial perfusion and function with PET and PET/CT. J. Nucl. Cardiol. 17: 498–513. 21 Strauss, H.W., Harrison, K., Langan, J.K. et al. (1975). Thallium-201 for myocardial imaging. Relation of thallium-201 to regional myocardial perfusion. Circulation 51: 641–645. 22 Wackers, F.J., Berman, D.S., Maddahi, J. et al. (1989). Technetium-99m hexakis 2-methoxyisobutyl isonitrile: human biodistribution, dosimetry, safety, and preliminary comparison to thallium-201 for myocardial perfusion imaging. J. Nucl. Med. 30: 301–311.

275

276

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

23 Leppo, J.A. and Meerdink, D.J. (1989). Comparison of the myocardial uptake of a technetium-labeled isonitrile analogue and thallium. Circ. Res. 65: 632–639. 24 Dowsley, T., Al-Mallah, M., Ananthasubramaniam, K. et al. (2013). The role of noninvasive imaging in coronary artery disease detection, prognosis, and clinical decision making. Can. J. Cardiol. 29: 285–296. 25 Lortie, M., Beanlands, R.S., Yoshinaga, K. et al. (2007). Quantification of myocardial blood flow with 82Rb dynamic PET imaging. Eur. J. Nucl. Med. Mol. Imaging 34: 1765–1774. 26 Jamiel, A., Ebid, M., Ahmed, A.M. et al. (2017). The role of myocardial viability in contemporary cardiac practice. Heart Fail. Rev. 22: 401–413. 27 Khalaf, S., Chamsi-Pasha, M., and Al-Mallah, M.H. (2019). Assessment of myocardial viability by PET. Curr. Opin. Cardiol. 34: 466–472. 28 Al Badarin, F., Aljizeeri, A., Almasoudi, F., and Al-Mallah, M.H. (2017). Assessment of myocardial blood flow and coronary flow reserve with positron emission tomography in ischemic heart disease: current state and future directions. Heart Fail. Rev. 22: 441–453. 29 Parker, M.W., Iskandar, A., Limone, B. et al. (2012). Diagnostic accuracy of cardiac positron emission tomography versus single photon emission computed tomography for coronary artery disease: a bivariate meta-analysis. Circ Cardiovasc Imaging. 5: 700–707. 30 Mc Ardle, B.A. and Dowsley, T.F. (2012). deKemp RA, Wells GA and Beanlands RS. Does rubidium-82 PET have superior accuracy to SPECT perfusion imaging for the diagnosis of obstructive coronary disease?: a systematic review and meta-analysis. J. Am. Coll. Cardiol. 60: 1828–1837. 31 Danad, I., Raijmakers, P.G., Driessen, R.S. et al. (2017). Comparison of coronary CT angiography, SPECT, PET, and hybrid imaging for diagnosis of ischemic Heart disease determined by fractional flow reserve. JAMA Cardiol. 2: 1100–1107. 32 Takx, R.A., Blomberg, B.A., El Aidi, H. et al. (2015). Diagnostic accuracy of stress myocardial perfusion imaging compared to invasive coronary angiography with fractional flow reserve meta-analysis. Circ Cardiovasc Imaging. 8: e002666. https://doi.org/10.1161/ CIRCIMAGING.114.002666. 33 Bateman, T.M., Heller, G.V., McGhie, A.I. et al. (2006). Diagnostic accuracy of rest/stress ECG-gated Rb-82 myocardial perfusion PET: comparison with ECG-gated Tc-99m sestamibi SPECT. J. Nucl. Cardiol. 13: 24–33. 34 Bravo, P.E., Di Carli, M.F., and Dorbala, S. (2017). Role of PET to evaluate coronary microvascular dysfunction in non-ischemic cardiomyopathies. Heart Fail. Rev. 22: 455–464.

35 Naya, M., Murthy, V.L., Taqueti, V.R. et al. (2014). Preserved coronary flow reserve effectively excludes high-risk coronary artery disease on angiography. J. Nucl. Med. 55: 248–255. 36 Patel, M.R., Peterson, E.D., Dai, D. et al. (2010). Low diagnostic yield of elective coronary angiography. N. Engl. J. Med. 362: 886–895. 37 Schindler, T.H. and Dilsizian, V. (2020). Coronary microvascular dysfunction: clinical considerations and noninvasive diagnosis. JACC Cardiovasc. Imaging 13: 140–155. 38 Dorbala, S., Vangala, D., Sampson, U. et al. (2007). Value of vasodilator left ventricular ejection fraction reserve in evaluating the magnitude of myocardium at risk and the extent of angiographic coronary artery disease: a 82Rb PET/CT study. J. Nucl. Med. 48: 349–358. 39 Navare, S.M., Mather, J.F., Shaw, L.J. et al. (2004). Comparison of risk stratification with pharmacologic and exercise stress myocardial perfusion imaging: a metaanalysis. J. Nucl. Cardiol. 11: 551–561. 40 Iskander, S. and Iskandrian, A.E. (1998). Risk assessment using single-photon emission computed tomographic technetium-99m sestamibi imaging. J. Am. Coll. Cardiol. 32: 57–62. 41 Gibbons, R.J., Hodge, D.O., Berman, D.S. et al. (1999). Long-term outcome of patients with intermediate-risk exercise electrocardiograms who do not have myocardial perfusion defects on radionuclide imaging. Circulation 100: 2140–2145. 42 Dorbala, S., Di Carli, M.F., Beanlands, R.S. et al. (2013). Prognostic value of stress myocardial perfusion positron emission tomography: results from a multicenter observational registry. J. Am. Coll. Cardiol. 61: 176–184. 43 Dorbala, S., Hachamovitch, R., Curillova, Z. et al. (2009). Incremental prognostic value of gated Rb-82 positron emission tomography myocardial perfusion imaging over clinical variables and rest LVEF. JACC Cardiovasc. Imaging 2: 846–854. 44 Al-Mallah, M.H., Hachamovitch, R., Dorbala, S., and Di Carli, M.F. (2009). Incremental prognostic value of myocardial perfusion imaging in patients referred to stress single-photon emission computed tomography with renal dysfunction. Circ Cardiovasc Imaging. 2: 429–436. 45 Sharir, T., Germano, G., Kavanagh, P.B. et al. (1999). Incremental prognostic value of post-stress left ventricular ejection fraction and volume by gated myocardial perfusion single photon emission computed tomography. Circulation 100: 1035–1042. 46 Travin, M.I., Heller, G.V., Johnson, L.L. et al. (2004). The prognostic value of ECG-gated SPECT imaging in patients undergoing stress Tc-99m sestamibi myocardial perfusion imaging. J. Nucl. Cardiol. 11: 253–262.

The Role of Noninvasive Cardiac Imaging in the Management of Diseases of the Cardiovascular System

47 Abidov, A., Bax, J.J., Hayes, S.W. et al. (2003). Transient ischemic dilation ratio of the left ventricle is a significant predictor of future cardiac events in patients with otherwise normal myocardial perfusion SPECT. J. Am. Coll. Cardiol. 42: 1818–1825. 48 Valdiviezo, C., Motivala, A.A., Hachamovitch, R. et al. (2011). The significance of transient ischemic dilation in the setting of otherwise normal SPECT radionuclide myocardial perfusion images. J. Nucl. Cardiol. 18: 220–229. 49 Gill, J.B., Ruddy, T.D., Newell, J.B. et al. (1987). Prognostic importance of thallium uptake by the lungs during exercise in coronary artery disease. N. Engl. J. Med. 317: 1485–1489. 50 Homma, S., Kaul, S., and Boucher, C.A. (1987). Correlates of lung/heart ratio of thallium-201 in coronary artery disease. J. Nucl. Med. 28: 1531–1535. 51 Pazhenkottil, A.P., Buechel, R.R., Husmann, L. et al. (2011). Long-term prognostic value of left ventricular dyssynchrony assessment by phase analysis from myocardial perfusion imaging. Heart 97: 33–37. 52 Abbott, B.G., Afshar, M., Berger, A.K., and Wackers, F.J. (2003). Prognostic significance of ischemic electrocardiographic changes during adenosine infusion in patients with normal myocardial perfusion imaging. J. Nucl. Cardiol. 10: 9–16. 53 Aljizeeri, A., Ahmed, D., Alali Alfaris, M. et al. (2018). Prognostic value of vasodilator-induced ST segment changes in patients with normal RBUDIUM-82 myocardial perfusion imaging. J. Am. Coll. Cardiol.: 71. 54 Ziadi, M.C., Dekemp, R.A., Williams, K.A. et al. (2011). Impaired myocardial flow reserve on rubidium-82 positron emission tomography imaging predicts adverse outcomes in patients assessed for myocardial ischemia. J. Am. Coll. Cardiol. 58: 740–748. 55 Wang, L., Jerosch-Herold, M., Jacobs, D.R. Jr. et al. (2006). Coronary risk factors and myocardial perfusion in asymptomatic adults: the multi-ethnic study of atherosclerosis (MESA). J. Am. Coll. Cardiol. 47: 565–572. 56 Murthy, V.L., Naya, M., Foster, C.R. et al. (2011). Improved cardiac risk assessment with noninvasive measures of coronary flow reserve. Circulation 124: 2215–2224. 57 Patel, K.K., Spertus, J.A., Chan, P.S. et al. (2020). Myocardial blood flow reserve assessed by positron emission tomography myocardial perfusion imaging identifies patients with a survival benefit from early revascularization. Eur. Heart J. 41: 759–768. 58 Hachamovitch, R., Hayes, S.W., Friedman, J.D. et al. (2003). Comparison of the short-term survival benefit associated with revascularization compared with medical

59

60

61

62

63

64

65

66

67

68

69

70

therapy in patients with no prior coronary artery disease undergoing stress myocardial perfusion single photon emission computed tomography. Circulation 107: 2900–2907. Hachamovitch, R., Rozanski, A., Shaw, L.J. et al. (2011). Impact of ischaemia and scar on the therapeutic benefit derived from myocardial revascularization vs. medical therapy among patients undergoing stress-rest myocardial perfusion scintigraphy. Eur. Heart J. 32: 1012–1024. Maron, D.J., Hochman, J.S., Reynolds, H.R. et al. (2020). Initial Invasive or Conservative Strategy for Stable Coronary Disease. N. Engl. J. Med. 382: 1395–1407. Hochman, J.S., Reynolds, H.R., Bangalore, S. et al. (2019). Baseline characteristics and risk profiles of participants in the ISCHEMIA randomized clinical trial. JAMA Cardiol. 4: 273–286. Ambrosio, G., Betocchi, S., Pace, L. et al. (1996). Prolonged impairment of regional contractile function after resolution of exercise-induced angina. Evidence of myocardial stunning in patients with coronary artery disease. Circulation 94: 2455–2464. Braunwald, E. and Kloner, R.A. (1982). The stunned myocardium: prolonged, postischemic ventricular dysfunction. Circulation 66: 1146–1149. Dilsizian, V., Rocco, T.P., Freedman, N.M. et al. (1990). Enhanced detection of ischemic but viable myocardium by the reinjection of thallium after stress-redistribution imaging. N. Engl. J. Med. 323: 141–146. Perrone-Filardi, P., Pace, L., Prastaro, M. et al. (1996). Assessment of myocardial viability in patients with chronic coronary artery disease. Rest-4-hour-24-hour 201Tl tomography versus dobutamine echocardiography. Circulation 94: 2712–2719. Medrano, R., Lowry, R.W., Young, J.B. et al. (1996). Assessment of myocardial viability with 99mTc sestamibi in patients undergoing cardiac transplantation. A scintigraphic/pathological study. Circulation 94: 1010–1017. Schinkel, A.F., Bax, J.J., Poldermans, D. et al. (2007). Hibernating myocardium: diagnosis and patient outcomes. Curr. Probl. Cardiol. 32: 375–410. Partington, S.L., Kwong, R.Y., and Dorbala, S. (2011). Multimodality imaging in the assessment of myocardial viability. Heart Fail. Rev. 16: 381–395. Allman, K.C., Shaw, L.J., Hachamovitch, R., and Udelson, J.E. (2002). Myocardial viability testing and impact of revascularization on prognosis in patients with coronary artery disease and left ventricular dysfunction: a metaanalysis. J. Am. Coll. Cardiol. 39: 1151–1158. Dilsizian, V., Bacharach, S.L., Beanlands, R.S. et al. (2016). ASNC imaging guidelines/SNMMI procedure

277

278

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

71

72

73

74

75

76

77

78

79

80

standard for positron emission tomography (PET) nuclear cardiology procedures. J. Nucl. Cardiol. 23: 1187–1226. Tamaki, N., Kawamoto, M., Tadamura, E. et al. (1995). Prediction of reversible ischemia after revascularization. Perfusion and metabolic studies with positron emission tomography. Circulation 91: 1697–1705. Di Carli, M.F., Asgarzadie, F., Schelbert, H.R. et al. (1995). Quantitative relation between myocardial viability and improvement in heart failure symptoms after revascularization in patients with ischemic cardiomyopathy. Circulation 92: 3436–3444. Mc Ardle, B., Shukla, T., Nichol, G. et al. (2016). Longterm follow-up of outcomes with F-18fluorodeoxyglucose positron emission tomography imaging-assisted management of patients with severe left ventricular dysfunction secondary to coronary disease. Circ Cardiovasc Imaging.: 9, e004331. https://doi. org/10.1161/CIRCIMAGING.115.004331. Beanlands, R.S., Hendry, P.J., Masters, R.G. et al. (1998). Delay in revascularization is associated with increased mortality rate in patients with severe left ventricular dysfunction and viable myocardium on fluorine 18-fluorodeoxyglucose positron emission tomography imaging. Circulation 98: 1151–1156. Mc Ardle, B.A., Davies, R.A., Chen, L. et al. (2014). Prognostic value of rubidium-82 positron emission tomography in patients after heart transplant. Circ Cardiovasc Imaging. 7: 930–937. Feher, A., Srivastava, A., Quail, M.A. et al. (2020). Serial assessment of coronary flow reserve by rubidium-82 positron emission tomography predicts mortality in Heart transplant recipients. JACC Cardiovasc. Imaging 13: 109–120. Konerman, M.C., Lazarus, J.J., Weinberg, R.L. et al. (2018). Reduced myocardial flow reserve by positron emission tomography predicts cardiovascular events after cardiac transplantation. Circ. Heart Fail. 11: e004473. Bamberg, F., Mayrhofer, T., Ferencik, M. et al. (2018). Age- and sex-based resource utilisation and costs in patients with acute chest pain undergoing cardiac CT angiography: pooled evidence from ROMICAT II and ACRIN-PA trials. Eur. Radiol. 28: 851–860. Hachamovitch, R., Hayes, S., Friedman, J.D. et al. (2003). Determinants of risk and its temporal variation in patients with normal stress myocardial perfusion scans: what is the warranty period of a normal scan? J. Am. Coll. Cardiol. 41: 1329–1340. Erthal, F., Hossain, A., Dorbala, S. et al. (2018). Is there an age when myocardial perfusion imaging may no longer be Prognostically useful? Circ Cardiovasc Imaging. 11: e007322.

81 Pepine, C.J., Ferdinand, K.C., Shaw, L.J. et al. (2015). Emergence of nonobstructive coronary artery disease: a Woman’s problem and need for change in definition on angiography. J. Am. Coll. Cardiol. 66: 1918–1933. 82 Melin, J.A., Wijns, W., Vanbutsele, R.J. et al. (1985). Alternative diagnostic strategies for coronary artery disease in women: demonstration of the usefulness and efficiency of probability analysis. Circulation 71: 535–542. 83 Taqueti, V.R., Dorbala, S., Wolinsky, D. et al. (2017). Myocardial perfusion imaging in women for the evaluation of stable ischemic heart disease-state-of-theevidence and clinical recommendations. J. Nucl. Cardiol. 24: 1402–1426. 84 Metz, L.D., Beattie, M., Hom, R. et al. (2007). The prognostic value of normal exercise myocardial perfusion imaging and exercise echocardiography: a meta-analysis. J. Am. Coll. Cardiol. 49: 227–237. 85 Hachamovitch, R., Berman, D.S., Kiat, H. et al. (1996). Effective risk stratification using exercise myocardial perfusion SPECT in women: gender-related differences in prognostic nuclear testing. J. Am. Coll. Cardiol. 28: 34–44. 86 von Mering, G.O., Arant, C.B., Wessel, T.R. et al. (2004). Abnormal coronary vasomotion as a prognostic indicator of cardiovascular events in women: results from the National Heart, Lung, and Blood Institute-sponsored Women’s Ischemia Syndrome Evaluation (WISE). Circulation 109: 722–725. 87 Taqueti, V.R., Shaw, L.J., Cook, N.R. et al. (2017). Excess cardiovascular risk in women relative to men referred for coronary angiography is associated with severely impaired coronary flow reserve, not obstructive disease. Circulation 135: 566–577. 88 Murthy, V.L., Naya, M., Taqueti, V.R. et al. (2014). Effects of sex on coronary microvascular dysfunction and cardiac outcomes. Circulation 129: 2518–2527. 89 Chiariello, M. and Indolfi, C. (1996). Silent myocardial ischemia in patients with diabetes mellitus. Circulation 93: 2089–2091. 90 Shaw, L.J. and Iskandrian, A.E. (2004). Prognostic value of gated myocardial perfusion SPECT. J. Nucl. Cardiol. 11: 171–185. 91 Murthy, V.L., Naya, M., Foster, C.R. et al. (2012). Association between coronary vascular dysfunction and cardiac mortality in patients with and without diabetes mellitus. Circulation 126: 1858–1868. 92 Shaw, L.J., Hendel, R.C., Cerquiera, M. et al. (2005). Ethnic differences in the prognostic value of stress technetium-99m tetrofosmin gated single-photon emission computed tomography myocardial perfusion imaging. J. Am. Coll. Cardiol. 45: 1494–1504.

The Role of Noninvasive Cardiac Imaging in the Management of Diseases of the Cardiovascular System

93 Hecht, H., Blaha, M.J., Berman, D.S. et al. Clinical indications for coronary artery calcium scoring in asymptomatic patients: expert consensus statement from the Society of Cardiovascular Computed Tomography. J. Cardiovasc. Comput. Tomogr. 11: 157–168. 94 Detrano, R., Guerci, A.D., Carr, J.J. et al. (2008). Coronary calcium as a predictor of coronary events in four racial or ethnic groups. N. Engl. J. Med. 358: 1336–1345. 95 Shaw, L.J., Giambrone, A.E., and Blaha, M.J., K.J.T., B.D.S., B.N., Q.A., B.M., C.T.Q.a.M.J.K. (2015). Longterm prognosis after coronary artery calcification testing in asymptomatic patients: a cohort study. Ann. Intern. Med. 163: 14–21. 96 Erbel, R., Möhlenkamp, S., Moebus, S. et al. (2010). Coronary risk stratification, discrimination, and reclassification improvement based on quantification of subclinical coronary atherosclerosis: the Heinz Nixdorf recall study. J. Am. Coll. Cardiol. 56: 1397–1406. 97 Blaha, M.J., Whelton, S.P., Al Rifai, M. et al. (2020). Comparing risk scores in the prediction of coronary and cardiovascular deaths: coronary artery calcium consortium. JACC Cardiovasc. Imaging https://doi. org/10.1016/j.jcmg.2019.12.010. 98 Dzaye, O., Al Rifai, M., Dardari, Z. et al. (2020). Coronary artery calcium as a synergistic tool for the age- and sex-specific risk of cardiovascular and cancer mortality: the coronary artery calcium consortium. J. Am. Heart Assoc. 9: e015306. 99 Al-Mallah, M.H. (2018). Coronary artery calcium scoring: do we need more prognostic data prior to adoption in clinical practice? JACC Cardiovasc. Imaging 11: 1807–1809. 100 Al’Aref, S.J., Maliakal, G., Singh, G. et al. (2020). Machine learning of clinical variables and coronary artery calcium scoring for the prediction of obstructive coronary artery disease on coronary computed tomography angiography: analysis from the CONFIRM registry. Eur. Heart J. 41: 359–367. 101 Al-Mallah, M.H., Aljizeeri, A., Villines, T.C. et al. (2015). Cardiac computed tomography in current cardiology guidelines. J. Cardiovasc. Comput. Tomogr. 9: 514–523. 102 Budoff, M.J., Dowe, D., Jollis, J.G. et al. (2008). Diagnostic performance of 64-multidetector row coronary computed tomographic angiography for evaluation of coronary artery stenosis in individuals without known coronary artery disease: results from the prospective multicenter ACCURACY (Assessment by Coronary Computed Tomographic Angiography of Individuals Undergoing Invasive Coronary Angiography) trial. J. Am. Coll. Cardiol. 52: 1724–1732.

103 Meijboom, W.B., Meijs, M.F., Schuijf, J.D. et al. (2008). Diagnostic accuracy of 64-slice computed tomography coronary angiography: a prospective, multicenter, multivendor study. J. Am. Coll. Cardiol. 52: 2135–2144. 104 Stein, P.D., Yaekoub, A.Y., Matta, F., and Sostman, H.D. (2008). 64-slice CT for diagnosis of coronary artery disease: a systematic review. Am. J. Med. 121: 715–725. 105 Andreini, D., Pontone, G., Mushtaq, S. et al. (2010). Multidetector computed tomography coronary angiography for the assessment of coronary in-stent restenosis. Am. J. Cardiol. 105: 645–655. 106 Rixe, J., Achenbach, S., Ropers, D. et al. (2006). Assessment of coronary artery stent restenosis by 64-slice multi-detector computed tomography. Eur. Heart J. 27: 2567–2572. 107 Sheth, T., Dodd, J.D., Hoffmann, U. et al. (2007). Coronary stent assessability by 64 slice multi-detector computed tomography. Catheter. Cardiovasc. Interv. 69: 933–938. 108 Van Mieghem, C.A., Cademartiri, F., Mollet, N.R. et al. (2006). Multislice spiral computed tomography for the evaluation of stent patency after left main coronary artery stenting: a comparison with conventional coronary angiography and intravascular ultrasound. Circulation 114: 645–653. 109 Norgaard, B.L., Leipsic, J., Gaur, S. et al. (2014). Diagnostic performance of noninvasive fractional flow reserve derived from coronary computed tomography angiography in suspected coronary artery disease: the NXT trial (analysis of coronary Blood flow using CT angiography: next steps). J. Am. Coll. Cardiol. 63: 1145–1155. 110 Koo, B.K., Erglis, A., Doh, J.H. et al. (2011). Diagnosis of ischemia-causing coronary stenoses by noninvasive fractional flow reserve computed from coronary computed tomographic angiograms. Results from the prospective multicenter DISCOVER-FLOW (Diagnosis of Ischemia-causing Stenoses Obtained Via Noninvasive Fractional Flow Reserve) study. J. Am. Coll. Cardiol. 58: 1989–1997. 111 Coenen, A., Lubbers, M.M., Kurata, A. et al. (2015). Fractional flow reserve computed from noninvasive CT angiography data: diagnostic performance of an on-site clinician-operated computational fluid dynamics algorithm. Radiology 274: 674–683. 112 Al-Mallah, M.H. and Ahmed, A.M. (2016). Controversies in the use of fractional flow reserve form computed tomography (FFRCT) vs. coronary angiography. Curr. Cardiovasc. Imaging Rep. 9: 34. 113 Gaemperli, O., Saraste, A., and Knuuti, J. (2012). Cardiac hybrid imaging. Eur. Heart J. Cardiovasc. Imaging 13: 51–60.

279

280

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

114 Cho, I., Al’Aref, S.J., Berger, A. et al. (2018). Prognostic value of coronary computed tomographic angiography findings in asymptomatic individuals: a 6-year follow-up from the prospective multicentre international CONFIRM study. Eur. Heart J. 39: 934–941. 115 Otaki, Y., Arsanjani, R., Gransar, H. et al. (2012). What have we learned from CONFIRM? Prognostic implications from a prospective multicenter international observational cohort study of consecutive patients undergoing coronary computed tomographic angiography. J. Nucl. Cardiol. 19: 787–795. 116 Schulman-Marcus, J., ó Hartaigh, B., Gransar, H. et al. (2016). Sex-specific associations between coronary artery plaque extent and risk of major adverse cardiovascular events. The CONFIRM Long-Term Registry. JACC Cardiovasc. Imaging 9: 364–372. 117 Blanke, P., Naoum, C., Ahmadi, A. et al. (2016). Long-term prognostic utility of coronary CT angiography in stable patients with diabetes mellitus. JACC Cardiovasc. Imaging 9: 1280–1288. 118 Cheruvu, C., Precious, B., Naoum, C. et al. (2016). Long term prognostic utility of coronary CT angiography in patients with no modifiable coronary artery disease risk factors: results from the 5 year follow-up of the CONFIRM international Multicenter registry. J Cardiovasc Comput Tomogr 10: 22–27. 119 Nielsen, L.H., Botker, H.E., Sorensen, H.T. et al. (2017). Prognostic assessment of stable coronary artery disease as determined by coronary computed tomography angiography: a Danish multicentre cohort study. Eur. Heart J. 38: 413–421. 120 Al-Mallah, M.H., Qureshi, W., Lin, F.Y. et al. (2014). Does coronary CT angiography improve risk stratification over coronary calcium scoring in symptomatic patients with suspected coronary artery disease? Results from the prospective multicenter international CONFIRM registry. Eur. Heart J. Cardiovasc. Imaging 15: 267–274. 121 van Rosendael, A.R., Bax, A.M., Smit, J.M. et al. (2020). Clinical risk factors and atherosclerotic plaque extent to define risk for major events in patients without obstructive coronary artery disease: the long-term coronary computed tomography angiography CONFIRM registry. Eur. Heart J. Cardiovasc. Imaging 21: 479–488. 122 Mirbolouk, M., Kianoush, S., Dardari, Z. et al. (2020). The association of coronary artery calcium score and mortality risk among smokers: the coronary artery calcium consortium. Atherosclerosis 294: 33–40. 123 Peng, A.W., Mirbolouk, M., Orimoloye, O.A. et al. (2020). Long-term all-cause and cause-specific mortality in asymptomatic patients with CAC >/=1,000: results

124

125

126

127

128

129

130

131

132

133

134

135

from the CAC consortium. JACC Cardiovasc. Imaging 13: 83–93. van Rosendael, A.R., Lin, F.Y., Ma, X. et al. (2020). Percent atheroma volume: optimal variable to report whole-heart atherosclerotic plaque burden with coronary CTA, the PARADIGM study. J. Cardiovasc. Comput. Tomogr. https://doi.org/10.1016/j. jcct.2020.01.012. Thomsen, C. and Abdulla, J. (2016). Characteristics of high-risk coronary plaques identified by computed tomographic angiography and associated prognosis: a systematic review and meta-analysis. Eur. Heart J. Cardiovasc. Imaging 17: 120–129. Cordeiro, M.A. and Lima, J.A. (2006). Atherosclerotic plaque characterization by multidetector row computed tomography angiography. J. Am. Coll. Cardiol. 47: C40–C47. Maroules, C.D., Hamilton-Craig, C., Branch, K. et al. (2018). Coronary artery disease reporting and data system (CAD-RADS[TM]): inter-observer agreement for assessment categories and modifiers. J Cardiovasc Comput Tomogr. 12: 125–130. Nam, K., Hur, J., Han, K. et al. (2019). Prognostic value of coronary artery disease-reporting and data system (CAD-RADS) score for cardiovascular events in ischemic stroke. Atherosclerosis 287: 1–7. SCOT-HEART Investigators, Newby, D.E., Adamson, P.D. et al. (2018). Coronary CT angiography and 5-year risk of myocardial infarction. N. Engl. J. Med. 379: 924–933. Ihdayhid, A.R., Norgaard, B.L., Gaur, S. et al. (2019). Prognostic value and risk continuum of noninvasive fractional flow reserve derived from coronary CT angiography. Radiology 292: 343–351. van Assen, M., De Cecco, C.N., Eid, M. et al. (2019). Prognostic value of CT myocardial perfusion imaging and CT-derived fractional flow reserve for major adverse cardiac events in patients with coronary artery disease. J Cardiovasc Comput Tomogr. 13: 26–33. Patel, M.R., Norgaard, B.L., Fairbairn, T.A. et al. (2020). 1-year impact on medical practice and clinical outcomes of FFRCT: the ADVANCE registry. JACC Cardiovasc. Imaging 13: 97–105. Aljizeeri, A., Sulaiman, A., Alhulaimi, N. et al. (2017). Cardiac magnetic resonance imaging in heart failure: where the alphabet begins! Heart Fail. Rev. 22: 385–399. Al-Mallah, M.H. and Shareef, M.N. (2011). The role of cardiac magnetic resonance imaging in the assessment of non-ischemic cardiomyopathy. Heart Fail. Rev. 16: 369–380. Greenwood, J.P., Maredia, N., Younger, J.F. et al. (2012). Cardiovascular magnetic resonance and single-photon

The Role of Noninvasive Cardiac Imaging in the Management of Diseases of the Cardiovascular System

136

137

138

139

140

141

142

143

144

145

emission computed tomography for diagnosis of coronary heart disease (CE-MARC): a prospective trial. Lancet 379: 453–460. Schwitter, J., Wacker, C.M., Wilke, N. et al. (2013). Magnetic resonance imaging for myocardial perfusion assessment in coronary artery disease trial: perfusioncardiac magnetic resonance vs. single-photon emission computed tomography for the detection of coronary artery disease: a comparative multicentre, multivendor trial. Eur. Heart J. 34: 775–781. Schwitter, J., Wacker, C.M., van Rossum, A.C. et al. (2008). MR-IMPACT: comparison of perfusion-cardiac magnetic resonance with single-photon emission computed tomography for the detection of coronary artery disease in a multicentre, multivendor, randomized trial. Eur. Heart J. 29: 480–489. Greenwood, J.P., Ripley, D.P., Berry, C. et al. (2016). Effect of care guided by cardiovascular magnetic resonance, myocardial perfusion Scintigraphy, or NICE guidelines on subsequent unnecessary angiography rates: the CE-MARC 2 randomized clinical trial. JAMA 316: 1051–1060. Watkins, S., McGeoch, R., Lyne, J. et al. (2009). Validation of magnetic resonance myocardial perfusion imaging with fractional flow reserve for the detection of significant coronary heart disease. Circulation 120: 2207–2213. Kim, R.J., Fieno, D.S., Parrish, T.B. et al. (1999). Relationship of MRI delayed contrast enhancement to irreversible injury, infarct age, and contractile function. Circulation 100: 1992–2002. Tamis-Holland, J.E., Jneid, H., Reynolds, H.R. et al. (2019). Contemporary diagnosis and management of patients with myocardial infarction in the absence of obstructive coronary artery disease: a scientific statement from the American Heart Association. Circulation 139: e891–e908. Liu, A., Wijesurendra, R.S., Liu, J.M. et al. (2018). Diagnosis of microvascular angina using cardiac magnetic resonance. J. Am. Coll. Cardiol. 71: 969–979. Nagel, E., Greenwood, J.P., McCann, G.P. et al. (2019). Magnetic resonance perfusion or fractional flow reserve in coronary disease. N. Engl. J. Med. 380: 2418–2428. Lipinski, M.J., McVey, C.M., Berger, J.S. et al. (2013). Prognostic value of stress cardiac magnetic resonance imaging in patients with known or suspected coronary artery disease: a systematic review and meta-analysis. J. Am. Coll. Cardiol. 62: 826–838. Greenwood, J.P., Herzog, B.A., Brown, J.M. et al. (2016). Prognostic value of cardiovascular magnetic resonance and single-photon emission computed tomography in suspected coronary Heart disease: long-term follow-up

146

147

148

149

150

151

152

153

154

155

156

of a prospective, diagnostic accuracy cohort study. Ann. Intern. Med. 165: 1–9. Kwong, R.Y., Ge, Y., Steel, K. et al. (2019). Cardiac magnetic resonance stress perfusion imaging for evaluation of patients with chest pain. J. Am. Coll. Cardiol. 74: 1741–1755. Heydari, B., Juan, Y.H., Liu, H. et al. (2016). Stress perfusion cardiac magnetic resonance imaging effectively risk stratifies diabetic patients with suspected myocardial ischemia. Circ Cardiovasc Imaging. 9: e004136. Vincenti, G., Masci, P.G., Monney, P. et al. (2017). Stress perfusion CMR in patients with known and suspected CAD: prognostic value and optimal ischemic threshold for revascularization. JACC Cardiovasc. Imaging 10: 526–537. Esteban-Fernandez, A., Bastarrika, G., Castanon, E. et al. (2020). Prognostic role of stress cardiac magnetic resonance in the elderly. Rev. Esp. Cardiol. (Engl. Ed.) 73: 241–247. Conrad, N., Judge, A., Tran, J. et al. (2018). Temporal trends and patterns in heart failure incidence: a population-based study of 4 million individuals. Lancet 391: 572–580. Yancy, C.W., Jessup, M., Bozkurt, B. et al. (2017). 2017 ACC/AHA/HFSA focused update of the 2013 ACCF/ AHA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines and the Heart Failure Society of America. J. Am. Coll. Cardiol. 70: 776–803. Bellenger, N.G., Burgess, M.I., Ray, S.G. et al. (2000). Comparison of left ventricular ejection fraction and volumes in heart failure by echocardiography, radionuclide ventriculography and cardiovascular magnetic resonance; are they interchangeable? Eur. Heart J. 21: 1387–1396. Karamitsos, T.D., Francis, J.M., Myerson, S. et al. (2009). The role of cardiovascular magnetic resonance imaging in heart failure. J. Am. Coll. Cardiol. 54: 1407–1424. Walsh, T.F. and Hundley, W.G. (2007). Assessment of ventricular function with cardiovascular magnetic resonance. Cardiol. Clin. 25: 15–33. v. Williams, K.A. and Taillon, L.A. (1996). Left ventricular function in patients with coronary artery disease assessed by gated tomographic myocardial perfusion images. Comparison with assessment by contrast ventriculography and first-pass radionuclide angiography. J. Am. Coll. Cardiol. 27: 173–181. McCrohon, J.A., Moon, J.C., Prasad, S.K. et al. (2003). Differentiation of heart failure related to dilated cardiomyopathy and coronary artery disease using

281

282

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

157

158

159

160

161

162

163

164

165

166

167

gadolinium-enhanced cardiovascular magnetic resonance. Circulation 108: 54–59. Friedrich, M.G., Sechtem, U., Schulz-Menger, J. et al. (2009). Cardiovascular magnetic resonance in myocarditis: a JACC White paper. J. Am. Coll. Cardiol. 53: 1475–1487. Alotaibi, A.M., Aljizeeri, A., Al-Mallah, M., and Alsaileek, A. (2017). Utility of cardiac magnetic resonance in recurrent myocarditis. J. Saudi Heart Assoc. 29: 214–218. Petersen, S.E., Selvanayagam, J.B., Wiesmann, F. et al. (2005). Left ventricular non-compaction: insights from cardiovascular magnetic resonance imaging. J. Am. Coll. Cardiol. 46: 101–105. Epstein, A.E., JP, D.M., Ellenbogen, K.A. et al. (2013). 2012 ACCF/AHA/HRS focused update incorporated into the ACCF/AHA/HRS 2008 guidelines for devicebased therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology Foundation/ American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J. Am. Coll. Cardiol. 61: e6–e75. Scott, P.A., Morgan, J.M., Carroll, N. et al. (2011). The extent of left ventricular scar quantified by late gadolinium enhancement MRI is associated with spontaneous ventricular arrhythmias in patients with coronary artery disease and implantable cardioverterdefibrillators. Circ. Arrhythm. Electrophysiol. 4: 324–330. Kim, R.J., Wu, E., Rafael, A. et al. (2000). The use of contrast-enhanced magnetic resonance imaging to identify reversible myocardial dysfunction. N. Engl. J. Med. 343: 1445–1453. Becker, M.A.J., Cornel, J.H., van de Ven, P.M. et al. (2018). The prognostic value of late gadoliniumenhanced cardiac magnetic resonance imaging in nonischemic dilated cardiomyopathy: a review and meta-analysis. JACC Cardiovasc. Imaging 11: 1274–1284. Jacobson, A.F., Lombard, J., Banerjee, G., and Camici, P.G. (2009). 123I-mIBG scintigraphy to predict risk for adverse cardiac outcomes in heart failure patients: design of two prospective multicenter international trials. J. Nucl. Cardiol. 16: 113–121. Owan, T.E., Hodge, D.O., Herges, R.M. et al. (2006). Trends in prevalence and outcome of heart failure with preserved ejection fraction. N. Engl. J. Med. 355: 251–259. Sharma, K. and Kass, D.A. (2014). Heart failure with preserved ejection fraction: mechanisms, clinical features, and therapies. Circ. Res. 115: 79–96. Nagueh, S.F. (2020). Left ventricular diastolic function: understanding pathophysiology, diagnosis, and prognosis with echocardiography. JACC Cardiovasc. Imaging 13: 228–244.

168 Akincioglu, C., Berman, D.S., Nishina, H. et al. (2005). Assessment of diastolic function using 16-frame 99mTc-sestamibi gated myocardial perfusion SPECT: normal values. J. Nucl. Med. 46: 1102–1108. 169 Talreja, D.R., Edwards, W.D., Danielson, G.K. et al. (2003). Constrictive pericarditis in 26 patients with histologically normal pericardial thickness. Circulation 108: 1852–1857. 170 Feng, D., Glockner, J., Kim, K. et al. (2011). Cardiac magnetic resonance imaging pericardial late gadolinium enhancement and elevated inflammatory markers can predict the reversibility of constrictive pericarditis after antiinflammatory medical therapy: a pilot study. Circulation 124: 1830–1837. 171 Young, P.M., Glockner, J.F., Williamson, E.E. et al. (2012). MR imaging findings in 76 consecutive surgically proven cases of pericardial disease with CT and pathologic correlation. Int. J. Card. Imaging 28: 1099–1109. 172 Chang, S.A., Choi, J.Y., Kim, E.K. et al. (2017). [(18)F] Fluorodeoxyglucose PET/CT predicts response to steroid therapy in constrictive pericarditis. J. Am. Coll. Cardiol. 69: 750–752. 173 Kremastinos, D.T. and Farmakis, D. (2011). Iron overload cardiomyopathy in clinical practice. Circulation 124: 2253–2263. 174 Kirk, P., Roughton, M., Porter, J.B. et al. (2009). Cardiac T2* magnetic resonance for prediction of cardiac complications in thalassemia major. Circulation 120: 1961–1968. 175 Maron, B.J., Maron, M.S., and Semsarian, C. (2012). Genetics of hypertrophic cardiomyopathy after 20 years: clinical perspectives. J. Am. Coll. Cardiol. 60: 705–715. 176 Marian, A.J. and Braunwald, E. (2017). Hypertrophic cardiomyopathy: genetics, pathogenesis, clinical manifestations, diagnosis, and therapy. Circ. Res. 121: 749–770. 177 Al-Sabeq, B., Soliman, A., Chamsi-Pasha, M. et al. (2020). Positrons, protons, and pulse waves: multimodality characterization of newly diagnosed hypertrophic cardiomyopathy. J. Nucl. Cardiol. https:// doi.org/10.1007/s12350-019-02012-3. 178 Chan, R.H., Maron, B.J., Olivotto, I. et al. (2014). Prognostic value of quantitative contrast-enhanced cardiovascular magnetic resonance for the evaluation of sudden death risk in patients with hypertrophic cardiomyopathy. Circulation 130: 484–495. 179 Adabag, A.S., Maron, B.J., Appelbaum, E. et al. (2008). Occurrence and frequency of arrhythmias in hypertrophic cardiomyopathy in relation to delayed enhancement on cardiovascular magnetic resonance. J. Am. Coll. Cardiol. 51: 1369–1374.

The Role of Noninvasive Cardiac Imaging in the Management of Diseases of the Cardiovascular System

180 Terai, H., Shimizu, M., Ino, H. et al. (2003). Cardiac sympathetic nerve activity in patients with hypertrophic cardiomyopathy with malignant ventricular tachyarrhythmias. J. Nucl. Cardiol. 10: 304–310. 181 Sorajja, P., Chareonthaitawee, P., Ommen, S.R. et al. (2006). Prognostic utility of single-photon emission computed tomography in adult patients with hypertrophic cardiomyopathy. Am. Heart J. 151: 426–435. 182 Marcus, F.I., McKenna, W.J., Sherrill, D. et al. (2010). Diagnosis of arrhythmogenic right ventricular cardiomyopathy/dysplasia: proposed modification of the task force criteria. Circulation 121: 1533–1541. 183 Gertz, M.A., Benson, M.D., Dyck, P.J. et al. (2015). Diagnosis, prognosis, and therapy of transthyretin amyloidosis. J. Am. Coll. Cardiol. 66: 2451–2466. 184 Liao, R., Jain, M., Teller, P. et al. (2001). Infusion of light chains from patients with cardiac amyloidosis causes diastolic dysfunction in isolated mouse hearts. Circulation 104: 1594–1597. 185 Dorbala, S., Cuddy, S., and Falk, R.H. (20209). How to image cardiac amyloidosis: a practical approach. JACC Cardiovasc. Imaging 13 (6): 1368–1383. 186 Kwong, R.Y., Heydari, B., Abbasi, S. et al. (2015). Characterization of cardiac amyloidosis by atrial late gadolinium enhancement using contrast-enhanced cardiac magnetic resonance imaging and correlation with left atrial conduit and contractile function. Am. J. Cardiol. 116: 622–629. 187 Fontana, M., Banypersad, S.M., Treibel, T.A. et al. (2014). Native T1 mapping in transthyretin amyloidosis. JACC Cardiovasc. Imaging 7: 157–165. 188 Karamitsos, T.D., Piechnik, S.K., Banypersad, S.M. et al. (2013). Noncontrast T1 mapping for the diagnosis of cardiac amyloidosis. JACC Cardiovasc. Imaging 6: 488–497. 189 Fontana, M., Pica, S., Reant, P. et al. (2015). Prognostic value of late gadolinium enhancement cardiovascular magnetic resonance in cardiac amyloidosis. Circulation 132: 1570–1579. 190 Martinez-Naharro, A., Treibel, T.A., Abdel-Gadir, A. et al. (2017). Magnetic resonance in Transthyretin cardiac amyloidosis. J. Am. Coll. Cardiol. 70: 466–477. 191 Gillmore, J.D., Maurer, M.S., Falk, R.H. et al. (2016). Nonbiopsy diagnosis of cardiac Transthyretin amyloidosis. Circulation 133: 2404–2412. 192 Castano, A., Haq, M., Narotsky, D.L. et al. (2016). Multicenter study of planar technetium 99m pyrophosphate cardiac imaging: predicting survival for patients with ATTR cardiac amyloidosis. JAMA Cardiol. 1: 880–889. 193 Sperry, B.W., Vranian, M.N., Tower-Rader, A. et al. (2018). Regional variation in technetium pyrophosphate

194

195

196

197

198

199

200

201

202

203

204

205

uptake in Transthyretin cardiac amyloidosis and impact on mortality. JACC Cardiovasc. Imaging 11: 234–242. Pilebro, B., Arvidsson, S., Lindqvist, P. et al. (2018). Positron emission tomography (PET) utilizing Pittsburgh compound B (PIB) for detection of amyloid heart deposits in hereditary transthyretin amyloidosis (ATTR). J. Nucl. Cardiol. 25: 240–248. Roberts, W.C., Chung, M.S., Ko, J.M. et al. (2014). Morphologic features of cardiac sarcoidosis in native hearts of patients having cardiac transplantation. Am. J. Cardiol. 113: 706–712. Kandolin, R., Lehtonen, J., Airaksinen, J. et al. (2015). Cardiac sarcoidosis: epidemiology, characteristics, and outcome over 25 years in a nationwide study. Circulation 131: 624–632. Birnie, D.H., Sauer, W.H., Bogun, F. et al. (2014). HRS expert consensus statement on the diagnosis and management of arrhythmias associated with cardiac sarcoidosis. Heart Rhythm. 11: 1305–1323. Ardehali, H., Howard, D.L., Hariri, A. et al. (2005). A positive endomyocardial biopsy result for sarcoid is associated with poor prognosis in patients with initially unexplained cardiomyopathy. Am. Heart J. 150: 459–463. Greulich, S., Deluigi, C.C., Gloekler, S. et al. (2013). CMR imaging predicts death and other adverse events in suspected cardiac sarcoidosis. JACC Cardiovasc. Imaging 6: 501–511. Crawford, T., Mueller, G., Sarsam, S. et al. (2014). Magnetic resonance imaging for identifying patients with cardiac sarcoidosis and preserved or mildly reduced left ventricular function at risk of ventricular arrhythmias. Circ. Arrhythm. Electrophysiol. 7: 1109–1115. Hulten, E., Agarwal, V., Cahill, M. et al. (2016). Presence of late gadolinium enhancement by cardiac magnetic resonance among patients with suspected cardiac sarcoidosis is associated with adverse cardiovascular prognosis: a systematic review and meta-analysis. Circ Cardiovasc Imaging. 9: e005001. Coleman, G.C., Shaw, P.W., Balfour, P.C. Jr. et al. (2017). Prognostic value of myocardial scarring on CMR in patients with cardiac sarcoidosis. JACC Cardiovasc. Imaging 10: 411–420. Blankstein, R., Osborne, M., Naya, M. et al. (2014). Cardiac positron emission tomography enhances prognostic assessments of patients with suspected cardiac sarcoidosis. J. Am. Coll. Cardiol. 63: 329–336. Youssef, G., Leung, E., Mylonas, I. et al. (2012). The use of 18F-FDG PET in the diagnosis of cardiac sarcoidosis: a systematic review and metaanalysis including the Ontario experience. J. Nucl. Med. 53: 241–248. Ohira, H., Tsujino, I., Ishimaru, S. et al. (2008). Myocardial imaging with 18F-fluoro-2-deoxyglucose

283

284

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

206

207 208

209

210

211

212

213

214

215

216

217

positron emission tomography and magnetic resonance imaging in sarcoidosis. Eur. J. Nucl. Med. Mol. Imaging 35: 933–941. Vita, T., Okada, D.R., Veillet-Chowdhury, M. et al. (2018). Complementary value of cardiac magnetic resonance imaging and positron emission tomography/ computed tomography in the assessment of cardiac sarcoidosis. Circ. Cardiovasc. Imaging 11: e007030. Lindman, B.R., Clavel, M.A., Mathieu, P. et al. (2016). Calcific aortic stenosis. Nat. Rev. Dis. Primers 2: 16006. Baumgartner, H., Hung, J., Bermejo, J. et al. (2009). American Society of E and European Association of E. echocardiographic assessment of valve stenosis: EAE/ ASE recommendations for clinical practice. J. Am. Soc. Echocardiogr. 22: 1–23. quiz 101-2. Baumgartner, H., Falk, V., Bax, J.J. et al. (2017). 2017 ESC/EACTS guidelines for the management of valvular heart disease. Eur. Heart J. 38: 2739–2791. Kon, M.W., Myerson, S.G., Moat, N.E., and Pennell, D.J. (2004). Quantification of regurgitant fraction in mitral regurgitation by cardiovascular magnetic resonance: comparison of techniques. J. Heart Valve Dis. 13: 600–607. Harris, A.W., Krieger, E.V., Kim, M. et al. (2017). Cardiac magnetic resonance imaging versus transthoracic echocardiography for prediction of outcomes in chronic aortic or mitral regurgitation. Am. J. Cardiol. 119: 1074–1081. Cahill, T.J., Baddour, L.M., Habib, G. et al. (2017). Challenges in infective endocarditis. J. Am. Coll. Cardiol. 69: 325–344. Mahmood, M., Kendi, A.T., Ajmal, S. et al. (2019). Meta-analysis of 18F-FDG PET/CT in the diagnosis of infective endocarditis. J. Nucl. Cardiol. 26: 922–935. Calais, J., Touati, A., Grall, N. et al. (2019). Diagnostic impact of (18)F-fluorodeoxyglucose positron emission tomography/computed tomography and white blood cell SPECT/computed tomography in patients with suspected cardiac implantable electronic device chronic infection. Circ Cardiovasc Imaging. 12: e007188. Granados, U., Fuster, D., Pericas, J.M. et al. (2016). Diagnostic accuracy of 18F-FDG PET/CT in infective endocarditis and implantable cardiac electronic device infection: a cross-sectional study. J. Nucl. Med. 57: 1726–1732. Erba, P.A., Conti, U., Lazzeri, E. et al. (2012). Added value of 99mTc-HMPAO-labeled leukocyte SPECT/CT in the characterization and management of patients with infectious endocarditis. J. Nucl. Med. 53: 1235–1243. San, S., Ravis, E., Tessonier, L. et al. (2019). Prognostic value of (18)F-fluorodeoxyglucose positron emission tomography/computed tomography in infective endocarditis. J. Am. Coll. Cardiol. 74: 1031–1040.

218 Di Eusanio, M., Trimarchi, S., Patel, H.J. et al. (2013). Clinical presentation, management, and short-term outcome of patients with type a acute dissection complicated by mesenteric malperfusion: observations from the international registry of acute aortic dissection. J. Thorac. Cardiovasc. Surg. 145: 385–390e1. 219 Shiga, T., Wajima, Z., Apfel, C.C. et al. (2006). Diagnostic accuracy of transesophageal echocardiography, helical computed tomography, and magnetic resonance imaging for suspected thoracic aortic dissection: systematic review and meta-analysis. Arch. Intern. Med. 166: 1350–1356. 220 Golledge, J. and Eagle, K.A. (2008). Acute aortic dissection. Lancet 372: 55–66. 221 Sun, Z. and Cao, Y. (2010). Multislice CT virtual intravascular endoscopy of aortic dissection: a pictorial essay. World J. Radiol. 2: 440–448. 222 Kontopodis, N., Metaxa, E., Gionis, M. et al. (2013). Discrepancies in determination of abdominal aortic aneurysms maximum diameter and growth rate, using axial and orhtogonal computed tomography measurements. Eur. J. Radiol. 82: 1398–1403. 223 Janssen, S.P., Comans, E.H., Voskuyl, A.E. et al. (2008). Giant cell arteritis: heterogeneity in clinical presentation and imaging results. J. Vasc. Surg. 48: 1025–1031. 224 Seyahi, E. (2017). Takayasu arteritis: an update. Curr. Opin. Rheumatol. 29: 51–56. 225 Salvarani, C., Pipitone, N., Versari, A. et al. (2005). Positron emission tomography (PET): evaluation of chronic periaortitis. Arthritis Rheum. 53: 298–303. 226 Walter, M.A., Melzer, R.A., Schindler, C. et al. (2005). The value of [18F]FDG-PET in the diagnosis of largevessel vasculitis and the assessment of activity and extent of disease. Eur. J. Nucl. Med. Mol. Imaging 32: 674–681. 227 Besson, F.L., Parienti, J.J., Bienvenu, B. et al. (2011). Diagnostic performance of (1)(8)F-fluorodeoxyglucose positron emission tomography in giant cell arteritis: a systematic review and meta-analysis. Eur. J. Nucl. Med. Mol. Imaging 38: 1764–1772. 228 Zeina, A.R., Slobodin, G., Naschitz, J.E. et al. (2007). Isolated periaortitis: clinical and imaging characteristics. Vasc. Health Risk Manag. 3: 1083–1086. 229 Blockmans, D., Coudyzer, W., Vanderschueren, S. et al. (2008). Relationship between fluorodeoxyglucose uptake in the large vessels and late aortic diameter in giant cell arteritis. Rheumatology (Oxford) 47: 1179–1184. 230 Restrepo, C.S., Ocazionez, D., Suri, R., and Vargas, D. (2011). Aortitis: imaging spectrum of the infectious and inflammatory conditions of the aorta. Radiographics 31: 435–451. 231 Raman, S.V., Aneja, A., and Jarjour, W.N. (2012). CMR in inflammatory vasculitis. J. Cardiovasc. Magn. Reson. 14: 82.

285

9 Vascular System Ahmad Shariftabrizi1,2, Khalid Balawi1, and Janet H. Pollard1 1 2

University of Iowa Carver College of Medicine, Iowa City, IA, USA Veterans Affair Medical Center, Iowa City, IA, USA

Introduction Nuclear medicine offers an array of methods for imaging the pathophysiology of diseases affecting the vascular system. The spectrum of vascular disorders ranges from acute, life-threatening situations to chronic processes, and spans the entities of infection, inflammation, trauma, and oncologic pathologies. Diagnostic imaging of the vascular system is dominated by ultrasound (US), computed tomography (CT), magnetic resonance imaging (MRI), and conventional angiography. Nuclear medicine with planar imaging, single-photon emission computed tomography (SPECT), and positron emission tomography (PET) provides important qualitative and quantitative information about biologic processes, as well as dynamic information about vascular flow, tissue perfusion, and trends in tissue uptake of radiotracer. These modalities are highly sensitive and able to detect minute quantities of radiotracer in the nano- to picomolar range. Nuclear imaging may detect disease processes before morphologic changes are evident on other imaging modalities, allowing for earlier opportunities for therapy in some cases. PET and non-PET radionuclides differ in their production methods and decay times. Fluorinated PET radiopharmaceuticals such fluorine-18 (18F)-fluorodeoxyglucose (18F-FDG) and 18F-sodium fluoride (18F-NaF) require cyclotron isotope production. Given the relatively short half-life of 18F (110 minutes), imaging centers must be within an accessible distance of radiopharmacies with cyclotrons. The longer half-lives of the non-PET radiotracers such as technetium-99 m (6 hours) and indium-111 (67 hours) offer greater flexibility in availability, timing, and coordination.

Compared to planar and SPECT imaging, PET has the advantages of a higher limit of resolution (2 mm vs. 10 mm), attenuation correction, and quantitative capability [1]. At this time hybrid PET/CT systems are in widespread clinical practice. Hybrid SPECT/CT systems are also widely available. Hybrid PET/MRI systems are of growing interest in research and clinical settings, but their availability is still limited. In this chapter cases of vascular pathology with correlative nuclear and anatomic imaging will be discussed. Areas covered include inflammation, infection, neoplasia, molecular receptor expression, and physiologic processes such as osteogenesis.

­Imaging of Vasculitis Vasculitis is a heterogeneous group of multisystem diseases characterized by inflammation of the blood vessels. Noninfectious vasculitides are classified based on vessel size into categories of large, medium, and small vessel vasculitides (SVV). Vasculitides usually affect vessels in their respective category for size, however, any size vessel can be affected. For example, large vessel vasculitis affects the large vessels more often than other vasculitides. Examples include the large vessel vasculitides (LVV) Takayasu arteritis (TAK) and giant cell arteritis (GCA), the medium vessel vasculitides (MVV) polyarteritis nodosa (PAN) and Kawasaki disease, and the SVV granulomatosis with polyangiitis (GPA) (also known as Wegener’s), pauci-immune antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis, and immune complex small vessel vasculitis [2].

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach, First Edition. Edited by Ali Gholamrezanezhad, Majid Assadi, and Hossein Jadvar. © 2023 John Wiley & Sons Ltd. Published 2023 by John Wiley & Sons Ltd.

286

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Imaging of vasculitis includes US, CT, MRI, and MRI angiography (MRA), conventional catheter angiography, and nuclear imaging. US is a targeted imaging modality with a limited field of view that measures intima-media thickness and luminal changes in large and medium-sized vessels. Contrast-enhanced CT (CECT) offers greater anatomic coverage with excellent spatial resolution and multiplanar and three-dimensional viewing capabilities. CT can show changes in thickness of the vessel wall and lesion extent. CT also offers more global assessment of various organs, which can help with differential diagnosis and complications. MRI/MRA can identify vessel abnormalities such as wall edema, mural enhancement, and luminal changes, and can detect early wall thickening even before luminal narrowing occurs  [3]. Conventional angiography detects later stages of disease after stenosis or other luminal abnormalities have developed. As a whole-body imaging modality, nuclear imaging can visualize disease distribution in various anatomic regions simultaneously. In certain types of vasculitis, 18F-FDG PET uptake correlates with disease severity and has a role in the detection of disease, assessment of extent of disease, and noninvasive monitoring of response to treatment for some types of vasculitis  [4]. 18F-FDG PET has a limited role in the direct detection and management of small vessel vasculitis but can aid in the diagnosis of systemic manifestations and direct biopsy. Technical factors and patient preparation can affect the ability to detect vasculitis. Compared to other imaging modalities, FDG PET/CT gives limited anatomic information about the vessel lumen (e.g. stenosis, occlusion, or aneurysm) and has limited resolution in small vessels. Although high-resolution PET scanners are available that can detect vascular inflammation in relatively small arteries, such as the temporal, facial, and maxillary arteries, clinical availability is variable  [5]. Additional challenges with FDG PET are interference from glucocorticoids and need for controlled serum glucose. FDG PET should be performed before or no more than 3 days after initiation of glucocorticoid therapy. Sensitivity declines to one-third after 10 days of glucocorticoid therapy. Alternatively, steroids should be discontinued for at least 2 weeks prior to FDG PET if the patient is treated with >10 mg per day [4]. Technical parameters such as uptake time can affect the sensitivity of FDG PET as well. Image acquisition at 120 minutes rather than 60 minutes post-injection improves the target-to-background of vascular wall uptake  [5] (Figure 9.1). FDG uptake in the aorta and branch vessels that is equal to or greater than the liver background, or in the extracranial vessels of the head and neck that is greater than adjacent tissue, is suggestive of vasculitis. High FDG

60 minutes

120 minutes

180 minutes

Figure 9.1 Changes in aortic wall and luminal blood FDG activity at different imaging time points as seen on sagittal FDG PET images of the thoracic aorta. With time, luminal blood activity decreases while the aortic wall activity increases, which improves the arterial wall-to-blood contrast (superior target-to-back ground ratio). (Source: Reproduced from: Moghbel et al. [27].)

uptake in a continuous linear or homogenous segmental and often circumferential pattern is consistent with vasculitis. False positives can occur with atherosclerosis. Although the uptake patterns can overlap, in atherosclerosis uptake tends to be inhomogeneous, focal, or patchy with overall milder intensity compared to vasculitis. With higher resolution PET scanners, inflammation can be detected in the small cranial arteries such as temporal, facial, and maxillary arteries (Figure 9.2). PET can detect pathologic mimics of vasculitis such as infection, malignancy, and atherosclerosis. High FDG uptake in a homogenous segmental and often circumferential pattern is consistent with LVV. Atherosclerosis also causes FDG uptake and should be distinguished from LVV. However, atherosclerosis often has milder and more inhomogeneous focal or patchy uptake compared to LVV, which usually appears as intense, smooth linear often circumferential uptake (Figure  9.3). Although the two patterns can overlap, FDG uptake is usually symmetric in GCA. PET shows greater sensitivity for inflammation than MRI and can be used for disease monitoring [5]. Disadvantages of PET/CT include lack of anatomical information about luminal changes, radiation exposure, the need to be relatively free of glucocorticoids, and the need for optimal glucose concentration [5, 6].

Vascular System

(a)

(b)

Axillary artery

Aorta

Subclavian artery

Vertebral artery

(c) Maxillary artery External carotid artery (ECA)

Maxillary artery Temporal artery

Temporal artery Internal carotid artery (ICA) Occipital artery

Figure 9.2 The 18F-FDG PET/CT maximum intensity projection images of GCA patients. (a) Large vessel-giant cell arteritis (LV-GCA) with increased FDG uptake consistent with vasculitis in aorta, subclavian, axillary, and vertebral arteries. In the lower limb arteries, FDG intensity is not higher than liver FDG uptake and can be due to either vasculitis or arteriosclerosis. (b) Vertebral artery highintensity FDG uptake in a GCA patient. (c) A GCA patient with widespread cranial artery involvement. Increased FDG uptake higher than surrounding tissue is recognized in temporal, maxillary, occipital, internal, and external carotid artery. (Source: Reproduced from: Schmidt and Nielsen [5].)

­Imaging of Large Vessel Vasculitis US is currently the recommended modality for the initial diagnosis of GCA. A standard US examination in GCA should include temporal and axillary arteries. Measuring the intima-media thickness in the temporal and axillary arteries allows rapid diagnosis of GCA, with reported sensitivity and specificity of 77% and 96%, respectively. The European League Against Rheumatism (EULAR) recommends that if US is unavailable or inconclusive, MRI of the cranial arteries maybe used. In a recent metanalysis of six studies, the pooled sensitivity and specificity for MRA were 73% and 88%, respectively, when using the clinical diagnosis as the reference standard [7]. Use of the MRA of cranial arteries is justified since it is now well-documented that vasculitis is not limited to extracerebral arteries but may also extend to intracerebral arteries [6]. EULAR proposes MRA as the first imaging modality for the diagnosis of TKA. MRA assesses the vessel wall and luminal changes and distribution of the involved vessels. MRA sensitivity and specificity can be as high as 98% and 100%, respectively. Most commonly the left subclavian and common carotid arteries are involved. US can be used to examine the carotid, subclavian, and vertebral arteries, and

abdominal aorta. Meta-analysis of the role of US reported sensitivity and specificity of 81% and >90%, respectively, using clinical criteria and/or angiography as standard. In both TAK and GCA, MRA delineates the typical arterial mural abnormalities, provides an overview of involved arteries (both intracranially and extracranially), and is superior to US in examining the aorta [7]. Compared with MRA, FDG PET offers useful and complementary information about large vessel vasculitis. FDG PET/CT is considered a reliable imaging modality for the confirmation of LVV diagnosis. FDG PET detects inflammation in the affected arteries through detection of increased metabolism in the inflamed vessels. FDG PET/CT provides an overview of the disease extent in suspected cases. If the diagnosis is uncertain, PET/CT can evaluate alternative diagnosis such as tumors, lymphoma, or septic foci. On MRI, edema, thickening, and contrast enhancement of the vessel wall suggest active disease [6]; however, PET image findings seem to better correlate with the disease activity assessed clinically compared to the MRI findings. An additional advantage of PET is that both cranial and large vessels can be imaged in one session. On the other hand, luminal abnormalities such as stenosis, occlusion, and aneurysm can

287

288

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

Figure 9.3 (a) Axial PET and fused PET/CT images for a 34-year-old male patient with history of large vessel vasculitis of the aorta externally compressing the coronary ostia showing diffusely increased mural uptake in the anterior aortic arch (blue arrows) and ascending aorta (green arrows). (b) Axial PET and fused PET/CT images showing increased mural uptake in the proximal left common carotid artery (blue arrows) and to a lesser extent the proximal innominate artery (green arrows). (c) Axial CT angiography images of the same patient demonstrating diffuse wall thickening of the aortic arch (blue arrow), ascending aorta (green arrow), proximal left common carotid artery (yellow arrow), and proximal left innominate artery (red arrow).

Vascular System

be evaluated by MRI, but are not well-evaluated on noncontrast-enhanced FDG PET/CT. Combining FDG PET and MRI in the form of hybrid PET/MRI provides simultaneous detailed morphological information of mural abnormalities together with metabolic activity. Arterial FDG quantification in LVV patients is comparable on PET/CT and PET/MRI, but disease extent on PET/MRI correlates better with the blood markers of systemic inflammation [5]. At initial diagnosis, studies have shown that visual analysis alone is superior or equivalent to semiquantitative methods [7]. A scoring methodology – PETVAS score – has been proposed, and has been shown to correlate with the incidence of relapse. The score is derived from a visual

assessment of uptake intensity using a scale comparing to liver (0 = no uptake, 1 = less than liver, 2 = equal to liver, and 3  =  greater than the liver) in designated vascular segments (four segments of aorta, including ascending, arch, descending, thoracic and abdominal aorta, and in 11 branch arteries, including innominate, carotids, subclavians, axillaries, iliacs, and femorals). The final PETVAS score is then calculated as the sum of scores for each arterial segment [8]. The distribution of vessel involvement in GCA is the subclavian (73%), abdominal, and thoracic aorta (50%), with the axillary, carotid, iliac, and femoral arteries showing abnormal FDG uptake in 30–40% of cases (Figure 9.4). Specificity of FDG uptake is typically higher

(a)

(b)

(c)

Figure 9.4 (a) Maximum intensity projection (MIP) of an FDG PET/CT study in a patient with GCA showing diffusely increased mural uptake in the bilateral subclavian arteries (purple arrows), axillary arteries (green arrows), and brachial arteries (blue arrows). (b) Axial PET and fused PET/CT images in the same patient showing increased mural uptake in the aortic arch (purple arrows), ascending aorta (blue arrows), and descending aorta (green arrows). (c) Additional coronal PET and fused PET/CT images also showing diffuse mural uptake in the bilateral common carotid arteries (purple arrows). (d) Coronal PET and fused PET/CT images in a patient with GCA showing diffuse mural uptake in the descending thoracic aorta (arrows). (e) Axial PET and fused PET/CT images in the same patient showing patchy mural uptake in the right axillary artery (arrows). (f) Coronal PET and fused PET/CT images of a follow-up FDG study done 21 months later after receiving steroid therapy showing interval decrease in the intensity of uptake in the descending aorta wall indicating treatment response (arrows). (g) Axial PET and fused PET/CT images of the same follow-up FDG study showing resolution of the uptake in the proximal right axillary artery (arrows).

289

290

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(d)

(e)

(f)

(g)

Figure 9.4 (Continued)

in supra-aortic arteries and lower in the aorta and lower extremities as these arteries are more prone to atherosclerosis, which can also show uptake. The sensitivity and specificity of PET/CT in the initial diagnosis of GCA are reported to be as high as 92% and 85%, respectively, with temporal artery biopsy as a reference [6]. A significant number of patients with polymyalgia rheumatica (PMR) show increased large vessel vascular uptake suggestive of subclinical vasculitis. FDG PET imaging in PMR reveals perisynovitis with pathologic uptake in the soft tissues and ligaments around the joints and attachment sites of the shoulders and hips, lumbar and cervical spinous processes, and ischial tuberosities. In patients with GCA and PMR, increased uptake in the shoulders is seen and FDG uptake in the subclavian or axillary arteries is not

more intense. When GCA is excluded, increased FDG uptake is seen in most patients and in the spinous processes of cervical/lumbar spine in half of the patients [7] (Figure 9.5). As PMR may be paraneoplastic, FDG PET can also identify the underlying malignancy [9]. FDG PET has a role in both the initial diagnosis and disease activity assessment in TAK [7]. Biopsies are not usually possible in most arteries involved in TAK, which heightens the need for accurate noninvasive diagnostic techniques [6]. In a metanalysis, FDG PET showed sensitivity and specificity of 87% and 74%, respectively, in differentiating active from inactive TAK [7]. TAK and GCA can have similar vascular distribution  [10]. In TAK, the subclavian and carotid arteries and the aorta are most affected; involvement of abdominal aorta, and mesenteric

Vascular System

Figure 9.5 A 71-year-old woman was examined for nuchal pain, fatigue, and 10 kg weight loss in 5.5 months. She also complained of pain in the upper arms and shoulders, and she had noticed loss of strength in the shoulder and pelvic girdle, and jaw claudication. On clinical examination, blood pressure was equal in both arms, the left temporal artery was thickened with absent pulsations, the strength of the shoulder muscles was markedly diminished, and abduction of shoulders and hips was painful and difficult. Blood examination revealed a CRP level of 74 mg/L, ESR of 31 mm/h, hemoglobin level of 10.5 g/dL, and slightly increased alkaline phosphatase. Chest X-ray and abdominal ultrasound results were normal. Clinical and biochemical features were suspicious for GCA and PMR. Vasculitis was confirmed by a positive temporal artery biopsy. FDG PET showed the typical features of both closely related conditions in this single patient: a strongly increased FDG uptake in the vessel walls of the carotid and subclavian arteries, and the aorta, revealing GCA, as well as an increased FDG uptake around both shoulders, hips, and between the spinous processes of the cervical and lumbar spine, revealing PMR. (Source: Reproduced from: Schmidt and Blockmans [7])

and renal arteries is common [11]. In a recent metanalysis of FDG PET analyzing 21 studies with 413 patients in total, pooled sensitivity and specificity for GCA were 90% and 98%, respectively, and the sensitivity for TAK was lower at 84%  [12]. The lower sensitivity for TAK may be because these patients tend to present in more chronic phase of disease when vascular wall inflammation has subsided. Intensity of FDG uptake in the thoracic aorta is an independent risk factor for complications of vasculitis such as aortic dilatation and aneurysm formation. In GCA, large vessel involvement is the strongest predictor of aortic dilatation. Patients with aortic complications of dilatation, aneurysm, and dissection nearly always have inflammation evident on the preceding FDG PET [7]. FDG PET/CT can be used for monitoring response to therapy in various vasculitides. FDG uptake decreases in response to successful treatment 4–12 weeks post therapy initiation [12]. Intensity of vascular uptake may decrease with glucocorticoid treatment; however, long-term vascular uptake may still be seen despite clinical remission, which has been ascribed to vascular remodeling. Investigating the role of FDG PET for assessing disease activity in TAK is difficult considering the lack of gold standard in this disease condition [6].

FDG PET/MR is an attractive option for monitoring inflammatory processes in LVV. Because the two modalities detect different manifestations of inflammation, together PET/MR offers a comprehensive means of scoring the total burden of vasculitis. In TAK, combining PET and MR information has shown to determine disease stage based on mural and lumen morphology and inflammatory features [4].

I­ maging of Medium and Small Vessel Vasculitis PAN is a systemic necrotizing vasculitis involving medium and, to a lesser extent, small arteries. PAN affects multiple systems including the kidneys, nervous system, gastrointestinal tract, and skin. FDG PET can detect vasculitis in PAN and also nonvascular findings. For example, in cutaneous PAN FDG PET can sensitively detect disseminated foci throughout the subcutaneous and muscle tissue. FDG PET can detect chronic peri-aortitis as an isolated finding or as part of IgG4 disease in which other organs such as pancreas, bile ducts, and lacrimal and salivary glands may  also show abnormal uptake. In Behcet’s disease,

291

292

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(c)

(b)

(d)

pulmonary artery aneurysm and vasculitis of other large arteries can be seen on PET [7]. PET/CT and PET/MR have a limited role in the initial diagnosis of small vessel vasculitis, but have significant potential in the assessment of inflammation and tissue damage caused by these diseases. For example, in GPA, FDG PET/CT imaging improves the detection of upper respiratory tract and lung lesions, aids in determining the site for biopsy, and detects undiagnosed involvement of large vessels (Figure  9.6). In ANCA-associated vasculitis from malignancy and infection FDG PET/CT can aid in differential diagnosis of disease relapse [4, 7].

­Imaging of Aortitis Aortitis, the condition of inflammation in the aortic wall, can be divided into two categories: infectious and noninfectious. Salmonellae species and Staphylococcus aureus are the most common pathogens causing aortitis  [13]. Anatomic imaging with CT or MRI is the first-line imaging choice in emergent and nonemergent evaluation of the aortic wall and lumen, and evaluation of extent of disease. CT in particular can rapidly distinguish conditions that mimic aortitis, such as aortic dissection (AD), intramural

Figure 9.6 A 67-year-old woman with a granulomatosis with polyangiitis. FDG PET/ CT scan shows increased FDG uptake in sinonasal and kidney locations (a and b, arrows). A follow-up FDG PET/CT scan, although the patient achieved remission, showed resolution of hypermetabolic activities in both locations (c and d).

hematoma, and penetrating aortic ulcer. Imaging findings on CT include aortic wall thickening with or without soft tissue mass, fluid collections, fat stranding, saccular aneurysms, and air within the aortic wall  [14]. When clinical suspicion is high, FDG PET/CT can reliably distinguish between infected and noninfected cases  [15], with a suggested SUVmax of 4.5  in infected cases and SUVmax 1–2 cm solid tumor

T1c

2.1–3 cm solid part within part-solid tumor total size 3 cm >2–3 cm solid tumor

T2a

3.1–4 cm

Involves main bronchus without involvement of carina

T2b

4.1–5 cm

Total/partial atelectasis Total/partial pneumonitis Involves hilar fat Involves visceral pleura (PL1 or PL2)

T3

5.1–7 cm

Table 11.2  N classification for appearance on chest CT for TNM, 8th edition. N classification

N component on CT

N0

No lymph node metastasis

N1

Ipsilateral peripheral, intrapulmonary or hilar nodes metastasis

N2

Ipsilateral mediastinal (upper, porticopulmonary, lower), subcarinal nodes metastasis

N3

Ipsilateral or contralateral supraclavicular/ scalene lymph node or contralateral mediastinal, hilar/interlobar, or peripheral nodes metastasis

Table 11.3  M classification for appearance on chest CT for TNM, 8th edition. M classification

M0 M1

M1a

Intrathoracic metastasis Pericardial effusion

Separate tumor nodules in the same lobe as the primary

Contralateral lung nodules/ pleural nodules

Parietal pericardium Chest wall Phrenic nerve >7 cm

No distal metastasis Pleural effusion

Involves parietal pleura (PL3)

T4

M component on CT

M1b

Single extrathoracic metastasis in a single organ

M1c

Multiple extrathoracic metastasis

Involves diaphragm Mediastinal fat or other mediastinal structures (trachea, great vessels, heart, recurrent laryngeal nerve, esophagus) Carina Vertebral body Visceral pericardium Separate tumor nodules in the same lung but different lobes as the primary

CT, computed tomography; AIS, adenocarcinoma in situ; GGN, ground-glass nodules; mi, minimally invasive; PL1, tumor invasion of the elastic layer of visceral pleura without reaching the visceral pleural surface; PL2, tumor invasion of the visceral pleural surface; PL3, tumor invasion of the parietal pleura or chest wall. T classification for appearance on chest CT for TNM 8th edition.

CT and FDG PET/CT for identifying mediastinal and chest wall invasion, and also evaluating the heart, pericardium, and vascular structures  [22, 23]. Some of the characteristic parameters of chest wall invasion in MRI include infiltration of the normal extrapleural fat plane

(on T1-weighted view), hyperintensity of the parietal pleura (on T2-weighted view), tumor fixation to the thoracic wall (on cine MRI), and evidence of rib destruction (on short tau inversion recovery [STIR] sequences). MRI is also superior to CT in differentiating the primary neoplasm and postobstructive atelectasis/pneumonitis as usually shows a higher signal on T2-weighted imaging than the central tumor. N Staging

The N descriptor in the lung cancer TNM staging system is based on the metastatic lymph node locations. In this regard, CT and PET are the mainstays of noninvasive assessment of mediastinal nodes regarding tumoral involvement in lung cancers. On the structural imaging (CT and MRI), nodal metastases are primarily defined as enlarged lymph nodes measuring more than 1 cm in the short axis or presenting abnormal shape and configuration, attenuation, or central necrosis [24, 25]. However, based on this criterion, CT is reported to be relatively inaccurate for the detection of pathologic

Thoracic Malignancies

(a)

(b)

(c)

(d)

Figure 11.2  Lung cancer staging. A 57-year-old man with histopathologically proven squamous cell carcinoma of the right lung. Staging FDG PET/CT shows a large heterogeneously hypermetabolic irregular soft tissue mass with SUVmax of 10.94 in the upper lobe of the right lung. Medially, the mass is infiltrating the right main bronchus and mediastinum (a and b, arrow) and laterally encasing the right first to fourth ribs with sclerosis and lysis of the ribs (c and d, arrows).

Figure 11.3  Lung cancer staging. FDG PET/CT in a known case of lung adenocarcinoma shows a large hypermetabolic heterogenous soft tissue mass in the left lung abutting the left chest wall; however, chest wall or rib infiltration cannot be accurately commented on in FDG PET/CT. (Teaching point: local disease infiltration cannot be accurately assessed on FDG PET/CT.)

337

338

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

(d)

(e)

(f)

(g)

Figure 11.4  Lung cancer staging. A 65-year-old male with adenocarcinoma of the right lung (a). FDG PET/CT demonstrates an intensely hypermetabolic soft tissue density mass lesion in the right upper lobe lung (b and c, arrows), abutting the right upper lobe bronchus with associated distal collapse of the right upper lobe lung (b and c, arrowheads). The mediastinum is shifted toward the right side. There is associated metastatic hypermetabolic pleural-based and parenchymal left pulmonary nodules (not shown), mediastinal nodes (b, c), and bilateral supraclavicular lymph nodes (d, e). Furthermore, a hypermetabolic lesion is seen in the L2 vertebral body with SUVmax 22.49, indicative of distant metastatic disease, with no associated CT abnormality (f, g, arrows). (Teaching point: FDG PET/CT helps in differentiating tumor from the atelectatic changes and improves patient management in radiotherapy planning. Furthermore, metastatic disease can be detected earlier than with conventional modalities like CT).

nodes, with sensitivity and specificity of 40–84% and 57–94%, respectively. Forty percent of CT-positive lymph nodes have been proved to represent benign processes, while 20% of normal-shaped nodes on CT scan have been found to be of a metastatic nature. PET is superior to CT for the identification of mediastinal nodal metastasis (Figures 11.4–11.6). In a meta-analysis with 2865 patients, PET has a sensitivity of 74% and specificity of 85% for the detection of nodal metastatic disease in NSCLC [26]. Nevertheless, PET has a lower spatial resolution than CT. Moreover, PET alone does not provide accurate anatomic localization. Contrast-CT and FDG PET/CT imaging disclose the complementary aspects of N staging in lung cancer. While CT illustrates the detailed anatomic information of the pathologic nodes, the functional results of PET scans illustrate their metabolic activity. Therefore, the accuracy

of integrated PET/CT is at least the same as PET alone for N staging in lung cancers. A study has found that 18FFDG PET/CT has higher diagnostic accuracy than CT or MRI to identify the mediastinal lymph node metastasis (with sensitivity and specificity being 58–94% and 76–96%, respectively)  [27]. However, benign entities (including inflammatory, granulomatous, and infectious) can also cause positive PET results. Thus, while integrated PET/ CT is routinely used in the clinical staging of NSCLC [28], tissue biopsy confirmation of suspicious lesions is recommended for patients being considered for surgery [29]. Diffusion-weighted MRI has also been suggested as an alternative to FDG PET in identifying metastatic mediastinal lymph nodes in patients with NSCLC [30], but further research is required before drawing a definite conclusion in this regard.

(a)

(b)

(c)

(d)

Figure 11.5  Lung cancer N staging. A 67-year-old male with left lung carcinoma. On FDG PET/CT, there is a hypermetabolic spiculated soft tissue mass in the apico-posterior segment of the left upper lobe lung (a and b, arrows). There are a few enlarged mediastinal lymph nodes with no significant hypermetabolism (c and d, arrows); however, metastatic involvement of these nodes cannot be ruled out.

(a)

(b)

(c)

(d)

(e)

(f)

Figure 11.6  Lung cancer staging. A 63-year-old female with poorly differentiated adenocarcinoma of the lung. FDG PET/CT for initial staging shows a hypermetabolic malignant irregular soft tissue density mass in the right hilar region in the middle lobe of the right lung (a and b, arrows), as well as few hypermetabolic metastatic mediastinal lymph nodes (a and b, arrowheads). In addition, there is a 3.7× 2.5 cm mass in the right adrenal gland with hypermetabolism, confirming metastatic disease (c and d, arrows). Furthermore, there is a small lesion in the left supraspinatus muscle showing focal FDG uptake, indicative of metastatic muscle deposits (e and f, arrows).

340

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

M Staging

Distant metastasis (M1) classifies the lung cancer patient as stage IV, which removes the consideration for curative surgical treatment. The most common metastatic sites in NSCLC are bones, adrenal glands, brain, and liver. Initial CT scan can detect the liver and adrenal metastases with sensitivity and specificity of 93% and 75%, and 41% and 84.5%, respectively. According to NICE guidelines, if the initial staging CT investigations suggest a curable disease, FDG PET/CT scan is indicated to fully stage the cancer, detect metastases outside the range of the staging CT scan, and identify the  FDG-avidity of any equivocal findings on the staging CT which would alter the stage, such as additional lung lesions, adrenal or nodal metastases etc. [31] (Figures 11.4 and  11.6). PET/CT has the advantage of evaluating the whole body in one examination, which makes it a desired method for M staging, particularly for soft tissue and pleural metastases. In addition, it is superior to nuclear medicine bone scintigraphy for detecting osseous metastases [32]. However, FDG PET is of limited ability to detect brain metastasis. For brain and liver metastases, MRI is preferred over other imaging modalities. MRI can also be used to diag(a)

(b)

(c)

(d)

nose adrenal metastasis. MRI is potentially able to differentiate adrenal metastases from adenomas, with reported sensitivity and specificity of 100% and 81%, respectively  [33]. Other studies have also proposed whole-body MRI as an alternative to FDG PET/CT with a slightly increased sensitivity and specificity in the identification of head and neck, and bone metastases [31]. Furthermore, false-positive nonspecific uptake should be differentiated from metastatic disease in whole-body FDG PET study (Figure 11.7).

Evaluation of Response to Treatment Conventional imaging (such as CT) has been traditionally applied to evaluate tumor response after chemotherapy or radiotherapy, based on changing size before and after treatment using WHO [34] and RECIST [35] criterion. However, FDG PET is potentially able to evaluate the early therapy response before structural changes occur (Figures  11.8 and  11.9). Moreover, the pattern of FDG uptake on PET examination might be a better indicator for monitoring of the therapy response, as it is able to distinguish necrotic or fibrotic tissue from viable tumor cells. Indeed, only viable tumor cells display FDG accumulation, and not necrotic

Figure 11.7  Lung cancer staging. A known case of adenocarcinoma of the right lung. FDG PET/CT shows a hypermetabolic ill-defined soft tissue density mass in right side of the mediastinum extending from the superior mediastinum across the right paratracheal region up to the precarinal region (a, b), indicative of the primary site. Hypermetabolism is also seen in the left vocal cord (arrow in c, d), mostly due to right vocal cord palsy, secondary to the involvement of the right recurrent laryngeal nerve by the right lung mass. (Teaching point: False-positive uptake may be seen in unilateral vocal cord due to involvement of the opposite side recurrent laryngeal nerve by the primary and should be interpreted cautiously.)

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

(j)

Figure 11.8  Lung cancer, response assessment. Known metastatic left lung carcinoma, status post chemotherapy. The left panel shows pretreatment FDG PET/CT images and the right panel shows post-treatment images. (a–d) The left lower lobe lung mass shows no significant interval change in size (36 × 28 mm vs. 36 × 30 mm); however, there is interval increased metabolic activity (SUVmax 3.8 vs. 7.5). (e, f) Enlarged paraaortic lymph node with interval increase in metabolic activity (SUVmax 4.1 vs. 7.1), but no significant change in size (38 × 26 mm vs. 39 × 28 mm). (g–j) Increase in size and metabolic activity in the right third rib anteriorly (SUVmax 5.5 vs. 10.4). Overall, metabolic disease progression is suggested, whereas morphologically there is no significant change in size of lesions.

342

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

(d)

Figure 11.9  Lung cancer, response assessment. Known squamous cell carcinoma of the right lung, treated with concurrent chemoradiation. The left panel shows pretreatment FDG PET/CT images and the right panel shows post-treatment images. (a, b) Pre- and post-treatment images show significant regression in size and metabolic activity of the primary due to local radiation treatment (arrows). (c, d) A new mildly hypermetabolic 9-mm paraaortic lymph node is noted (arrow), concerning for metastasis that may have been missed by CT size criteria.

tissues  [36, 37]. This issue makes FDG PET a desirable method for response evaluation in the majority of patients. It is reported that a reduction in metabolic activity after one cycle of chemotherapy is closely correlated with the final therapy outcome. Better prediction of therapy outcomes could help prevent unnecessary toxic treatments in  early respondent individuals and prompt second-line therapy in nonresponders.

Residual Disease and Recurrence Post-therapy anatomical distortion found in conventional radiological studies could be difficult to characterize as viable recurrent tumor versus fibrotic tissue. However, FDG PET imaging has a high diagnostic ability in differentiating nonspecific post-treatment changes from local recurrent, as the scar tissue reveals insignificant metabolic activity on PET images. The SUV values were significantly higher in tumoral tissue compared to the post-therapy changes [38, 39].

Small Cell Lung Cancer SCLCs make up about 20% of all lung cancers. A distinct two-stage system has been established for SCLC: limited stage, defined as a disease in only one hemithorax, and

extensive stage, defined as disease outside one hemithorax. However, the TNM staging system can still be utilized in SCLC, with better differentiation of stage-specific survival  [40]. Indeed, the prognostic value of TNM staging in  SCLC patients might offer beneficial information for therapy management. Some crucial radiologic information  should be assessed in the evaluation of SCLC cases, including the number of distant metastatic locations and involved organs (Figure 11.10), the diameter of metastatic lesions, and whether patients with brain metastases are symptomatic or not [41].

Summary After the initial diagnosis of NSCLC, most patients are further assessed with contrast-enhanced CT. Additional imaging modalities such as FDG PET/CT and MRI may be obtained depending on the suspected clinical stage and/or further evaluation of specific findings. Integrated PET/CT improves the staging of lung cancers through the detection of T4 and M1 disease. Moreover, PET/CT offers a higher diagnostic value to detect metastatic lymph nodes (N staging). In terms of prognostic implications, the SUV holds additional information. FDG PET provides information on tumor metabolic activity, while the CT component references the corresponding anatomic map. Therefore, PET/

Thoracic Malignancies

(a)

(b)

(c)

(d)

(e)

(f)

(g)

Figure 11.10  Small cell lung cancer. A 60-year-old male with small cell carcinoma of the left lung. The maximum intensity projection image of FDG PET (a) shows multiple foci of uptake in the body. Some of the representative axial sections show a hypermetabolic irregular soft tissue mass lesion in the left lower lobe lung (SUVmax 11.45) abutting the left major fissure and left hilum, suggestive of the primary malignancy (b, c), accompanied by a hypermetabolic small soft tissue nodule measuring 10 mm in the lateral limb of the left adrenal gland with SUVmax of 6.73 confirming metastatic involvement (d, e), and hypermetabolic foci in the pelvic bones with no corresponding CT changes indicative of marrow metastases (f, g).

CT yields high diagnostic accuracy and should be applied as a routine preoperative diagnostic care in patients with NSCLS.

Pleural Diseases Most pleural malignancies are of metastatic origins, likely secondary to lung and breast cancers, lymphoma, and gastric carcinoma. Primary neoplasm of the pleura is less common, with the most common being malignant mesothelioma. Chest radiography: In most patients with malignant pleural disease, CRs reveal pleural effusion with or without pleural thickening. Radiographic features of pleural involvement by cancers include circumferential lobulated pleural thickening, rib crowding, and elevation of the hemidiaphragm (because of volume loss due to neoplastic airway obstruction) [42]. Thoracic ultrasonography (TUS): TUS has been offered to identify malignant pleural effusion with acceptable

accuracy (sensitivity 73%, specificity 100%) [43]. On TUS, hypoechoic pleural thickening >1 cm, especially with irregular or unclear borders, pleural nodularity, diaphragmatic thickening >7 mm, and invasion of pleural masses into adjacent structures is highly suggestive of malignant pleural disease. Moreover, the echogenic swirling patterns within an effusion on real-time chest sonography indicate cellular debris and predict malignant effusion in patients with underlying malignancies  [44]. Whereas the hypoechoic pleural thickening with undefined irregular borders suggests mesothelioma, the metastatic pleural lesions might present as nodular, circular, hemispheric, or broadbased lesions, with frond-like protrusions into the pleural space. CT scan: Contrast-enhanced chest CT is currently the imaging modality of choice to evaluate pleural effusion for  malignancy. The characteristic features of malignant pleural disease on CT scan include nodular or circumferential pleural thickening, parietal pleural thickening >1 cm, mediastinal pleural involvement, or evidence of a primary  tumor  [45]. Chest CT abnormalities predictive of

343

344

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

mesothelioma include pleural thickening >1 cm and interlobar fissure involvement. In patients with lung cancer, any subtle pleural thickening or nodularity on chest CT should raise the possibility of pleural metastases, even in the absence of pleural effusion [46]. MRI: The ability of MRI is at least the same as chest CT for diagnosing malignant pleural disease  [47]. MRI signal intensity is a valuable additional radiologic feature when encountering equivocal pleural cases. Furthermore, MRI is potentially able to provide excellent soft tissue imaging, which enables the evaluation of diaphragmatic and chest wall invasion. MRI with triple echo pulse sequences is also highly sensitive for identifying small pleural effusions and can differentiate between exudates and transudates  [48]. However, despite these strong capacities, contrast-enhanced CT remains the preferred initial study in suspected pleural malignancy, and chest MRI is reserved for challenging and complex cases. Indeed, in the subgroup of lesions misinterpreted on CT images, MR signal intensity on long-repetition time images is able to differentiate the benign and malignant entities correctly. FDG PET: FDG PET has also been utilized for identifying malignant pleural effusion and thickening. A metaanalysis by Parcel et al. reported a moderate ability for FDG PET/CT using semiquantitative parameters for categorizing the pleural effusion as malignant nature [49]. The ability of FDG PET to discriminate malignant from benign

(a)

pleural effusion relies on differential FDG uptake in malignant pleural lesions. The focal increased FDG accumulation in the pleura usually corresponds to solid pleural abnormalities on CT views (Figure  11.11). However, it is worth noting that false-positive FDG PET results caused by  hypermetabolic inflammatory or infectious processes, such as those seen in para-pneumonic effusions, can be misleading, which may warrant further tissue sampling.

Mediastinal Tumors Tumors in the anterior mediastinum (a prevascular component of the mediastinum) are rare, and imaging plays a pivotal role in evaluating such tumors (detection, differential diagnosis, staging, and follow-up). The most common causes of anterior mediastinal (prevascular) masses include thymoma, thyroid disease, lymphoma, and teratoma. In many cases, an anterior mediastinal lesion is only detected as an incidental finding on either CR or CT scans. Hence, it  is of paramount importance to recognize the specific radiologic features of such masses seen in routine practice. Chest radiography: As the most available imaging method, simple chest radiography is the initial imaging tool in patients presenting with newly developed symptoms. Due to insufficient sensitivity and specificity of chest X-rays, once the patients have been detected to have

(b)

(c)

(d)

(e)

Figure 11.11  Mesothelioma. (a) FDG PET/CT shows hypermetabolic circumferential and nodular left pleural thickening with mediastinal and left chest wall invasion (arrows; b, c, d, e) and SUVmax of 8.64, suggestive of mesothelioma.

Thoracic Malignancies

mediastinal masses, they should undergo further imaging methods, usually CT scans. CT scan: In patients with initial suspicious findings of a mediastinal lesion on chest X-ray, further investigation with CT scan is warranted. A contrast-enhanced CT scan is the imaging method of choice to further investigate such masses due to its high spatial and temporal resolution. Additionally, a CT scan enables the characterization of tissue components in mediastinal masses, which is important for the differential diagnosis of these tumors. In a study by Tomiyama on 127 cases with pathologically verified anterior mediastinal tumors, the first-choice CT diagnosis was correct in 61% of patients, whereas the first-choice MRI diagnosis was correct in only 56% [50]. Furthermore, a CT scan allows for the staging of anterior mediastinal tumors, as it could determine the local invasion and lymph node status. MRI: While CT scan is the imaging modality of choice in evaluating anterior mediastinal tumors, MRI has been shown to be a useful problem-solving method in challenging cases [51]. MRI is particularly valuable in the characterization of thymic cysts with hemorrhage or inflammation (which could mimic solid tumors despite low enhancement). Indeed, further evaluation with MRI or a biopsy is indicated when there are no pathognomonic CT features. Given its superior contrast resolution and lack of ionizing radiation, MRI is particularly helpful when encountering indeterminate lesions on CT and CR, particularly in detecting cystic lesions, fat in a tumor, and vascular diseases, which CT struggles to explain. MRI is potentially useful in distinguishing the normal thymus, thymic hyperplasia, and neoplastic lesions, as well as vascular invasion. It is also helpful for detecting and characterization of cystic lesions and discerning thymic hyperplasia from other thymic masses. For example, whereas thymic hyperplasia shows reduced signal intensity on chemical-shift MRI, other thymic tumors do not show this decrease in signal. Furthermore, in mature teratoma, MRI may follow CT when more information is required. FDG PET/CT: The diagnostic role of FDG PET/CT in the evaluation of anterior mediastinal tumors has not yet been fully established. FDG PET/CT is not routinely performed to characterize an anterior mediastinal mass, but it may be applied to further stage some specific malignancies and to monitor the therapy response. However, it should be kept in mind that, although higher FDG uptake indicates malignancy, diagnostic tissue biopsy is still warranted due to false-positive and false-negative results seen on FDG PET imaging. Some low FDG-avid masses are still malignant, while a mass showing high FDG uptake might be benign. FDG PET may also be misleading given the high FDG uptake frequently seen in the normal or benign

thymic hyperplasia and inflammatory lesions in the mediastinum. Therefore, PET/CT is mainly reserved for patients diagnosed with lymphoma and for the detection of distant metastases. Currently, FDG PET/CT has become the modality of choice for staging lymphoma [52].

Thymoma The thymus gland consists of two cell types: epithelial and lymphoid cells. Thymic epithelial tumors are primarily classified into thymoma, thymic carcinoma, and neuroendocrine tumors of the thymus. Thymoma is a rare cancer but is still the most common primary neoplasm of the anterior mediastinum, accounting for more than 35% of anterior mediastinal masses [53]. They present as slow-growing neoplasms and most commonly occur in the fourth or fifth decades of life. With the increasing use of chest imaging, a growing number of asymptomatic patients have been discovered incidentally. While several thymoma cases present with systemic syndromes (such as myasthenia gravis), thymic carcinoma is clinically silent for a long time before being diagnosed. They usually become symptomatic in later stages, manifested by local compression or distant metastasis. Although there is no pathognomonic feature for definite differentiation between thymoma and thymic carcinoma, some imaging features may help to narrow the differential diagnosis or be suggestive of a certain element in each pathology. Chest radiography: Initial investigation of thymoma commonly starts with a CR, followed by a CT scan for further characterization and staging. Rarely, in selected cases, additional imaging is used, such as MRI or nuclear medicine studies. Thymomas are frequently visible on chest X-rays, as a well-marginated lobulated or ovoid mass projecting over the mediastinum [54]. CT scan: After the clinical suspicion on CRs, the patient must undergo further workup with CT. On noncontrast CT scans, thymomas have rather typical appearances highly suggestive of such diagnosis. They usually appear as welldefined homogenous round or lobulated masses, predominately fatty in attenuation, with a variable extent of soft tissue components. On the contrary, thymic carcinoma present as a rather ill-defined heterogeneous lesion, frequently presenting with local compression on adjacent normal structures (especially superior vena cava). Furthermore, lymph node and/or visceral metastases might be seen in thymic carcinomas (Figure  11.12). Contrast-enhanced CT provides additional data regarding the local infiltration of thymoma in the neighboring structures, which is helpful for preoperative staging and therapeutic planning. Local nodal status for TNM staging of such masses could be obtained using either CT or PET/CT imaging.

345

346

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

(d)

Figure 11.12  Thymic carcinoma. (a, b) Mild heterogenous FDG uptake is seen in the large irregular anterior mediastinal mass encasing the mediastinal structures with areas of necrosis and calcification, suggestive of malignant anterior mediastinal mass, which on histopathology showed thymic carcinoma (c, d). Furthermore, an FDG-avid well-defined soft tissue mass with areas of calcification is noted in the left parietal lobe of the brain with adjacent edema, suggestive of metastatic disease.

MRI: In patients with renal failure or other contraindications for contrast-enhanced CT, MRI is an alternative imaging modality for differentiating malignant mediastinal tumors such as thymic epithelial neoplasms from benign entities, which can also be helpful to characterize and stage thymic tumors [55]. In addition, the lack of ionizing radiation enables MRI to be used safely in young patients. Furthermore, MRI is superior for distinguishing normal thymus and thymic hyperplasia from small thymomas thanks to high sensitivity  of  T1-weighted gradient-echo images for detecting microscopic fat. Like CT, several methods have been proposed for preoperative staging of thymomas using MRI. Abdel Raze et  al. found that lobulated margins on MRI were more frequently seen in low-risk thymoma than in highrisk thymoma or thymic carcinoma [56]. They also introduced some quantitative assessment of ADC values in diffusion-weighted images (DWI), demonstrating that there was a significant difference of ADC values between low-risk and high-risk thymomas. Indeed, ADC values were higher in low-risk thymomas than in high-risk thymomas and carcinomas. In another study, Seki et al. demonstrated that both ADC values on DWI MRI and

multidetector-row CT were equally effective for differentiating thymic neoplasms from other anterior mediastinal tumors and for defining thymic WHO categories and clinical staging (early versus advanced). Hence, DWI techniques may be utilized as a complementary method of mediastinal imaging in patients with equivocal CT findings [57]. FDG PET/CT: Although CT is essential for detecting and evaluating anterior mediastinal masses, it may be difficult to accurately differentiate benign and malignant tumors using only CT. Additional imaging with MRI and PET is often obtained for the preoperative assessment of such masses. The potential added value of FDG uptake in thymic pathologies has been studied in several studies. PET provides additional data to predict the malignant nature and invasiveness of thymomas. Many studies have demonstrated the usefulness of SUVmax analysis on FDG PET to predict the degree of malignancy of thymomas. They have frequently reported higher FDG activity in highrisk thymoma, and particularly thymic carcinomas and carcinoids, than in low-risk thymomas  [58]. In a metaanalysis by Treglia [59], a statistically significant difference of SUVmax was demonstrated between the different thymic

Thoracic Malignancies

epithelioid tumors (low-grade thymomas, high-grade thymomas, and thymic carcinomas). Park et  al.  [60] reported a mean SUVmax of 3.43  in low-risk thymomas, 4.42  in high-risk thymomas, and 8.23  in thymic carcinomas. Similar results have been found by Kitami et al., with the SUVmax values of low-grade thymomas, high-grade thymomas, and thymic carcinoma being 3.14, 4.34, and 8.59, respectively, with significant differences between low- and high-grade thymomas. Eguchi et  al. reported a SUVmax threshold of 3.5 to distinguish between low-risk (A, AB, B1) and high-risk (B2, B3, C) thymomas with sensitivity and specificity of 92.3% and 83.3%, respectively [61]. Additional parameters such as tumor to mediastinal (T/M) ratio (defined as the ratio of the SUVmax of the tumor to  the aorta) have also been proposed for discriminating high-risk thymoma from low-risk thymoma. Very recently, Yajima et al. showed that the SUVmax and T/M ratio were significantly greater in high-risk thymomas than in lowrisk thymomas [62]. They also suggested that the optimal cut-off value of SUVmax was 5.15 for detecting malignant lesions, indicating a need for a surgical approach in anterior mediastinal tumors. Lastly, in many cases with equivocal imaging findings, additional tissue sampling is indicated to draw a definite conclusion in the management of mediastinal neoplasms.

Teratoma Mediastinal teratomas are germ cell tumors, accounting for approximately 10–15% of all mediastinal masses [63]. They are generally subdivided into mature teratoma (most common), immature teratoma, and teratoma with malignant transformation (rare). More than 50% of patients with mediastinal teratoma are asymptomatic and detected incidentally on a routine CR. Chest radiography: Both immature and mature teratomas typically occur as a round and sharply marginated anterior mediastinal mass on chest X-ray. Dystrophic calcification, which presents in about 25% of teratomas, would aid in the differential diagnosis of these masses [64, 65]. CT scan: After a suspicious radiograph, CT or MRI would be applied for better characterization. CT scan is the modality of choice to ascertain the diagnosis and study the extent of such tumors. Teratomas usually have some distinct imaging features. The hallmarks of both benign and malignant teratomas are fluid, fat, and calcific components on CT or MRI  [66]. Most of the mediastinal teratomas are mature (benign) and typically present as well-circumscribed heterogeneous masses in the anterior mediastinum, accompanied by variable degrees of multilocular cystic degeneration, with a peripheral capsule surrounding multiple fat-density lobules. Intralobular calcifications  – related to teeth and

skeletal elements  – are also commonly seen  [67, 68]. Suggestive elements for immature teratomas are multiple cysts with areas of necrosis-related hypodensity, separated by thick ill-defined septa, which can invade the adjacent structures. Nonetheless, it should be noted that imaging alone cannot always differentiate between mature and immature teratoma, with clinical and laboratory data being mandatory in such situations. MRI: As with CT, teratomas can be easily defined by MRI, except for calcification, which is better assessed using CT imaging. On MRI, gross intralesional fat often has a high signal intensity on T1- and T2-weighted images, and a reduced signal in fat-suppression sequences. The fat-fluid level is another diagnostic clue  [69]. Radiologic features suggestive of malignant teratomas include focal contrastenhancing areas or invasion to the adjacent normal mediastinal structures. FDG PET/CT: FDG PET/CT is not a first-line modality in this entity, and there is still insufficient data on its usefulness for differentiation between mature and immature teratomas. Nonetheless, it is generally believed that mature and immature teratomas reveal low/mild FDG uptake, although no discrete SUV threshold has been established yet. As a tangible example, in a study by Kaira et al. [70] with two mature teratoma cases, they found a tumor/ mediastinum (T/M) ratio of 1.58 and 3.12, respectively. A median value of 2.5 was used as the cut-off T/M ratio in their study, indicating that a T/M ratio of more than 2.5 was defined as high uptake. They concluded that FDG uptake was helpful for predicting the grade of malignancy in mediastinal nonthymic neoplasms. Nevertheless, a large-scale study is still necessary to draw a definite conclusion in this regard.

Summary Correlative imaging is crucial for the diagnostic evaluation and management of mediastinal tumors. While chest radiography continues to be the most common initial modality for imaging mediastinal masses, CT is the modality of choice in such an entity. Patients with equivocal CR and CT results may be further investigated with MRI. Both CT and MRI studies provide anatomic data regarding size, location, and adjacent organ involvement. Although PET/ CT is not currently applied as the first-line modality in the routine clinical practice of mediastinal masses, it has become a first-line tool for staging, restaging, and response evaluation of lymphoma. Metabolic mapping using FDG PET/CT is also effective for distinguishing between benign and malignant lesions. Furthermore, high FDG uptake in  various mediastinal masses may represent a clinical

347

348

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

biomarker of tumor behavior, predicting its malignant nature and invasiveness preoperatively. Given these advantages, FDG PET/CT may be a promising objective modality in detecting and managing mediastinal tumors, particularly in thymic carcinomas, lymphomas, and invasive

thymomas. The role of PET-MRI in mediastinal tumors is still under investigation. Lastly, it should be noted that although imaging studies are mandatory for preoperative management of mediastinal tumors, the final diagnosis in most cases still relies on surgical pathology.

References 1 Patel, V.K., Naik, S.K., Naidich, D.P. et al. (2013). A practical algorithmic approach to the diagnosis and management of solitary pulmonary nodules: Part 1: Radiologic characteristics and imaging modalities. Chest 143 (3): 825–823. 2 Xu, Y., Ma, D., and He, W. (2012). Assessing the use of digital radiography and a real-time interactive pulmonary nodule analysis system for large population lung cancer screening. Eur. J. Radiol. 81 (4): e451–e456. 3 Al-Ameri, A., Malhotra, P., Thygesen, H. et al. (2015). Risk of malignancy in pulmonary nodules: a validation study of four prediction models. Lung Cancer 89 (1): 27–30. https:// doi.org/10.1016/j.lungcan.2015.03.018. Epub 28 Mar 2015. 4 Fletcher, J.W., Kymes, S.M., Gould, M.K. et al. (2008). A comparison of the diagnostic accuracy of 18F-FDG PET and CT in the characterization of solitary pulmonary nodules. J. Nucl. Med. 49: 179–185. 5 Tang, K., Wang, L., Lin, J. et al. (2019). The value of 18F-FDG PET/CT in the diagnosis of different size of solitary pulmonary nodules. Medicine (Baltimore) 98 (11): e14813. 6 Yu, M., Wang, Z., Yang, G., and Cheng, Y. (2020). A model of malignant risk prediction for solitary pulmonary nodules on 18 F-FDG PET/CT: building and estimating. Thorac. Cancer 11 (5): 1211. 7 Hickeson, M., Yun, M., Matthies, A. et al. (2002). Use of a corrected standardized uptake value based on the lesion size on CT permits accurate characterization of lung nodules on FDG-PET. Eur. J. Nucl. Med. Mol. Imaging 29 (12): 1639–1647. 8 Fatihoğlu, E., Biri, S., Aydın, S. et al. (2019). MRI in evaluation of solitary pulmonary nodules. Turk. Thorac. J. 20 (2): 90–96. 9 Feng, F., Qiang, F., Shen, A. et al. (2018). Dynamic contrast-enhanced MRI versus 18F-FDG PET/CT: which is better in differentiation between malignant and benign solitary pulmonary nodules? Chin. J. Cancer Res. 30 (1): 21–30. 10 Fujimoto, K. (2008). Usefulness of contrast-enhanced magnetic resonance imaging for evaluating solitary pulmonary nodules. Cancer Imaging 8 (1): 36–44. 11 Barta, J.A., Powell, C.A., and Wisnivesky, J.P. (2019). Global epidemiology of lung cancer. Ann. Glob. Health 85 (1): 8.

12 Siegel, R.L., Miller, K.D., and Jemal, A. (2018). Cancer statistics, 2018. CA Cancer J. Clin. 68 (1): 7–30. 13 Agrawal, A. and Rangarajan, V. (ed.) (2018). PET/CT in Lung Cancer. Springer. 14 Xing, P.Y., Zhu, Y.X., Wang, L. et al. (2019). What are the clinical symptoms and physical signs for non-small cell lung cancer before diagnosis is made? A nation-wide multicenter 10-year retrospective study in China. Cancer Med. 8 (8): 4055–4069. 15 Chowienczyk, S., Price, S., and Hamilton, W. (2020). Changes in the presenting symptoms of lung cancer from 2000-2017: a serial cross-sectional study of observational records in UK primary care. Br. J. Gen. Pract. 70 (692): e193–e199. 16 de Groot, P.M., Wu, C.C., Carter, B.W., and Munden, R.F. (2018). The epidemiology of lung cancer. Transl. Lung Cancer Res. 7 (3): 220. 17 Bradley, S.H., Abraham, S., Callister, M.E. et al. (2019). Sensitivity of chest X-ray for detecting lung cancer in people presenting with symptoms: a systematic review. Br. J. Gen. Pract. 69 (689): e827–e835. 18 Hoffman, R.M. and Sanchez, R. (2017). Lung cancer screening. Med. Clin. North Am. 101 (4): 769–785. 19 Wielpütz, M. and Kauczor, H.U. (2012). MRI of the lung: state of the art. Diagn. Interv. Radiol. 18 (4): 344–353. 20 Feng, S.H. and Yang, S.T. (2019). The new 8th TNM staging system of lung cancer and its potential imaging interpretation pitfalls and limitations with CT image demonstrations. Diagn. Interv. Radiol. 25 (4): 270. 21 Amin, M.B. (2017). American Joint Committee on Cancer and American Cancer Society AJCC Cancer Staging Manual, 8e. Chicago IL: American Joint Committee on Cancer, Springer. 22 Ohno, Y., Koyama, H., Dinkel, J., and Hintze, C. (2009). Lung cancer. In: MRI of the Lung (ed. H.U. Kauczor), 179–216. Heidelberg: Springer. 23 Ohno, Y., Adachi, S., Motoyama, A. et al. (2001). Multiphase ECG triggered 3D contrast-enhanced MR angiography: utility for evaluation of hilar and mediastinal invasion of bronchogenic carcinoma. J. Magn. Reson. Imaging 13: 215–224. 24 Glazer, G.M., Gross, B.H., Quint, L.E. et al. (1985). Normal mediastinal lymph nodes: number and size

Thoracic Malignancies

25

26

27

28

29

30

31

32

33

34

35

36

37

according to American Thoracic Society mapping. Am. J. Roentgenol. 144: 261–265. Walker, C.M., Chung, J.H., Abbott, G.F. et al. (2012). Mediastinal lymph node staging: from noninvasive to surgical. Am. J. Roentgenol. 199: W54–W64. Silvestri, G.A., Gould, M.K., Margolis, M.L. et al. (2007). Non-invasive staging of non-small cell lung cancer: ACCP evidenced-based clinical practice guidelines (2nd edition). Chest 132 (3 suppl): 178S–201S. Schimmer, C., Neukam, K., and Elert, O. (2006). Staging of non-small cell lung cancer: clinical value of positron emission tomography and mediastinoscopy. Interact. Cardiovasc. Thorac. Surg. 5: 418. Kay, F.U., Kandathil, A., Batra, K. et al. (2017). Revisions to the tumor, node, metastasis staging of lung cancer: rationale, radiologic findings and clinical implications. World J. Radiol. 9 (6): 269. van Tinteren, H., Hoekstra, O.S., Smit, E.F. et al. Effectiveness of positron emission tomography in the preoperative assessment of patients with suspected non-small-cell lung cancer: the PLUS multicentre randomised trial. Lancet 359 (9315): 1388–1393. Xu, L., Tian, J., Liu, Y., and Li, C. (2014). Accuracy of diffusion-weighted (DW) MRI with background signal suppression (MR-DWIBS) in diagnosis of mediastinal lymph node metastasis of nonsmall-cell lung cancer (NSCLC). J. Magn. Reson. Imaging 40 (1): 200–205. Ohno, Y., Koyama, H., Nogami, M. et al. (2007). STIR turbo SE MR imaging vs. coregistered FDG-PET/ CT: quantitative and qualitative assessment of N-stage in non-small-cell lung cancer patients. J. Magn. Reson. Imaging 26 (4): 1071–1080. Chao, F. and Zhang, H. (2012). PET/CT in the staging of the non-small-cell lung cancer. J. Biomed. Biotechnol. 2012: 783739. Carter, B.W. and Erasmus, J.J. (2019). Current concepts in the diagnosis and staging of lung cancer. In: Diseases of the Chest, Breast, Heart and Vessels 2019-2022, 79–93. Cham: Springer. Miller, A.B., Hoogstraten, B.F., Staquet, M.F., and Winkler, A. (1981). Reporting results of cancer treatment. Cancer 47 (1): 207–214. Therasse, P., Arbuck, S.G., Eisenhauer, E.A. et al. (2000). New guidelines to evaluate the response to treatment in solid tumors. J. Natl. Cancer Inst. 92 (3): 205–216. Weber, W.A., Petersen, V., Schmidt, B. et al. (2003). Positron emission tomography in non-small-cell lung cancer: prediction of response to chemotherapy by quantitative assessment of glucose use. J. Clin. Oncol. 21 (14): 2651–2657. Ichiya, Y., Kuwabara, Y., Sasaki, M. et al. (1996). A clinical evaluation of FDG-PET to assess the response

38

39

40

41

42

43

44

45

46

47

48

49

50

in radiation therapy for bronchogenic carcinoma. Ann. Nucl. Med. 10 (2): 193–200. Vansteenkiste, J., Fischer, B.M., Dooms, C., and Mortensen, J. (2004). Positron-emission tomography in prognostic and therapeutic assessment of lung cancer: systematic review. Lancet Oncol. 5 (9): 531–540. Hicks, R.J., Kalff, V., MacManus, M.P. et al. (2001). The utility of 18F-FDG PET for suspected recurrent non–small cell lung cancer after potentially curative therapy: impact on management and prognostic stratification. J. Nucl. Med. 42 (11): 1605–1613. Ignatius Ou, S.H. and Zell, J.A. (2009). The applicability of the proposed IASLC staging revisions to small cell lung cancer (SCLC) with comparison to the current UICC 6th TNM edition. J. Thorac. Oncol. 4 (3): 300–310. Nicholson, A.G., Chansky, K., Crowley, J. et al. (2016). The International Association for the Study of Lung Cancer lung cancer staging project: proposals for the revision of the clinical and pathologic staging of small cell lung cancer in the forthcoming eighth edition of the TNM classification for lung cancer. J. Thorac. Oncol. 11 (3): 300–311. Fenton, K.N. and Richardson, J.D. (1995). Diagnosis and management of malignant pleural effusions. Am. J. Surg. 170 (1): 69–74. Qureshi, N.R., Rahman, N.M., and Gleeson, F.V. (2009). Thoracic ultrasound in the diagnosis of malignant pleural effusion. Thorax 64 (2): 139–143. Chian, C.F., Su, W.L., Soh, L.H. et al. (2004). Echogenic swirling pattern as a predictor of malignant pleural effusions in patients with malignancies. Chest 126: 129–134. Yilmaz, U., Polat, G., Sahin, N. et al. (2005). CT in differential diagnosis of benign and malignant pleural disease. Monaldi Arch. Chest Dis. 63: 17–22. Hwang, J.H., Song, K.S., Park, S.I. et al. (2005). Subtle pleural metastasis without large effusion in lung cancer patients: preoperative detection on CT. Korean J. Radiol. 6: 94–101. Falaschi, F., Battolla, L., Zampa, V. et al. (1996). Comparison of computerized tomography and magnetic resonance in the assessment of benign and malignant pleural diseases. Radiol. Med. (Torino) 92: 713–718. Davis, S.D., Henschke, C.L., Yankelevitz, D.F. et al. (1990). MR imaging of pleural effusions. J. Comput. Assist. Tomogr. 14: 192–198. Porcel, J.M., Hernández, P., Martínez-Alonso, M. et al. (2015). Accuracy of fluorodeoxyglucose-PET imaging for differentiating benign from malignant pleural effusions: a meta-analysis. Chest 147 (2): 502–512. Tomiyama, N., Honda, O., Tsubamoto, M. et al. (2009). Anterior mediastinal tumors: diagnostic accuracy of CT and MRI. Eur. J. Radiol. 69 (2): 280–288.

349

350

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

51 Prosch, H., Röhrich, S., Tekin, Z.N., and Ebner, L. (2020). The role of radiological imaging for masses in the prevascular mediastinum in clinical practice. J. Thorac. Dis. 12 (12): 7591–7597. 52 Carter, B.W., Okumura, M., Detterbeck, F.C., and Marom, E.M. (2014). Approaching the patient with an anterior mediastinal mass: a guide for radiologists. J. Thorac. Oncol. 9 (Suppl 2): S110. 53 Duwe, B.V., Sterman, D.H., and Musani, A.I. (2005). Tumors of the mediastinum. Chest 128 (4): 2893–2909. 54 Marom, E.M. (2010). Imaging thymoma. J. Thorac. Oncol. 5 (10 Suppl 4): S296–S303. 55 Gentili, F., Pelini, V., Lucii, G. et al. (2019). Update in diagnostic imaging of the thymus and anterior mediastinal masses. Gland Surg. 8 (Suppl 3): S188–S207. 56 Abdel Razek, A.A.K., Khairy, M., and Nada, N. (2014). Diffusion-weighted MR imaging in thymic epithelial tumors: correlation with World Health Organization classification and clinical staging. Radiology 273: 268–275. 57 Seki, S., Koyama, H., Ohno, Y. et al. (2014). Diffusionweighted MR imaging vs. multi-detector row CT:direct comparison of capability for assessment of management needs for anterior mediastinal solitary tumors. Eur. J. Radiol. 83: 835–842. 58 Benveniste, M.F., Moran, C.A., Mawlawi, O. et al. (2013). FDG PET/CT aids in the preoperative assessment of patients with newly diagnosed thymic epithelial malignancies. J. Thorac. Oncol. 8 (4): 502–510. 59 Treglia, G., Sadeghi, R., Giovanella, L. et al. (2014). Is (18) F-FDG PET useful in predicting the WHO grade of malignancy in thymic epithelial tumors? A meta-analysis. Lung Cancer 86 (1): 5–13. 60 Park, S.Y., Cho, A., Bae, M.K. et al. (2016). Value of 18F-FDG PET/CT for predicting the World Health Organization malignant grade of thymic epithelial tumors: focused in volume-dependent parameters. Clin. Nucl. Med. 41 (1): 15–20.

61 Eguchi, T., Yoshida, K., Hamanaka, K. et al. (2012). Utility of 18F-fluorodeoxyglucose positron emission tomography for distinguishing between the histological types of early stage thymic epithelial tumours. Eur. J. Cardiothorac. Surg. 41 (5): 1059–1062. 62 Yajima, T., Mogi, A., Shimizu, K. et al. (2020). Quantitative analysis of metabolic parameters at 18F-fluorodeoxyglucose positron emission tomography in predicting malignant potential of anterior mediastinal tumors. Oncol. Lett. 19 (3): 1865–1871. 63 Shahrzad, M., Le, T.S., Silva, M. et al. (2014). Anterior mediastinal masses. Am. J. Roentgenol. 203 (2): W128–W138. 64 Quint, L.E. (2007). Imaging of anterior mediastinal masses. Cancer Imaging 7 (Special issue A): S56–S62. 65 Ranganath, S.H., Lee, E.Y., Restrepo, R., and Eisenberg, R.L. (2012). Mediastinal masses in children. Am. J. Roentgenol. 198 (3): W197–W216. 66 Lee, E.Y. (2009). Evaluation of non-vascular mediastinal masses in infants and children: an evidence-based practical approach. Pediatr. Radiol. 39 (Suppl 2): S184–S190. 67 Thacker, P.G., Mahani, M.G., Heider, A., and Lee, E.Y. (2015). Imaging evaluation of mediastinal masses in children and adults: practical diagnostic approach based on a new classification system. J. Thorac. Imaging 30 (4): 247–267. 68 Mikail, N., Khalil, A., and Rouzet, F. (2021). Mediastinal masses: 18F-FDG-PET/CT features based on the International Thymic Malignancy Interest Group Classification. Semin. Nucl Med. 51 (1): 79–97. 69 Park, J.W., Jeong, W.G., Lee, J.E. et al. (2021). Pictorial review of mediastinal masses with an emphasis on magnetic resonance imaging. Korean J. Radiol. 22 (1): 139–154. 70 Kaira, K., Abe, M., Nakagawa, K. et al. (2012). 18F-FDG uptake on PET in primary mediastinal non-thymic neoplasm: a clinicopathological study. Eur. J. Radiol. 81 (9): 2423–2429.

351

12 A Correlative Approach to Breast Imaging Shabnam Mortazavi1, Sonya Khan2, Kathleen Ruchalski1, Cory Daignault3, and Jerry W. Froelich4 1

Radiology, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA Los Angeles and Veterans Administration, Greater Los Angeles Healthcare Systems, University of California, Los Angeles, CA, USA 3 Minneapolis VA Medical Center, Minneapolis, MN, USA 4 Radiology, University of Minnesota, Minneapolis, MN, USA 2

­ natomy of the Breast A and Locoregional Lymph Nodes

lymphatic drainage routes are axillary, interpectoral, and internal mammary.

The breast is a modified cutaneous gland located within the superficial pectoralis fascia, overlying the pectoralis major muscle superiorly, serratus anterior muscle laterally, and upper abdominal oblique muscles inferiorly. Adult female breast tissue typically extends from the second to seventh ribs, bounded by the sternum medially and midaxillary line laterally. Breast parenchyma consists of glandular, adipose, and connective tissue. A typical female breast consists of approximately 15–20 segments demarcated by lactiferous ducts converging at the nipple in a radial configuration. The functional unit of the breast is the terminal ductal lobular unit (TDLU). Each TDLU comprises a lobule and terminal duct connecting the TDLU to a lactiferous duct (Figure 12.1). Blood flow to the breast is typically supplied by the internal mammary, lateral thoracic, and intercostal arteries. Venous drainage of the breast parallels the arterial anatomy at the level of the deep breast tissues; however, superficial draining veins do not follow the path of arterial blood supply. Typical regional lymphatic drainage of the breast includes the axillary, supraclavicular, and internal mammary nodal chains. Axillary lymph nodes are subdivided into three named levels, defined in relation to the pectoralis minor muscle. Level I (low axillary) nodes are lateral to the muscle, level II (mid axillary) nodes are deep to the muscle, and level III (high axillary) nodes are superior and medial to the muscle (Figure 12.2). The three major breast

­Breast Cancer Classification Breast cancer is the most common cancer in women and about one out of eight women (roughly 12%) in the United States will develop invasive breast cancer in their lifetime [1]. The term “breast cancer” encompasses a heterogenous disease process, subclassified and described clinically by histopathology, biomarkers, and genomic profile. Histopathological subtypes: Breast carcinomas arise from the epithelial component of either lobules or the terminal duct. There are many different types of breast cancer  [2]. The most common histopathological subtype (approximately 80%) is invasive ductal carcinoma, followed by invasive lobular carcinoma (20%). Biomarker subtypes: Currently, there are three major clinically measured biomarkers: estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) expression. Accordingly, there are three clinical biomarker-based subclassifications of breast cancer: (ER and/or PR)-positive and HER2-negative, HER2-positive, and triple-negative. Approximately 70% of breast cancers are hormone receptor positive [3]. Molecular subtypes: There are currently four distinct, clinically diagnosed molecular subtypes of breast cancer, classified on the basis of gene-expression analysis and hormone receptor status: luminal A, luminal B, HER2enriched, and triple-negative/basal-like tumors [4].

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach, First Edition. Edited by Ali Gholamrezanezhad, Majid Assadi, and Hossein Jadvar. © 2023 John Wiley & Sons Ltd. Published 2023 by John Wiley & Sons Ltd.

352

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

m te ys s al ct Du

TDLU

Figure 12.1 (a) Lateral ductogram mammographic images obtained immediately following intraductal contrast administration via cannulation of a single periareolar lactiferous duct, illustrating the extent of a single lactiferous duct. A terminal ductal lobular unit (TDLU) is depicted (b).

(a)

Pectoralis minor muscle

(f)

(g)

(b)

III II I

(c) (e)

(d)

Figure 12.2 (a) Axillary lymph nodes used to determine N category according to the 8th AJCC Cancer Staging Manual. (b–g) Axial FDG PET-CT images demonstrate FDG-avid and/or enlarged (b) right axillary level I node, seen lateral to the pectoralis muscles (arrow), (c) left axillary level II lymph nodes (arrow), deep to the pectoralis minor muscle, (d) a left axillary level III lymph node (arrow), medial and superior to the pectoralis minor muscle, (e) left (arrow), greater than right, internal mammary chain lymph nodes in a patient with mediastinal and axillary level I lymphadenopathy, (f) a right chest wall/intramammary lymph node (arrow), and (g) a left interpectoral (Rotter) lymph node (arrow), by definition located between the pectoralis major and minor muscles in a patient with enlarged left axillary level I and II and neck lymph nodes.

­ olecular Techniques in Breast M Imaging: A Correlative Approach to the BI-­RADS Paradigm A general familiarity with the highly regulated and algorithmic paradigm of conventional breast imaging, which centers on the diagnosis and sequelae of breast cancer, and

an understanding of the varied and crucial tasks nuclear imaging examinations fulfill within this complex system are essential for clinically meaningful correlation between molecular and nonmolecular imaging modalities. The term “breast imaging” encompasses multiple imaging modalities. Detailed guidelines for the clinical uses of these modalities are outlined in the American College of Radiology Appropriateness Criteria and Practice Guidelines [5].

A Correlative Approach to Breast Imaging

Conventional Nonmolecular Breast Imaging Modalities Conventional mammography and digital breast tomosynthesis (DBT) utilize ionizing radiation, captured on a detector, to depict breast anatomy in multiple orthogonal or quasi-orthogonal imaging planes, with the goal of identifying/localizing alterations in typical anatomy suspicious for malignancy. DBT uses a moving X-ray source to image the breast at different angles, obtaining multiple digital images that can be reconstructed to obtain a quasi-tridimensional (3D) representation of the breast and allowing overlapping breast tissue to be viewed separately, increasing breast cancer detection and decreasing the false-positive rate, particularly in women with radiographically dense breast tissue [6]. As the relative amount of fat decreases and glandular tissue increases, the breast becomes denser on mammography and cancer detection becomes more challenging. Figure  12.3 demonstrates quasi-orthogonal, standard (a)

(b)

(c)

(d)

Figure 12.3 Right breast craniocaudal, aka CC (a), and mediolateral oblique, aka MLO (b), standard screening mammogram views in a high-risk patient demonstrates heterogeneously dense breast tissue. The parenchymal pattern was stable from previous mammograms, without mammographic evidence for malignancy. Subsequent imaging with MBI in the CC (c) and MLO (d) projections demonstrates mass-like uptake in the lower outer quadrant that yielded invasive lobular carcinoma on tissue sampling. Source: Reprinted from Huppe et al. [7], with permission from Elsevier.

mediolateral oblique (MLO), and craniocaudal (CC) screening mammography views as performed for the purposes of localizing breast findings to one of four quadrants via triangulation. These standard image views have been adopted for dedicated molecular breast imaging (MBI), as discussed later, and can be used to localize MBI findings as well, as in the case of the right breast findings shown in Figure 12.3. Breast ultrasound is a preferred modality for imageguided breast procedures and has been investigated as a breast cancer screening modality [8]. Dynamic, gadolinium-based contrast-enhanced magnetic resonance imaging (MRI or CE-MRI) is the most sensitive imaging tool for detection of breast cancer, providing functional physiologic data via dynamic contrast enhancement analysis in addition to superb anatomic detail due to MRI’s high soft tissue contrast resolution, in some cases detecting cancer occult on clinical examination or conventional modalities (mammography and ultrasound). Breast MRI utilizes magnetic fields (no ionizing radiation) to create multiplanar cross-sectional images through the selected field of view, requiring a dedicated breast coil to obtain images of diagnostic quality while the patient lies in the prone position with both or one breast(s) hanging free in the coil cavity. Dedicated breast positron emission tomography systems (dbPET), discussed later, employ similar prone positioning. Intravenous gadolinium-based contrast agents are needed to reliably detect cancers, extent of malignant disease, and some common benign lesions. Additional physiological data is obtained through kinetic analysis of the rate of contrast uptake and washout by the background breast parenchyma and any lesions of interest, a process that is typically performed by the interpreting physician at a separate workstation using specialized post-processing software. Analogous kinetic analysis techniques have been devised and are being studied for the relative rate of various nuclear imaging radiotracer uptake, particularly FDG, in the setting of potentially malignant lesions, including breast lesions. One relative advantage to CE-MRI is the relative swiftness of gadolinium washout from target lesions and the background compared to 18F-fluorodeoxyglucose (FDG), allowing the overlay of both the vascular uptake and washout characteristics of lesions to be studied with CE-MRI. Breast MRI can be used to supplement screening mammography in women with an increased risk for breast cancer, evaluate extent of disease in selected cases, and monitor response to systemic therapy in the neoadjuvant setting  [9]. In selected cases, it may be used for problem solving. Figures  12.10, 12.14, 12.16, and  12.17 depict clinical examples of breast MRI.

353

354

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Molecular Breast Imaging MBI, a term that includes breast-specific gamma imaging (BSGI), positron emission mammography (PEM), and dbPET, is a physiologic approach to breast imaging, performed following intravenous injection of radiopharmaceuticals, and imaged with dedicated single-photon gamma imaging or annihilation coincidence detection systems (i.e.  PET-CT). Radiopharmaceutical uptake is dependent on physiologic factors, such as increased vascularity (blood flow) and mitochondrial activity/concentration in malignant cells, resulting in images that depict functional activity localizing to malignant, and a subset of benign, breast lesions  [7]. Molecular imaging techniques (BSGI, PEM, dbPET) therefore are not affected by anatomic characteristics, such as breast density or postsurgical distortion, unlike mammography or ultrasound, but do suffer from lower anatomic resolution and precise localization.

Limitations of Early Scintimammography Imagers first recognized the potential for molecular imaging of breast cancer in the 1970s, when single-photonemitting radiopharmaceuticals administered for bone and cardiac imaging were noted to incidentally localize to breast malignancies [10–12], resulting in the development of scintimammography, a term referring to single-photonemitting radiopharmaceutical breast imaging performed with scintillation gamma cameras. Despite early promise, researchers found that early planar and SPECT scintimammography techniques, performed with large field-of-view whole-body detectors, had low sensitivity for detection of small (0.2 ng/mL 6 weeks after prostatectomy 6 weeks, confirmed by a second PSA measurement (AUA) or (b) PSA increase greater than 2 ng/mL above nadir after radiation therapy or brachytherapy (ASTRO Phoenix) [8, 9]

Oligometastatic disease

Limited number of metastatic lesions (up to three to five lesions) in up to two organs

Progressive disease despite androgen Castrationresistant prostate deprivation therapy and low levels of testosterone cancer

the sensitivity of these conventional imaging studies decreases markedly, necessitating the use of more sophisticated modalities. Early and accurate detection of the sites of recurrent disease is of paramount importance for patient management and treatment, since local salvage therapy is  used for pelvic recurrence and systematic treatments are used for metastatic disease [10, 11]. Locoregional salvage therapies are most effective with serum PSA 30 mL), heterogeneous TZ, bladder wall hypertrophy and trabeculation, and post-micturition urine residue in the urinary bladder. Gray-scale transrectal ultrasound (TRUS) was introduced in 1968, but despite the development of higher frequency

Correlative Approach to Prostate Imaging

transducers and improved signal reception, it is still not an optimal modality for detection of prostate cancer at early stages, with many prostate cancer foci appearing isoechoic on TRUS. Color Doppler ultrasound improves the diagnostic performance of gray-scale TRUS. In 1989, the first TRUSguided systematic sextant biopsy protocol was described, and gradually the 12-sysmetic biopsy turned into the standard of care for the diagnosis of prostate cancer. Biopsy is performed after initiation of prophylactic antibiotics (fluroquinolones by choice, a widely used regimen is 500 mg of ciprofloxacin 1 day prior to biopsy and for 2 days after) and under local anesthesia. Contrast-enhanced TRUS and elasto-sonography can augment visualization of suspicious foci for targeted TRUS biopsy. Biopsy is carried out using a guided biopsy system attached to the transrectal probe.

CT Scan Overall, CT scan is an insufficient modality for assessment of prostatic disease. Prostate volume measurement might be overestimated on CT scan. In daily practice, a prostate gland with a transverse diameter of greater than 4.5 cm in the axial plane is considered enlarged. Prostate calcifications are frequently seen on CT scan, without clear clinical significance. In the setting of BPH, enlarged prostate gland, heterogenous appearance of TZ with visible high- or lowdensity nodules, and, in severe cases, protrusion of the gland into the urinary bladder might be seen. Bladder wall thickening and trabeculation are indirect clues for the presence of BPH. With acute prostatitis, the prostate gland becomes enlarged and edematous, and if a prostate abscess develops, a rim-enhancing unilocular or multilocular hypodense area in the PZ is expected on contrast-enhanced CT scan. CT scan is limited for the primary detection of prostate cancer, but focal increased enhancement in the PZ of prostate gland is described in some patients with prostate cancer. In high-grade disease or when there is suspicion for locoregional or distant metastasis, chest and abdominopelvic CT scan are commonly used for detection of nodal and soft tissue metastasis, and planning for radiotherapy.

The potential application of MRI in prostate cancer was first described in 1982 [15]. With remarkable advances in image acquisition techniques, postprocessing software, and standardized interpretation guidelines during the past decade, multiparametric prostate MRI (mpMRI) has turned into the imaging modality of choice when assessment of intraprostatic malignancy or image-guided prostate biopsy is indicated. Prostate mpMRI is generally recommended when there is either (i) clinically high suspicion for intermediate or high-grade prostate cancer (based on serum PSA measures and findings of DRE) in patients who had a prior negative transrectal ultrasound guided biopsy or (ii) restaging or re-biopsy planning in patients with known low-grade prostate cancer under active surveillance  [16]. The American Urologic Association currently does not recommend mpMRI for the purpose of prostate cancer screening or surveillance [17]. Adapted from a similar scoring system in breast cancer, the European Society of Urogenital Radiology (ESUR) introduced the first Prostate Imaging-Reporting and Data System (PI-RADS) in 2012, which included guidelines for prostate MRI acquisition, interpretation, and reporting [18]. PI-RADS version 1 (v1) was succeeded by recommendations of a steering committee from the American College of Radiology (ACR), ESUR, and the AdMeTech Foundation, formulated as PI-RADS v2. in 2016  [19] followed by PI-RADS v2.1  in 2019  [16]. Since their introduction, PIRADS guidelines have been widely validated and right now are widely applied to daily practice in the United States and Europe. PI-RADS guidelines assess the probability of clinically significant prostate cancer on a five-point scale for each lesion on a standard prostate mpMRI (Table 19.2). For the acquiring prostate mpMRI, most experts recommend using a 3 T MRI scanner when available. Compared to 1.5 T scanner, a 3 T scanner is more prone to Table 19.2 The likelihood of clinically significant prostate cancer with each PI-RADS score and the related recommendation.

PI-RADS score

Likelihood of clinically significant prostate cancer

Recommendation

Multiparametric MRI and MRI-directed Biopsy

PI-RADS 1

Very low

Follow-up

MRI is not routinely used for the diagnosis of benign prostate conditions. MRI, however, is superior to ultrasound for classification of BPH subtypes and estimation of glandular/stromal ratio, findings which can help to tailor treatment. On MRI, chronic prostatitis and granulomatous prostatitis can cause signal abnormality and mimic prostate cancer [14].

PI-RADS 2

Low

Follow-up

PI-RADS 3

Intermediate

Depends on pre-test probability, most guidelines recommend MRI-directed biopsy

PI-RADS 4

High

MRI-directed biopsy

PI-RADS 5

Very high

MRI-directed biopsy

535

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach PZ

PIRADSv2.1 Score

TZ

1

T2WI 1

DWI 2

2

T2WI 2

DWI 3

3

If D

W

I≥4

DWI 1

T2WI 3

+

W

CE

I=5

If D

the susceptibility artifact, which might compromise the image quality (particularly in diffusion-weighted imaging [DWI]), and therefore in specific situations (such as the presence of metallic hip prosthesis) a 1.5 T scanner might be preferred. An external phased array coil is used with or without an endorectal coil. An endorectal coil increases signal-to-noise ratio (SNR), particularly in larger patients, but it increases the time and cost of the scan, and is associated with patient discomfort. Rectal evacuation before examination is recommended. The use of antispasmodic agents (such as glucagon and hyoscine bromide) to decrease bowel peristalsis and rectal gas is controversial. Prostate multiparametric MRI initially included T1weighted imaging (T1WI), T2-weighted imaging (T2WI), DWI with its corresponding apparent diffusion coefficient (ADC) map, dynamic contrast-enhanced imaging (DCE), and magnetic resonance spectroscopy (MRS). With the widespread use of DWI, and based on the recommendations of PIRADS v.2, MRS has now fallen out of favor. For interpretation of prostate mpMRI, the PZ and TZ should be separately scrutinized for the presence of focal lesions, with the assessments primarily based on DWI/ADC at PZ and T2WI at TZ. DCE has a secondary role in PZ lesions with PI-RADS score 3 (Figure 19.1). T1-weighted imaging is not part of the PIRADS criteria but it is useful for identification of intraprostatic hemorrhage and for detection of pelvic lymph node or bone involvement. The details of PIRADS guidelines for the acquisition, interpretation, and reporting of prostate mpMRI are available at the American College of Radiology website (https://www.acr.org/-/ media/ACR/Files/RADS/Pi-RADS/PIRADS-V2-1.pdf). In a meta-analysis of 21 studies (3857 patients), PIRADS v2 had pooled sensitivity and pooled specificity of 89% and 73%, respectively, for detection of prostate cancer, using prostatectomy or prostate biopsy as the reference standard [20]. Prostate mpMRI has significantly higher sensitivity and negative predictive value for the detection of clinically significant prostate cancer compared to TRUS biopsy, which indicates that acquiring prostate mpMRI could potentially avoid biopsy in up to 27% of men at risk  [21]. Systematic TRUS-guided biopsy commonly oversamples clinically insignificant disease and has a limited role for the diagnosis of prostate cancer in TZ. Based on the results of the PRECISION trial (which included 500 biopsy-naïve patients), performing MRI with or without MRI-targeted biopsy results in 12% higher detection of clinically significant prostate cancer and 13% lower detection of clinically insignificant cancers compared to TRUS-guided biopsy [22]. MRI-targeted biopsy is usually indicated with prostate lesions PIRADS categories 3, 4, and 5. Several approaches are used for performing MRI-targeted biopsy of prostate without

If D

536

DWI 4

4

T2WI 4

DWI 5

5

T2WI 5

Figure 19.1 PI-RADS v2.1 scoring schematic in the peripheral zone (PZ) and transition zone (TZ).

an absolute consensus on the preferred method. MRItargeted prostate biopsy might be performed via three approaches: MRI-ultrasound fusion biopsy, cognitive MRIguided biopsy, and direct in-bore MRI-guided biopsy. In MRI-ultrasound fusion biopsy, the previously acquired MRI data are fused on the real-time TRUS on a fusion platform and sampling is performed under the tracking system. In cognitive MRI-guided biopsy, the operator mentally maps (i.e. cognitively fuses) the previously acquired MRI target to the real-time TRUS and guides the needle to this location. Finally, in-bore MRI-guided biopsy is performed under direct MRI guidance while the patient is lying in the MRI gantry. Overall, with MRI-targeted approaches there is a significantly higher detection rate of clinically significant prostate cancer (relative risk [RR] 1.16) and a significantly lower detection rate of clinically insignificant cancer (RR 0.47) over TRUS-guided biopsy [23]. There is not any significant difference in detection of clinically significant prostate cancer between MRI-ultrasound fusion

Correlative Approach to Prostate Imaging

biopsy, cognitive MRI-ultrasound fusion biopsy, and direct in-bore MRI-guided biopsy. Direct MRI-guided biopsy, however, allows direct visualization of suspicious lesions with higher core positivity rate (47.7%) compared to cognitive MRI-guided biopsy (33.3%) and MRIultrasound fusion biopsy (31.3%) [24]. Cost-effectiveness

studies have found both MRI-based approaches to have higher quality-adjusted life-year benefits  [25, 26]. The optimal biopsy approach should be chosen per patient and per center basis, considering the availability of biopsy platforms and the experience of the performing operators.

Non-molecular Prostate Imaging: Key Concepts ●

●

Ultrasound and CT scan are limited for detection of prostate malignancy. With remarkable advances in multiparametric MRI, it is now turned into the imaging modality of choice when assessment of intraprostatic malignancy or guided prostate biopsy is indicated. Prostate mpMRI has significantly higher sensitivity and negative predictive value for detection of clinically significant prostate cancer compared to TRUSguided biopsy.

Molecular Prostate Imaging Fluorodeoxyglucose Fluorodeoxyglucose (FDG) is a synthetic glucose analog that is taken up by glucose transporters (GLUTs), phosphorylated by hexokinase, and trapped in the cells. Most malignant tumors demonstrate enhanced glucose metabolism, increased hexokinase enzymatic activity, and upregulation of GLUTs (Warburg effect) compared to normal tissue. With relatively low glucose metabolism in the majority of prostate cancers, FDG PET/CT has a limited role in the primary detection, staging, and restaging of prostate cancer  [27]. Castration-resistant prostate cancer is typically FDG avid, and therefore FDG PET/CT might be useful for assessment of its treatment response and prognostication (time to hormonal treatment failure in castrate-sensitive disease and overall survival in castrate-resistant metastatic disease) (Figure 19.2) [28]. PET Response Criteria in Solid Tumors (PERCIST) 1.0, particularly in combination with PSA response criteria [29], sum of SUVmax, and the number of FDG-avid lesions  [30], can independently provide prognostic information in this group of patients. Incidental focal intraprostatic FDG uptake is nonspecific but may carry a risk of malignancy. The prevalence of incidental high intraprostatic activity is 1.8%, with at least 17% risk of malignancy [31]. Some investigators have suggested that focal incidental intraprostatic FDG activity with SUV more than 6 should be further assessed by prostate mpMRI or serum PSA level [32].

●

●

Interpretation of multiparametric MRI is performed by using PI-RADS criteria and lesions with PI-RADS score 4–5 and most with PI-RADS score 3 undergo MRI-directed biopsy, which can be performed via cognitive fusion, MRI-ultrasound fusion or direct in-bore approaches. Twelve-core TRUS-guided biopsy remains the standard of care for initial assessment of suspected prostate cancer, with mpMRI and MRI-directed biopsy typically considered for inconclusive cases.

Choline Choline, as a component of the cytoplasmic membrane phospholipid layer, is phosphorylated by the enzyme choline kinase and is trapped in the cell membrane in the form of phosphatidylcholine  [33]. The upregulation of  choline kinase in prostate cancer is the most likely mechanism leading to enhanced expression of choline in prostate cancer. The most available radiolabeled choline ligands are 11C-choline, 18F-fluoroethylcholine, and 18 F-fluoromethylcholine (latter two are commonly known as 18F-fluorocholine or FCH). Normal biodistribution of 11 C-choline demonstrates relatively high accumulation in the pancreas, liver, kidneys, and salivary glands, and variable uptake in the bowel, with minimal urinary excretion (latter only with 11C-choline) [34]. The short half-life (~20 minutes) of 11C limits its utility to institutions with an on-site cyclotron. 18F-labeled choline radiotracers have a longer half-life (~110 minutes) and allow transportation from off-site cyclotrons. With a shorter position range of 18F-labeled choline ligands, the spatial resolution of images improves, but the higher urine activity with 18F-choline in contrast to 11C-choline is disadvantageous. Physiologic ureteric or urinary bladder activity might limit assessment of prostate bed or pelvic nodal involvement. With 11C-choline, imaging starts immediately after injection, while with 18F-choline imaging starts at, most commonly, 60 minutes post injection. When 18F-choline is used, additional post-micturition images may help distinguish pathologic activity from physiologic urine activity.

537

Baseline

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

SUV=24.5

PSA=223.3

CTHU=837

SUV=21.7

PSA=284

CTHU=1084

SUV=16.8

PSA=119

CTHU=1121

SUV=8.1

PSA=52.5

8 months

4 months

CTHU=772

12 months

538

Figure 19.2 Treatment response evaluation with FDG PET/CT in metastatic castration-resistant prostate cancer.The patient was scanned at baseline and at 4, 8, and 12 months (rows) after the start of docetaxel chemotherapy. From left to right columns: Axial CT at bone window level, axial FDG PET, axial fused FDG PET/CT, sagittal CT at bone window level, FDG MIP images, and bone scan. The FDG MIP images clearly demonstrate decline in the total metallically active tumor burden as result of favorable response to docetaxel chemotherapy in concordance with the decline in the serum PSA level while CT and bone scan are inconclusive. (Courtesy of Hossein Jadvar, MD, PhD, University of Southern California, Los Angeles, CA, USA; NIH R01-CA111613.)

Choline PET/CT is not indicated for the detection of primary prostate cancer or routine initial staging. Tracer uptake in prostatic intraepithelial neoplasia and benign conditions such as prostatitis and BPH lowers its specificity for assessment of intraprostatic malignancy. 18F-fluorocholine PET/ CT has slightly higher sensitivity and specificity compared to 11C-choline. Choline PET/CT is more accurate than conventional imaging for detection of pelvic nodal disease at initial staging with pooled sensitivity and specificity of 62% and 92%, respectively [35]. When there is biochemical recurrence of treated prostate cancer, choline PET/CT has higher accuracy than conventional imaging modalities; however, the disease detection is highly influenced by serum PSA level and the sensitivity of the scan drops sharply, with 80% of scans being negative at

PSA levels −2 SD

Below expected for age

Z-score −2 SD

BMD, bone mineral density; SD, standard deviation. Target group for utilizing T-score: postmenopausal women and men 50 years of age or older. Target age group for utilizing Z-score: premenopausal patients or men under 50 years of age.

that level should be excluded from assessment (Figure 23.4). The presence of artifacts and acquisition quality must be reported, as well as any excluded site followed by the rationale behind such decision. Finally, the addition of the FRAX score has proved to be helpful in decision-making, especially

(a)

for patients with osteopenia that could benefit from initiation of treatment [8]. DXA can also be used to assess treatment response, but the data are less robust than fracture risk prediction. Patients with stable or improving BMD are encouraged to be on the same treatment, whereas nonresponders are considered for alternate medications.

CT and Dual-energy CT The high spatial resolution makes computerized tomography (CT) ideal for the evaluation of bone integrity. Low attenuation and qualitative loss of cortical and trabecular volume throughout the bone can suggest a diagnosis of osteoporosis. Despite the lack of consensus in official thresholds, a linear decreased attenuation, measured in Hounsfield units (HU), in the trabecular region of interest (ROI) at a single lumbar level seems to correlate with osteoporosis in the elderly [9]. Features like anterior wedging, retropulsion of the fracture, and vacuum phenomenon in the spine are classic signs for osteoporotic fractures (Figure 23.6). More advanced techniques such as bone marrow analysis with dual-energy CT can also be valuable and comparable to MR evaluation to determine the age of the fracture (Figure 23.6) [10].

(b)

1

(c)

2

1

3

Figure 23.3 Examples of optimally obtained DXA studies. (a) The lumbar spine is centered and straight, the field of view includes 12th ribs with the accurate numbering of the lumbar levels, bone edges (yellow outline) include only the outlines of the vertebral bodies, and horizontal lines must cross each disc level. (b) Radius evaluation must include a third (33%) of the nondominant radius, the reference line (1) is at the tip of the ulnar styloid, the distal side of the ultradistal (UD) region of interest (ROI) (2) should not include the radial endplate, and the vertical line (3) is located between radius and ulna. (c) Femur assessment requires internal rotation of the hip such that lesser trochanter is barely visible, the ward’s area (yellow rectangle) should cover about half of the femoral neck, and the neck ROI rectangle (1) is perpendicular to femoral neck, contains soft tissue at either side of the femur neck, and does not include ischium or greater trochanter. Approximately 2 in. of the femur shaft is included in the field of view.

Osteoporosis

Table 23.2

Summary of quality control consideration in dual-energy X-ray absorptiometry studies.

Lumbar spine

Hip

Radius

Patient is centered and straight

Patient and femoral shaft should be straight on table

Use distal third of nondominant radius

Ribs appear at T12, middle portion of L5, and T12 vertebral bodies must be included

Leg in internal rotation to hide the lesser trochanter

Forearm centered with straight ulna and radius

Automatic segmentation outlines vertebral bodies and does not include soft tissues Horizontal lines go through disc spaces

Femur neck ROIa does not intersect with ilium or ischium

Automatic segmentation outlines radial and ulnar shafts, and does not include soft tissues

Exclude altered vertebrae (e.g. fractured, prior surgery) from final assessment

Ward’s area covers about half the femoral neck

Reference line is located at the distal tip of the ulnar styloid

L4 at top of iliac crests

Ischium and greater trochanter included in the field of view

Ultra distal ROI does not include the radial endplate The vertical line is between the center of ultra distal and 33% ROIs

Disturbances in soft tissues manually excluded from the image (e.g. metallic staples from bowel surgery) a

ROI, region of interest.

(a)

Ap Spine Bone density

YA T-score 2

1.32

1

1.20

0

1.08

–1

0.96

–2

Osteopenia

0.84

–3

0.72

–4 Osteoporosis

0.60 20

30

40

50 60 70 Age (years)

AP Spine (L2-L3) Results 1 BMD (g/cm2) 2 T-score (SD of young-adult BMD) 3 Z-score (SD of age-matched BMD)

Region

BMD1 (g/cm2)

L1 L2 L3 L4 L2-L3

1.061 0.862 0.895 1.093 0.879

(b)

Densitometry Ref: L2-L3 (BMD) BMD (g/cm2) 1.44 Normal

80

–5

90 100

0.879 –2.7 –1.5

Figure 23.4 Example of the value of the multimodality approach evaluating BMD. (a) Screening DXA study in a 67-year-old healthy woman with unremarkable clinical history and no known history of fragility fracture. DXA study shows an abnormally elevated BMD of L1 when compared to L2. Linear sclerosis in the superior endplate of L1 is suspicious for fracture and likely the reason for falsely elevated BMD in L1. The lumbar CT scan was recommended. (b) Lumbar CT scan confirms the suspected diagnosis on the DXA study, and a recent asymptomatic L1 vertebral body compression fracture is identified. L1 is therefore excluded from the assessment on the DXA study, BMD of L2–3 level was reported, and the impression of osteoporosis is established. Given the combination of osteoporosis and the fragility fracture, the final diagnosis in this case is severe osteoporosis.

663

664

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

Figure 23.5 Value of comparison with clinical history and prior study. (a) DXA study of a recently diagnosed prostate cancer in a 75-year-old man shows abnormally increased BMD and T-scores. Note diffuse sclerosis of vertebral bodies. (b) Prior DXA study from 4 years ago shows approximately 50% interval increase of BMD measurements between the two studies, which is an unusual observation. (a) Given the patient history of prostate cancer, osteoblastic metastasis is suspected and confirmed with a radiograph of the lumbar spine. Source: Image courtesy of Luke Hiller, MD, VA Medical Center, San Diego, CA, USA.

(a)

(b)

Figure 23.6 Identifying the age of compression fracture utilizing dual-energy CT technique. A 65-year-old female with pain. (a) Sagittal CT of the thoracic spine shows a compression fracture at two adjacent thoracic levels with sclerosis, anterior wedging, vertebral body height loss, and retropulsion of the posterior cortices into the central canal. (b) Sagittal virtual noncalcium technique with a color overlay where marrow edema is depicted in green. Detection of bone marrow edema in the involved vertebral bodies confirms their acuity.

Quantitative Computerized Tomography Three-dimensional evaluation of cortical and trabecular bone to assess skeletal macro and microarchitecture is possible using CT. Quantitative computerized tomography (QCT) can be performed in two ways. The most common method is

to use a standard CT scanner to perform low-dose examination of the hip or spine and acquire volumetric threedimensional (3D) measurements. The other method requires a dedicated extremity scanner performing high-resolutionperipheral quantitative computed tomography (HR-)pQCT.

Osteoporosis

Diagnosis of osteoporosis on qCT is based on measurement at the hips or spine. The femoral neck qCT BMD measurement can also be used to determine fracture risk using the FRAX tool (FRAX score). However, T-score values for spine qCT are not standardized and hence are not recommended for routine use. Instead, absolute numeric values of 120 and 80 mg/cm3 should be used as a cutoff. A  value 20% loss of signal intensity in opposed images confirming intravoxel lipid, while malignant fracture will demonstrate 20% signal loss in the opposed-phase images relative to the in-phase image. However, the signal drop on L3 and L5 in the opposed-phase images is minimal and less than 20% (blue dashed lines) in keeping with a pathological fracture from underlying metastasis.

In the clinical setting, with advanced age, fat fraction increases (Figure 23.12), and perfusion enhancement peak and slope indexes decrease. Study of enhancement slopes and maximum enhancement provides valuable information about bone quality and BMD in osteoporotic patients [23].

Quantitative Ultrasound Accessibility, cost-effectiveness, and lack of radiation risk make quantitative ultrasound (QUS) a potentially attractive method for the routine evaluation of osteoporosis. QUS utilizes mechanical wave propagation through the bone to characterize physical properties in healthy and osteoporotic patients. In clinical practice, calcaneus bone has been widely validated. Data available demonstrated that velocity and attenuation values are lower in postmenopausal than premenopausal women  [25]. Results are controversial when compared to DXA performance due to the heterogeneity of methodologies and devices. A large meta-analysis showed that the variability in sensitivity

and specificity results could not exclude or confirm osteoporosis diagnosis [26]. Nonetheless, assessing fracture risk in both women and men seems promising and reproducible [27, 28]. The ISCD does not recommend QUS for diagnosis nor as a decision-making tool to initiate treatment. However, the ISCD recognizes the individual benefit of QUS for fracture risk assessment and its value to predict fractures in postmenopausal women and men over the age of 65, specifically through heel evaluation and only in individually validated devices [29].

Summary The increasing incidence of osteoporosis among the population represents a significant and debilitating burden on society, as reflected by high healthcare costs and low quality of life for the patients. Early detection and management focused on prevention might lessen the effects of this disease in the healthcare system. A systematic and

Osteoporosis

(a)

Young subject

6

2 1 2 14 8 2 0 10 ) (ms δ (ppm) TE

4

Fitted fat peaks Restricted fat peaks

multidisciplinary approach to this diagnosis is necessary. The application of different imaging modalities can enhance diagnostic sensitivity, aiding clinicians to target interventions in a timely manner. DXA remains the gold standard for the detection of low bone density. However, robust evidence supports opportunistic evaluation of bone quality

(b)

Old subject

Signal (a.u.)

Signal (a.u.)

Figure 23.12 Vertebral bone marrow spectra (measured spectra in black) with fitted fat peaks (in red), fitted water peak (in blue), and fat peaks whose peak areas were restricted to the main fat peak area (in green) in (a) a young (23-year-old female) and (b) an old (64-year-old male) subject. Source: Reproduced from Karampinos et al. [23]/with permission of John Wiley & Sons.

6

4

2 δ (ppm)

2 2 5 15 0 0 11 ) (ms TE

Fitted water peaks Measured signal

with a radiograph, CT, and MRI, which must be considered in selected patients when available to the radiologist. Finally, our understanding of bone microarchitecture through the development of new imaging techniques continues to advance, providing encouraging perspectives in disease monitoring and possible clinical applications.

References 1 Kanis, J.A. (1994). Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report. Osteoporos. Int. 4: 368–381. 2 Wright, N.C., Looker, A.C., Saag, K.G. et al. (2014). The recent prevalence of osteoporosis and low bone mass in the United States based on bone mineral density at the femoral neck or lumbar spine. J. Bone Miner. Res. 29 (11): 2520–2526. 3 Black, D.M. and Rose, C.J. (2016). Postmenopausal osteoporosis. N. Engl. J. Med. 374: 254–262. 4 Burge, R., Dawson-Hughes, B., Solomon, D.H. et al. (2007). Incidence and economic burden of osteoporosis-related fractures in the United States, 2005–2025. J. Bone Miner. Res. 22 (3): 465–475. 5 National Osteoporosis Foundation (1998). Osteoporosis: review of the evidence for prevention, diagnosis and treatment and cost-effective analysis. Osteoporos. Int. 8: S7–S80. 6 Cosman, F., deBeur, S.J., MS, L.B. et al. (2014). Clinician’s guide to prevention and treatment of osteoporosis. Osteoporos. Int. 25 (10): 2359–2381. 7 Lenchik, L., Rogers, L.F., Delmas, P.D., and Genant, H.K. (2004). Diagnosis of osteoporotic vertebral fractures: importance of recognition and description by radiologists. Am. J. Roentgenol. 183 (4): 949–958. 8 Lewiecki, E.M., Binkley, N., Morgan, S.L. et al. (2016). Best practices for dual-energy X-ray absorptiometry measurement and reporting: International Society for Clinical Densitometry guidance. J. Clin. Densitom. 19 (2): 127–140.

9 Jang, S., Graffy, P.M., Ziemlewicz, T.J. et al. (2019). Opportunistic osteoporosis screening at routine abdominal and thoracic CT: normative L1 trabecular attenuation values in more than 20000 adults. Radiology 291 (2): 360–367. 10 Kaup, M., Wichmann, J.L., Scholtz, J.E. et al. (2016). Dual-energy CT-based display of bone marrow edema in osteoporotic vertebral compression fractures: impact on diagnostic accuracy of radiologists with varying levels of experience in correlation to MR imaging. Radiology 280 (2): 510–519. 11 Engelke, K., Lang, T., Khosla, S. et al. (2015). Clinical use of quantitative computed tomography (QCT) of the hip in the management of osteoporosis in adults: the 2015 ISCD official positions–part I. J. Clin. Densitom. 18 (3): 338–358. 12 Zysset, P., Qin, L., Lang, T. et al. (2015). Clinical use of quantitative computed tomography-based finite element analysis of the hip and spine in the management of osteoporosis in adults: the 2015 ISCD official positions– part II. J. Clin. Densitom. 18 (3): 359–392. 13 Nishiyama, K.K. and Shane, E. (2013). Clinical imaging of bone microarchitecture with HR-pQCT. Curr. Osteoporos. Rep. 11 (2): 147–155. 14 Engelke, K., Lang, T., Khosla, S. et al. (2015). Clinical use of quantitative computed tomography-based advanced techniques in the management of osteoporosis in adults: the 2015 ISCD official positions–part III. J. Clin. Densitom. 18 (3): 393–407. 15 Schmitz, A., Risse, J.H., Textor, J. et al. (2002). FDG-PET findings of vertebral compression fractures in

669

670

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

16

17

18

19

20

21

22

osteoporosis: preliminary results. Osteoporos. Int. 13 (9): 755–761. Shin, D.S., Shon, O.J., Byun, S.J. et al. (2008). Differentiation between malignant and benign pathologic fractures with F-18-fluoro-2-deoxy-D-glucose positron emission tomography/computed tomography. Skeletal Radiol. 37 (5): 415–421. Cho, W.I. and Chang, U.K. (2011). Comparison of MR imaging and FDG-PET/CT in the differential diagnosis of benign and malignant vertebral compression fractures. J. Neurosurg. Spine 14 (2): 177–183. Zhuang, H., Sam, J.W., Chacko, T.K. et al. (2003). Rapid normalization of osseous FDG uptake following traumatic or surgical fractures. Eur. J. Nucl. Med. Mol. Imaging 30 (8): 1906–2103. Schwaiger, B.J., Kopperdahl, D.L., Nardo, L. et al. (2017). Vertebral and femoral bone mineral density and bone strength in prostate cancer patients assessed in phantomless PET/CT examinations. Bone 101: 62–69. Kay, F.U., Ho, V., Dosunmu, E.B. et al. (2021). Quantitative CT detects undiagnosed low bone mineral density in oncologic patients imaged with 18F-FDG PET/ CT. Clin. Nucl. Med. 46 (1): 8–15. Reilly, C.C., Raynor, W.Y., Hong, A.L. et al. (2018). Diagnosis and monitoring of osteoporosis with 18F-sodium fluoride PET: an unavoidable path for the foreseeable future. Semin. Nucl. Med. 48 (6): 535–540. Mauch, J.T., Carr, C.M., Cloft, H., and Diehn, F.E. (2018). Review of the imaging features of benign osteoporotic

23

24

25

26

27

28

29

and malignant vertebral compression fractures. Am. J. Neuroradiol. 39 (9): 1584–1592. Karampinos, D.C., Rushke, S., Dieckmeyer, M. et al. (2018). Quantitative MRI and spectroscopy of bone marrow. J. Magn. Reson. Imaging 47 (2): 332–353. Sollman, N., Loffler, M.T., Kronthaler, S. et al. (2020). MRI-based quantitative osteoporosis imaging at the spine and femur. J. Magn. Reson. Imaging https://doi.org/ 10.1002/jmri.27260. Epub ahead of print. Nairuss, J., Ahmadi, S., Baker, S., and Baran, D. (2000). Quantitative ultrasound: an indicator of osteoporosis in perimenopausal women. J. Clin. Densitom. 3 (2): 141–147. Nayak, S., Olkin, I., Liu, H. et al. (2006). Meta-analysis: accuracy of quantitative ultrasound for identifying patients with osteoporosis. Ann. Intern. Med. 144 (11): 832–841. Gluer, C.C., Eastell, R., Reid, D.M. et al. (2004). Association of five quantitative ultrasound devices and bone densitometry with osteoporotic vertebral fractures in a population-based sample: the OPUS study. J. Bone Miner. Res. 19 (5): 782–793. Khaw, K.T., Reeve, J., Luben, R. et al. (2004). Prediction of total and hip fracture risk in men and women by quantitative ultrasound of the calcaneus: EPIC-Norfolk prospective population study. Lancet 363 (9404): 197–202. Krieg, M.A., Barkmann, R., Gonnelli, S. et al. (2008). Quantitative ultrasound in the management of osteoporosis: the 2007 ISCD official positions. J. Clin. Densitom. 11 (1): 163–187.

671

24 Emergency Radiology Sean K. Johnston1, Russell Flato2, Peter Hu2, Peter Henry Joyce2, and Andrew Chong2 1 2

Department of Radiology, Division of Emergency Radiology, Keck School of Medicine of USC, LAC+USC Medical Center, Los Angeles, CA, USA Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA

Emergency Imaging The role of nuclear medicine in the emergency setting is varied and has changed a lot in the last few decades. While the increasing use of computed tomography (CT) imaging in the emergency room (ER) has led to a drop in the number of emergency nuclear medicine exams such as ventilation/perfusion (V/Q) or tagged red blood cell (RBC) scans ordered around the country, the usefulness of nuclear medicine in the ER remains. Thus, it is even more important now that we educate our colleagues about the important and vital role nuclear medicine imaging can play in the emergency setting. In this chapter we will review some of the more common nuclear medicine tests utilized in the ER and illustrate how they can be used in a complimentary fashion with more commonly used imaging modalities like CT and ultrasonography (US) as well as common pathologies that are encountered. The different examinations discussed in the chapter are covered in greater detail separately within this book and can be referred to if further detail is needed.

Myocardial Perfusion Imaging Chest pain is the underlying reason for approximately 6.5  million emergency department visits in the United States per year [1]. Acute chest pain necessitates a prompt and thorough evaluation in the acute clinical setting to triage patients based on their risk factors and probability of unstable angina or infarct. Patients with a high likelihood of myocardial ischemia or infarction, as confirmed by positive electrocardiography (EKG) findings and/or serum cardiac enzyme levels, will often proceed immediately to

coronary angiography. When there is adequate clinical suspicion but routine diagnostic tests are equivocal, myocardial perfusion imaging is often the next step in evaluation to determine whether a patient needs further workup or can be safely discharged. Myocardial perfusion scintigraphy has been previously shown to accurately predict adverse cardiac outcomes in ER patients presenting with typical angina and nondiagnostic EKG findings [2]. Myocardial perfusion scintigraphy provides a map of regional myocardial perfusion, with ischemia or infarction represented as a “cold” region or perfusion defect on the images. The study is usually limited to the acquisition of only resting (nonstress) images for patients with acute chest pain. The examination is commonly performed using electrocardiographically gated SPECT, while 2D planar imaging is reserved for large patients who exceed weight restrictions on the single-photon emission computer tomography (SPECT) imaging table. The radiotracer typically used is technetium-99m sestamibi (99mTc-MIBI) or 99mTc-tetrofosmin, although thallium-201 chloride was used historically and newer agents are becoming available. The radiotracer is optimally injected while the patient is actively experiencing angina symptoms, since radiotracer administration after the resolution of chest pain can decrease diagnostic accuracy. Both 99mTc-sestamibi and 99m Tc-tetrofosmin can be injected in the ER, and since their pharmacokinetics are such that the radiotracers remain fixed within the myocardium, delayed imaging will reflect perfusion at the time of injection. In a normal study myocardial perfusion rest images will demonstrate uniform radiotracer uptake throughout the left ventricular myocardium (see Figure 24.1). The left ventricle will appear doughnut-shaped on short-axis views, and the heart will appear horseshoe or U-shaped on horizontal and

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach, First Edition. Edited by Ali Gholamrezanezhad, Majid Assadi, and Hossein Jadvar. © 2023 John Wiley & Sons Ltd. Published 2023 by John Wiley & Sons Ltd.

672

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 24.1  Normal regadenoson sestamibi perfusion study. 10.4 mCi of 99mTc-MIBI was administered and rest SPECT images were obtained. Subsequently 0.4 mg of regadenoson was administered intravenously followed by 26.2 mCi of 99mTc-MIBI and stress SPECT images were obtained. The rest and stress images are identical and normal.

vertical long-axis views. The myocardium will also normally appear thinner at the apex. The right ventricle will have less uptake given its lesser myocardial mass, and the atria are not visualized. Physiologic lung uptake can also be seen, although increased uptake can be seen in chronic smokers and patients with lung disease and heart failure [3]. Given that only resting myocardial perfusion images are acquired in an acute setting, it is not possible to reliably distinguish between myocardial ischemia and infarction.

Nonetheless, important features to note when describing a perfusion abnormality include the size of the defect, severity (subendocardial or transmural), location (i.e. coronary artery territory), and presence of ventricular dilatation or wall motion abnormalities. An abnormality seen on rest-only scintigraphy has been shown to have sensitivity and specificity of 94% and 83%, respectively, for predicting an adverse cardiac event in the setting of acute angina  [4]. In other words, a normal study is highly

Emergency Radiology

Figure 24.2  Fixed and reversible defects seen on regadenoson sestamibi perfusion study. 9.8 mCi of 99mTc-MIBI was administered and rest SPECT images were obtained. Subsequently 0.4 mg regadenoson was administered intravenously followed by 28.4 mCi 99m Tc-MIBI and stress SPECT images were obtained. The stress images are displayed above the rest images. There is a moderate-sized fixed defect involving the basal inferior and inferolateral wall (red arrows), consistent with infarct. Additionally, there is a moderatesized partially reversible defect of moderate intensity involving the inferoapical and inferolateral wall (blue arrows), compatible with peri-infarct ischemia.

predictive of a good prognosis. If no abnormality is seen at rest-only imaging, a complete rest-stress study may be performed at a later time, after the patient’s acute symptoms have resolved (see Figure 24.2). Anticipating the potential pitfalls to accurate study interpretation requires an understanding of the distinct imaging artifacts that may be encountered. For example, soft tissue attenuation by large breasts or breast implants can simulate an anterior or lateral myocardial wall defect, whereas adipose tissue in the lateral chest wall of obese patients can similarly produce an apparent lateral wall defect (see Figure 24.3). In addition, an overlying left hemidiaphragm may cause attenuation of the inferior myocardial wall and simulate a defect. A method to resolve suspected diaphragmatic attenuation is to reposition the patient prone and repeat the study, which should produce an increase in measured activity if the apparent defect was due to attenuation. Other factors such as superimposed bowel, liver, or spleen activity can also confound myocardial perfusion findings. Therefore, it is always prudent to review the raw data for any potential source of artifact prior to image interpretation. Myocardial perfusion radiotracers are taken up in all metabolically active tissues within the body, except for the brain. Structures that accumulate radiotracer within the routine field of view include thyroid, salivary glands, liver, gallbladder, kidneys, bowel, and skeletal muscle. It is

important to recognize these physiologic uptake patterns, since an “extra” focus of activity may represent a metabolically active tumor, such as a parathyroid adenoma or lung cancer. Clearly, an incidental tumor should not be missed and requires further workup.

Testicular Torsion Acute scrotal pain represents a diagnostic challenge for the clinician, requiring careful evaluation to differentiate testicular torsion from inflammatory etiologies, such as epididymitis. Testicular torsion is a medical emergency affecting 4.5  in 100 000  males younger than 25 years annually  [5]. Trauma-related torsion accounts for only 4–8% of cases [6], while the majority tend to occur spontaneously [7]. One predisposing factor involves faulty attachment of the testis to the scrotal wall, most commonly due to a congenital anomaly called the bell-clapper deformity, in which the tunica vaginalis completely surrounds the testis, resulting in a testis that is solely attached to the spermatic cord and otherwise able to rotate freely. The viability of the testis depends on the duration of torsion and degree of spermatic cord rotation. Proper diagnosis and treatment is ideally achieved within 6 hours of symptom onset, as the rate of testicular salvage is over 90% in

673

674

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 24.3  Breast attenuation. SPECT images from a regadenoson sestamibi perfusion study show a mild central area of hypoactivity within the anteroseptal wall (white arrows) with normal activity distally. This area demonstrates normal wall motion (not pictured) and was attributed to breast attenuation.

the first 12 hours after symptom onset but falls to 18% after 24 hours [8].

Ultrasonography US has become the main imaging modality for evaluating acute scrotal pain given its widespread availability, ability to be performed quickly, and lack of ionizing radiation. At

gray-scale US, the normal testes appear as ovoid, homogenous structures with medium echogenicity. An echogenic line surrounds each testis, corresponding with the tunica albuginea. In the first few hours of torsion, the testis can appear normal in echogenicity. Over time, the torsed testis may demonstrate increased or decreased echogenicity and a heterogeneous appearance. An enlarged testis with heterogeneous echotexture is suggestive of infarction (see Figure 24.4).

Figure 24.4  Testicular infarct. Transverse ultrasound of the right testis shows a wedge-shaped heterogeneously hypoechoic area extending to the periphery of the right testis. Transverse color Doppler demonstrates diminished vascularity in this area, compatible with infarct.

Emergency Radiology

Color Doppler US is a valuable tool to evaluate for ischemia by comparing vascular flow between the normal and affected testes. Complete testicular torsion is demonstrated by the absence of blood flow on the affected side and normal flow on the contralateral side. Less pronounced cases may show asymmetrically decreased flow on the affected side when compared to the normal side. When presented with less definitive findings, it is important to utilize the combination of gray-scale and Doppler US findings in making the diagnosis (see Figure 24.5). Despite its high diagnostic accuracy, false positives and negatives can occur with gray-scale and color Doppler US. Technical skill is a limitation, as there is certainly variability based on sonographer experience and equipment. Testicular perfusion can be difficult to detect in prepubertal boys with small volume testes, especially with older US equipment  [9]. False negatives can also occur when

Figure 24.5  Testicular torsion. Sagittal color Doppler ultrasound of the right testis shows asymmetric enlargement and heterogeneous echotexture with absent vascular flow, consistent with torsion. Sagittal color Doppler of the left testis demonstrates normal echotexture and vascular flow.

testicular torsion is mild or intermittent, and there is minimal to no decrease in blood flow at Doppler imaging (see Figure 24.6).

Scintigraphy Evaluation of acute scrotal pain using scintigraphy was first described by Nadel et  al.  [10]. The basic premise behind scintigraphy is that a testis which has undergone torsion should have decreased perfusion and demonstrate a “cold” defect on radionuclide scan. Conversely, inflammatory processes, such as epididymitis, should cause hyperemia and exhibit an increased radiotracer uptake “hot” area on imaging. Technetium-99m sodium pertechnetate is the commonly used radiopharmaceutical agent for scrotal scintigraphy. The typical adult dose is 15 mCi given intravenously, while pediatric patients receive a proportionately smaller dose given differences in body weight [11]. Scrotal scintigraphy can be divided into three phases: early dynamic flow, late dynamic flow, and postdynamic static images. Interpretation requires comparison of the normal and symptomatic sides in a systematic manner. On early dynamic images, flow through the spermatic cord and pudendal vessels is compared with that on the contralateral side, while also comparing cord flow relative to pudendal vessel flow on the symptomatic side. Scrotal wall perfusion is optimally assessed on late dynamic images. Lastly, on static images the appearance of the cord, testis, and epididymis on the normal side is compared with that on the symptomatic side, looking for areas of increased or decreased activity. Scintigraphic findings in testicular torsion can vary based on the degree of ischemia, but generally the findings can be grouped into early, mid phase, and late torsion. In

Figure 24.6  Normal testes after detorsion and orchiopexy. Sagittal color Doppler ultrasound of the right testis and transverse color Doppler of bilateral testes in the same patient as Figure 24.5 show normal echotexture and symmetric blood flow after surgical detorsion and orchiopexy.

675

676

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

early torsion, relatively increased activity may be seen in the testicular artery proximal to the site of the twist (“nubbin sign”) on dynamic images. On static images, decreased or absent activity may be seen in the region of the affected testis. In the mid phase of torsion, prolonged testicular ischemia leads to infarction, and the scrotal wall subsequently becomes hyperemic. Scintigraphic findings during this stage may demonstrate increased activity in the region of the pudendal vessels and scrotum with relatively decreased or absent activity in the affected testis. This appearance of a “cold” testis with a surrounding halo of increased activity becomes more prominent in late torsion. In the late phase of torsion, persistent ischemia leads to increasing pudendal flow and scrotal hyperemia. As one would expect, the intensity of scrotal activity surrounding the ischemic/infarcted testis increases further. Other causes of acute scrotal pain should be entertained when the imaging findings are equivocal or inconsistent with testicular torsion, such as inflammatory disease and torsion of the appendix testis. Scrotal inflammatory disease, most commonly epididymitis and epididymo-orchitis, classically shows enlargement of the affected structures on gray-scale US and asymmetric hyperemia on color Doppler  [12] (see Figure  24.7). Scintigraphy may show increased activity in the spermatic cord vessels on dynamic phase images, as well as increased activity in the epididymitis in addition to the testis (in epididymo-orchitis) on static images. Torsion of the appendix testis classically reveals a round, hyperechoic or mixed echogenicity mass at the upper pole of the epididymis without internal vascular flow  [13]. Reactive hypervascularity in the adjacent epididymis and scrotal wall may be seen at color Doppler US. Scintigraphic findings are usually normal or sometimes may show a focal area of increased activity above the testis on static images.

Both Doppler sonography and scintigraphy are valuable techniques for evaluating acute scrotal pain, with reported accuracies around 88% and 95%, respectively [14]. Rather than taking a standalone approach, supplementing both imaging modalities can be advantageous in reducing their respective limitations. For example, scintigraphy may detect a photopenic defect overlying the testis caused by a large hydrocele, leading to an incorrect diagnosis of testicular torsion  [15]. By correlating with sonography, which would readily visualize a large hydrocele, a potentially false-positive scintigraphic scan can be avoided. Ultimately, the goal is to correctly identify patients needing urgent surgical intervention while avoiding unnecessary surgery.

Renal Scintigraphy While US and CT have emerged as the preferred imaging modalities for evaluation of renal anatomy, the use of renal scintigraphy in the acute clinical setting is typically reserved for functional analysis. The most common acute indications for radionuclide renal imaging include urinary tract obstruction, urine leak, and vascular compromise. Renal scintigraphy can be particularly valuable when urgent interventions are being considered, such as percutaneous nephrostomy, ureteral stent placement, or surgery.

Urinary Tract Obstruction US is often the initial imaging modality used to assess the urinary system, given its lack of ionizing radiation and ability to be performed at the patient’s bedside. The renal collecting system appears as anechoic spaces that conform to the expected shape of the renal calyces and pelvis [16]. On

Figure 24.7  Epididymitis. Sagittal ultrasound of the left testicle and epididymis shows enlargement of the left epididymal tail. Sagittal color Doppler of the left epididymal tail demonstrates markedly increased flow, consistent with focal epididymitis.

Emergency Radiology

Figure 24.8  Severe hydronephrosis. Sagittal grayscale ultrasound of the kidney shows marked dilatation of the collecting system and cortical thinning, consistent with severe hydronephrosis.

US, hydronephrosis is defined by dilatation of the collecting system to varying degrees (see Figure 24.8). Mild cases can be identified by modest distention of the renal pelvis and calyces without parenchymal atrophy. Moderate hydronephrosis is demonstrated by progressive dilatation of the renal collecting system with blunting and flattening of the renal calyces and papillae, respectively; slight cortical thinning may also be present. Finally, severe hydronephrosis can be identified by marked dilatation or a balloon appearance of the renal pelvis and calyces with loss of their intervening borders; cortical thinning is also usually seen. Once hydronephrosis is identified, further investigation should be performed to reveal the underlying cause, such as stones, ureteropelvic junction (UPJ) obstruction, or malignancy. Alternatively, unenhanced CT may be performed for suspected obstruction caused by renal or ureteral stones. The sensitivity and specificity of unenhanced CT for detecting ureteral calculi has been shown to be around 95–96% and 98%, respectively [17]. CT images frequently reveal varying degrees of hydronephrosis and/or hydroureter upstream of the obstructing calculus, which can be anywhere along the urinary tract. Most stones are radiopaque and commonly contain calcium, although they will vary in density based on their composition. Of note, protease inhibitor (indinavir) stones and other radiolucent stones are usually undetectable on unenhanced CT. Magnetic resonance imaging (MRI) is rarely performed for renal evaluation in the acute setting but may be considered on subsequent workup on a case-by-case basis. While US and CT can provide good visualization of renal anatomy, they are not well-suited for functional analysis. Therefore, renal scintigraphy can be performed for functional evaluation of urinary tract obstruction  [18].

Diuretic renal scintigraphy has become a key diagnostic test to help differentiate obstructive hydronephrosis from a nonobstructive cause of collecting system dilatation. The most commonly used radiopharmaceutical agent is technetium-99m mertiatide (99mTc-MAG3). Since 99mTcMAG3 is cleared primarily by tubular secretion (95%) with minimal glomerular filtration (24 months) and can be further classified, based on duration of symptoms, as acute (6 weeks). Acute infection presents as septic arthritis with erythema, swelling, pain, and abnormal laboratory test results and is readily diagnosed. Diagnosing chronic infection is more challenging and arthrocentesis is required for definitive diagnosis [61].

Spondylodiscitis is an infection of the vertebral body and/ or disc and accounts for about 1% of all cases of osteomyelitis. The infection may extend into the epidural space, posterior elements, and paraspinal soft tissues. The lumbar, thoracic, and cervical spine, in decreasing order of prevalence, are the major sites of involvement. In about 65% of cases, the infection is limited to a single spinal segment: two contiguous vertebral bodies and the intervening disc. Multilevel contiguous infection occurs in about 20% of cases and noncontiguous infection in about 10% of cases. Predisposing risk factors include advanced age, diabetes mellitus, coronary artery disease, spinal interventions, immunosuppression, intravenous drug use, and comorbidities such as endocarditis and urinary tract infection. Staphylococcus aureus (60%) followed by Enterobacter species (30%) are the most common offending organisms [64]. In developing countries and among HIV-infected patients, Mycobacterium tuberculosis is an important cause of spondylodiscitis, affecting the thoracic spine more frequently than the rest of the spine and with a propensity for multilevel involvement [65]. Approximately 95% of pyogenic cases of spondylodiscitis involve the vertebral body, with only 5% involving the posterior elements. Posterior element involvement is more frequent in tuberculous and fungal than in pyogenic infections [64]. Spondylodiscitis frequently is an indolent disease. There may be a long interval between onset of symptoms and diagnosis [66]. Back pain, which is present in a myriad of other conditions affecting the spine, followed by fever are the most common presenting symptoms. C-reactive protein levels and erythrocyte sedimentation rate can be

Infection/Inflammation Imaging

elevated, but are not specific, and the peripheral white blood cell count is not sensitive [64]. Bone scintigraphy frequently is used as a screening test in suspected spondylodiscitis. However, false-negative results have been reported in elderly individuals, presumably due to atherosclerosis-induced ischemia. In addition, bone scintigraphy is not useful for detecting the soft tissue infections that often accompany, or mimic, spondylodiscitis. The bone scan can remain abnormal for some time after the infection has resolved due to ongoing bony remodeling during healing. Performing the test as a three-phase bone scan does not improve accuracy because although it improves specificity, it does so at the expense of sensitivity [2]. 67 Ga, alone and in combination with bone scintigraphy has been used to diagnose spondylodiscitis. 67Ga improves the bone scan specificity, may detect infection sooner than the bone scan, and can identify accompanying soft tissue infections, which frequently go unrecognized on bone scintigraphy  [2]. Incorporating SPECT/CT into the 67Ga imaging protocol has incremental value by precisely localizing radiopharmaceutical uptake and improving disease detection (Figure 26.16) [64]. In a retrospective investigation, combined bone/67Ga SPECT/CT detected 17/18 cases of spondylodiscitis and was negative in all 16 patients without infection (97% accuracy) [67]. WBC imaging is not useful for diagnosing spondylodiscitis because, for reasons that are not well understood, 50% or more of all cases present as areas of decreased or absent activity. The

Coronal

photopenic presentation is associated with several conditions in addition to active infection, including treated spondylodiscitis, tumor, infarction, compression fracture, and Paget’s disease [64]. The role of 18F-FDG in the diagnosis of spondylodiscitis has been extensively investigated and the results have been uniformly excellent. In one meta-analysis, the pooled sensitivity and specificity of 18F-FDG PET/PET-CT were 97% and 88%  [68]. In another, more recent, meta-analysis, 18 F-FDG PET/CT had a pooled sensitivity of 94.8% and a  pooled specificity of 91.4% (95% CI, 78.2–96.9%)  [69]. 18 F-FDG is superior to single-photon emitting radiopharmaceuticals for detecting both bone and soft tissue infections and for differentiating severe arthritic changes from infection [64]. Postoperative spondylodiscitis has a prevalence ranging from 0.5% to approximately 19% and often has an indolent, nonspecific presentation. Superficial infections are easily diagnosed, while diagnosis of deeper infections is more difficult. Nonspecific back pain and constitutional symptoms are the most common presentation. Fever is present in about half of the cases and laboratory tests are of limited value. Prompt diagnosis is imperative because a delay may lead to involvement of the bone, epidural space, and paravertebral soft tissues, and may necessitate hardware removal, which can lead to instability and pseudoarthrosis [70]. There are few data on bone, gallium, and labeled leukocyte imaging in the setting of postoperative

Sagittal

Figure 26.16 Infected spinal hardware. There is focally increased accumulation of gallium-67 in the upper thoracic spine (arrows) in a patient with extensive spinal hardware.

729

730

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

spondylodiscitis. In one series 67Ga scintigraphy failed to differentiate postoperative changes from infection [71]. In a meta-analysis of 18F-FDG for diagnosing postoperative spondylodiscitis, the summary AUC for spondylodiscitis was 0.92  in patients with versus 0.98  in patients without spinal hardware. False-positive results were more common in patients with than in patients without hardware (12.8 vs. 7%), presumably due to hardware-induced aseptic inflammation. Performing PET/CT rather than PET alone reduces hardware-associated false-positive results  [68]. Analyzing uptake patterns may facilitate the differentiation of aseptic inflammation from infection. In an investigation of children and young adults, confluent increased 18F-FDG uptake in soft tissue and bone immediately adjacent to the hardware at multiple contiguous levels was associated with infection, while noninfectious hardware complications were characterized by focal 18F- FDG uptake adjacent to one or two hooks, screws or anchors, usually at the upper or lower aspects of the spinal hardware. The intense contiguous areas of uptake in bone and soft tissue surrounding infected hardware was not present in these patients [72]. 18 F-FDG has shown promise for monitoring response to treatment in patients with spondylodiscitis (Figure 26.17). Some investigators have reported that changes in maximum standardized uptake value (SUVmax) reliably differentiate responders from nonresponders while other investigators have found that changes in uptake patterns (a)

are equally useful. 18F-FDG is more accurate than MRI for this purpose [64, 73, 74]. Diabetic Foot Osteomyelitis

Diabetic foot infections are infections of the soft tissues or bone below the malleoli in diabetic individuals. Major predisposing factors include peripheral arterial disease, peripheral neuropathy, and compromised immunity [75–77]. These infections usually occur at sites of skin trauma or ulceration. Sixty percent of the ulcers become infected during treatment, and about 20% progress to osteomyelitis. The diabetic foot ulcer is one of the most common reasons for hospitalizations. Two-thirds of diabetic patients with foot infections severe enough to require hospitalization have underlying osteomyelitis. These patients have worse outcomes, more surgeries and amputations, longer hospitalizations, and higher rates of recurrent infection and readmission for infection than do patients with soft tissue infection alone [78, 79]. Unfortunately, diabetics can have a significant foot infection with few or no signs or symptoms and without mounting a systemic inflammatory response, and the diagnosis of osteomyelitis can easily be overlooked [79]. Most investigations have found that the bone scan has high sensitivity, but a poor specificity for diagnosing diabetic foot osteomyelitis. In a meta-analysis, bone scintigraphy had a pooled sensitivity of 81% and a pooled specificity

(b)

Figure 26.17 Spondylodiscitis. On the pretreatment image (a) there is increased 18F-FDG uptake in T2–T4 vertebrae (arrows). There also is increased 18F-FDG uptake around the esophageal stent (arrowhead) in this patient with metastatic lung carcinoma. On the follow-up study (b) performed approximately 3 months later there is complete resolution of the vertebral body/disc space uptake (arrows). There is persistent activity around the esophageal stent (arrowhead) thought to represent foreign body reaction and/or metastatic disease. 18F-FDG has shown promise for monitoring response to treatment in patients with spondylodiscitis. (Source: reproduced with permission from Palestro [4].)

Infection/Inflammation Imaging

of 28% for diabetic foot osteomyelitis versus a pooled sensitivity of 74% and a pooled specificity of 68% for 111Inlabeled leukocyte scintigraphy  [80]. Evaluating arterial hyperperfusion and performing next day imaging (fourphase bone scintigraphy) have improved specificity, but at the expense of sensitivity [79]. Using bone scintigraphy as a screening test in diabetic foot infections is of limited value because the vast majority are positive whether or not the patient has osteomyelitis. 67 Ga scintigraphy, alone and in combination with bone scintigraphy, is not useful for diagnosing diabetic foot osteomyelitis. In one investigation, the sensitivity, specificity, and accuracy of 67Ga scintigraphy were 100%, 40%, and 73%, respectively. The test was more accurate than threephase bone scintigraphy (59%), but less accurate than 111 WBC scintigraphy (86%) and combined bone/WBC scintigraphy (91%). The sensitivity, specificity, and accuracy of combined bone/67Ga scintigraphy were identical to those of 67Ga scintigraphy alone  [81]. In another investigation, the sensitivity and specificity of combined bone/67Ga scintigraphy were 44% and 77%, respectively [82]. Performing SPECT/CT does not improve the accuracy of the test. In an investigation of 42 patients, the sensitivity, specificity, and accuracy of 67Ga SPECT/CT were 100%, 45%, and 62% respectively [83]. WBC scintigraphy has been used to diagnose diabetic foot osteomyelitis for many years. In one meta-analysis 111 In WBC scintigraphy had a pooled sensitivity of 74% and a pooled specificity of 68% for diagnosing diabetic foot osteomyelitis [81]. In another meta-analysis, although the pooled sensitivities of 99mTc WBC and 111In WBC scintigraphy were similar, 91% versus 92%, respectively, the pooled

(a)

specificity 99mTc WBC scintigraphy was higher, 92% vs. 75%. Four studies performed with 99mTc WBCs evaluated scintigraphy with SPECT or SPECT/CT. In these studies, the sensitivity ranged from 0.88% to 1.00% and the specificity from 0.35% to 1.00% (Figure 26.18) [84]. Combined bone/WBC scintigraphy offers a modest improvement in accuracy compared to WBC scintigraphy alone. In a prospective investigation using 111In WBCs, the accuracy of the combined test was 91% versus 86% for WBC scintigraphy alone  [81]. In a prospective investigation using 99mTc WBCs the sensitivity and specificity of the combined test were 88% and 97%, respectively  [85]. In another prospective investigation using 99mTc WBCs the sensitivity and specificity of the combined test were 93% and 98%, respectively  [86]. Unfortunately, the combined study was not compared to WBC imaging alone in either of these two investigations. One retrospective investigation using 111In WBCs evaluated combined bone/WBC SPECT/CT for diagnosing diabetic foot osteomyelitis. The sensitivities of bone SPECT/ CT, WBC SPECT/CT, and combined bone/WBC SPECT/ CT were 94%, 87%, and 95%, respectively. The specificity of combined bone/WBC SPECT/CT was 94%, significantly higher than that of bone SPECT/CT (47%) and WBC SPECT/CT (68%) individually [87]. Although these results are better than what has been reported for planar In WBC imaging, they are similar to what has been reported for planar Tc WBC and for combined bone/WBC imaging with both 111In and 99mTc WBCs [81, 84–86]. 18 F-FDG also is useful for diagnosing diabetic foot osteomyelitis. In one meta-analysis the pooled sensitivity and specificity of 18F-FDG were 89% and 92%, respectively,

(b)

Figure 26.18 Osteomyelitis left first metatarsal. On the planar image (a) there is abnormal accumulation of technetium-99m-labeled leukocytes along the medial aspect of the distal left foot (arrow). It is not possible to determine if this activity is confined to the soft tissues or involves the bone as well. On the axial SPECT/CT image (b) it is clear that the abnormality involves both bone and soft tissue.

731

732

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

Figure 26.19 Osteomyelitis right second toe. There is bony destruction of the second digit on the CT component of the examination (a). There is focal 18F-FDG activity in this toe on the PET image (b). The fused image (c) confirms that the bone is involved (Source: Courtesy Dr. G. Abikhzer.)

similar to the pooled sensitivity and specificity of 99mTcscintigraphy: 91% and 92%, respectively (Figure 26.19) [84]. In another meta-analysis, the pooled sensitivity and specificity of 18F-FDG were 74% and 91%, respectively [88]. Although most diabetics undergoing molecular imaging for osteomyelitis present with an ulcer in the distal forefoot, another less frequent complication is the neuropathic, or Charcot, joint. Differentiating between the neuropathic joint and osteomyelitis, or diagnosing osteomyelitis superimposed on the neuropathic joint is challenging. Threephase bone scintigraphy usually is positive in both situations. Labeled WBCs accumulate in both the infected and the uninfected neuropathic joint. Performing complementary marrow imaging facilitates the differentiation of WBC uptake due to infection from uptake due to bone marrow [79, 87]. Septic Arthritis

Septic arthritis has an annual incidence of 10/100000 individuals in the United States. Patients with joint prostheses, rheumatoid arthritis, diabetes mellitus, and HIV are at increased risk for this entity  [89, 90]. Septic arthritis can involve multiple joints, causing rapid joint destruction. Prompt diagnosis to facilitate appropriate antibiotic management is essential, since cartilage can be destroyed within days and in-hospital mortality of untreated infections can be as high as 15% [91]. The classic presentation of septic arthritis on three-phase bone scintigraphy consists of hyperperfusion, hyperemia, and increased activity limited to the articular surfaces of the involved bones on the skeletal phase. This presentation is associated with both septic and aseptic arthritis. Osteomyelitis and acute arthritis are not mutually exclusive, and bone scan findings consistent with septic arthritis do not exclude underlying osteomyelitis. 67Ga and WBC

imaging are sensitive, but they cannot reliably differentiate between septic arthritis and aseptic inflammatory arthritis [92]. There are few data on the use of 18F-FDG for diagnosing of septic arthritis. 18F-FDG accumulates in a variety of inflammatory arthridities, including rheumatoid arthritis, psoriatic arthritis, and ankylosing spondylitis, and therefore, as with other molecular imaging agents, it may not be possible to differentiate septic from inflammatory arthritis with this agent [93].

Pulmonary Infections Pneumonia

The role of molecular imaging in the diagnosis of pneumonia is limited because most cases are easily and accurately diagnosed with radiologic imaging. In the early stages of the HIV/AIDS epidemic the ability of 67Ga to detect opportunistic pneumonias was invaluable. Over time, however, as clinicians became more familiar with AIDS-associated diseases, along with the increased sophistication of other imaging modalities and improved treatments, 67Ga imaging in this population all but disappeared. At the present time molecular imaging is most useful in patients with tuberculosis and in patients with sarcoid. Tuberculosis

Tuberculosis, the leading cause of infectious disease-related mortality worldwide, is caused by M. tuberculosis, a slowgrowing complex acid-fast bacillus. Approximately 20–25% of individuals exposed to M. tuberculosis develop infection. One-fourth of the world’s population is latently infected and 3–5% of these individuals develop active tuberculosis disease during their lifetime. The lungs are the most common site for the development of tuberculosis and pulmonary disease is present in more than 80% of cases. The infection can

Infection/Inflammation Imaging

spread to other parts of the body via the lymphatics, bloodstream, and direct extension. The most common sites of extrapulmonary disease are thoracic and cervical lymph nodes, spine, adrenal glands, meninges, and gastrointestinal and genitourinary tracts. In the immunocompetent population, extrapulmonary tuberculosis is present in about 20% of cases. In immunosuppressed populations, however, extrapulmonary disease can be present in more than 50% of cases [94, 95]. The presentation of active tuberculosis varies from asymptomatic to life-threatening, depending on the systems involved. Early, accurate diagnosis with prompt initiation of treatment is important to minimize the morbidity and mortality, and to reduce the likelihood of transmission. Screening tests like the Mantoux tuberculin skin test and interferon gamma release assay confirm exposure, but do not confirm the presence of active disease. The acid-fast bacilli stain is fast but requires a high organism load for positivity. Culture is the gold standard for diagnosis, but it may take up to 10 weeks for results to be available and the sensitivity has been reported to be as low as 80% [95]. Polymerase chain reaction assays have improved the diagnosis and can help identify drug resistance. The sensitivity of these tests varies from 67% in smear-negative, culture-positive patients to 98% in smear-positive patients. As a result, imaging tests assume a very important role in patients with suspected Figure 26.20 Pulmonary tuberculosis. There is intensely increased 18F-FDG in the walls of the right and left upper lobe cavities, and in the lung parenchyma bilaterally. (Source: Courtesy Professor M. M. Sathekge.)

tuberculosis who are sputum negative, unable to produce sputum, or have extrapulmonary infection [96]. There is no role for WBC imaging in the diagnostic workup of tuberculosis. 67Ga imaging is helpful, especially in extrapulmonary disease, but has never engendered widespread use [97, 98]. 18 F-FDG, in contrast, has come to play an increasingly important role in tuberculosis. This test is useful for identifying both pulmonary and extrapulmonary disease, measuring disease activity, identifying individuals with latent tuberculous infection who are at risk of developing active infection, and monitoring response to treatment. In patients with active infection, there are two general patterns of 18 F-FDG uptake: lung and lymphatic. The lung pattern is associated with pulmonary tuberculosis. Patients have predominantly pulmonary symptoms with pulmonary parenchymal involvement, with 18F-FDG uptake in areas of pulmonary consolidation (Figure 26.20). Mediastinal lymph nodes can be slightly enlarged and demonstrate moderate 18 F-FDG uptake. The lymphatic pattern is associated with systemic infection; patients have predominantly systemic, extra-thoracic disease. Mediastinal lymph nodes are larger and have higher 18F-FDG uptake than what is seen in patients with the lung pattern. Immunocompetent patients tend to develop the lung pattern, while immunocompromised

733

734

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

patients are more likely to develop the lymphatic pattern  [99]. Lesion activity as measured by standardized uptake value (SUV) correlates with disease activity. In one investigation it was possible, using dual time-point imaging, to distinguish active from inactive pulmonary tuberculomas with 100% accuracy. Active pulmonary tuberculomas had a higher SUVmax at 1 and 2 hours and a greater increase in SUVmax from the early to the late imaging compared to inactive pulmonary tuberculomas  [100]. It should be noted, however, that 18F-FDG uptake can be seen in clinically cured patients who do not develop disease on follow-up. This may represent a post-treatment state of equilibrium in which the immune system is able to contain replicating bacilli and prevent overt disease  [101]. Interpretation of metabolic activity in lesions with morphologic evidence of healed or old TB lesions must be correlated with the patient’s clinical status. Latent tuberculosis infection, which is characterized by an absence of symptoms in an individual with dormant, but live, bacilli, progresses to active infection in about 5–10% of cases. The ability to identify individuals with latent tuberculosis infection at greatest risk of progression to active infection is important because these individuals should be treated [94]. In an investigation of 35 asymptomatic adults with latent tuberculosis, there were 10 subjects with pulmonary abnormalities suggestive of subclinical active disease who were substantially more likely to progress to clinical disease. These 10 subjects had an initial negative screen for active disease by sputum culture, chest X-ray, and symptom screening. 18F-FDG showed either infiltrates and/or fibrotic scars or active nodules. These subjects were significantly more likely to have 18FFDG uptake within mediastinal lymph nodes whereas the 25 subjects with either normal lung parenchyma or discrete small nodules showed no evidence of subclinical pathology  [102]. In another investigation four of five asymptomatic subjects with normal chest radiographs and positive QuantiFERON gold assays had positive 18F-FDGPET/CT scans involving mediastinal lymph nodes, none of which met the radiological criteria for enlarged nodes. Nodal activity regressed with treatment [103]. The standard treatment duration for simple pulmonary or lymph node disease is 6 months, while the treatment duration for extrapulmonary tuberculosis is variable. Early response assessment is important, especially for patients with no bacteriological proof or with multidrug or extensively drug-resistant infection. 18F-FDG can assess early treatment response when radiological features may remain unchanged, with consequent significant impact on patient management. In an investigation of 28 subjects with multidrug resistant tuberculosis, 18F-FDG-PET/CT performed 2 months into treatment was the best method for early

prediction of treatment results and long-term outcome, with a 96% sensitivity for predicting treatment success and 79% specificity for predicting treatment failure. Similar results were achieved with CT, but not until the 6 months scan  [104]. Similar results have been reported by other investigators [105–107]. There is no absolute measure for determining if a patient is cured of tuberculosis. In an investigation of 113 HIVnegative patients who underwent 18F-FDG-PET/CT before, during, and after therapy, individuals who had achieved a clinical cure had various patterns of 18F-FDG uptake ranging from complete resolution to partial resolution to appearance of new lesions. 18F-FDG findings must be correlated with clinical data when interpreting end-of-therapy scans  [101]. While high uptake may represent persistent active disease, it may also just reflect the host immune system activity. In summary, 18F-FDG is valuable for staging tuberculosis, locating extrapulmonary disease, identifying patients with subclinical tuberculosis, and assessing early treatment response. Sarcoidosis

Sarcoidosis is a multisystemic disease that most often affects the lungs and intrathoracic lymph nodes but can involve any organ in the body. It is a world-wide disease with variable prevalence and incidence, and develops in genetically predisposed individuals after exposure to environmental triggers which incite a granulomatous inflammatory response. The sarcoid granuloma consists of a core composed of multinucleated giant cells, surrounded by CD4+ T cells and some CD8+ T cells. The granuloma in sarcoidosis usually is not necrotizing [108]. The clinical presentation of sarcoidosis depends on the organs involved and the degree of involvement. The diagnosis is based on a combination of history, physical examination, radiologic and pathologic findings, and the exclusion of other causes. Lofgren syndrome is a clinically distinct phenotype of sarcoidosis that is most common in younger patients who present with an acute onset of erythema nodosum, hilar lymphadenopathy, fever, and migratory polyarthritis, without granulomatous skin involvement. Lofgren’s syndrome has an excellent prognosis [108, 109]. The clinical presentation of pulmonary sarcoidosis, the most common form of the disease, is variable. It can be an incidental diagnosis on chest imaging performed for other reasons or can present with nonspecific symptoms including cough, dyspnea, and fatigue. Chest radiographs are often graded on a scale of 0 (normal) to 4 (pulmonary fibrosis, conglomerate mass formation). On CT of the chest, common findings include bilateral hilar and/or mediastinal lymphadenopathy, and parenchymal nodules

Infection/Inflammation Imaging

in a peribronchovascular, subpleural, and/or interlobular distribution [110]. Not surprisingly, given the histopathology of sarcoid, WBC imaging has no role for evaluating the patient with suspected or known sarcoidosis [6]. 67Ga imaging, in contrast, has been very useful in these patients. Pulmonary and extrapulmonary uptake of 67Ga in sarcoidosis is a wellknown phenomenon  [111, 112]. Certain uptake patterns increase the likelihood that the cause of an abnormal scan is sarcoidosis. One pattern is the “panda” sign because the abnormal activity in the lacrimal and parotid glands present on facial images bears a resemblance to the panda bear. Another pattern is the “lambda sign” because 67Ga uptake in the infrahilar, perihilar, and right paratracheal lymph nodes has the appearance of the Greek letter lambda. The presence of both the lambda and panda signs further increases the likelihood that the patient has sarcoidosis  [111, 113]. In patients with known sarcoid, 67Ga imaging can be used to assess pulmonary parenchymal activity and to measure response to therapy. Although 67Ga is useful, 18F-FDG has become the molecular imaging study of choice for sarcoid. 18F-FDG is more sensitive than 67Ga for detecting the extent and activity of disease  [114–116]. Additional advantages of 18F-FDG

include less radiation exposure, shorter interval between injection and imaging, superior image quality, and the ability to quantify the degree of inflammation with the SUV. The overall sensitivity of 18F-FDG for detecting sarcoid is quite high, ranging from 89% to 100%. 18F-FDG is more sensitive than the ACE and soluble interleukin-2 receptor tests. Conventional chest radiographic staging of sarcoid frequently underestimates the extent of active inflammation compared to 18F-FDG. Sarcoid is a systemic disease and whole-body imaging facilitates the identification of unsuspected sites of disease and can help guide management in these patients. This is especially important in patients with ocular and cardiac sarcoid, which if left untreated can have serious consequences (Figures 26.21 and 26.22) [117, 118]. 18 F-FDG is useful for monitoring treatment response (Figure 26.21). The decreased 18F-FDG avidity of a lesion after the initiation of treatment correlates with clinical signs of improvement, while persistent activity identifies nonresponders. In one investigation, positive 18FDG studies led to a change in management in 80% of the patients [119]. In another investigation, patients with sarcoidosis who were treated with systemic corticosteroids and had a metabolic response on 18F-FDG had significantly fewer relapses [120].

(a)

(b)

Figure 26.21 Sarcoid. On the pretreatment images (a) there is intense 18F-FDG uptake in multiple cervical, mediastinal, and mesenteric lymph nodes. In addition, there are multiple skeletal lesions in the sternum, spine, and pelvis. On the posttreatment images (b), there is complete resolution of both the nodal and the skeletal disease.

735

736

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

Pulmonary parenchymal uptake of 18F-FDG uptake correlates with active pulmonary disease and predicts response to anti-inflammatory treatment. In a single-blind prospective investigation of patients with chronic pulmonary sarcoidosis, there was considerable correlation between 18 F-FDG findings and clinical improvement. In contrast, there was no correlation between chest radiographic findings and clinical improvement  [121]. 18F-FDG may also have a prognostic role in patients with sarcoidosis. A high number of neutrophils in bronchoalveolar lavage fluid is a poor prognostic indicator in these patients. 18F-FDG uptake correlates with the bronchoalveolar lavage fluid neutrophil count and may serve as a noninvasive prognostic tool [122]. In patients with pulmonary sarcoid it can be difficult to distinguish between pure end-stage fibrosis and fibrosis with active inflammation. This is an important differentiation to make because patients with active inflammation might benefit from a change in therapy. In advanced pulmonary sarcoidosis, CT is limited in its ability to differentiate irreversible fibrosis from active inflammation. Published data suggest that 18F-FDG can facilitate the differentiation between pure fibrosis and fibrosis plus inflammation. Inactive pulmonary fibrotic changes do not demonstrate uptake while active lesions do. In one investigation, 85% of the patients with pulmonary sarcoidosis and fibrosis on high resolution CT had abnormal pulmonary uptake of 18F-FDG, even in the absence of serological evidence inflammation [122]. In another investigation, nearly two-thirds of the patients with stage IV (fibrosis) pulmonary disease and persistent disabling symptoms had abnormal pulmonary parenchymal uptake of 18F-FDG. In 20% of

Figure 26.22 Cardiac sarcoid. On the maximum intensity projection (a) there are multiple 18F-FDG-avid lymph nodes in the left supraclavicular region, mediastinum, both hila and the heart. There also is a solitary hypermetabolic focus in the liver (arrow). On the axial PET/CT image (b), there is patchy uptake in the left and right ventricular myocardium and the interventricular septum.

the patients with abnormal 18F-FDG uptake, there was no serological evidence of active inflammation [123].

Intraabdominal Infections Intraabdominal infections develop when there is a breach of the normal anatomic barriers of the abdomen due to surgery, trauma, or inflammatory processes such as appendicitis and pancreatitis. These infections include several different pathological conditions and are classified as uncomplicated and complicated. In uncomplicated intraabdominal infections, infection is confined to a single organ and does not involve the peritoneum. Uncomplicated intraabdominal infections can be managed with either surgery or with antibiotics alone. In complicated intraabdominal infections, the infection extends beyond the organ and causes either localized or diffuse peritonitis, which if left untreated leads to frank abscess formation. Effective treatment of these infections requires prompt diagnosis, early initiation of appropriate antimicrobial therapy, and prompt and effective control of the source [124, 125]. Diagnostic imaging is routinely performed as part of the diagnostic workup of intrabdominal infections, with CT being the test of choice. CT scans are very sensitive for detecting small quantities of fluid, areas of inflammation, and abscesses. Management of intrabdominal infections, especially abscesses, includes percutaneous drainage, which usually is performed under CT guidance [124, 126]. Molecular imaging studies have an adjunctive, but nevertheless important, role in the diagnostic workup of these infections. 67Ga has been used to diagnose intraabdominal

Infection/Inflammation Imaging

(a) (b)

Anterior

Axial

Sagittal

Figure 26.23 Endometritis. On the planar image (a) there is faintly increased indium-111 labeled leukocyte activity in the pelvis (arrow). On the SPECT/CT images (b) there is increased activity throughout much of the uterus (arrows). At surgery, a friable, infected uterus with abscess formation was removed.

infections, and although sensitive is not specific for infection  [127]. The normal gastrointestinal and genitourinary excretion of 67Ga makes evaluation of the abdomen difficult. Persistent uptake in normally healing surgical incisions for up to a few weeks can confound the interpretation of these studies in the postoperative patient. 67Ga accumulates in tumor as well as infection and it is not always possible to differentiate between the two. The relatively low target to background ratio in abscesses and overall suboptimal image quality are additional disadvantages of 67Ga [128, 129]. There are few data on the role of 18F-FDG in the workup of intraabdominal infections. 18F-FDG has been used to detect abscesses, infected hepatic cysts, and infected renal cysts in patients with polycystic kidney disease [130, 131]. Disadvantages of 18F-FDG are similar to those of 67Ga: uptake in tumors and in a variety of postoperative settings in the absence of infection, including surgical incisions, fat necrosis, seromas, and hematomas [132]. WBC imaging is both sensitive and specific for intraabdominal infections, including abscesses, appendicitis, and urinary tract infections (Figure 26.23) [133–135]. Labeled leukocytes generally do not accumulate in normally healing incisions or in uninfected tumors, which are important advantages over both 67Ga and 18F-FDG [6]. One group of investigators found that WBC imaging, whether positive or negative, provided useful information upon which therapeutic decisions could be made [135].

Fever of Unknown Origin Fever, or pyrexia, of unknown origin (FUO) is a fever that exceeds 38.3 °C (101 °F) on several occasions, with more than 3 weeks’ duration of illness and a failure to

obtain a diagnosis after an appropriate inpatient or outpatient workup. FUO poses a diagnostic challenge. There are few if any published guidelines or standardized approaches to the diagnosis. In about half of the cases a definitive diagnosis is never established. Subcategories of the classical FUO include nosocomial, neutropenic, human immunodeficiency virus, and organ transplant associated FUOs. There are more than 200 causes of FUO, which can be divided into four principle categories: infection (20–40%), malignancy (20–30%), noninfectious inflammatory diseases, (10–30%), and miscellaneous (10–20%). In developed countries, noninfectious inflammatory diseases and undiagnosed groups comprise a higher proportion of FUOs, while infection and neoplasm more commonly are the cause in underdeveloped countries [136]. The typical FUO workup consists of several first-line investigations: comprehensive history and physical examination, complete blood count with differential, chest X-ray, urinalysis and culture, electrolyte panel, liver enzymes, erythrocyte sedimentation rate, and C-reactive protein level. Additional imaging procedures, including molecular imaging tests, serum antibodies, and tissue biopsies, are included as second-line procedures [136]. At one time 67Ga and WBC imaging were mainstays of molecular imaging in patients with FUO, but these tests were only modestly successful. In a meta-analysis the pooled sensitivity and specificity of 67Ga scintigraphy were 60% and 63%, respectively, and the pooled sensitivity and specificity of WBC scintigraphy were 33% and 83%, respectively. The diagnostic yield, defined as the proportion of patients in whom the imaging results were reported to contribute to the diagnosis of FUO, was 35% for 67Ga and 20% for WBC imaging [137].

737

738

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 26.24 Giant cell arteritis. There is diffusely increased 18 F-FDG in the aorta (arrows) and the subclavian arteries (arrowheads). 18F-FDG is very sensitive for detecting large vessel vasculitis, which is a well-recognized cause of fever of unknown origin. (Source: reproduced with permission from Palestro [4].) 18

F-FDG, the most extensively investigated radiopharmaceutical in patients with FUO, has rapidly assumed an increasingly important role in the diagnostic workup of this entity, and is now the molecular imaging test of choice for FUO. 18F-FDG permits accurate localization of foci of hypermetabolism that may be due to inflammation, infection, or neoplasia, all of which are causes of FUO. Abnormalities identified with 18F-FDG can guide additional investigations that may yield a final diagnosis (Figure  26.24). Equally important, a negative study excludes these conditions with a reasonable degree of certainty, thereby avoiding unnecessary additional testing  [138]. In one meta-analysis the pooled sensitivity, specificity, and diagnostic yield of 18F-FDG were 86%, 52%, and 58%, respectively [137]. In another meta-analysis the

diagnostic yield was 56% [139]. In intraindividual comparisons, 18F-FDG has outperformed 67Ga and WBC imaging in patients with FUO [137, 140–142]. Although most investigations have focused on adults with FUO, 18F-FDG also is very helpful in children with FUO [143– 146]. This is in contrast to the disappointing results that have been reported for 67Ga and WBC imaging in this population  [136]. In a retrospective investigation of 110 children, the sensitivity and specificity of 18F-FDG were 85.5 and 79.2%, respectively. 18F-FDG detected the cause of FUO in 48% of the cases and altered management in 53% [146]. There is mounting evidence that 18F-FDG should be a first-line, rather than a second-line diagnostic procedure in the workup of FUO. In a multicenter trial, 18F-FDG PET was diagnostically helpful in 33% of cases, while chest CT, abdominal CT, abdominal ultrasound, and chest X-ray were diagnostically helpful in 33%, 20%, 20%, 10%, and 8% of cases, respectively  [147]. In an investigation of 79 patients with FUO, 18F-FDG contributed useful information in about 74% of patients compared to about 62% for chest/abdominal CT  [148]. In another investigation of FUO, 18F-FDG was more sensitive (90% vs. 43.5%) and more specific (97% vs.67.6%) than anatomic imaging [149]. In a retrospective investigation of 79 patients with FUO, 18 F-FDG PET/CT contributed useful information in about 74% of the patients versus 62% for chest/abdominal CT. Performing 18F-FDG imaging earlier in the diagnostic workup of the patient with FUO may be very cost-effective, expediting the diagnosis, facilitating prompt institution of appropriate treatment, decreasing morbidity and mortality, decreasing the length of hospitalization, and reducing the number of tests performed [148, 149].

The Future The clinically available molecular imaging agents, though useful, reflect the physiological changes that are part of the inflammatory process and the host response to infection, rather than the infection itself, therefore these agents cannot consistently differentiate infections from other diseases. Future efforts need to focus on the development of infectionspecific molecular imaging agents that can be used for diagnosis, prognosis, and treatment response monitoring.

References 1 Ordonez, A.A., Sellmyer, M.A., Gowrishankar, G. et al. (2019). Molecular imaging of bacterial infections: overcoming the barriers to clinical translation. Sci. Transl. Med. 11 (508): eaax8251. https://doi.org/10.1126/scitranslmed.aax8251. PMID: 31484790; PMCID: PMC6743081.

2 Palestro, C.J. (2016). Radionuclide imaging of musculoskeletal infection: a review. J. Nucl. Med. 57 (9): 1406–1412. https://doi.org/10.2967/jnumed.115.157297. Epub 7 Jul 2016. PMID: 27390160.

Infection/Inflammation Imaging

3 Palestro, C.J. (1994). The current role of gallium imaging in infection. Semin. Nucl. Med. 24 (2): 128–141. https:// doi.org/10.1016/s0001-2998(05)80227-2. PMID: 8023169. 4 Palestro, C.J. (2020). Molecular imaging of infection: the first 50 years. Semin. Nucl. Med. 50 (1): 23–34. https://doi. org/10.1053/j.semnuclmed.2019.10.002. Epub 25 Oct 2019. PMID: 31843059. 5 Thakur, M.L., Lavender, J.P., Arnot, R.N. et al. (1977). Indium-111-labeled autologous leukocytes in man. J. Nucl. Med. 18 (10): 1014–1021. PMID: 409745. 6 Palestro, C.J., Love, C., and Bhargava, K.K. (2009). Labeled leukocyte imaging: current status and future directions. Q. J. Nucl. Med. Mol. Imaging 53 (1): 105–123. PMID: 19182734. 7 Love, C., Opoku-Agyemang, P., Tomas, M.B. et al. (2002). Pulmonary activity on labeled leukocyte images: physiologic, pathologic, and imaging correlation. Radiographics 22 (6): 1385–1393. https://doi.org/10.1148/ rg.226025038. PMID: 12432109. 8 Meller, J., Sahlmann, C.O., and Scheel, A.K. (2007). 18F-FDG PET and PET/CT in fever of unknown origin. J. Nucl. Med. 48 (1): 35–45. PMID: 17204697. 9 Palestro, C.J. (2013). FDG-PET in musculoskeletal infections. Semin. Nucl. Med. 43 (5): 367–376. https://doi. org/10.1053/j.semnuclmed.2013.04.006. PMID: 23905618. 10 Love, C., Tomas, M.B., Tronco, G.G. et al. (2005). FDG PET of infection and inflammation. Radiographics 25 (5): 1357–1368. https://doi.org/10.1148/rg.255045122. PMID: 16160116. 11 Wilson, W., Taubert, K.A., Gewitz, M. et al. American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee; American Heart Association Council on Cardiovascular Disease in the Young; American Heart Association Council on Clinical Cardiology; American Heart Association Council on Cardiovascular Surgery and Anesthesia; Quality of Care and Outcomes Research Interdisciplinary Working Group (2007 Oct 9). Prevention of infective endocarditis: guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation 116 (15): 1736–1754. https:// doi.org/10.1161/CIRCULATIONAHA.106.183095. Epub 2007 Apr 19. Erratum in: Circulation. 2007;116(15):e3767. PMID: 17446442. 12 Holland, T.L., Baddour, L.M., Bayer, A.S. et al. (2016). Infective endocarditis. Nat. Rev. Dis. Primers. 2: 16059. https://doi.org/10.1038/nrdp.2016.59. PMID: 27582414; PMCID: PMC5240923.

13 Erba, P.A., Conti, U., Lazzeri, E. et al. (2012). Added value of 99mTc-HMPAO-labeled leukocyte SPECT/CT in the characterization and management of patients with infectious endocarditis. J. Nucl. Med. 53 (8): 1235–1243. https://doi.org/10.2967/jnumed.111.099424. Epub 2012 Jul 11. PMID: 22787109. 14 Habib, G., Derumeaux, G., Avierinos, J.F. et al. (1999). Value and limitations of the Duke criteria for the diagnosis of infective endocarditis. J. Am. Coll. Cardiol. 33 (7): 2023–2029. https://doi.org/10.1016/s07351097(99)00116-3. PMID: 10362209. 15 Yan, J., Zhang, C., Niu, Y. et al. (2016). The role of 18F-FDG PET/CT in infectious endocarditis: a systematic review and meta-analysis. Int. J. Clin. Pharmacol. Ther. 54 (5): 337–342. https://doi.org/10.5414/CP202569. PMID: 27008000. 16 Rouzet, F., Chequer, R., Benali, K. et al. (2014). Respective performance of 18F-FDG PET and radiolabeled leukocyte scintigraphy for the diagnosis of prosthetic valve endocarditis. J. Nucl. Med. 55 (12): 1980–1985. https://doi.org/10.2967/jnumed.114.141895. Epub 2014 Nov 13. PMID: 25453046. 17 Baddour, L.M., Epstein, A.E., Erickson, C.C. et al. American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee; Council on Cardiovascular Disease in Young; Council on Cardiovascular Surgery and Anesthesia; Council on Cardiovascular Nursing; Council on Clinical Cardiology; Interdisciplinary Council on Quality of Care; American Heart Association (2010). Update on cardiovascular implantable electronic device infections and their management: a scientific statement from the American Heart Association. Circulation 121 (3): 458–477. https:// doi.org/10.1161/CIRCULATIONAHA.109.192665. Epub 4 Jan 2010. PMID: 20048212. 18 Erba, P.A., Sollini, M., Conti, U. et al. (2013). Radiolabeled WBC scintigraphy in the diagnostic workup of patients with suspected device-related infections. JACC Cardiovasc. Imaging 6 (10): 1075–1086. https://doi. org/10.1016/j.jcmg.2013.08.001. Epub 4 Sep 2013. PMID: 24011775. 19 Litzler, P.Y., Manrique, A., Etienne, M. et al. (2010). Leukocyte SPECT/CT for detecting infection of leftventricular-assist devices: preliminary results. J. Nucl. Med. 51 (7): 1044–1048. https://doi.org/10.2967/ jnumed.109.070664. Epub 16 Jun 2010. PMID: 20554736. 20 Hyafil, F., Rouzet, F., Lepage, L. et al. (2013). Role of radiolabelled leucocyte scintigraphy in patients with a suspicion of prosthetic valve endocarditis and inconclusive echocardiography. Eur. Heart J. Cardiovasc. Imaging 14 (6): 586–594. https://doi.org/10.1093/ehjci/ jet029. Epub 1 Mar 2013. PMID: 23456094.

739

740

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

21 Pizzi, M.N., Roque, A., Fernández-Hidalgo, N. et al. (2015). Improving the diagnosis of infective endocarditis in prosthetic valves and intracardiac devices with 18F-fluordeoxyglucose positron emission tomography/ computed tomography angiography: initial results at an infective endocarditis referral center. Circulation 132 (12): 1113–1126. https://doi.org/10.1161/ CIRCULATIONAHA.115.015316. Epub 14 Aug 2015. PMID: 26276890. 22 Bensimhon, L., Lavergne, T., Hugonnet, F. et al. (2011). Whole body [(18) F]fluorodeoxyglucose positron emission tomography imaging for the diagnosis of pacemaker or implantable cardioverter defibrillator infection: a preliminary prospective study. Clin. Microbiol. Infect. 17 (6): 836–844. https://doi.org/10.1111/j.1469-0691. 2010.03312.x. Epub 15 Jul 2010. PMID: 20636421. 23 Sarrazin, J.F., Philippon, F., Tessier, M. et al. (2012). Usefulness of fluorine-18 positron emission tomography/ computed tomography for identification of cardiovascular implantable electronic device infections. J. Am. Coll. Cardiol. 59 (18): 1616–1625. https://doi.org/10.1016/j. jacc.2011.11.059. PMID: 22538331. 24 Dell’Aquila, A.M., Mastrobuoni, S., Alles, S. et al. (2016). Contributory role of fluorine 18-fluorodeoxyglucose positron emission tomography/computed tomography in the diagnosis and clinical management of infections in patients supported with a continuous-flow left ventricular assist device. Ann. Thorac. Surg. 101 (1): 87–94; discussion 94. https://doi.org/10.1016/j. athoracsur.2015.06.066. Epub 1 oct 2015. PMID: 26433521. 25 Mahmood, M., Kendi, A.T., Farid, S. et al. (2019). Role of 18F-FDG PET/CT in the diagnosis of cardiovascular implantable electronic device infections: a meta-analysis. J. Nucl. Cardiol. 26 (3): 958–970. https://doi.org/10.1007/ s12350-017-1063-0. Epub 14 Sep 2017. PMID: 28913626. 26 Juneau, D., Golfam, M., Hazra, S. et al. (2017). Positron emission tomography and single-photon emission computed tomography imaging in the diagnosis of cardiac implantable electronic device infection: a systematic review and meta-analysis. Circ. Cardiovasc. Imaging 10 (4): e005772. https://doi.org/10.1161/ CIRCIMAGING.116.005772. PMID: 28377468. 27 Wilson, W.R., Bower, T.C., Creager, M.A. et al. American Heart Association Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease of the Council on Cardiovascular Disease in the Young; Council on Cardiovascular and Stroke Nursing; Council on Cardiovascular Radiology and Intervention; Council on Cardiovascular Surgery and Anesthesia; Council on Peripheral Vascular Disease; and Stroke Council (2016). Vascular graft infections, mycotic aneurysms, and

28

29

30

31

32

33

34

35

36

37

endovascular infections: a scientific statement from the American Heart Association. Circulation 134 (20): e412–e460. https://doi.org/10.1161/ CIR.0000000000000457. Epub 13 Oct 2016. PMID: 27737955. Brunner, M.C., Mitchell, R.S., Baldwin, J.C. et al. (1986). Prosthetic graft infection: limitations of indium white blood cell scanning. J. Vasc. Surg. 3 (1): 42–48. https://doi. org/10.1067/mva.1986.avs0030042. PMID: 3079840. Liberatore, M., Iurilli, A.P., Ponzo, F. et al. (1998). Clinical usefulness of technetium-99m-HMPAO-labeled leukocyte scan in prosthetic vascular graft infection. J. Nucl. Med. 39 (5): 875–879. PMID: 9591592. Palestro, C.J., Vega, A., Kim, C.K. et al. (1990). Indium111-labeled leukocyte scintigraphy in hemodialysis access-site infection. J. Nucl. Med. 31 (3): 319–324. PMID: 2106577. Shahidi, S., Eskil, A., Lundof, E. et al. (2007). Detection of abdominal aortic graft infection: comparison of magnetic resonance imaging and indium-labeled white blood cell scanning. Ann. Vasc. Surg. 21 (5): 586–592. https://doi. org/10.1016/j.avsg.2007.03.018. PMID: 17823040. Williamson, M.R., Boyd, C.M., Read, R.C. et al. (1986). 111 In-labeled leukocytes in the detection of prosthetic vascular graft infections. Am. J. Roentgenol. 147 (1): 173–176. https://doi.org/10.2214/ajr.147.1.173. PMID: 3487211. Lou, L., Alibhai, K.N., Winkelaar, G.B. et al. (2010). 99mTc-WBC scintigraphy with SPECT/CT in the evaluation of arterial graft infection. Nucl. Med. Commun. 31 (5): 411–416. https://doi.org/10.1097/ MNM.0b013e328337142c. PMID: 20134358. Erba, P.A., Leo, G., Sollini, M. et al. (2014). Radiolabelled leucocyte scintigraphy versus conventional radiological imaging for the management of late, low-grade vascular prosthesis infections. Eur. J. Nucl. Med. Mol. Imaging 41 (2): 357–368. https://doi.org/10.1007/s00259-013-2582-9. Epub 19 Oct 2013. PMID: 24142027. Keidar, Z., Engel, A., Hoffman, A. et al. (2007). Prosthetic vascular graft infection: the role of 18F-FDG PET/CT. J. Nucl. Med. 48 (8): 1230–1236. https://doi.org/10.2967/ jnumed.107.040253. Epub 13 Jul 2007. PMID: 17631553. Sah, B.R., Husmann, L., Mayer, D. et al. VASGRA Cohort (2015). Diagnostic performance of 18F-FDG-PET/CT in vascular graft infections. Eur. J. Vasc. Endovasc. Surg. 49 (4): 455–464. https://doi.org/10.1016/j.ejvs.2014.12.024. Epub 31 Jan 2015. PMID: 25648371. Keidar, Z., Pirmisashvili, N., Leiderman, M. et al. (2014). 18F-FDG uptake in noninfected prosthetic vascular grafts: incidence, patterns, and changes over time. J. Nucl. Med. 55 (3): 392–395. https://doi.org/10.2967/ jnumed.113.128173. Epub 10 Feb 2014. PMID: 24516259.

Infection/Inflammation Imaging

38 Saleem, B.R., Pol, R.A., Slart, R.H. et al. (2014). 18F-Fluorodeoxyglucose positron emission tomography/ CT scanning in diagnosing vascular prosthetic graft infection. Biomed. Res. Int. 2014: 471971. https://doi. org/10.1155/2014/471971. Epub 19 Aug 2014. PMID: 25210712; PMCID: PMC4156987. 39 Bowles, H., Ambrosioni, J., Mestres, G. et al. Hospital Clinic Endocarditis Study Group (2020). Diagnostic yield of 18F-FDG PET/CT in suspected diagnosis of vascular graft infection: a prospective cohort study. J. Nucl. Cardiol. 27 (1): 294–302. https://doi.org/10.1007/ s12350-018-1337-1. Epub 15 Jun 2018. PMID: 29907934. 40 Reinders Folmer, E.I., Von Meijenfeldt, G.C.I., Van der Laan, M.J. et al. (2018). Diagnostic imaging in vascular graft infection: a systematic review and meta-analysis. Eur. J. Vasc. Endovasc. Surg. 56 (5): 719–729. https://doi. org/10.1016/j.ejvs.2018.07.010. Epub 16 Aug 2018. PMID: 30122333. 41 Puges, M., Bérard, X., Ruiz, J.B. et al. (2019). Retrospective study comparing WBC scan and 18F-FDG PET/CT in patients with suspected prosthetic vascular graft infection. Eur. J. Vasc. Endovasc. Surg. 57 (6): 876–884. https://doi.org/10.1016/j.ejvs.2018.12.032. Epub 23 May 2019. PMID: 31130421. 42 Husmann, L., Sah, B.R., Scherrer, A. et al. VASGRA Cohort (2015). 18F-FDG PET/CT for therapy control in vascular graft infections: a first feasibility study. J. Nucl. Med. 56 (7): 1024–1029. https://doi.org/10.2967/ jnumed.115.156265. Epub 14 May 2015. PMID: 25977463. 43 Husmann, L., Ledergerber, B., Anagnostopoulos, A. et al. VASGRA Cohort Study (2018). The role of FDG PET/CT in therapy control of aortic graft infection. Eur. J. Nucl. Med. Mol. Imaging 45 (11): 1987–1997. https://doi.org/10.1007/ s00259-018-4069-1. Epub 11 Jun 2018. PMID: 29948106. 44 Palestro, C.J. (2013). Nuclear medicine imaging of osteomyelitis: white blood cell, monoclonal antibody, or bacterial imaging? In: Diagnostic Imaging of Infections and Inflammatory Diseases: A Multidisciplinary Approach (ed. A. Signore and A.M. Quintero), 168–186. New York, NY: Wiley. 45 Schauwecker, D.S. (1992). The scintigraphic diagnosis of osteomyelitis. Am. J. Roentgenol. 158 (1): 9–18. https:// doi.org/10.2214/ajr.158.1.1727365. PMID: 1727365. 46 Tande, A.J. and Patel, R. (2014). Prosthetic joint infection. Clin. Microbiol. Rev. 27 (2): 302–345. https://doi. org/10.1128/CMR.00111-13. PMID: 24696437; PMCID: PMC3993098. 47 Ong, K.L., Kurtz, S.M., Lau, E. et al. (2009). Prosthetic joint infection risk after total hip arthroplasty in the Medicare population. J. Arthroplasty 24 (6 Suppl): 105–109. https://doi.org/10.1016/j.arth.2009.04.027. Epub 2 Jun 2009. PMID: 19493644.

48 Kurtz, S.M., Ong, K.L., Lau, E. et al. (2010). Prosthetic joint infection risk after TKA in the Medicare population. Clin. Orthop. Relat. Res. 468 (1): 52–56. https://doi. org/10.1007/s11999-009-1013-5. Epub 8 Aug 2009. PMID: 19669386; PMCID: PMC2795807. 49 Kurtz, S.M., Lau, E., Schmier, J. et al. (2008). Infection burden for hip and knee arthroplasty in the United States. J. Arthroplasty 23 (7): 984–991. https://doi.org/10.1016/j. arth.2007.10.017. Epub 10 Apr 2008. PMID: 18534466. 50 Del Arco, A. and Bertrand, M.L. (2013). The diagnosis of periprosthetic infection. Open Orthop. J. 7: 178–183. https://doi.org/10.2174/1874325001307010178. PMID: 23898349; PMCID: PMC3722546. 51 Love, C., Marwin, S.E., and Palestro, C.J. (2009). Nuclear medicine and the infected joint replacement. Semin. Nucl. Med. 39 (1): 66–78. https://doi.org/10.1053/j. semnuclmed.2008.08.007. PMID: 19038601. 52 Palestro, C.J. and Love, C. (2017). Role of nuclear medicine for diagnosing infection of recently implanted lower extremity arthroplasties. Semin. Nucl. Med. 47 (6): 630–638. https://doi.org/10.1053/j.semnuclmed.2017.07.008. Epub 26 Jul 2017. PMID: 28969761. 53 Verberne, S.J., Raijmakers, P.G., and Temmerman, O.P. (2016). The accuracy of imaging techniques in the assessment of periprosthetic hip infection: a systematic review and meta-analysis. J. Bone Joint Surg. Am. 98 (19): 1638–1645. https://doi.org/10.2106/JBJS.15.00898. PMID: 27707850. 54 Verberne, S.J., Sonnega, R.J., Temmerman, O.P. et al. (2017). What is the accuracy of nuclear imaging in the assessment of periprosthetic knee infection? A metaanalysis. Clin. Orthop. Relat. Res. 475 (5): 1395–1410. https://doi.org/10.1007/s11999-016-5218-0. Epub 3 Jan 2017. Erratum in: Clin Orthop Relat Res. 2017;475(6):17531754. PMID: 28050818; PMCID: PMC5384926. 55 Palestro, C.J. (2014). Nuclear medicine and the failed joint replacement: past, present, and future. World J. Radiol. 6 (7): 446–458. https://doi.org/10.4329/wjr. v6.i7.446. PMID: 25071885; PMCID: PMC4109096. 56 Arıcan, P., Okudan Tekin, B., Şefizade, R. et al. (2015). The role of bone SPECT/CT in the evaluation of painful joint prostheses. Nucl. Med. Commun. 36 (9): 931–940. https://doi.org/10.1097/MNM.0000000000000348. PMID: 26049374. 57 Al-Nabhani, K., Michopoulou, S., Allie, R. et al. (2014). Painful knee prosthesis: can we help with bone SPECT/ CT? Nucl. Med. Commun. 35 (2): 182–188. https://doi. org/10.1097/MNM.0000000000000028. PMID: 24152807. 58 Jin, H., Yuan, L., Li, C. et al. (2014). Diagnostic performance of FDG PET or PET/CT in prosthetic infection after arthroplasty: a meta-analysis. Q. J. Nucl. Med. Mol. Imaging 58: 85–93.

741

742

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

59 Kwee, T.C., Kwee, R.M., and Alavi, A. (2008). FDG-PET for diagnosing prosthetic joint infection: systematic review and metaanalysis. Eur. J. Nucl. Med. Mol. Imaging 35: 2122–2132. https://doi.org/10.1007/s00259-008-0887-x. 60 Suárez, J.P., Domínguez, M.L., Zeidán, N. et al. (2019). Complications of shoulder arthroplasty: added value of SPECT/CT imaging. Rev. Esp. Med. Nucl. Imagen Mol. (Engl. Ed.) 38 (4): 234–237. English, Spanish. https://doi. org/10.1016/j.remn.2018.12.004. Epub 25 Jan 2019. PMID: 30692042. 61 Wiater, B.P., Moravek, J.E. Jr., and Wiater, J.M. (2014). The evaluation of the failed shoulder arthroplasty. J. Shoulder Elbow Surg. 23 (5): 745–758. https://doi. org/10.1016/j.jse.2013.12.003. Epub 4 Mar 2014. PMID: 24618199. 62 Falstie-Jensen, T., Daugaard, H., Søballe, K. et al. ROSA study group(2019). Labeled white blood cell/bone marrow single-photon emission computed tomography with computed tomography fails in diagnosing chronic periprosthetic shoulder joint infection. J. Shoulder Elbow Surg. 28 (6): 1040–1048. https://doi.org/10.1016/j. jse.2018.10.024. Epub 31 Jan 2019. PMID: 30713059. 63 Falstie-Jensen, T., Lange, J., Daugaard, H. et al. ROSA study-group (2019). 18F FDG-PET/CT has poor diagnostic accuracy in diagnosing shoulder PJI. Eur. J. Nucl. Med. Mol. Imaging 46 (10): 2013–2022. https://doi. org/10.1007/s00259-019-04381-w. Epub 11 Jul 2019. PMID: 31292698. 64 Raghavan, M., Lazzeri, E., and Palestro, C.J. (2018). Imaging of spondylodiscitis. Semin. Nucl. Med. 48 (2): 131–147. https://doi.org/10.1053/j.semnuclmed.2017.11.001. Epub 6 Feb 2018. PMID: 29452617. 65 Dunn, R.N., Castelein, S., and Held, M. (2019). Impact of HIV on spontaneous spondylodiscitis. Bone Joint J. 101-B: 617–620. 66 Legrand, E., Flipo, R.M., Guggenbuhl, P. et al. Rheumatology Network Organization (2001). Management of nontuberculous infectious discitis. Treatments used in 110 patients admitted to 12 teaching hospitals in France. Joint Bone Spine 68 (6): 504–509. https://doi.org/10.1016/s1297-319x(01)00315-3. PMID: 11808988. 67 Tamm, A.S. and Abele, J.T. (2017). Bone and gallium single-photon emission computed tomography-computed tomography is equivalent to magnetic resonance imaging in the diagnosis of infectious spondylodiscitis: a retrospective study. Can. Assoc. Radiol. J. 68 (1): 41–46. https://doi.org/10.1016/j.carj.2016.02.003. Epub 11 Aug 2016. PMID: 27523445. 68 Prodromou, M.L., Ziakas, P.D., Poulou, L.S. et al. (2014). FDG PET is a robust tool for the diagnosis of spondylodiscitis: a meta-analysis of diagnostic data. Clin.

69

70

71

72

73

74

75

76

77

78

Nucl. Med. 39 (4): 330–335. https://doi.org/10.1097/ RLU.0000000000000336. PMID: 24445277. Treglia, G., Pascale, M., Lazzeri, E. et al. (2020). Diagnostic performance of 18F-FDG PET/CT in patients with spinal infection: a systematic review and a bivariate meta-analysis. Eur. J. Nucl. Med. Mol. Imaging 47 (5): 1287–1301. https://doi.org/10.1007/s00259-019-04571-6. Epub 15 Nov 2019. PMID: 31729539. Frenkel Rutenberg, T., Baruch, Y., Ohana, N. et al. (2019). The role of 18F-fluorodeoxyglucose positron-emission tomography/computed tomography in the diagnosis of postoperative hardware-related spinal infections. Isr. Med. Assoc. J. 21 (8): 532–537. PMID: 31474015. Hadjipavlou, A.G., Cesani-Vazquez, F., Villaneuva-Meyer, J. et al. (1998). The effectiveness of gallium citrate Ga 67 radionuclide imaging in vertebral osteomyelitis revisited. Am. J. Orthop. (Belle Mead N.J.) 27 (3): 179–183. PMID: 9544359. Bagrosky, B.M., Hayes, K.L., Koo, P.J. et al. (2013). 18F-FDG PET/CT evaluation of children and young adults with suspected spinal fusion hardware infection. Pediatr. Radiol. 43 (8): 991–1000. https://doi.org/10.1007/ s00247-013-2654-9. Epub 2 Mar 2013. PMID: 23455308. Skanjeti, A., Penna, D., Douroukas, A. et al. (2012). PET in the clinical work-up of patients with spondylodiscitis: a new tool for the clinician? Q. J. Nucl. Med. Mol. Imaging 56 (6): 569–576. PMID: 23358410. Dauchy, F.A., Dutertre, A., Lawson-Ayayi, S. et al. (2016). Interest of [(18)F]fluorodeoxyglucose positron emission tomography/computed tomography for the diagnosis of relapse in patients with spinal infection: a prospective study. Clin. Microbiol. Infect. 22 (5): 438–443. https://doi. org/10.1016/j.cmi.2015.12.028. Epub 21 Jan 2016. PMID: 26802215. Gemechu, F.W., Seemant, F., and Curley, C.A. (2013). Diabetic foot infections. Am. Fam. Physician 88 (3): 177–184. PMID: 23939696. Lipsky, B.A., Berendt, A.R., Cornia, P.B. et al. (2012). 2012 Infectious Diseases Society of America clinical practice guideline for the diagnosis and treatment of diabetic foot infections. Clin. Infect. Dis. 54 (12): e132–e173. https://doi.org/10.1093/cid/cis346. PMID: 22619242. Boulton, A.J., Vileikyte, L., Ragnarson-Tennvall, G. et al. (2005). The global burden of diabetic foot disease. Lancet 366 (9498): 1719–1724. https://doi.org/10.1016/S01406736(05)67698-2. PMID: 16291066. Lavery, L.A., Ryan, E.C., Ahn, J. et al. (2020). The infected diabetic foot: re-evaluating the Infectious Diseases Society of America diabetic foot infection classification. Clin. Infect. Dis. 70 (8): 1573–1579. https://doi. org/10.1093/cid/ciz489. PMID: 31179491.

Infection/Inflammation Imaging

79 Palestro, C.J. and Love, C. (2009). Nuclear medicine and diabetic foot infections. Semin. Nucl. Med. 39 (1): 52–65. https://doi.org/10.1053/j.semnuclmed.2008.08.006. PMID: 19038600. 80 Dinh, M.T., Abad, C.L., and Safdar, N. (2008). Diagnostic accuracy of the physical examination and imaging tests for osteomyelitis underlying diabetic foot ulcers: metaanalysis. Clin. Infect. Dis. 47 (4): 519–527. https://doi. org/10.1086/590011. PMID: 18611152; PMCID: PMC7450707. 81 Johnson, J.E., Kennedy, E.J., Shereff, M.J. et al. (1996). Prospective study of bone, indium-111-labeled white blood cell, and gallium-67 scanning for the evaluation of osteomyelitis in the diabetic foot. Foot Ankle Int. 17 (1): 10–16. https://doi.org/10.1177/107110079601700103. PMID: 8821280. 82 Delcourt, A., Huglo, D., Prangere, T. et al. (2005). Comparison between leukoscan (Sulesomab) and gallium-67 for the diagnosis of osteomyelitis in the diabetic foot. Diabetes Metab. 31 (2): 125–133. https://doi. org/10.1016/s1262-3636(07)70178-7. PMID: 15959418. 83 Aslangul, E., M’bemba, J., Caillat-Vigneron, N. et al. (2013). Diagnosing diabetic foot osteomyelitis in patients without signs of soft tissue infection by coupling hybrid 67Ga SPECT/CT with bedside percutaneous bone puncture. Diabetes Care 36 (8): 2203–2210. https://doi. org/10.2337/dc12-2108. Epub 20 Mar 2013. PMID: 23514729; PMCID: PMC3714532. 84 Lauri, C., Tamminga, M., Glaudemans, A.W.J.M. et al. (2017). Detection of osteomyelitis in the diabetic foot by imaging techniques: a systematic review and metaanalysis comparing MRI, white blood cell scintigraphy, and FDG-PET. Diabetes Care 40 (8): 1111–1120. https:// doi.org/10.2337/dc17-0532. PMID: 28733376. 85 Devillers, A., Moisan, A., Hennion, F. et al. (1998). Contribution of technetium-99m hexamethylpropylene amine oxime labelled leucocyte scintigraphy to the diagnosis of diabetic foot infection. Eur. J. Nucl. Med. 25 (2): 132–138. https://doi.org/10.1007/s002590050205. PMID: 9473260. 86 Poirier, J.Y., Garin, E., Derrien, C. et al. (2002). Diagnosis of osteomyelitis in the diabetic foot with a 99mTcHMPAO leucocyte scintigraphy combined with a 99mTc-MDP bone scintigraphy. Diabetes Metab. 28 (6 Pt 1): 485–490. PMID: 12522329. 87 Heiba, S.I., Kolker, D., Mocherla, B. et al. (2010). The optimized evaluation of diabetic foot infection by dual isotope SPECT/CT imaging protocol. J. Foot Ankle Surg. 49 (6): 529–536. https://doi.org/10.1053/j.jfas.2010.07.010. Epub 18 Sep 2010. PMID: 20851003. 88 Treglia, G., Sadeghi, R., Annunziata, S. et al. (2013). Diagnostic performance of

89

90

91

92

93

94

95

96

97

98

Fluorine-18-Fluorodeoxyglucose positron emission tomography for the diagnosis of osteomyelitis related to diabetic foot: a systematic review and a meta-analysis. Foot (Edinb.) 23 (4): 140–148. https://doi.org/10.1016/j. foot.2013.07.002. Epub 29 Jul 2013. PMID: 23906976. Perno, J.R., Gordon, A., and Roach, N. (2011). Osteomyelitis. In: Radiology Secrets Plus, 3e (ed. E.S. Pretorius and J.A. Solomon), 283–290. Philadelphia, PA: Mosby Elsevier. George, J., Chandy, V.J., Premnath, J. et al. (2019). Microbiological profile of septic arthritis in adults: lessons learnt and treatment strategies. Indian J. Med. Microbiol. 37 (1): 29–33. https://doi.org/10.4103/ijmm. IJMM_19_134. PMID: 31424007. Gupta, M.N., Sturrock, R.D., and Field, M. (2001). A prospective 2-year study of 75 patients with adult-onset septic arthritis. Rheumatology (Oxford) 40 (1): 24–30. https:// doi.org/10.1093/rheumatology/40.1.24. PMID: 11157138. Palestro, C.J., Love, C., and Miller, T.T. (2006). Infection and musculoskeletal conditions: imaging of musculoskeletal infections. Best Pract. Res. Clin. Rheumatol. 20 (6): 1197–1218. https://doi.org/10.1016/j. berh.2006.08.009. PMID: 17127204. Bruijnen, S.T., Gent, Y.Y., Voskuyl, A.E. et al. (2014). Present role of positron emission tomography in the diagnosis and monitoring of peripheral inflammatory arthritis: a systematic review. Arthritis Care Res. (Hoboken) 66 (1): 120–130. https://doi.org/10.1002/ acr.22184. PMID: 24124027. Yu, W.Y., Lu, P.X., Assadi, M. et al. (2019). Updates on 18F-FDG-PET/CT as a clinical tool for tuberculosis evaluation and therapeutic monitoring. Quant. Imaging Med. Surg. 9 (6): 1132–1146. https://doi.org/10.21037/ qims.2019.05.24. PMID: 31367568; PMCID: PMC6629575. Ankrah, A.O., Glaudemans, A.W.J.M., Maes, A. et al. (2018). Tuberculosis. Semin. Nucl. Med. 48 (2): 108–130. https://doi.org/10.1053/j.semnuclmed.2017.10.005. Epub 16 Nov 2017. PMID: 29452616. Horne, D.J., Kohli, M., Zifodya, J.S. et al. (2019). Xpert MTB/RIF and Xpert MTB/RIF Ultra for pulmonary tuberculosis and rifampicin resistance in adults. Cochrane Database Syst. Rev. 6 (6): CD009593. https://doi. org/10.1002/14651858.CD009593.pub4. Update in: Cochrane Database Syst Rev. 2021 Feb 22;2:CD009593. PMID: 31173647; PMCID: PMC6555588. Palestro, C.J., Love, C., and Miller, T.T. (2007). Diagnostic imaging tests and microbial infections. Cell. Microbiol. 9 (10): 2323–2333. https://doi.org/10.1111/j.1462-5822.2007.01013.x. Epub 20 Jul 2007. PMID: 17645552. Yang, S.O., Lee, Y.I., Chung, D.H. et al. (1992). Detection of extrapulmonary tuberculosis with gallium-67 scan and computed tomography. J. Nucl. Med. 33 (12): 2118–2123. PMID: 1460503.

743

744

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

99 Soussan, M., Brillet, P.Y., Mekinian, A. et al. (2012). Patterns of pulmonary tuberculosis on FDG-PET/CT. Eur. J. Radiol. 81 (10): 2872–2876. https://doi. org/10.1016/j.ejrad.2011.09.002. Epub 25 Sep 2011. PMID: 21945403. 100 Kim, I.J., Lee, J.S., Kim, S.J. et al. (2008). Double-phase 18F-FDG PET-CT for determination of pulmonary tuberculoma activity. Eur. J. Nucl. Med. Mol. Imaging 35 (4): 808–814. https://doi.org/10.1007/s00259-007-0585-0. Epub 18 Dec 2007. PMID: 18097664. 101 Malherbe, S.T., Shenai, S., Ronacher, K. et al. (2016). Persisting positron emission tomography lesion activity and Mycobacterium tuberculosis mRNA after tuberculosis cure. Nat. Med. 22 (10): 1094–1100. https://doi. org/10.1038/nm.4177. Epub 5 Sep 2016. PMID: 27595324. 102 Esmail, H., Lai, R.P., Lesosky, M. et al. (2016). Characterization of progressive HIV-associated tuberculosis using 2-deoxy-2-[18F]fluoro-D-glucose positron emission and computed tomography. Nat. Med. 22 (10): 1090–1093. https://doi.org/10.1038/nm.4161. Epub 5 Sep 2016. PMID: 27595321; PMCID: PMC5055809. 103 Ghesani, N., Patrawalla, A., Lardizabal, A. et al. (2014). Increased cellular activity in thoracic lymph nodes in early human latent tuberculosis infection. Am. J. Respir. Crit. Care Med. 189 (6): 748–750. https://doi. org/10.1164/rccm.201311-1976LE. PMID: 24628316. 104 Chen, R.Y., Dodd, L.E., Lee, M. et al. (2014). PET/CT imaging correlates with treatment outcome in patients with multidrug-resistant tuberculosis. Sci. Transl. Med. 6 (265): 265ra166. https://doi.org/10.1126/ scitranslmed.3009501. PMID: 25473034; PMCID: PMC5567784. 105 Lee, H.S., Oh, J.Y., Lee, J.H. et al. (2004). Response of pulmonary tuberculomas to anti-tuberculous treatment. Eur. Respir. J. 23 (3): 452–455. https://doi.org/10.1183/09 031936.04.00087304. PMID: 15065838. 106 Sathekge, M., Maes, A., Kgomo, M. et al. (2011). Use of 18F-FDG PET to predict response to first line tuberculostatics in HIV-associated tuberculosis. J. Nucl. Med. 52 (6): 880–885. https://doi.org/10.2967/ jnumed.110.083709. Epub 13 May 2011. PMID: 21571788. 107 Sathekge, M., Maes, A., D’Asseler, Y. et al. (2012). Tuberculous lymphadenitis: FDG PET and CT findings in responsive and nonresponsive disease. Eur. J. Nucl. Med. Mol. Imaging 39 (7): 1184–1190. https://doi. org/10.1007/s00259-012-2115-y. Epub 18 Apr 2012. PMID: 22526957. 108 Llanos, O. and Hamzeh, N. (2019). Sarcoidosis. Med. Clin. North Am. 103 (3): 527–534. https://doi. org/10.1016/j.mcna.2018.12.011. Epub 21 Feb 2019. PMID: 30955519.

109 Brown, F., Modi, P., and Tanner, L.S. (2022). Lofgren syndrome. In: StatPearls [Internet]. Treasure Island, FL: StatPearls Publishing. https://www.ncbi.nlm.nih.gov/ books/NBK482315/. PMID: 29493940. 110 Scadding, J.G. (1950). Sarcoidosis, with special reference to lung changes. Br. Med. J. 1 (4656): 745–753. https:// doi.org/10.1136/bmj.1.4656.745. PMID: 20787829; PMCID: PMC2037093. 111 Sulavik, S.B., Spencer, R.P., Weed, D.A. et al. (1990). Recognition of distinctive patterns of gallium-67 distribution in sarcoidosis. J. Nucl. Med. 31 (12): 1909–1914. PMID: 2266386. 112 Rohatgi, P.K., Singh, R., and Vieras, F. (1987). Extrapulmonary localization of gallium in sarcoidosis. Clin. Nucl. Med. 12 (1): 9–16. https://doi. org/10.1097/00003072-198701000-00003. PMID: 3469056. 113 Sulavik, S.B., Palestro, C.J., Spencer, R.P. et al. (1990). Extrapulmonary sites of radiogallium accumulation in sarcoidosis. Clin. Nucl. Med. 15 (12): 876–878. https:// doi.org/10.1097/00003072-199012000-00005. PMID: 2276228. 114 Nishiyama, Y., Yamamoto, Y., Fukunaga, K. et al. (2006). Comparative evaluation of 18F-FDG PET and 67Ga scintigraphy in patients with sarcoidosis. J. Nucl. Med. 47 (10): 1571–1576. PMID: 17015889. 115 Keijsers, R.G., Grutters, J.C., Thomeer, M. et al. (2011). Imaging the inflammatory activity of sarcoidosis: sensitivity and inter observer agreement of (67)Ga imaging and (18)F-FDG PET. Q. J. Nucl. Med. Mol. Imaging 55 (1): 66–71. Epub 18 Jan 2011. PMID: 21242947. 116 Braun, J.J., Kessler, R., Constantinesco, A. et al. (2008). 18F-FDG PET/CT in sarcoidosis management: review and report of 20 cases. Eur. J. Nucl. Med. Mol. Imaging 35 (8): 1537–1543. https://doi.org/10.1007/s00259-0080770-9. Epub 17 Apr 2008. PMID: 18418595. 117 Keijsers, R.G.M. and Grutters, J.C. (2020). In which patients with sarcoidosis is FDG PET/CT indicated? J. Clin. Med. 9 (3): 890. https://doi.org/10.3390/ jcm9030890. PMID: 32213991; PMCID: PMC7141490. 118 Akaike, G., Itani, M., Shah, H. et al. (2018). PET/CT in the diagnosis and workup of sarcoidosis: focus on atypical manifestations. Radiographics 38 (5): 1536– 1549. https://doi.org/10.1148/rg.2018180053. Epub 17 Aug 2018. PMID: 30118393. 119 Sobic-Saranovic, D., Grozdic, I., Videnovic-Ivanov, J. et al. (2012). The utility of 18F-FDG PET/CT for diagnosis and adjustment of therapy in patients with active chronic sarcoidosis. J. Nucl. Med. 53 (10): 1543–1549. https://doi.org/10.2967/jnumed.112.104380. Epub 9 Aug 2012. PMID: 22879080.

Infection/Inflammation Imaging

120 Maturu, V.N., Rayamajhi, S.J., Agarwal, R. et al. (2016). Role of serial F-18 FDG PET/CT scans in assessing treatment response and predicting relapses in patients with symptomatic sarcoidosis. Sarcoidosis Vasc. Diffuse Lung Dis. 33 (4): 372–380. PMID: 28079849. 121 Keijsers, R.G., Verzijlbergen, J.F., van Diepen, D.M. et al. (2008). 18F-FDG PET in sarcoidosis: an observational study in 12 patients treated with infliximab. Sarcoidosis Vasc. Diffuse Lung Dis. 25 (2): 143–149. PMID: 19382534. 122 Guleria, R., Jyothidasan, A., Madan, K. et al. (2014). Utility of FDG-PET-CT scanning in assessing the extent of disease activity and response to treatment in sarcoidosis. Lung India 31 (4): 323–330. https://doi. org/10.4103/0970-2113.142092. PMID: 25378838; PMCID: PMC4220312. 123 Mostard, R.L., Vöö, S., van Kroonenburgh, M.J. et al. (2011). Inflammatory activity assessment by F18 FDGPET/CT in persistent symptomatic sarcoidosis. Respir. Med. 105 (12): 1917–1924. https://doi.org/10.1016/j. rmed.2011.08.012. Epub 6 Sep 2011. PMID: 21899998. 124 Sartelli, M., Chichom-Mefire, A., Labricciosa, F.M. et al. (2017). The management of intra-abdominal infections from a global perspective: 2017 WSES guidelines for management of intra-abdominal infections. World J. Emerg. Surg. 12: 29. https://doi.org/10.1186/s13017-0170141-6. Erratum in: World J Emerg Surg. 2017;12: 36. PMID: 28702076; PMCID: PMC5504840. 125 Mazuski, J.E., Tessier, J.M., May, A.K. et al. (2017 Jan). The Surgical Infection Society revised guidelines on the management of intra-abdominal infection. Surg. Infect. (Larchmt.) 18 (1): 1–76. https://doi.org/10.1089/ sur.2016.261. PMID: 28085573. 126 Bulander, R.E., Dunn, D.L., and Beilman, G.J. Surgical infections. In: Schwartz’s Principles of Surgery, 11e (ed. F. Brunicardi, D.K. Andersen, T.R. Billiar, et al.). McGrawHill https://accessmedicine.mhmedical.com/content. aspx?bookid=2576§ionid= accessed 21 June 2021. 127 Lin, C.M., Hung, G.U., Chao, T.H. et al. (2005). The limited use of ultrasound in the detection of abdominal abscesses in patients after colorectal surgery: compared with gallium scan and computed tomography. Hepatogastroenterology 52 (61): 79–81. PMID: 15782999. 128 Datz, F.L. (1996). Abdominal abscess detection: gallium, 111In-, and 99mTc-labeled leukocytes, and polyclonal and monoclonal antibodies. Semin. Nucl. Med. 26 (1): 51–64. https://doi.org/10.1016/s0001-2998(96)80016-x. PMID: 8623052. 129 Love, C. and Palestro, C.J. (2010). Altered biodistribution and incidental findings on gallium and labeled leukocyte/bone marrow scans. Semin. Nucl. Med. 40 (4): 271–282. https://doi.org/10.1053/j. semnuclmed.2010.03.004. PMID: 20513449.

130 Tahara, T., Ichiya, Y., Kuwabara, Y. et al. (1989). High [18F]-fluorodeoxyglucose uptake in abdominal abscesses: a PET study. J. Comput. Assist. Tomogr. 13 (5): 829–831. https://doi.org/10.1097/00004728198909000-00014. PMID: 2789238. 131 Bleeker-Rovers, C.P., de Sévaux, R.G., van Hamersvelt, H.W. et al. (2003). Diagnosis of renal and hepatic cyst infections by 18-F-fluorodeoxyglucose positron emission tomography in autosomal dominant polycystic kidney disease. Am. J. Kidney Dis. 41 (6): E18–E21. https://doi. org/10.1016/s0272-6386(03)00368-8. PMID: 12776306. 132 Garg, G., Benchekroun, M.T., and Abraham, T. (2017). FDG-PET/CT in the postoperative period: utility, expected findings, complications, and pitfalls. Semin. Nucl. Med. 47 (6): 579–594. https://doi.org/10.1053/j.semnuclmed. 2017.07.005. Epub 24 Jul 2017. PMID: 28969758. 133 Lantto, E.H., Lantto, T.J., and Vorne, M. (1991). Fast diagnosis of abdominal infections and inflammations with technetium-99m-HMPAO labeled leukocytes. J. Nucl. Med. 32 (11): 2029–2034. PMID: 1941134. 134 Rypins, E.B., Evans, D.G., Hinrichs, W. et al. (1997). Tc-99m-HMPAO white blood cell scan for diagnosis of acute appendicitis in patients with equivocal clinical presentation. Ann. Surg. 226 (1): 58–65. https://doi. org/10.1097/00000658-199707000-00008. PMID: 9242338; PMCID: PMC1190907. 135 Baba, A.A., McKillop, J.H., Cuthbert, G.F. et al. (1990). Indium 111 leucocyte scintigraphy in abdominal sepsis. Do the results affect management? Eur. J. Nucl. Med. 16 (4–6): 307–309. https://doi.org/10.1007/BF00842785. PMID: 2112472. 136 Palestro, C.J. and Love, C. (2018). Nuclear medicine imaging in fever of unknown origin: the new paradigm. Curr. Pharm. Des. 24 (7): 814–820. https://doi.org/10.217 4/1381612824666171129194507. PMID: 29189130. 137 Takeuchi, M., Dahabreh, I.J., Nihashi, T. et al. (2016). Nuclear imaging for classic fever of unknown origin: meta-analysis. J. Nucl. Med. 57 (12): 1913–1919. https:// doi.org/10.2967/jnumed.116.174391. Epub 23 Jun 2016. PMID: 27339873. 138 Wright, W.F., Auwaerter, P.G., Dibble, E.H. et al. (2021). Imaging a fever-redefining the role of 2-deoxy-2-[18F] Fluoro-D-Glucose-positron emission tomography/ computed tomography in fever of unknown origin investigations. Clin. Infect. Dis. 72 (7): 1279–1286. https://doi.org/10.1093/cid/ciaa1220. PMID: 32829386. 139 Bharucha, T., Rutherford, A., Skeoch, S. et al. (2017). Diagnostic yield of FDG-PET/CT in fever of unknown origin: a systematic review, meta-analysis, and Delphi exercise. Clin. Radiol. 72 (9): 764–771. https://doi. org/10.1016/j.crad.2017.04.014. Epub 7 Jun 2017. PMID: 28600002.

745

746

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

140 Hung, B.T., Wang, P.W., Su, Y.J. et al. (2017). The efficacy of 18F-FDG PET/CT and 67Ga SPECT/CT in diagnosing fever of unknown origin. Int. J. Infect. Dis. 62: 10–17. https://doi.org/10.1016/j.ijid.2017.06.019. Epub 23 Jun 2017. PMID: 28652213. 141 Dibble, E.H., Yoo, D.C., Baird, G.L. et al. (2019). FDG PET/CT of infection: should it replace labeled leukocyte scintigraphy of inpatients? Am. J. Roentgenol. 213 (6): 1358–1365. https://doi.org/10.2214/AJR.18.20475. Epub 28 Aug 2019. PMID: 31461320. 142 Seshadri, N., Sonoda, L.I., Lever, A.M. et al. (2012). Superiority of 18F-FDG PET compared to 111In-labelled leucocyte scintigraphy in the evaluation of fever of unknown origin. J. Infect. 65 (1): 71–79. https://doi. org/10.1016/j.jinf.2012.02.008. Epub 24 Feb 2012. PMID: 22369860. 143 Jasper, N., Däbritz, J., Frosch, M. et al. (2010). Diagnostic value of [(18)F]-FDG PET/CT in children with fever of unknown origin or unexplained signs of inflammation. Eur. J. Nucl. Med. Mol. Imaging 37 (1): 136–145. https://doi.org/10.1007/s00259-009-1185-y. PMID: 19526234. 144 Blokhuis, G.J., Bleeker-Rovers, C.P., Diender, M.G. et al. (2014). Diagnostic value of FDG-PET/(CT) in children with fever of unknown origin and unexplained fever during immune suppression. Eur. J. Nucl. Med. Mol. Imaging 41 (10): 1916–1923. https://doi.org/10.1007/ s00259-014-2801-z. Epub 29 May 2014. PMID: 24869631.

145 Sturm, E., Rings, E.H., Schölvinck, E.H. et al. (2006). Fluordeoxyglucose positron emission tomography contributes to management of pediatric liver transplantation candidates with fever of unknown origin. Liver Transpl. 12 (11): 1698–1704. https://doi. org/10.1002/lt.20922. PMID: 17058252. 146 Pijl, J.P., Kwee, T.C., Legger, G.E. et al. (2020). Role of FDG-PET/CT in children with fever of unknown origin. Eur. J. Nucl. Med. Mol. Imaging 47 (6): 1596–1604. https://doi.org/10.1007/s00259-020-04707-z. Epub 7 Feb 2020. PMID: 32030452; PMCID: PMC7188721. 147 Bleeker-Rovers, C.P., Vos, F.J., Mudde, A.H. et al. (2007). A prospective multi-centre study of the value of FDG-PET as part of a structured diagnostic protocol in patients with fever of unknown origin. Eur. J. Nucl. Med. Mol. Imaging 34 (5): 694–703. https://doi.org/10.1007/ s00259-006-0295-z. Epub 14 Dec 2006. PMID: 17171357. 148 Crouzet, J., Boudousq, V., Lechiche, C. et al. (2012). Place of (18)F-FDG-PET with computed tomography in the diagnostic algorithm of patients with fever of unknown origin. Eur. J. Clin. Microbiol. Infect. Dis. 31 (8): 1727–1733. https://doi.org/10.1007/s10096-0111440-6. Epub 27 Jun 2012. PMID: 22735898. 149 Manohar, K., Mittal, B.R., Jain, S. et al. (2013). F-18 FDG-PET/CT in evaluation of patients with fever of unknown origin. Jpn. J. Radiol. 31 (5): 320–327. https:// doi.org/10.1007/s11604-013-0190-z. Epub 1 Mar 2013. PMID: 23456545.

747

27 Imaging the Lymphatic System Girolamo Tartaglione1, Marco Pagan1, Francesco Pio Ieria1, Giuseppe Visconti2, and Tommaso Tartaglione3 1

Nuclear Medicine, Cristo Re Hospital, Rome, Italy Plastic Surgery, Lymphedema Center, A. Gemelli Hospital, Sacro Cuore Catholic University, Rome, Italy 3 Radiology, IDI- IRCCS, Rome, Italy 2

Lymphatic System The lymphatic system is a collection of vessels and lymph nodes (LNs) that is similar yet distinct from the vascular system. It is a complex network composed of lymphatic capillaries, lymphatic vessels, LNs, and other organs (tonsils, adenoids, spleen, thymus, bone marrow) distributed throughout the body. The primary function of the lymphatic system is to maintain fluid balance in the body. It is also an absorptive apparatus of fats and a part of the body’s immune system [1]. Lymph, a clear fluid that derives from plasma filtration across the permeable capillary endothelium, is transported within lymphatic vessels. It contains interstitial fluids, fats, few proteins, unwanted materials (waste and bacteria), and lymphocytes. Lymphatic capillaries are numerous throughout the body, especially in the papillary and reticular dermis of the skin where the superficial and deep plexuses are located  [2]. However, lymphatic capillaries are not present in the epidermis. Lymph is transported within lymphatic capillaries at a very low pressure, up to 2 mmHg. The function of lymphatic capillaries is to drain excessive interstitial fluids to lymphatic vessels. Superficial lymphatic vessels are found in subcutaneous tissues, where they run parallel to larger vessels, conducting lymph from the lymphatic capillaries to the large veins of the neck. Lymphatic vessels, in contrast to lymphatic capillaries, have abundant unidirectional valves, as well as thin endothelial walls and a low intraluminal pressure [3]. The lymphatic system lacks a central pump. Lymphatic flow is created by a pressure gradient and aided by unidirectional valves along the lymphatic collectors to prevent backflow. At rest, the peristaltic movement of lymphatic vessels is slow and depends on the intrinsic pumping mechanism of

the lymphangion. Lymphatic vessel pressure in a heathy subject is around 25 mmHg in the legs when supine and 30 mmHg when seated [4]. Lymph drainage is favored by a mechanism such as the surrounding skeletal muscle pump and thoracic pump. With muscular exercise, the pressure within lymphatic vessels increases, allowing lymph to move quickly toward the great veins of the neck. Before returning to the bloodstream, lymph passes through LNs. LNs are little bean-shaped structures and numbered in the hundreds. They can be found throughout the body, mainly concentrated in the neck, axilla, chest, abdomen, and groin. Lymphatic vessels connect LNs to one another. The lymphatic system plays a key role in the body’s immune system, helping the body to fight infection and disease. The LNs perform a “filtering” action, removing waste, cells, and pathogens from the lymph and releasing lymphocytes, before returning lymph back to the bloodstream.

Lymphedema Lymphedema is a condition of swelling caused by lymphatic obstruction, most commonly occurring in the lower limbs. It is produced by a blockage in or damage to the lymphatic system, causing an imbalance between capillary filtration and lymphatic drainage. Moreover, lymphedema negatively affects the quality of life and self-perception of the patient. The consequence of lymph accumulation in the tissues provokes swelling downstream to the site of obstruction. Edema indicates the presence of excess fluids in the surrounding tissues. In most cases, edema is the collection of extracellular fluids, but in extreme situations of metabolic tissue imbalances, it can involve the intracellular space.

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach, First Edition. Edited by Ali Gholamrezanezhad, Majid Assadi, and Hossein Jadvar. © 2023 John Wiley & Sons Ltd. Published 2023 by John Wiley & Sons Ltd.

748

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Lymphedema is classified into primary and secondary. Primary lymphedema is a hereditary condition caused by poorly developed lymphatic vessels or regional LNs. It is characteristically found in congenital dysplasia, genetic mutations, and hereditary syndromes. Primary lymphedema is relatively uncommon (30  million cases/ year worldwide, 1/100 000 people under the age of 20) and  can occur at birth, puberty or, more rarely, in adulthood. The specific causes of primary lymphedema include: ●

●

●

Milroy disease (congenital lymphedema), which is a familial condition occurring in infancy characterized by abnormal connections between the lymphatic and venous system Meige’s disease (lymphedema praecox), which is a disorder often causing lymphedema before the age of 35, typically during puberty or pregnancy late-onset lymphedema (lymphedema tarda), which is rare and usually begins after the age of 35.

Secondary lymphedema is far more common than primary lymphedema. In developing countries, filariasis, a parasitic infection of LNs that can limit lymph flow, is the most frequent cause of secondary lymphedema. In the industrialized world, secondary lymphedema is mainly iatrogenic. Secondary lymphedema may result from surgical cancer treatments, radiation, vascular trauma, or tumor cell invasion of LNs. Surgical dissection and radiation of LNs due to cancer are the more common causes of secondary lymphedema. Most frequently, lymphedema begins 12–36 months after a damage to lymphatic vessels. Threefourths of patients may develop swelling within 3 years after the injury with a 1% increase risk of lymphedema each year thereafter  [5]. Regional LN radiation may increase the risk of lymphedema, particularly radiation of the supraclavicular and posterior axillary LNs. Secondary lymphedema may develop during radiotherapy or a few years after treatment. Tumor metastases may infiltrate LNs, block the lymphatic vessels, and cause lymphedema. Older age, obesity, and rheumatoid or psoriatic arthritis are risk factors that increase the risk of developing lymphedema. Lymphedema is clinically observed as swelling of an arm or leg and, in severe cases, associated with swelling of the fingers and/or foot. Subjective symptoms may include a sense of heaviness or tension, limited movement, discomfort, recurrent infections, and hardening and thickening of the skin. Lymphedema may lead to serious complications. Even the smallest injury of an arm or leg may be an entry point for infection. Possible infections include serious bacterial infections of the skin (cellulitis) and lymph vessel infection (lymphangitis). Lymphangiosarcoma is a rare form of soft

tissue cancer that may arise from the most severe cases of lymphedema if left untreated. At clinical examination, lymphedema is typically characterized by a nonpitting edema (negative fovea sign) or swelling of one or more limbs. Physical examination parameters taken include the circumferential measurement of the affected limb compared to the contralateral (if healthy) in proportion to the body weight. The diagnosis of lymphatic insufficiency is confirmed with imaging methods (lymphoscintigraphy).

Lymphography Clinical imaging progressively advanced with the discovery of lymphography in the 1950s  [6]. Conventional oilcontrast lymphography had long been the mainstay for lymphatic imaging. Lymphography is a radiograph of lymphatic channels and LNs, obtained after the injection of a radiopaque contrast agent (ethiodol, the ethyl ester of iodized fatty acids, in 60–90 minutes) in surgically isolated small lymphatic vessels after a previous visualization with subcutaneous injections of patent blue dye (15 minutes earlier). In time, this method was largely abandoned due to complications related to the vital dye and contrast medium: pulmonary embolism, hypersensitivity, intra-alveolar hemorrhage, and hypothyroidism.

Subcutaneous Lymphoscintigraphy in Lymphedema In 1953, Sherman et al. first reported that radioactive colloidal gold (198Au), a beta emitter, had a potential for lymphatic system imaging and lymph-node identification [7]. The interstitial administration of 198Au was followed by a dose of radiation at the site of injection. In time, 198Au was replaced by radiocolloids marked with 99mTc to reduce the dosage of radiation. Lymphoscintigraphy is the gold standard for diagnosis of lymphedema but lacks procedural standardization. Several 99mTc-based radiopharmaceutical agents have been used for lymphatic imaging worldwide. Many factors may influence the visualization of the lymphatic vessels and the radiotracer uptake by loco-regional LNs, such as particle size, the amount and volume of injected radiotracer, and particle concentration. The choice of radiopharmaceutical is based on local availability. 99m Tc-antimony trisulfide, containing radioactive particles smaller than 20 nm and an acidic pH, is available in Australia and Canada. 99mTc-sulfur colloid, with a maximum size of 350–5000 nm and pH of 5.5, is the preferred agent in the United States. After the preparation of the

Imaging the Lymphatic System

radiopharmaceutical, it should be filtered with a 100– 200 nm membrane to select smaller particles. Rhenium sulfide nanocolloid, with a particle size 50–200 nm and pH of 5.5–6, is available in Europe. Because the dermis has abundant neuroreceptors sensitive to pH alterations, the radiopharmaceuticals should be injected subcutaneously to avoid pain at injection site. The radiotracer is drained by lymphatic vessels to LNs, where it is captured by macrophage phagocytosis and/or retained due to particle size. A fraction of the radiopharmaceutical moves on to secondand third-echelon nodes downstream. Smaller particles are drained more quickly, whereas larger particles migrate more slowly. For several years, lymphoscintigraphy has been performed using a subcutaneous injection of the radiotracer. It is a minimally invasive method, derived from the radiological experience of lymphography, used to assess lymphatic drainage. It consists of injecting a dose of 99mTclabeled colloids subcutaneously in the first interdigital space. Following the subcutaneous injection, the absorption of the radiocolloid is slow and requires prolonged static and total body acquisitions at 2–3 hours or more. In the lower limbs, normal superficial lymphatic drainage occurs via lymphatic vessels running within the superficial venous plexus, draining into the inguinal, iliac, and paraaortic nodes (the axillary, clavicular, and neck LNs in the upper limbs). Traditional lymphoscintigraphy offers a picture of the lymphatic system, showing how it functions by following the tracer as it moves up from the limb toward the central circulation. The tracer may identify the location where lymph is slowed or stopped to determine the extent of obstruction and diffusion of the tracer from the lymphatic vessels. When normal routes are blocked, lymphatic drainage is diverted either into the skin as dermal backflow and/or into the deep subfascial planes, resulting in popliteal (or antecubital) nodal uptake. Lymphoscintigraphy also includes the calculation of several semiquantitative parameters of lymphatic function: ● ● ● ●

tracer appearance time (TAT) to regional LNs % clearance at the injection site % of uptake of inguinal or axillary LNs transport index, TI = (K + D + 0.04T + N + V) (normal value 35 mSv) [56]. Interest in using this technique as an alternative to standard radiography is expanding, so that since 2011 the IMWG has proposed a full-body CT for use with MM instead of radiography. Due to the absence of radiation exposure, MRI has become the preferred imaging method for evaluating bone marrow involvement. Several sequences have been proposed for determining focal or diffuse bone marrow lesions, including T1-weighted imaging, T2-weighted imaging, and short-time inversion recovery (STIR)  [57]. Intravenous contrast enhancement requires additional fast T1-weighted imaging before and after contrast [53]. The main limitation of MRI is the long acquisition time, which is about 40 minutes [58]. Given that WB-MRI has not yet been widely used and several studies have been conducted to compare this procedure with other methods  [54], clinical experience with this imaging technique as a screening tool in patients with MM has yet to be clearly established. The appearance of PET scanners has changed the approach to the diagnosis of MM toward functional evaluation. 18F-FDG-PET in MM is the only procedure capable of detecting both bone marrow lesions and extramedullary lesions with high sensitivity and specificity  [59]. The appearance of hybrid PET/CT scanners offered the possibility of obtaining both metabolic and morphological information: while 18F-FDG-PET detects areas with intense hypermetabolic activity (active lesions), a CT can visualize lytic lesions [54]. In 2006, to improve the standardization of treatment and disease setting, Durie [60] revised the original 1975 setting system of Durie and Salmon [61], and developed the diagnosis system Durie/Salmon plus, which combines functional and anatomical diagnostic methods. The recently

updated IMWG criteria recommend the use of WB-LDCT as a newer imaging method in the basic treatment of patients with MM. WB-LDCT is considered an ideal tool for detecting early bone destruction due to its high resolution in both cortical and spongy bone [58]. WB-LDCT can also detect the presence of comorbid disease  [56]. Given that WB-LDCT protocols have not yet found wide application in radiological practice and WB-MDCT suggests a high level of radiation exposure, CT of individual areas is currently used in individual cases: (i) to visualize certain areas of the body that are poorly visible on X-rays, (ii) to evaluate certain vertebrae, (iii) to navigate for biopsy, (iv) to assess the risk of fracture, and (v) in patients who are contraindicated by MRI. Recently, WB-MRI, in contrast to MRI of individual body regions, has been the most sensitive imaging technique for detecting MM involvement not only in the spine but also in the whole skeleton [62]. As recently reported in IMWG guidelines, MRI is the gold standard imaging technique for detecting bone marrow involvement in MM (degree A), before osteolysis foci appear  [51]. Spine and pelvic MRI can detect 90% of focal lesions in MM. MRI is often the only method to detect spinal cord compression  [54] as well as the soft tissue component in spinal fractures [63]. The most interesting feature of MRI is the ability to evaluate the characteristics of bone marrow, given that it has a predictive value and correlates with overall survival [60]. In MM, combined with an increase in plasma cells in the bone marrow, the amount of fat decreases and this is characteristic of typical myeloma lesions, which have a hypo-intensity signal on T1w and a hyper-intensity signal on STIR/T2w images. Four different types of bone marrow can be detected: (i) normal, typical of MGUS and 50–75% of patients with clinical multiple myeloma, (ii) diffuse, observed in almost 80% of patients with progressive disease or high tumor load, which differs from the previous model because the infiltration of plasma cells diffusely affects the entire bone medullary compartments, (iii) micronodular (also called motley or “salt and pepper”), which reflects the heterogeneous structure of the bone marrow with the presence of fat islets and weak plasma cell infiltration (usually 5 mm) of high plasma cell infiltration [54]. MRI may also play a role in asymptomatic myeloma, given that patients with more than one focal area (diameter >5 mm) should be considered in accordance with IMWG as a symptomatic myeloma disease that requires treatment. Patients with dubious focal lesions should repeat the IMWG after 3–6 months and if there is an MRI progression

Lymphoma and Myeloma Correlative Imaging

they should be treated as symptomatic patients requiring treatment [51]. Several studies have considered using 18F-FDG-PET/CT as a reliable method for diagnosing and planning treatment of MM patients [64]. This imaging technique does provide accurate anatomical and functional information in a single full-body scan. It has been shown that 18F-FDG-PET/CT is able to detect hidden areas of bone and/or volume soft tissue formation in 30–50% of patients with newly diagnosed symptomatic MM  [65]. Since 18F-FDG-PET/CT measures tumor tissue metabolic activity, it can detect bone marrow lesions with high sensitivity and specificity and can detect at an early stage of MM before any detectable bone changes from CT data. In some cases, plasma cell tumors exhibit low proliferative activity, resulting in less glucose fixation compared with, say, lymphoma  [63]. There are assumptions that the level of metabolic activity correlates with tumor prognosis (51). In addition, 18F-FDG-PET seems to have a certain prognostic role in the newly diagnosed symptomatic MM, as patients with more than three focal lesions have a lower survival rate  [63]. False-negative results may be caused by small foci (less than a centimeter) and low metabolic activity. The most common cause of false negatives is inflammatory processes. The main advantage of 18F-FDG-PET/CT is the ability to distinguish the active process of myeloma disease (18F-FDG-positive) from its predecessors. Some authors [67] report that the risk of progression of active MM increases with the appearance of new focal or diffuse lesion zones.

What’s New in Myeloma Diagnosis? Relatively new diagnostic methods include WB-MRI combined with DWI. Some studies [68] have shown that adding WB-MRI to standard MRI pulsed sequences provides additional information. In particular, stage enhancement by Durie/Salmon plus is possible in more than a third of patients [69]. New PET radiopharmaceuticals have been investigated  [70] to overcome FDG limitations (18F). Each PET radiopharmaceutical allows different metabolic pathways and biological features to be evaluated. 18F-sodium fluoride (18F-NaF) deserves special mention because it is osteotropic and allows the detection of minimal osteoblastic activity accompanying osteolysis lesion (Figures  28.6 and 28.7) [71]. The absorption of 18F-NaF by bone is twice as high as that of 99mTc-MDP. One should not forget about the higher spatial resolution of PET in comparison with the gamma camera. Since 18F-FDG-PET/CT and 18F-NaF PET/CT provide different molecular information, the

Figure 28.6 CT of pelvic bones demonstrating multiple osteolytic lesions.

Figure 28.7 18F-FDG-PET/CT of pelvic bones demonstrates multiple osteolytic lesions with hypermetabolism.

combined use of 18F-NaF and 18F-FDG in one PET/CT scan [72] has been proposed. This is relevant for patients with MM, given that 18F-NaF can accurately localize bone involvement, while 18F-FDG can detect early bone marrowbased diseases as well as extramedullary manifestations of the disease. The role of PET/CT 3′-18F-fluoro-3′-deoxy-l-thymidine 18 ( F-FLT) in patients with MM is also the subject of study. 18 F-FLT is a precursor of DNA. The absorption of 18F-FLT is directly related to the DNA synthesis rate depending on the activity of the enzyme thymidinkinase-1, which is more pronounced in tumor cells. The other two radiopharmaceuticals tested for MM are 11 C-methionine and 11C-choline (11C-CHO). The first drug is also an amino acid labeled with a radioactivity label; the drug exhibits higher uptake in plasma cells, resulting in good sensitivity when detecting lesions associated with MM. 11C-CHO is a precursor of phospholipids, which participate in the formation of cell membranes, and its absorption increases in proliferating cells, where the mitotic process is faster. Preliminary studies have shown that

783

784

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

11

C-CHO-PET/CT is more sensitive than 18F-FDG-PET/CT to detect bone tissue involvement associated with MM [70]. Unlike the other radiopharmaceuticals described above, 11 C-CHO is not excreted in the urine and this property allows a better assessment of the urinary tract. A new frontier in imaging technology is the introduction of PET/MRI scanners. The combination of 18F-FDG-PET/ CT and WB-MRI can theoretically be the best diagnostic tool for MM. PET/MRI provides more information and at the same time reduces diagnostic time, long-term costs, and patient exposure due to the absence of CT (Figure 28.8). Unfortunately, PET/MRI scanners are expensive and due to their recent introduction diagnostic protocols are still being investigated and their feasibility should await the results of an ongoing clinical study.

Conclusion

Figure 28.8 (a) Map of apparent diffusion coefficient (ADC) demonstrates multiple areas of diffusion restriction which correspond to metastatic bone lesions. (b) Diffusion-weighted image demonstrates multiple areas of diffusion restriction which correspond to metastatic bone lesions.

WB-XR is still used in the initial examination of patients with suspected MM, but is increasingly being replaced by new and more sensitive imaging techniques. In particular, WB-LDCT is a selection procedure at present, although WB-MRI should be preferred for detecting bone marrow infiltration and extramedullary manifestations of the disease. WB-MRI and 18F-FDG-PET/CT represent the most reliable and effective imaging methods for developing individual therapy for patients with MM.

References 1 Swerdlow, S.H., Campo, E., Pileri, S.A. et al. (2016). The 2016 revision of the World Health Organization classification of lymphoid neoplasms. Blood 127 (20): 2375–2390. 2 Stone, M.J. (2005). Thomas Hodgkin: medical immortal and uncompromising idealist. Baylor Univ. Med. Cent. Proc. 18 (4): 368–375. 3 Shanbhag, S. and Ambinder, R.F. (2018). Hodgkin lymphoma: a review and update on recent progress. CA Cancer J. Clin. 68 (2): 116–132. 4 Wahed, A., Dasgupta, A., Wahed, A., and Dasgupta, A. (2015). Chapter 13 – Hodgkin lymphoma. In: Hematology and Coagulation, 215–223. Elsevier. 5 Brierley, J.D., Gospodarowicz, M.K., and Wittekind, C. (2017). TNM Classification of Malignant Tumours, 8e. Wiley. 6 Cheson, B.D., Fisher, R.I., Barrington, S.F. et al. (2014). Recommendations for initial evaluation, staging, and response assessment of hodgkin and non-hodgkin lymphoma: the Lugano classification. J. Clin. Oncol. 32 (27): 3059–3067.

7 Esau, D. (2017). Viral causes of lymphoma: the history of Epstein-Barr virus and human T-lymphotropic virus 1. Virol.: Res. Treat. 8. SAGE Publications Ltd: 1–5. 8 Zhang, Y., Dai, Y., Zheng, T. et al. (2011). Risk factors of non-Hodgkin’s lymphoma. Expert Opin. Med. Diagn. 5 (6): 539–550. 9 Flowers, C.R., Shenoy, P., Maggioncalda, A. et al. (2011). Incidence patterns and outcomes for Hodgkin lymphoma patients in the United States. Adv. Hematol. 2011, 1–2011. 10 DeSantis, C.E., Lin, C.C., Mariotto, A.B. et al. (2014). Cancer treatment and survivorship statistics, 2014. CA Cancer J. Clin. 64 (4): 252–271. 11 Metzger, M.L. and Mauz-Körholz, C. (2019). Epidemiology, outcome, targeted agents and immunotherapy in adolescent and young adult non-Hodgkin and Hodgkin lymphoma. Br. J. Haematol. 185 (6): 1142–1157. 12 Eichenauer, D.A., Aleman, B.M.P., André, M. et al. (2018). Hodgkin lymphoma: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann. Oncol. 29 (February 2002): iv19–iv29.

Lymphoma and Myeloma Correlative Imaging

13 Tanaka, Y. et al. (2018). Expression pattern of PD-L1 and PD-L2 in classical Hodgkin lymphoma, primary mediastinal large B-cell lymphoma, and gray zone lymphoma. Eur. J. Haematol. 100 (5): 511–517. 14 Younes, A. et al. (2012). Results of a pivotal phase II study of brentuximab vedotin for patients with relapsed or refractory Hodgkin’s lymphoma. J. Clin. Oncol. 30 (18): 2183–2189. 15 Harrison, A.M., Thalji, N.M., Greenberg, A.J. et al. (2014). Rituximab for non-Hodgkin’s lymphoma: a story of rapid success in translation. Clin. Transl. Sci. 7 (1): 82–86. 16 Zimmermann, M., Oehler, C., Mey, U. et al. (2016). Radiotherapy for non-Hodgkin’s lymphoma: still standard practice and not an outdated treatment option. Radiat. Oncol. 11 (1): 110. 17 Barrington, S.F. et al. (2014). Role of imaging in the staging and response assessment of lymphoma: Consensus of the International Conference on Malignant Lymphomas Imaging Working Group. J. Clin. Oncol. 32 (27): 3048–3058. 18 Shen, B., Huang, T., Sun, Y. et al. (2017). Revisit 18 F-fluorodeoxyglucose oncology positron emission tomography: ‘systems molecular imaging’ of glucose metabolism. Oncotarget 8 (26) Impact Journals LLC: 43536–43542. 19 Boellaard, R., Delgado-Bolton, R., Oyen, W.J.G. et al. (2014). FDG PET/CT: EANM procedure guidelines for tumour imaging: version 2.0. Eur. J. Nucl. Med. Mol. Imaging 42 (2): 336–338. 20 Edwards, C.L. and Hayes, R.L. (1969). Tumor scanning with 67Ga citrate. J. Nucl. Med. 10 (2): 103–105. 21 Divgi, C. (2005). Imaging: staging and evaluation of lymphoma using nuclear medicine. Semin. Oncol. 32 (SUPPL. 1): 11–18. 22 Tsan, M.F. and Scheffel, U. (1986). Mechanism of gallium-67 accumulation in tumors. J. Nucl. Med 27 (7): 1215–1219. 23 Moog, F., Kotzerke, J., and Reske, S.N. (1999). FDG PET can replace bone scintigraphy in primary staging of malignant lymphoma. J. Nucl. Med. 40 (9): 1407–1413. 24 Love, C., Din, A.S., Tomas, M.B. et al. (2003). Radionuclide bone imaging: an illustrative review. Radiographics 23 (2) Radiological Society of North America Inc.: 341–358. 25 O’Sullivan, G.J. (2015). Imaging of bone metastasis: an update. World J. Radiol. 7 (8): 202. 26 Mawlawi, O., Erasmus, J.J., Munden, R.F. et al. (2006). Quantifying the effect of IV contrast media on integrated PET/CT: clinical evaluation. Am. J. Roentgenol. 186 (2): 308–319. 27 de Bonilla-Damiá, A., Fernández-López, R., CapoteHuelva, F.J. et al. (2017). Role of 18F-FDG PET/CT in

28

29

30

31

32

33

34

35

36

37

38

39

40

primary brain lymphoma. Rev. Esp. Med. Nucl. Imagen Mol. 36 (5): 298–303. Ngeow, J.Y.Y., Quek, R.H.H., Ng, D.C.E. et al. (2009). High SUV uptake on FDG-PET/CT predicts for an aggressive B-cell lymphoma in a prospective study of primary FDG-PET/CT staging in lymphoma. Ann. Oncol. 20 (9): 1543–1547. Broccoli, A., Nanni, C., Cappelli, A. et al. (2017). PET/ CT-guided biopsy for the diagnosis of lymphoma. Hematol. Oncol. 35: 425–425. Meignan, M., Gallamini, A., and Haioun, C. (2009). Report on the first international workshop on interimPET scan in lymphoma. Leuk. Lymphoma 50 (8): 1257–1260. Ma, Y., Wang, Q., Dong, Q. et al. (2019). How to differentiate pseudoprogression from true progression in cancer patients treated with immunotherapy. Am. J. Cancer Res. 9 (8): 1546–1553. Cheson, B.D., Ansell, S., Schwartz, L. et al. (2016). Refinement of the Lugano classification lymphoma response criteria in the era of immunomodulatory therapy. Blood 128 (21): 2489–2496. van Son, P.M., Evers, D., and Aarntzen, E.H.J.G. (2019). The value of FDG-PET CT scans to evaluate bone marrow in haemato-oncological conditions. Neth. J. Med. 77 (7): 234–242. Inoue, K., Okada, K., Harigae, H. et al. (2006). Diffuse bone marrow uptake on F-18 FDG PET in patients with myelodysplastic syndromes. Clin. Nucl. Med. 31 (11): 721–723. Long, N.M. and Smith, C.S. (2011). Causes and imaging features of false positives and false negatives on 18FPET/CT in oncologic imaging. Insights Imaging 2 (6): 679–698. Pessanha, L.B., de Oliveira, A.R.N., Guerra, L.F.A. et al. (2015). PET/CT e gordura marrom na avaliação da resposta terapêutica no linfoma de Hodgkin. Radiol. Bras. 48 (6) Colegio Brasileiro de Radiologia: 402–403. Steinberg, J.D., Vogel, W., and Vegt, E. (2017). Factors influencing brown fat activation in FDG PET/CT: a retrospective analysis of 15,0001 cases. Br. J. Radiol. 90 (1075): 1–8. Shetty, A. and Carter, J.D. (2011). Sarcoidosis mimicking lymphoma on FDG-PET imaging. Radiol. Case Rep. 6 (2): 409. Kis, A., Eszes, N., Tamasi, L. et al. (2013). Sarcoidosis lymphoma syndrome - the value of PET-CT in the diagnosis. World J. Surg. Oncol. 11: 1–6. Hou, S., Shen, J., and Tan, J. (2017). Case report: multiple systemic disseminated tuberculosis mimicking lymphoma on 18F-FDG PET/CT. Med. (United States) 96 (29): 1–8.

785

786

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

41 Barrington, S.F. and Johnson, P.W.M. (2017). 18F-FDG PET/CT in lymphoma: has imaging-directed personalized medicine become a reality? J. Nucl. Med. 58 (10): 1539–1544. 42 Skoura, E., Ardeshna, K., Halsey, R. et al. (2016). False-positive 18F-FDG PET/CT imaging dramatic flare response after rituximab administration. Clin. Nucl. Med. 41 (3): e171–e172. 43 Chang, J.M., Lee, H.J., Goo, J.M. et al. (2006). False positive and false negative FDG-PET scans in various thoracic diseases. Korean J. Radiol. 7 (1): 57–69. 44 Mayerhoefer, M.E., Raderer, M., Jaeger, U. et al. (2018). Ultra-early response assessment in lymphoma treatment: [18F]FDG PET/MR captures changes in glucose metabolism and cell density within the first 72 hours of treatment. Eur. J. Nucl. Med. Mol. Imaging 45 (6): 931–940. 45 Lin, C., Itti, E., Luciani, A. et al. (2010). Whole-body diffusion-weighted imaging in lymphoma. Cancer Imaging 10 (SPEC. ISS. A): 1–7. 46 Abdulqadhr, G., Buemi, F., Quagliozzi, M. et al. (2011). Whole-body diffusion-weighted imaging compared with FDG-PET/CT in staging of lymphoma patients. Acta Radiol. 52 (2): 173–180. 47 Littooij, A.S., Kwee, T., Barber, I. et al. (2014). Wholebody MRI for initial staging of paediatric lymphoma: prospective comparison to an FDG-PET/CT-based reference standard. Eur. Radiol. 24 (5): 1153–1165. 48 Sher, A.C., Seghers, V., Paldino, M.J. et al. (2016). Assessment of sequential PET/MRI in comparison with PET/CT of pediatric lymphoma: a prospective study. Am. J. Roentgenol. 206 (3): 623–631. 49 Afaq, A., Fraioli, F., Sidhu, H. et al. (2017). Comparison of PET/MRI with PET/CT in the evaluation of disease status in lymphoma. Clin. Nucl. Med. 42 (1): e1–e7. 50 Terpos, E. and Dimopoulos, M.-A. (2005). Myeloma bone disease: pathophysiology and management. Ann. Oncol. Off. J. Eur. Soc. Med. Oncol. 16 (8): 1223–1231. 51 Dimopoulos, M.A., Hillengass, J., Usmani, S. et al. (2015). Role of magnetic resonance imaging in the management of patients with multiple myeloma: a consensus statement. J. Clin. Oncol. 33 (6): 657–664. 52 D’Sa, S., Abildgaard, N., Tighe, J. et al. (2007). Guidelines for the use of imaging in the management of myeloma. Br. J. Haematol. 137 (1): 49–63. 53 Healy, C.F., Murray, J.G., Eustace, S.J. et al. (2011). Multiple myeloma: a review of imaging features and radiological techniques. Bone Marrow Res. 2011: 583439. 54 Zamagni, E. and Cavo, M. (2012). The role of imaging techniques in the management of multiple myeloma. Br. J. Haematol. 159 (5): 499–513.

55 Mahnken, A.H., Wildberger, J.E., Gehbauer, G. et al. (2002). Multidetector CT of the spine in multiple myeloma: comparison with MR imaging and radiography. Am. J. Roentgenol. 178 (6): 1429–1436. 56 Gleeson, T.G., Moriarty, J., Shortt, C.P. et al. (2009). Accuracy of whole-body low-dose multidetector CT (WBLDCT) versus skeletal survey in the detection of myelomatous lesions, and correlation of disease distribution with whole-body MRI (WBMRI). Skelet. Radiol. 38 (3): 225–236. 57 Weininger, M., Lauterbach, B., Knop, S. et al. (2009). Whole-body MRI of multiple myeloma: comparison of different MRI sequences in assessment of different growth patterns. Eur. J. Radiol. 69 (2): 339–345. 58 Lütje, S., Rooy, J.W.J., Croockewit, S. et al. (2009). Role of radiography, MRI and FDG-PET/CT in diagnosing, staging and therapeutical evaluation of patients with multiple myeloma. Ann. Hematol. 88 (12): 1161–1168. 59 Bredella, M.A., Steinbach, L., Caputo, G. et al. (2005). Value of FDG PET in the assessment of patients with multiple myeloma. Am. J. Roentgenol. 184 (4): 1199–1204. 60 Durie, B.G.M. (2006). The role of anatomic and functional staging in myeloma: description of Durie/ Salmon plus staging system. Eur. J. Cancer 42 (11): 1539–1543. 61 Durie, B.G.M. and Salmon, S.E. (1975). A clinical staging system for multiple myeloma correlation of measured myeloma cell mass with presenting clinical features, response to treatment, and survival. Cancer 36 (3): 842–854. 62 Baur-Melnyk, A., Buhmann, S., Dürr, H.R. et al. (2005). Role of MRI for the diagnosis and prognosis of multiple myeloma. Eur. J. Radiol. 55 (1): 56–63. 63 Tosi, P. (2013). Diagnosis and treatment of bone disease in multiple myeloma: spotlight on spinal involvement. Scientifica (Cairo). 2013: 104546. 64 Zamagni, E., Nanni, C., Patriarca, F. et al. (2007). A prospective comparison of 18F-fluorodeoxyglucose positron emission tomography-computed tomography, magnetic resonance imaging and whole-body planar radiographs in the assessment of bone disease in newly diagnosed multiple myeloma. Haematologica 92 (1): 50–55. 65 Nanni, C., Zamagni, E., Farsad, M. et al. (2006). Role of 18F-FDG PET/CT in the assessment of bone involvement in newly diagnosed multiple myeloma: preliminary results. Eur. J. Nucl. Med. Mol. Imaging 33 (5): 525–531. 66 Tan, E., Weiss, B.M., Mena, E. et al. (2011). Current and future imaging modalities for multiple myeloma and its precursor states. Leuk. Lymphoma 52 (9): 1630–1640. 67 Zamagni, E., Nanni, C., Gay, F. et al. (2016). 18F-FDG PET/CT focal, but not osteolytic, lesions predict the

Lymphoma and Myeloma Correlative Imaging

progression of smoldering myeloma to active disease. Leukemia 30 (2): 417–422. 68 Cafagna, D., Rubini, G., Luele, F. et al. (2012). Wholebody RM-DWIBS vs. 18F-FDG-PET/TC nelle neoplasie maligne: Studio retrospettivo di confronto. Radiol. Med. 117 (2): 293–311. 69 Narquin, S. (2013). Comparison of whole-body diffusion MRI and conventional radiological assessment in the staging of myeloma. Diagn. Interv. Imaging 94 (6): 629–636.

70 Nishizawa, M., Nakamoto, Y., Suga, T. et al. (2010). 11C-methionine PET/CT for multiple myeloma. Int. J. Hematol. 91 (5): 733–734. 71 Sood, A., Revannasiddaiah, S., and Kumar, R. Nuclear medicine in myeloma: the state of the science and emerging trends. Hell. J. Nucl. Med. 14 (1): 2–5. 72 Iagaru, A., Mittra, E., Yaghoubi, S.S. et al. (2009). Novel strategy for a cocktail 18F-fluoride and 18F-FDG PET/CT scan for evaluation of malignancy: results of the pilotphase study. J. Nucl. Med. 50 (4): 501–505.

787

788

29 Clinical Application of PET/MRI Laura Evangelista, Paolo Artoli, Paola Bartoletti, Antonio Bignotto, Federica Menegatti, Marco Frigo, Stefania Antonia Sperti, Laura Vendramin, and Diego Cecchin Nuclear Medicine Unit, Department of Medicine, University of Padua, Padua, Italy

PET/MRI General Introduction Positron emission tomography/magnetic resonance imaging (PET/MRI) represents a complex imaging modality able to combine the anatomical and functional information provided by MRI and the functional/molecular data provided by PET. However, its interpretation requires coordination of many professional figures, from technologists to physicians (both radiologists and nuclear medicine physicians). Furthermore, specific knowledge about the clinical information obtained from PET and MRI is important in order to select appropriate patients who may benefit from this hybrid imaging modality. Anamnesis and clinical data should be always accurately taken before the examination because a lot of patients can suffer from claustrophobia. Moreover, some conditions are not compatible with MRI scanners, such as the presence of metal or a pacemaker. Whole-body PET/MRI is usually performed in all patients with oncological requests. It is usually performed using the most common MRI sequences (T1-weighted and T2-weighted sequences), with or without the adjunction of contrast-enhancement (such as gadolinium-based agents). Moreover, additional sequences can be added, such as diffusion weighted imaging (DWI), for the assessment of the apparent diffusion coefficient (ADC). Often, additional segmental images are made in the case of specific clinical requests, for example for liver metastases, head and neck cancer, and similar. Nononcological disease, such as cardiac sarcoidosis and brain dementia, requires dedicated protocols for the analysis of the movements or for specific anatomical assessments, respectively.

Finally, for the correction of PET images, the Dixon or Caipirinha sequences are used. In the following sections we address the use of PET/ MRI, mainly with 18F-fluorodeoxyglucose (18F-FDG), in oncological and nononcological conditions based on the daily clinical experiences we have amassed.

­ pplication of PET/MRI in the A Oncological Setting PET/MRI in Brain Tumors Gliomas are the most common primary brain tumors in children and adults, consisting of a heterogeneous group of neoplastic diseases arising from the supporting cells of the central nervous system (CNS). Computed tomography (CT) is frequently obtained prior to MRI for the initial assessment of a suspected intracranial mass lesion. The limitations of CT compared to MRI include inferior soft tissue characterization, posterior fossa beam hardening artifact, and the use of ionizing radiation. Conventional gadolinium-enhanced MRI is the modality of choice for the evaluation and management of intracranial masses. It is pivotal for the differential diagnosis of brain tumors as it provides information about both their size and location, and also gives an estimate of blood–brain barrier breakdown by the degree of gadolinium contrast enhancement. However, it has some limitations in differentiating high- from low-grade tumors [1] and in distinguishing disease recurrence from post-therapy changes  [2]. Moreover, radiological assessment of the true extent of glioma by conventional MRI is unsatisfactory, as enhancing MR abnormalities

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach, First Edition. Edited by Ali Gholamrezanezhad, Majid Assadi, and Hossein Jadvar. © 2023 John Wiley & Sons Ltd. Published 2023 by John Wiley & Sons Ltd.

Clinical Application of PET/MRI

can reflect either oedema or tumor infiltration, therefore brain tumor tissue may extend beyond areas with gadolinium enhancement [3]. Advanced MRI modalities have been proposed to determine all information that cannot be obtained through standard anatomic MRI. The most widely studied and most commonly employed of these modalities are spectroscopy [4], perfusion, and DWI MRI. This latter sequence is sensitive to the motion of water molecules in three dimensions within tissue and diffusion tensor imaging acquires water diffusion information using a minimum of six diffusion-encoded directions. Diffusion tensor imaging (DTI) has been shown to differentiate between low- and high-grade glioma, and to distinguish glioblastoma from metastases. It also delineates the margins of primary brain tumors better than conventional MRI alone [5, 6]. Functional MRI (fMRI) is primarily used for presurgical evaluation to localize regions of motor activation and language activation that lie nearby or within a lesion [7]. The information provided by fMRI is often coupled with diffusion tractography to optimize presurgical planning with the goal of maximizing the extent and safety of tumor resection while minimizing postoperative neurological deficits [8]. Hybrid PET/MRI combines, in a single scan, the high spatial resolution and soft tissue characterization of gadolinium-enhanced and advanced MRI with metabolic information from PET. The acquisition is performed with simultaneous PET and MRI scanning in the magnetic resonance head coil. The first step is a localization scan followed by a T1-weighted sequence of volumetric interpolated breath-hold examination (VIBE) two-point-Dixon gradient echo, used to create a tissue model for attenuation correction and to reconstruct the attenuation-corrected PET images. Standard anatomic imaging of the brain typically includes two-dimensional precontrast T1- and T2-weighted spin-echo, fluid-attenuated inversion recovery (FLAIR), and postcontrast T1-weighted spin-echo sequences in at least two orthogonal planes. Many protocols also include advanced MRI modalities, such as DWI or DTI. Different radiotracers are available for PET brain imaging, and some papers have reported the advantages of PET/ MRI with them [9, 10]. 18F-FDG is the most commonly used radiopharmaceutical for its high specificity in differentiating the presence of high-grade tumors, characterized by pathologically increased accumulation of tracer above the level of gray matter and for the possibility of distinguishing postradiation necrosis from local recurrence, often characterized by accumulation at the same level of the surrounding unaffected white matter  [11]. Figure  29.1 shows a patient with a brain tumor detected by 18F-FDG PET/MRI. The major advantage of amino acid analog PET tracers is their relatively high uptake in tumor tissue and comparatively

low uptake in the normal brain, with significant tumor-tobackground contrast. The uptake of these radiolabeled amino acids is based on the expression of l-amino acid transporter (LAT)-1 and LAT2 on the surface of tumor cells with nondependence on blood–brain barrier permeability; for this reason, amino acid probes are able to depict noncontrast-enhancing brain tumor regions with a clear advantage over other PET tracers [12]. 11C-methionine (MET) is the most widely studied tracer in this group, but its use is restricted to centers with an on-site cyclotron by the short half-life of 11C (20minutes) compared to 18F (110minutes)  [13]. For this reason, 18F-labeled tracers have been developed, including 18F-fluoroethyl-ltyrosine (FET) and the amino acid analog tracer 3,4-dihydroxy-6-18F-fluoro- l-phenylalanine (FDOPA). FET PET has been studied for delineating the extent of gliomas in order to guide tissue sampling and treatment planning, for detection of tumor recurrence, and for prognostication in low-grade glioma. In a recent study including a small prospective cohort of high-grade glioma recurrences treated with bevacizumab and irinotecan, both standard and kinetic FET PET imaging were demonstrated to be superior to conventional MRI, using response assessment in neurooncology (RANO) criteria, for prediction of early treatment failure [14]. Figure 29.2 shows an example of PET/MRI with FET in a recurrent glioma. FDOPA PET has been shown to be more accurate than 18F-FDG PET for evaluation of lowgrade tumors and for distinguishing tumor recurrence from radiation necrosis [12]. However, the use of FDOPA imaging of gliomas is rare because it is associated with a high degree of accumulation in the dopaminergic system in the brain/ basal ganglia and also in the putamen or substantia nigra, which might interfere with tumor delineation [10]. 18 F-fluoromisonidazole (FMISO) represents a PET imaging agent for identification of hypoxia and has an improved ability to distinguish glioblastoma (GBM) from lower grades as compared with 18F-FDG. Similarly, 18F-fluoroazomycin arabinoside (FAZA) seems to detect the presence of a hypoxic region within the tumors, thus guiding a personalized radiotherapy planning.

PET/MRI in Head and Neck Cancer Head and neck cancer ranks within the 10 most common cancers in the human body  [15, 16]. The head and neck region is an anatomically complex site that performs various physiological functions and constitutes the facial morphology. Therefore, for patients with head and neck cancer, accurate localization of primary and recurrent tumor is an essential for planning optimal treatment to preserve its function and morphology as much as possible. PET MRI offers many advantages in the characterization and prognostication of head and neck malignancies thanks

789

790

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 29.1 18F-FDG-PET/MRI of an intracranial mass in the left parietal lobe. The PET scan shows a high 18F-FDG uptake in the left parietal lobe. The postcontrast T1-weighted MRI shows an inhomogeneous postgadolinium enhancement in the same region, with a low signal for DWI images.

to the sensitivity and the negative predictive value of PET combined with the high soft tissue contrast of MRI. In the initial staging of primary tumors, 18F-FDG PET/MRI performed better than PET/CT. In particular, the superiority of T2-weighted and nonenhanced T1-weighted sequences over PET/CT in local staging  [17] and the advantage of contrast-enhanced T1-weighted sequences in evaluating perineural tumor spread that is a negative prognostic risk factor have been demonstrated. However, the assessment of primitive tumors can be affected by imaging artifacts, depending on location: tumors arising in the oral cavity and oropharynx are more affected by dental artifacts (beam hardening), while tumors arising in the hypopharynx and larynx are more affected by motion artifacts from swallowing. The first artifacts can significantly compromise the interpretation of PET/ CT and the second one can compromise the interpretation of PET/MRI. The appropriate hybrid imaging should therefore be assessed on the basis of clinical information. In the assessment of lymph node involvement, a slight advantage of T2-weighted imaging over T1 has been reported,

although many studies have confirmed an equal performance of PET/MRI in comparison to PET/CT [16–19]. The main challenge for PET/MRI is the evaluation of lung metastasis. Due to breathing and the duration of MRI sequences, motion can affect image quality. However, this limitation can be overcome by using a gated lung technique and/or a dedicated MR pulse sequence for the lungs that allows the detection of nodules only 3 mm in size. Nevertheless, there is evidence that PET/MRI might surpass PET/CT in the detection of concurrent occult primaries, particularly in other head and neck regions [20, 21]. The main disadvantages of PET/MRI in head and neck tumors, but also in other malignancies, are the longer acquisition time and the higher costs compared to CT. However, a full diagnostic PET/MRI scan in the head and neck can be obtained in 30–40 minutes. The acquisition protocol should be optimized using T1-weighted axial, T1 fat saturation axial plus coronal or sagittal, and T2 fat sat axial sequences to evaluate the neck region, and evaluating the whole body using a T1 DIXON axial and a T2 fat sat coronal to assess the

Clinical Application of PET/MRI

Figure 29.2 18F FET-PET/MRI of GBM in the right temporal lobe. The PET scan shows a high FET uptake in the antero-lateral and medial side of the surgical cord in the right temporal lobe (TBRmax mean 3.8). The postcontrast T1-weighted MRI shows a lesion with an enhancement, associated with an alteration of the fluid-attenuated inversion recovery (FLAIR) signal in the T2 sequences.

whole body. Whenever possible, the scan should be gadolinium contrast-enhanced to obtain the highest diagnostic accuracy. DWI and perfusion-weighted imaging should be performed for the evaluation of the response to therapy and for the planning of radiation treatment. Figure 29.3 shows PET/MRI of a primitive head and neck tumor detected by 18 F-FDG, postcontrast T1, and DWI-MRI sequences.

PET/MRI in Hepatocellular Carcinoma Hepatocellular carcinoma (HCC) represents over a third of liver cancer. In recent years, 5-year survival rates have been considerably improved due to earlier detection and curative therapies [22]. Many studies have demonstrated that MRI has excellent sensitivity and specificity for the detection and characterization of HCC compared with CT and ultrasound [23–26]. The role of 18F-FDG PET in HCC diagnosis is not well established since the metabolism of HCC cells is highly variable. Low expression of glucose transport

(GLUT)1 and GLUT2, high expression of P-glycoprotein, and high activity of 18F-FDG-6-phosphatase were demonstrated in moderately and well-differentiated HCCs  [27]. However, many authors have reported on the high sensitivity of 18F-FDG PET in the detection of extrahepatic HCC metastases [28–32]. Moreover, 18F-FDG PET showed strong prognostic value in HCC patients, since less differentiated and more aggressive HCCs usually have higher rates of glucose consumption  [33, 34]. The combination of PET with MRI in patients with HCC seems very promising for many reasons: (i) detection of the primary tumor, thanks to the additional information provided by PET and MRI, (ii) assessment of local therapies (such as transarterial chemoembolization), and (iii) target lesion dosimetry in patients who are candidates for 90Y radioembolization. However, to date few studies using PET/MRI in patients with HCC have been available [35–38]. Figure 29.4 shows the presence of multiple areas of 18F-FDG uptake and specific MRI signs in a patient with a multifocal HCC.

791

792

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 29.3 18F-FDG PET showing two focal areas of uptake in the anterior left margin of the tongue and in a small left satellite lymph node. A postcontrast T1-VIBE-MRI sequence shows increased enhancement in the same regions. With DWI a high signal was visible, mainly in the left lymph node, while a small diffusion signal was demonstrated in the tongue.

PET/MRI in Pancreas Disease Pancreatic ductal adenocarcinoma (PDA) accounts for 80% of pancreatic malignancies. A complete resection is the only cure, but is limited to 15–30% of patients who received an early diagnosis in a late onset of the disease  [39]. Enhanced CT is the most used imaging modality for diagnosis and initial staging of PDA, but several studies have shown that MRI is superior in detecting smaller lesions. MRI using T1- and T2-weighted images, DWI, ADC, and a hepato-specific intravenous contrast is useful for the evaluation of morphologic features of primary and metastatic liver cancer. 18F-FDG PET/CT in preoperatory staging of PDA is useful in detecting metastatic disease, with some studies showing a change in management in up to 11% of patients, avoiding unnecessary surgery owing to the detection of occult metastases  [40]. Moreover, high standardized uptake values (SUVs) on pretreatment 18FFDG PET/CT scans correlated with a worse prognosis and a decrease in progression-free survival  [41]. After neoadjuvant therapy, anatomic imaging showed limited sensitivity for the evaluation of treatment response due to postradiation changes. PET has been found to be very helpful in this setting because a decrease in 18F-FDG uptake

may precede morphologic changes and therefore can better estimate the response to treatment. Similarly, also in the postoperative setting, the presence of surgical scarring limits the identification of PDA recurrence. Integrated PET/ MRI can potentially be superior to each modality alone, combining high soft tissue contrast for MRI and metabolic information from PET. For example, 18F-FDG PET/MRI had sensitivity equal to 93% versus 91% for MRI with hepato-specific contrast alone for the identification of liver metastases [42]. Therefore, in the initial staging of disease, 18 F-FDG PET/MR can be used, although the small amount of evidence available and the limited availability of PET/ MR scanners represent important limitations in clinical practice. Figure  29.5 shows an example of patient with pancreatic cancer staged with 18F-FDG PET/MRI. Intraductal papillary mucinous neoplasms (IPMNs) are considered premalignant lesions, characterized by the papillary growth of ductal epithelium with mucin production. They are classified into main duct type (main pancreatic duct dilatation >5 mm), associated with 60% risk for malignant transformation, branch duct type (cystic lesion in continuity with the main pancreatic duct), associated with 25% risk for malignant transformation, and mixed IPMN.

Clinical Application of PET/MRI

Figure 29.4 Top: The first three images on the left show multiple 18F-FDG uptake in the right liver lobe. T2, T1, and DWI MRI images show a inhomogeneous signal at the level of the liver segments (S8, 11 mm in diameter; S7, 20 mm; S4b, 6 mm; S5–S8, 7mm).

Based on the degree of dysplasia of ductal epithelium, they are classified from mild, intermediate, and high-grade dysplasia to invasive carcinoma [43]. MRI/MR cholangiopancreatography and endoscopic ultrasound imaging showed a similar diagnostic accuracy for the diagnosis of IPMN.

18

F-FDG PET/CT has demonstrated that an increased 18FFDG uptake is highly specific in differentiating between benign and invasive IPMNs with a suggested SUVmax cutoff value of 2.3 [44–47]. Pedrazzoli et al., in a trial published in 2011, showed a higher accuracy of 18F-FDG PET/CT in

793

794

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 29.5 80-year-old woman with a diagnosis of pancreatic ductal adenocarcinoma. Top, 18F-FDG PET/MR shows a solid nodule with a high 18F-FDG uptake in the pancreatic head. In the same region, hysointensity in T1-weighted and hypointensity in T2-weighted images was found. Bottom, PET/MRI images in the liver show the presence of a hyperintense lesion in T2 and a hypointense lesion in T1 of 13 mm in diameter in S3 with a moderate 18F-FDG uptake, compatible with a distant metastasis.

Clinical Application of PET/MRI

differentiating malignant from benign IPMN compared with 2006 international consensus guidelines criteria (sensitivity 83% vs. 93%, specificity 100% vs. 22%, accuracy 91% vs. 61) [46]. It is important to mention that 18-FDG PET can be liable to false-positive results due to inflammatory disease, as in pancreatitis. The timing of PET in patients under surveillance remains to be established, but it might reasonably be performed when lesions reveal morphological changes or patients become symptomatic. PET/MRI scanners have a possible role in the management of IPMN, but data in the literature are limited [48]. Pancreatic neuroendocrine tumors (PNETs) origin from endocrine cells and 80% of them express somatostatin receptors (SSRT) type 1–5. The expression of SSRT allows targeting with synthetic somatostatin for diagnostic and therapeutic purposes. Usually, PNETs are combined with small bowel neuroendocrine tumors and classified as gastroenteropancreatic NET (GEPNETs). The classification is based on mitotic count and Ki-67 proliferation index: G1 (Ki-67 < 3%), G2 (Ki-67 3–20%), and G3 (Ki-67 > 20%) [49]. Enhanced CT is usually the first study performed, showing a sensitivity of 70%. MRI with intravenous contrast triple phase (also with delayed postcontrast), DWI, and ADC sequences is useful to identify hepatic metastases. Recently, 68 Ga-labeled SSTR agonist (DOTA-TOC, DOTA-NOC, and DOTA-TATE) PET has been demonstrated to be useful for staging, restaging, and selecting candidates for therapy with radiolabeled somatostatin analogues. A recent metaanalysis of 22 studies showed a sensitivity of 97% and a specificity of 95%  [50]. The largest study evaluating how 68 Ga DOTA-TATE PET/CT impacts on the therapeutic decision of this tumor type showed that it changed management in 40.9% of patients [51]. PET/MRI seems a promising tool for evaluation of GEPNETs, combining the accuracy of SSTR PET and the soft tissue contrast of MRI, therefore it can be considered definitely better in finding liver metastases, but has a lower detection rate for lung lesions. Again, evidence about this topic is scarce in the literature [52].

PET/MRI in Prostate Cancer Prostate cancer (PCa) is the second most common cancer in men worldwide according to data from the WHO GLOBOCAN database. Acinar adenocarcinoma represents 90–95% of prostate cancers diagnosticated, usually in patients who are asymptomatic, based on abnormalities in a screening prostate-specific antigen (PSA). MRI is considered the main imaging modality for the detection of the extent of primary tumor and local recurrence in PCa patients, but the clinical availability of integrated PET/MRI scanners has changed the diagnostic approach. The integration of anatomical and functional

imaging by using different radiopharmaceuticals, such as radiolabeled choline or prostate-specific antigen membrane (PSMA) can change the staging and restaging of these patients. Indeed, PET imaging has significantly impacted on different clinical settings, from initial staging to biochemical recurrence (BCR), and has also guided personalized treatment planning [53]. For initial staging, it has been reported that PET/MRI alters therapeutic strategies in at least 30% of patients [54]. This latter advantage is mainly due to higher diagnostic value of PET/MRI compared with multiparametric MRI (mpMRI)  [55, 56] or PET imaging alone  [57] for the detection of PCa. PET/MRI is more sensitive than mpMRI in identifying primary tumors in the peripheral zone of the prostate gland, and in revealing extracapsular extension and seminal vesicle infiltration. Moreover, especially if PSMA is used, PET/MRI shows high lesion contrast and an excellent consistency in lesion detection [58] because an intense uptake on PET combined to mpMRI anatomical changes strongly correlates with the dominant lesion in the prostate glands of men undergoing imaging before surgery [59]. Some authors documented that 68Ga-PSMA-11 PET/MRI provides valuable diagnostic information and improves patient selection for extended pelvic lymph node dissection by comparison with the currently used clinical nomograms [60–62]. It has also been reported that PET and PET/ MRI have a considerably lower proportion of equivocal results in distinguishing aggressive from indolent disease and in the detection of extracapsular extension compared to mpMRI alone [63]. Also, in restaging settings PET/MRI proved to have a high detection rate for recurrent disease, particularly when radiolabeled PSMA was used. Moreover, it is able to identify oligometastatic disease and guide to an appropriate radiation treatment (extension of the radiation field, extension of lymph node adenectomy, etc.) allowing up to 75% of change in the patients’ management. PET/MRI in cases of biochemical disease recurrence has been demonstrated to achieve a detection rate of up to 97% [64] and is more accurate than PET/CT in detecting local recurrences. This improves the detection rate for lower PSA levels, thanks to the MRI component, which is crucial in identifying local recurrences otherwise masked by the accumulation of radiopharmaceuticals in the bladder. Figure  29.6 shows an example of a patient with recurrent PCa detected by 18F-choline PET/MRI. The choice of MRI sequences has an important influence on the detection of local and distant metastases: metastases from prostate cancer are best detected by DWI or gadolinium-enhanced T1-weighted VIBE sequences and the most time-efficient sequence with the highest lesion detection rate is gadolinium-enhanced T1-weighted VIBE.

795

796

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 29.6 A 62-year-old male with prostate cancer (histopathological characteristics: Gleason score = 6, pT3apNx, and positive margins). Three years after the radical prostatectomy, a biochemical recurrence occurred (PSA value increased from 0.01 to 5.2 ng/ml). T2_tse sequence shows a 5 × 2 mm area of recurrence in prostatic fossa with a moderate 18F-choline uptake.

The detection of bone metastasis is increased by using multiple MR sequences, especially in the case of bone marrow involvement.

PET/MRI in Rectal Cancer Rectal cancer accounts for about one in three cases of colorectal cancer tumor, representing the most common neoplasm of the gastrointestinal tract. Current guidelines recommend MRI for local staging and contrast-enhanced thoraco-abdominal CT for detecting distant metastases [65]. PET/CT with 18F-FDG is recommended in patients with a locally advanced or recurrent rectal cancer as this can identify the presence of distant metastases [66]. However, PET/MRI has shown some advantages in rectal cancer, mainly in its ability to assess the soft tissue and the relationship between the tumor and the surrounding tissue. In the definition of a primary tumor, PET/MRI can reveal mesorectal fascia involvement and also distinguish

between recurrence and fibrosis post-radiotherapy better than PET/CT [67, 68]. Figure 29.7 shows an example of a patient with rectal cancer detected by 18F-FDG PET/MRI. For the definition of lymph node involvement, PET/MRI is able to identify the presence of small pathological lymph nodes more accurately than MRI alone thanks to the metabolic information provided by PET  [68]. Lee et  al.  [69] found that PET/MRI significantly improves the detection of distant metastases in rectal cancer, mainly liver metastases  [70]. However, it has limited use in the detection of lung lesions because of the low-quality sequences in this region. Moreover, PET/MRI can help in the identification of response to local therapies in the liver metastases.

PET/MRI in Multiple Myeloma Multiple myeloma (MM) is a neoplastic disease characterized by the presence of 10% or more of clonal bone marrow plasma cells or the presence of extramedullary sites.

Clinical Application of PET/MRI

Figure 29.7 Sagittal and coronal 18F-FDG PET showing linear and intense uptake in the rectal wall. The core-gistrated PET/MRI trans-axial image shows intense 18F-FDG uptake in the right lateral wall of the rectum. In the same area, DWI-MRI (below, first image on the right) and post-contrast MRI show an increased signal and increased enhancement, respectively, compatible with an invasive rectal cancer.

797

798

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Moreover, at least one of the following calcium, renal, anemia and bone (CRAB) criteria must be present: (i) elevated serum calcium level (1 mg/dL above normal or over 11; C), (ii) signs of renal failure (R), with serum creatinine above 2 mg/dL or renal clearance less than 40 mL/min, (iii) hemoglobin level less than 100 g/L or greater than 20 g/L below the lower limit of normal (anemia, A), and (iv) one or more osteolytic lesions at CT, RX or PET/CT (bone, B) [71]. There are different forms of MM, such as monoclonal gammopathy of undetermined significance (MGUS), smoldering myeloma, solitary plasmacytoma of bone or extramedullary, and Waldenstrom macroglobulinemia. The definition of disease depends on serum and urinary monoclonal protein levels, clonal bone marrow plasma cells level, and the number of osteolytic lesions. In recent decades CT, 18F-FDGPET/CT, and MRI have increased their value in the management of MM patients regarding diagnosis, follow-up, planning of therapy, and response to treatment. The International Myeloma Working Group (IMWG) recommend MRI as the gold standard in the diagnosis of MM since it allows (i) the evaluation of the bone marrow involvement that is usually defined in terms of normal, focal, diffuse, and variegated/salt-and-pepper pattern, (ii) the identification of painful lesions for spinal cord compression, and (iii) for the differentiation between benign and malignant vertebral fractures [72]. In general, a positive MRI for MM shows the presence of lesions with a loss of signal on T1-weighted images, an increased signal on T2-weighted images, and a delayed contrast enhancement [73]. 18 F-FDG PET allows both marrow involvement (sensitivity 59–85%, specificity 79–92%) and extraosseous disease to be identified; moreover, it can be useful in the detection of occult lesions [74]. MRI, as well as PET/CT, can be helpful in differentiating active myeloma from smoldering myeloma. According to the National Comprehensive Cancer Network (NCCN) guidelines, an active MM is diagnosed when there is at least one focal lesion on the MRI greater than 5 mm or one focal hypermetabolic osteolytic lesion on PET/CT along with clonal cells of at least 10% in the bone marrow. Shortt et al. [75] showed superior sensitivity and specificity of MRI compared to 18F-FDG PET/CT (68% vs. 59% and 83% vs. 75%, respectively) for staging but also reported specificity and positive predictive values equal to 100% when both modalities were concordant, suggesting a possible synergy between them. The pattern of marrow involvement in MRI in patients affected by MM represents an important prognostic value: an elevated number of focal lesions, as well as a diffuse involvement instead of variegated or normal are bad prognostic factors  [76]. Even the appearance of new lesions in MRI of asymptomatic patients plays an important role in the

prognosis because it leads to greater probability of disease progression [77]. Also, PET/CT has shown prognostic ability: it has been proven that metabolic parameters of at least three 18F-FDG avid lesions [78], a SUVmax of more than 4.2, the presence of extramedullary disease  [79], higher metabolic tumor volume [80], and liver, lung, and muscle involvement  [81] have a bad impact on prognosis and overall survival. Both MRI and PET/CT are also indicated in the evaluation of the response to treatment after chemotherapy or radiotherapy or stem cell transplantation, especially when conventional imaging with radiography or CT is inconclusive. In patients without response to therapy, MRI allows new lesions to be detected, while PET can show an increase in size and/or avidity or number of lesions. Conversely, a complete response to treatment is usually characterized by regaining MRI intensity in bone lesions and the disappearance of 18F-FDG uptake at PET [73]. The combined use of MRI and PET is not considered a routine examination yet. Few academic works have explored PET/MRI use for MM, and most studies and reviews have considered the modalities as separate. A recent study including 30 patients with MM has compared SUVs from PET/MRI with those from PET/CT. The findings showed that SUVs on PET/MRI were lower although nonetheless tightly correlated with PET/CT [82]. In most cases, evidence suggest the utility of both procedures, with little difference in specificity and sensitivity. Therefore, considering the abovementioned abilities of each modality, their combination can contribute to the assessment of lesion distribution and activity, monitoring the recurrence or progression of disease either from a less advanced plasma cell dyscrasia or in already existing MM, evaluating disease burden in the whole body, and predicting the response to therapy. Figure  29.8 shows an example of a patient with MM at 18F-FDG PET/MRI.

PET/MRI in Pediatric Lymphoma Pediatric lymphoma is the third most common childhood cancer, consisting of 15% of all pediatric malignancies [83]. In children, Hodgkin lymphomas (HLs) and the majority of the different histologic subtypes of non-Hodgkin lymphomas (NHLs) are frequent and may have both nodal and extranodal involvement. For LH, most cases have intrathoracic nodal involvement [84]. Evaluation of the extent of disease is essential for appropriate risk stratification and treatment planning. More specifically, the stage of disease at presentation helps to define the patient’s chemotherapy regimen and the potential extent of field for radiation therapy [83]. All pediatric lymphomas have a high metabolism showing a high 18F-FDG uptake, thus benefitting from imaging

Clinical Application of PET/MRI

Figure 29.8 A 58-year-old woman with MM. PET/MRI with 18F-FDG was performed for staging of disease. MRI showed hypointense multiple lesions at T1 sequences in the axial and appendicular skeleton, compatible with bone marrow disease. PET showed inhomogeneous 18F-FDG uptake in several vertebrae, especially in the lumbar and sacral spine, and mild 18F-FDG avidity at the level of multiple anterior and posterior costal arches. Moreover, a focal 18F-FDG hypermetabolism the liver was reported at PET, reflecting the presence of extramedullary involvement.

799

800

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

assessment with PET  [84], from the diagnosis to the response to therapy. The Lugano classification, a modification of the Ann Arbor classification, formally incorporates the use of 18F-FDG PET/CT for staging and treatment response. 18F-FDG-PET demonstrates greater specificity and accuracy compared with conventional staging methods, with 100% versus 60% and 96.7% versus 85.2%, respectively [84]. Moreover, the Deauville score is used to assess the response to therapy in the interim or at the end. An interim PET scan defined a higher likelihood of treatment failure and relapse in lymphoma, mainly in the HL  [84]. The clinical value of PET/MRI has been studied in pediatric conditions, especially when there is a need for patients to undergo multiple imaging scans, such as in oncology. PET/MRI is an attractive imaging modality because of the radiation dose savings, but its scan length often requires anesthetic procedures [85]. PET/MRI offers a reduced radiation dose compared to PET/CT of up to 73% [84]. PET/MRI protocols have demonstrated significantly greater sensitivity, negative predictive value, and diagnostic accuracy compared with whole-body MRI in pediatric patients with lymphoma (sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy were 96%, 96.5%, 97%, 95%, and 96% for unenhanced PET/MRI, 97%, 96.5%, 97%, 96.5%, and 97% for contrast-enhanced PET/MRI, 97%, 96.5%, 97%, 96.5%, and 97% for contrast-enhanced PET/MRI with DWI PET/MRI, and 77%, 96%, 78.5%, and 86% for MRI-DWI) [83]. Although numerous acquisition protocols are currently possible, most pediatric whole-body PET/MRI protocols are similar in their fundamental components. A whole-body examination usually includes whole-body PET and MR acquisitions, with typically MR-based attenuation correction sequences and fat-suppressed T2-weighted and/or inversion recovery sequences acquired at each bed position (usually 3–4minutes at each position). The most commonly used method for attenuation correction is segmentation based, with a double-echo chemical shift gradient-echo sequence (Dixon technique) that provides separate fat and water images that are then segmented into four tissue compartments: soft tissue, fat, lung, and background air. Atlas-based attenuation correction methods may be inaccurate for use in pediatric patients because they are based on adult atlases. Additionally, dedicated imaging of a specific body part may be performed based on the indication, including administration of gadolinium-based contrast where needed [84]. The combination of 18F-FDG PET and MRI is useful for the detection of nodal and extranodal disease, and is effective in evaluating the extent of bone marrow involvement [84]. Few studies comparing PET/CT and PET/MRI have been performed. Sher and colleagues performed 40 prospective sequential PET/CT and PET/MRI studies in

children with known diagnoses of lymphoma (Figure 29.9). Between these two modalities there was no statistically significant difference in lesion detection or diagnostic accuracy, and no differences in Ann Arbor clinical staging [83]. Although 18F-FDG-PET/MRI is a promising new modality, there is not yet enough data to support its routine use for staging or surveillance of children with lymphoma. PET/ MRI protocols are still under development, and the availability of this technique globally is limited. In addition to the cost of performing PET/MRI fused imaging, PET/MRI can take 60–90 minutes or longer to acquire compared with approximately less than 30 minutes with 18F-FDG-PET/CT. In children, this additional time may also require extended periods of anesthesia to limit motion degradation [83]. Although primary pulmonary lymphoma is rare in children, secondary lung involvement occurs in up to 10% of children with HL [83]. Imaging of lung parenchyma is difficult with MRI due to low proton density, rapid loss of signal because of field inhomogeneity, and respiration motion. For these reasons, as well as the inferior spatial resolution of MRI compared with CT, it is challenging to detect small pulmonary nodules with PET/MRI, and chest CT maintains an important role in thoracic staging of pediatric malignancies [83].

­ pplication of PET/MRI in A Nononcological Setting PET/MRI in Brain Dementia The number of patients with dementia is rapidly increasing worldwide and it is estimated it will double every 20 years. Alzheimer’s disease (AD) is the most common dementia in the elderly population, whereas dementia with Lewy bodies (DLB) and frontotemporal dementia (FTD) account for approximately 15–25% of cases. Regional hypometabolism on 18F-FDG PET and volume loss of specific brain regions on MRI have been described as “neuronal injury” biomarkers and used as an adjunct to clinical diagnosis for over two decades for classifying various types of dementia  [86]. Recently, the finding of Aβ accumulation playing a critical role in the pathogenesis of patients with early-onset AD has put forth a structural basis of dementia and has introduced amyloid PET scan for the differential diagnosis of dementia and as a tool that may guide the design and timing of therapeutic interventions aimed at modifying the course of this illness [87, 88]. Functional neuroimaging with diverse tracers along with the structural and functional correlates of MRI is the current strategy to improve the diagnostic accuracy for classifying dementias  [89] and predicting mild cognitive impairment

Clinical Application of PET/MRI

Figure 29.9 Staging of nodular sclerosis LH in a 16-year-old male. MRI T1-weighted turbo-spin-echo (TSE). TSE coronal image (a) shows lymphadenopathy localized in the left supraclavicular region, emergence of the epi-aortic vessels, and an aorto-pulmonary window with intense 18F-FDG uptake in PET with attenuation correction image (c). In the fused PET/MRI image (b) the morphological aspect corresponds to the intense uptake of image (c).

(MCI) to AD conversion [90]. With the advent of integrated PET/MRI in the clinical setting, it is expected that morphologic as well as molecular information regarding neurodegenerative disorders will be obtainable in a single study with the highest attainable accuracy. The combination of PET and MRI information when acquired at the same time offers several potential advantages in neurosciences: (i) systematic addition of high-resolution MRI information to PET data provides accurate and consistent information on underlying structures with better anatomic localization and improved scan interpretation, (ii) anatomical abnormalities such as arachnoidal cysts or old hemorrhage, (iii) provides an opportunity to perform atrophy correction and partial volume effect correction of the PET data for both “cold spot” (18FFDG) and “hot spot” (amyloid PET) imaging using consistent simultaneous MRI information and mathematical algorithms  [91], (iv) detection and exclusion of nonneurodegenerative pathologies coexistent with neurodegenerative conditions using MRI information, (v) aids improved follow-up/treatment response evaluation by better quantification of PET tracer uptake using better attenuation correction sequences such as ultrashort time echo (UTE), and (vi)

reduced CT dose exposures are expected if PET/MRI becomes the routine investigational modality as most of these patients would require repeated follow-up evaluations, especially with the advent of newer treatment strategies. MRI can evaluate also morphological features better than CT, which could improve global evaluation. For example, by using an atrophy index, such as (i) the Fazekas scale for white matter lesions, in which the deep white matter component is used to assess the amount of chronic small vessel ischemic change, (ii) the posterior atrophy score of parietal atrophy (PA or PCA or Koedam score), which is useful in atypical (posterior cortical atrophy) or early-onset Alzheimer disease, (iii) the medial temporal lobe atrophy score (MTA score), or (iv) the global cortical atrophy scale (GCA scale), we can improve the diagnosis of the dementia. Figures  29.10 and 29.11 show two cases of dementia diagnosed by PET/MRI with 18F-FDG. Therefore, a composite report with PET data and MRI can be useful for the correct interpretation of imaging and can be useful to improve the positive and negative predictive values for the diagnosis of dementia, mainly with the use of 18F-FDG. Figure 29.10 shows an example of this.

801

802

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 29.10 Top, low uptake of 18F-FDG was observed in the occipital region, posterior parietal region, and precuneus, bilaterally. However, the posterior region of the cingulum was completely spared. Bottom, MRI shows a slight hypotropia in the same regions. The final diagnosis was compatible with Lewy body dementia.

PET/MRI in Retroperitoneal Fibrosis Retroperitoneal fibrosis (RPF) is a rare disease characterized by the presence of retroperitoneal proliferation of fibro-inflammatory tissue that surrounds retroperitoneal organs such vessels and ureters. The fibrous mass usually develops between the origin of the renal arteries and the pelvis. In most cases the disease is idiopathic, but it can be secondary to certain drugs, trauma, surgery, malignant tumors, infection, and abdominal aortic aneurysm. The most common symptoms are abdominal or back pain, fatigue, nausea, anorexia, weight loss, myalgias, lower extremity oedema, and deep vein thrombosis. In severe cases it may cause ureteral obstruction and renal failure. Medical treatment is mainly based on the use of corticosteroids, other options are immunosuppressant and tamoxifen, while surgery is undertaken to treat complications such as ureteral obstruction. Imaging has an important role in diagnosis, but a histological examination of the retroperitoneal mass is sometimes

required. CT and MRI are the modalities of choice for evaluating the extent of the process  [92]. With MRI, idiopathic RPF appears as a fibrotic plaque, hypointense on T1weighted images, while the intensity on T2-weighted images varies depending on the stage: in the early stage (active inflammatory stage) of disease the fibrous mass has high signal intensity due to acute oedema and hypercellularity, whereas in the nonacute stages (inactive fibrotic stage) a low signal intensity is reported due to mature fibrosis [93]. MRI is a well-established imaging modality for assessing treatment response. Different studies, e.g. by Ruhlmann et al. [94], have shown that patients with an active RPF have higher values for DWI and contrast enhancement compared to treated patients. DWI is helpful for functional assessment of acute inflammatory tissue, while contrast enhancement is useful for differentiating between acute RPF and potential vessel stenosis caused by disease [94]. In the evaluation of RPF, 18F-FDG PET provides additional metabolic information, such as index of severity and

Clinical Application of PET/MRI

Figure 29.11 Top, 18F-FDG PET shows a hypometabolism in the parieto-temporal cortex, bilaterally. Moreover, hypometabolism was found in the precuneum and the posterior region of the cingulum. Neurodegenerative indexs were GCA score = 1; Koedam score left = 1, right = 1; MTA score left = 0, right = 1; Fazekas score 0–1. Bottom, MRI images are negative. PET/MRI images are compatible with AD.

extent of disease. Furthermore, PET plays a useful role in assessing the response to therapy and identifying recurrent disease, especially in patients with normal acute-phase reactants and a stable residual mass on repeat CT scanning  [95–97]. The evaluation of response to treatment is often made by assessing erythrocyte sedimentation rate (ESR) and serum C-reactive protein (CRP) levels. Moreover, sonography can be used for monitoring ureteral obstruction, while CT and MRI are used for evaluating changes in the size of the fibrous mass [92]. However, it is demonstrated that acute-phase proteins are only poor predictors of therapy response. According to Magrey et  al. there is no significant difference between the baseline CRP level and

ESR between patients whose mass decreased in size during treatment and those in whom the mass did not decrease [98]. Ruhlmann et al. showed comparable mean SUVs derived from PET/MRI and PET/CT, as well as significant differences in the SUVs between the untreated and treated patients: untreated patients presented SUVmax and visual 18 F-FDG uptake score (the score was assessed in comparison with liver uptake) higher than treated patients. Moreover, they proved a potentially higher diagnostic value of PET over MRI parameters since PET data (i.e. 18F-FDG uptake score and SUVmax) showed a higher impact than MRI data (i.e. DWI, T2 signal intensity, and ADC values) for discriminating between the untreated and treated RPF [3].

803

804

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figures 29.12 and 29.13 show baseline and post-therapy PET/MRI scans in a patient with RPF. In conclusion, 18F-FDG PET combined with MRI can contribute to evaluating the activity of disease, assessing complications such as ureteral obstruction or vessel stenoses, monitoring therapy response, and demonstrating inflammatory relapse.

PET/MRI in Vasculitis Large-vessel vasculitis is defined as vasculitis affecting large arteries (the aorta and its main branches) and can lead to stenosis and aneurysms. The two main forms are giant cell arteritis and Takayasu’s arteritis, and their diagnosis can be challenging. In particular, in giant cell arteritis some patients can have symptoms but a negative temporal artery biopsy (the primary modality for establishing this diagnosis), while in Takayasu’s arteritis distinguishing fibrotic stenosis from active arterial lesions could be hard [99, 100]. Several studies have shown good sensitivity 18F-FDG PET/CT for the diagnosis of arterial involvement in largevessel vasculitis (around 89% for giant cell arteritis and 87% for Takayasu’s arteritis)  [101–105], but very few studies have focused on 18F-FDG PET/MRI. The advantage in using this latter modality of imaging is the ability of 18FFDG PET to define the presence of an active inflammation in the vessel’s wall and the opportunity to assess the thickening of the vessel wall by MRI. MRI images, using T1-weighted VIBE pre- and postcontrast, are important to evaluate thickening and enhancement of the aortic wall on delayed enhancement images and for luminal narrowing and dilation. Aortic wall thickening was considered pathologic with a wall thickness ≥2 mm [106, 107]. The visual interpretation of imaging by 18 F-FDG PET, co-registered with either CT or MRI, considers grade 3 positive for active large-vessel vasculitis [108] as follows: 0  =  no uptake ( mediastinum), 1  =  low-grade uptake (< liver), 2  =  intermediate-grade uptake (= liver), and 3 = high-grade uptake (> liver). Furthermore, based on PET and MRI images, three patterns can be defined: ●

●

●

inflammatory: positive PET (grade 3) and abnormal MRI (stenosis/wall thickening) fibrous: negative PET (grade 0–2) and abnormal MRI (stenosis/wall thickening) normal: both negative PET and MRI.

The inflammatory pattern suggests the presence of an inflammatory process, and the CRP level appeared to be correlated with SUVs, whereas the fibrous pattern suggests fibrotic lesions without “macroscopic” inflammation [99]. The inflammatory pattern is found in more than 70% of scans from patients with active disease versus none with

remission, whereas a normal or fibrous pattern is found in 27% of patients with active disease versus approximately 50% of patients in remission [99]. The inflammatory pattern seems highly associated with disease activity in Takayasu’s arteritis, and less so with giant cells arteritis (about half of patients). This could be explained by the prevalence of aortitis in each disease [99]. Compared with PET/CT, PET/MRI can reduce the radiation dose to patients, allowing repeated use to evaluate disease activity after treatment and during follow-up. Moreover, the simultaneous acquisition of PET and MRI offers a better analysis with gadolinium uptake and wall edema, and a good co-localization of anatomic structures and biological processes. MRI data would also be helpful in older patients, who often have atherosclerosis, because atherosclerotic plaques can accumulate 18F-FDG, leading to false-positive results (Figure 29.14) [99, 109]. The use of digital PET detectors and integrated PET/MRI analysis can increase sensitivity for the detection of vasculitis, making the use of invasive tests such as temporal artery biopsy not necessary (the estimated false-negative rate ranges from 6% to 17%, especially if biopsy is obtained in an arteritis-free segment) [110–114].

PET/MRI in Cardiac Sarcoidosis Sarcoidosis is a systemic inflammatory disease with an unclear etiology. It is pathologically characterized by noncaseating granulomas in multiple organs; lungs and lymph nodes are the most frequently interested sites of disease, while cardiac involvement is associated with an increased mortality rate [115, 116]. Clinical manifestations of cardiac sarcoidosis are not relative to the stage of disease, thus an early diagnosis and starting therapy with immunosuppressive agents are required to avoid complications such as conduction abnormalities, ventricular arrhythmias, and left ventricular dysfunction [117]. Endomyocardial biopsy is the gold standard for the diagnosis but it has low sensitivity, therefore an alternative diagnostic algorithm can be used, including a combination of nuclear imaging and radiological tests. 18F-FDG PET can be used to image the inflammation and cardiac magnetic resonance (CMR) with late gadolinium enhancement (LGE) can detect scar tissue  [115, 118–120]. 18 F-FDG is a glucose analogue that accumulates in metabolically active cells and is accumulated by inflammatory cells of the granulomas in the myocardium. Focal myocardial 18FFDG uptake has high sensitivity for active cardiac sarcoidosis, and can provide prognostic stratification and predict the risk of ventricular arrhythmia and death [120–122]. However, inadequate preparation of the patient can lead to falsepositive results, thus reducing its specificity. Indeed, PET with 18F-FDG should be performed after dietary

Clinical Application of PET/MRI

Figure 29.12 A 64-year-old man with retroperitoneal fibrosis and abdominal aortic aneurysm. PET shows an intense hypermetabolism (SUVmax 11) at the level of a fibrous mass which extends between the tract below the renal arteries and the iliac bifurcations. MRI shows a low signal intensity on T2-weighted images and a diffusion restriction (ADC 0.57–0.7).

805

806

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Figure 29.13 After medical treatment with steroids, MRI shows partial reduction of the thickness in wall mass with a significant reduction of 18F-FDG uptake at PET.

modifications such as a high-fat, high-protein, and lowcarbohydrate diet for at least 12–24 hours before the scan [123, 124]. Cardiac MRI has high sensitivity in identification of myocardial fibrosis, and can give an accurate assessment

of myocardial function. Conversely, it cannot distinguish between active or quiescent cardiac sarcoidosis. The combination of PET with MRI can overcome the limitations of each single imaging modality [125–130]. To

Clinical Application of PET/MRI

(a)

(b)

(c)

Figure 29.14 PET/MRI of a 66-year-old woman with a large vessel vasculitis showing an inflammatory pattern involving aorta and left subclavian artery: T1-weighted VIBE exhibits increased wall thickness (a), PET shows uptake higher than liver (c), fused PET/MRI (b).

(a)

(a)

(b)

(b)

(c)

Figure 29.15 PET/MRI of a 49-year-old man with cardiac sarcoidosis. Top, the MRI phase-sensitive inversion recovery sequence shows late gadolinium enhancement in the left ventricle wall (a) and PET confirms two areas of focal 18F-FDG uptake (b). Bottom, MRI PSIR sequence (a) and T2-weighted STIR (b) show LGE in the interventricular wall and PET confirms focal 18F-FDG uptake in this area (c).

807

808

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

date, few papers are available about the role of hybrid PET/ MRI for the diagnosis of cardiac sarcoidosis. Some of them show that PET/MR has some advantages compared to PET/CT in patients with confirmed or suspected cardiac sarcoidosis, for many reasons. First, PET/MRI has a higher diagnostic performance than PET/CT in identifying positive patients (90% vs. 60%, respectively). Second, PET/MRI exposes patients to a lower radiation dose. Third, PET/MRI has an improved image quality due to more advanced PET detectors, longer duration of acquisition, delayed acquisition, which guarantees a better blood clearance, and perfect co-registration of systems [115, 131, 132]. In patients with an active sarcoidosis, 18F-FDG PET/MRI is considered positive where there is an association between

focal tracer uptake and LGE in the ventricle wall. LGE has emerged as the dominant CMR sequence for the evaluation of cardiac sarcoidosis because it is associated with adverse events and cardiac death  [133]. Therefore, PET/ MRI can also provide prognostic information in cardiac sarcoidosis by combining metabolic and functional information [115, 123]. Based on available studies, PET/MRI is a valuable tool for the assessment of response to treatment. In case of glucocorticosteroid therapy, a reduction in 18F-FDG uptake and the persistence of LGE are considered the resolution of active inflammation [134, 135]. Figure 29.15 shows a confined cardiac sarcoidosis detected at PET/MRI with 18 F-FDG.

­References 1 Scott, J.N., Brasher, P.M.A., Sevick, R.J. et al. (2002). How often are nonenhancing supratentorial gliomas malignant? A population study. Neurology 59: 947–949. 2 Kruser, T.J., Mehta, M.P., and Robins, H.I. (2013). Pseudoprogression after glioma therapy: a comprehensive review. Expert Rev. Neurother. 13: 389–403. 3 Galldiks, N., Kracht, L.W., Dunkl, V. et al. (2011). Imaging of non- or very subtle contrast-enhancing malignant gliomas with [11C]-methionine positron emission tomography. Mol. Imaging 10: 453–459. 4 Kwock, L., Smith, J.K., Castillo, M. et al. (2002). Clinical applications of proton MR spectroscopy in oncology. Technol. Cancer Res. Treat. 1: 17–28. 5 Kono, K., Inoue, Y., Nakayama, K. et al. (2001). The role of diffusion-weighted imaging in patients with brain tumors. Am. J. Neuroradiol. 22: 1081–1088. 6 Ulmer, J.L., Salvan, C.V., Mueller, W.M. et al. (2004). The role of diffusion tensor imaging in establishing the proximity of tumor borders to functional brain systems: implications for preoperative risk assessments and postoperative outcomes. Technol. Cancer Res. Treat. 3: 567–576. 7 Bizzi, A., Blasi, V., Falini, A. et al. (2008). Presurgical functional MR imaging of language and motor functions: validation with intraoperative electrocortical mapping. Radiology 248: 579–589. 8 Bagadia, A., Purandare, H., Misra, B.K. et al. (2011). Application of magnetic resonance tractography in the perioperative planning of patients with eloquent region intra-axial brain lesions. J. Clin. Neurosci. 18: 633–639. 9 Fink, J.R., Muzi, M., Peck, M. et al. (2015). Continuing education: multi-modality brain tumor imaging – MRI, PET, and PET/MR. Department of Radiology, University of Washington, Seattle, WA, USA. J. Nucl. Med. 56: 1554–1561.

10 Ferda, J., Ferdová, E., Hes, O. et al. (2017). PET/MRI: multiparametric imaging of brain tumors. Eur. J. Radiol. 94: A14–A25. 11 Zukotynski, K.A., Vajapeyam, S., Fahey, F.H. et al. (2017). Correlation of 18F-FDG PET and MRI apparent diffusion coefficient histogram metrics with survival in diffuse intrinsic pontine glioma: a report from the pediatric brain tumor consortium. J. Nucl. Med. 58: 1264–1269. 12 Nowosielski, M., DiFranco, M.D., Putzer, D. et al. (2014). An intra-individual comparison of MRI, [18F]-FET and [18F]-FLT PET in patients with high-grade gliomas. PLoS One 9 (4): e95830. 13 Filss, C.P., Cicone, F., Shah, N.J. et al. (2017). Amino acid PET and MR perfusion imaging in brain tumours. Clin. Transl. Imaging 5: 209–223. 14 Galldiks, N., Rapp, M., Stoffels, G. et al. (2013). Response assessment of bevacizumab in patients with recurrent malignant glioma using [18F] fluoroethyl-L-tyrosine PET in comparison to MRI. Eur. J. Nucl. Med. Mol. Imaging 40: 22–33. 15 Jemal, A., Siegel, R., Xu, J., and Ward, E. (2010). Cancer statistics, 2010. CA Cancer J. Clin. 60: 277–300. 16 Sekine, T., de Galiza, B.F., Kuhn, F.P. et al. (2017). PET+MR versus PET/CT in the initial staging of head and neck cancer, using a trimodality PET/ CT+MR system. Clin. Imaging 42: 232–239. 17 Cheng, Y., Bai, L., Shang, J. et al. (2020). Preliminary clinical results for PET/MR compared with PET/CT in patients with nasopharyngeal carcinoma. Oncol. Rep. 43: 177–187. 18 Schaarschmidt, B.M., Heusch, P., Buchbender, C. et al. (2016). Locoregional tumour evaluation of squamous cell carcinoma in the head and neck area: a comparison between MRI, PET/CT and integrated PET/MRI. Eur. J. Nucl. Med. Mol. Imaging 43: 92–102.

Clinical Application of PET/MRI

19 Partovi, S., Kohan, A., Vercher-Conejero, J.L. et al. (2014). Qualitative and quantitative performance of 18F-FDGPET/MRI versus 18F-FDG-PET/CT in patients with head and neck cancer. AJNR Am. J. Neuroradiol. 35: 1970–1975. 20 Sekine, T., Barbosa, F.G., Sah, B.R. et al. (2017). PET/MR outperforms PET/CT in suspected occult tumors. Clin. Nucl. Med. 42: e88–e95. 21 Ruhlmann, V., Ruhlmann, M., Bellendorf, A. et al. (2016). Hybrid imaging for detection of carcinoma of unknown primary: a preliminary comparison trial of whole-body PET/MRI versus PET/CT. Eur. J. Radiol. 85: 1941–1947. 22 Tang, A., Hallouch, O., Chernyak, V. et al. (2018). Epidemiology of hepatocellular carcinoma: target population for surveillance and diagnosis. Abdom. Radiol. (NY) 43: 13–25. 23 Crissien, A.M. and Frenette, C. (2014). Current management of hepatocellular carcinoma. Gastroenterol. Hepatol. (NY) 10: 153–161. 24 European Association for The Study of The Liver, European Organisation for Research and Treatment of Cancer (2012). EASL-EORTC clinical practice guidelines: management of hepatocellular carcinoma. J. Hepatol. 56: 908–943. 25 Elsayes, K.M., Hooker, J.C., Agrons, M.M. et al. (2017). 2017 version of LI-RADS for CT and MR imaging: an update. Radiographics 37: 1994–2017. 26 Arif-Tiwari, H., Kalb, B., Chundru, S. et al. (2014). MRI of hepatocellular carcinoma: an update of current practices. Diagn. Interv. Radiol. 20: 209–221. 27 Li, Y.C., Yang, C.S., Zhou, W.L. et al. (2018). Low glucose metabolism in hepatocellular carcinoma with GPC3 expression. World J. Gastroenterol. 24: 494–503. 28 Bernstine, H., Braun, M., Yefremov, N. et al. (2011). FDG PET/CT early dynamic blood flow and late standardized uptake value determination in hepatocellular carcinoma. Radiology 260: 503–510. 29 Klasser, G.D., Echandi, L., and Shannon, M. (2014). Hepatocellular carcinoma metastasis to the condyle: a case report and review of the literature. J. Am. Dent. Assoc. 145: 1063–1067. 30 Cho, Y., Lee, D.H., Bin, L.Y. et al. (2014). Does 18F-FDG positron emission tomography computed tomography have a role in initial staging of hepatocellular carcinoma? PLoS One 9 (8): e105679. 31 Jadvar, H. (2012). Hepatocellular carcinoma and gastroenteropancreatic neuroendocrine tumors: potential role of other positron emission tomography radiotracers. Semin. Nucl. Med. 42: 247–254. 32 Lin, C.-Y., Chen, J.-H., Liang, J.-A. et al. (2012). 18F-FDG PET or PET/CT for detecting extrahepatic metastases or recurrent hepatocellular carcinoma: a systematic review and meta-analysis. Eur. J. Radiol. 81: 2417–2422.

33 Lee, J.W., Paeng, J.C., Kang, K.W. et al. (2009). Prediction of tumor recurrence by 18F-FDG PET in liver transplantation for hepatocellular carcinoma. J. Nucl. Med. 50: 682–687. 34 Cheng, G. and Huang, H. (2018). Prognostic value of 18 F-fluorodeoxyglucose PET/computed tomography in non small-cell lung cancer. PET Clin. 13: 59–72. 35 Çelebi, F., Yaghouti, K., Cindil, E. et al. (2021). The role of 18F-FDG PET/MRI in the assessment of primary intrahepatic neoplasms. Acad. Radiol. 28: 189–198. 36 Hectors, S.J., Wagner, M., Besa, C. et al. (2018). Multiparametric FDG-PET/MRI of hepatocellular carcinoma: initial experience. Contrast Media Mol. Imaging 2018: 5638283. 37 Fowler, K.J., Maughan, N.M., Laforest, R. et al. (2016). PET/MRI of hepatic 90Y microsphere deposition determines individual tumor response. Cardiovasc. Intervent. Radiol. 39: 855–864. 38 Kong, E., Chun, K.A., and Cho, I.H. (2017). Quantitative assessment of simultaneous F-18 FDG PET/MRI in patients with various types of hepatic tumors: correlation between glucose metabolism and apparent diffusion coefficient. PLoS One 12: e0180184. 39 Vincent, A., Herman, J., Schulick, R. et al. (2011). Pancreatic cancer. Lancet 378: 607–620. 40 Kim, R., Prithviraj, G., Kothari, N. et al. (2015). PET/CT fusion scan prevents futile laparotomy in early stage pancreatic cancer. Clin. Nucl. Med. 40: 501–505. 41 Schellenberg, D., Quon, A., Minn, A.Y. et al. (2010). 18Fluorodeoxyglucose PET is prognostic of progressionfree and overall survival in locally advanced pancreas cancer treated with stereotactic radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 77: 1420–1425. 42 Donati, O.F., Hany, T.F., Reiner, C.S. et al. (2010). Value of retrospective fusion of PET and MR images in detection of hepatic metastases: comparison with 18 F-FDG PET/CT and Gd-EOB-DTPA-enhanced MRI. J. Nucl. Med. 51: 692–699. 43 Pagliari, D., Saviano, A., and Serricchio, M.L. (2017). Update in the assessment and management of intraductal papillary mucinous neoplasms of the pancreas. Eur. Rev. Med. Pharmacol. Sci. 21: 2858–2874. 44 Baiocchi, G.L., Bretagna, F., Gheza, F. et al. (2012). Searching for indicators of malignancy in pancreatic IPMN; the value of (18) FDG-PET confirmed. Ann. Surg. Oncol. 19: 3574–3580. 45 Roch, A., Barron, M., Tann, M. et al. (2015). Does PET with CT have clinical utility in the management of patients with intraductal papillary mucinous neoplasm? J. Am. Coll. Surg. 221: 48–56. 46 Pedrazzoli, S., Sperti, C., Pasquali, C. et al. (2011). Comparison of international consensus guidelines versus 18-FDG PET in detecting malignancy of intraductal

809

810

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

47

48

49

50

51

52

53

54

55

56

57

58

papillary mucinous neoplasm of the pancreas. Ann. Surg. 254: 171–176. Sperti, C., Bissoli, S., Pasquali, C. et al. (2007). 18-fluorodeoxyglucose positron emission tomography enhances computed tomography diagnosis of malignant intraductal papillary mucinous neoplasms of the pancreas. Ann. Surg. 246: 932–937. Serafini, S., Sperti, C., Brazzale, A.R. et al. (2020). The role of positron emission tomography in clinical management of intraductal papillary mucinous neoplasms of the pancreas. Cancers (Basel) 27: 807. Bosman, F.T., Carneiro, F., Hruban, R.H. et al. (2018). WHO classification of tumours of the digestive system. Mod. Pathol. 31: 1770–1786. Geijer, H. and Breimer, L.H. (2013). Somatostatin receptor PET/CT in neuroendocrine tumours: update on systematic review and meta-analysis. Eur. J. Nucl. Med. Mol. Imaging 40: 1770–1780. Skoura, E., Michopoulou, S., Mohmaduvesh, M. et al. (2016). The impact of 68Ga-DOTATATE PET/CT imaging on management of patients with neuroendocrine tumors: experience from a national referral center in the United Kingdom. J. Nucl. Med. 57: 34–40. Evangelista, L., Ravelli, I., Bignotto, A. et al. (2020). Ga-68 DOTA-peptides and F-18 FDG PET/CT in patients with neuroendocrine tumor: a review. Clin. Imaging 67: 113–116. Lecouvet, F.E., Oprea-Lager, D.E., Liu, Y. et al. (2018). Use of modern imaging methods to facilitate trials of metastasis-directed therapy for oligometastatic disease in prostate cancer: a consensus recommendation from the EORTC imaging group. Lancet Oncol. 19: e534–e545. Grubmuller, B., Baltzer, P., Hartenbach, S. et al. (2018). PSMA ligand PET/MRI for primary prostate cancer: staging performance and clinical impact. Clin. Cancer Res. 24: 6300–6307. Wetter, A., Lipponer, C., Nensa, F. et al. (2014). Evaluation of the PET component of simultaneous [(18) F]choline PET/MRI in prostate cancer: comparison with [(18)F]choline PET/CT. Eur. J. Nucl. Med. Mol. Imaging 41: 79–88. Kim, Y.I., Cheon, G.J., Paeng, J.C. et al. (2015). Usefulness of MRI-assisted metabolic volumetric parameters provided by simultaneous (18)F-fluorocholine PET/MRI for primary prostate cancer characterization. Eur. J. Nucl. Med. Mol. Imaging 42: 1247–1256. Eiber, M., Weirich, G., Holzapfel, K. et al. (2016). Simultaneous (68)Ga-PSMA HBED-CC PET/MRI improves the localization of primary prostate cancer. Eur. Urol. 70: 829–836. Domachevsky, L., Bernstine, H., Goldberg, N. et al. (2017). Early (68)GA-PSMA PET/MRI acquisition: assessment of lesion detectability and PET metrics in

59

60

61

62

63

64

65

66

67

68

69

patients with prostate cancer undergoing same-day late PET/CT. Clin. Radiol. 72: 944–950. Bauman, G., Martin, P., Thiessen, J.D. et al. (2018). [(18) F]-DCFPyL positron emission tomography/magnetic resonance imaging for localization of dominant intraprostatic foci: first experience. Eur. Urol. Focus 4: 702–706. Park, S.Y., Zacharias, C., Harrison, C. et al. (2018). Gallium 68 PSMA-11 PET/MR imaging in patients with intermediateor high-risk prostate cancer. Radiology 288: 495–505. Thalgott, M., Duwel, C., Rauscher, I. et al. (2018). One-stop-shop whole-body (68)Ga-PSMA-11 PET/MRI compared with clinical nomograms for preoperative T and N staging of high-risk prostate cancer. J. Nucl. Med. 59: 1850–1856. Ferraro, D.A., Muehlematter, U.J., Garcia Schuler, H.I. et al. (2020). (68)Ga-PSMA-11 PET has the potential to improve patient selection for extended pelvic lymph node dissection in intermediate to high-risk prostate cancer. Eur. J. Nucl. Med. Mol. Imaging 47: 147–159. Al-Bayati, M., Grueneisen, J., Lutje, S. et al. (2018). Integrated 68gallium labelled prostate specific membrane antigen-11 positron emission tomography/ magnetic resonance imaging enhances discriminatory power of multi-parametric prostate magnetic resonance imaging. Urol. Int. 100: 164–171. Abufaraj, M., Grubmuller, B., Zeitlinger, M. et al. (2019). Prospective evaluation of the performance of [(68)Ga] Ga-PSMA-11 PET/CT(MRI) for lymph node staging in patients undergoing superextended salvage lymph node dissection after radical prostatectomy. Eur. J. Nucl. Med. Mol. Imaging 46: 2169–2177. Benson, A.B., Venook, A.P., Al-Hawary, M.M. et al. (2018). Rectal cancer, version 2.2018, NCCN clinical practice guidelines in oncology. J. Natl. Compr. Canc. Netw. 16: 874–901. Patel, S., McCall, M., Ohinmaa, A. et al. (2011). Positron emission tomography/computed tomographic scans compared to computed tomographic scans for detecting colorectal liver metastases: a systematic review. Ann. Surg. 253: 666–671. Crimì, F., Spolverato, G., Lacognata, C. et al. (2020). 18F-FDG PET/MRI for rectal cancer TNM restaging after preoperative chemoradiotherapy: initial experience. Dis. Colon Rectum 63: 310–318. Paspulati, R.M., Partovi, S., Herrmann, K.A. et al. (2015). Comparison of hybrid FDG PET/MRI compared with PET/CT in colorectal cancer staging and restaging: a pilot study. Abdom. Imaging 40: 1415–14125. Lee, S.J., Seo, H.J., Kang, K.W. et al. (2015). Clinical performance of whole-body 18F-FDG PET/Dixon-VIBE, T1-weighted, and T2-weighted MRI protocol in colorectal cancer. Clin. Nucl. Med. 40: e392–e398.

Clinical Application of PET/MRI

70 Brendle, C., Schwenzer, N.F., Rempp, H. et al. (2016). Assessment of metastatic colorectal cancer with hybrid imaging: comparison of reading performance using different combinations of anatomical and functional imaging techniques in PET/MRI and PET/CT in a short case series. Eur. J. Nucl. Med. Mol. Imaging 43: 123–132. 71 Rajkumar, S.V., Dimopoulos, M.A., Palumbo, A. et al. (2014). International Myeloma Working Group updated criteria for the diagnosis of multiple myeloma. Lancet Oncol. 15: e538–e548. 72 Dimopoulos, M.A., Hillengass, J., Usmani, S. et al. (2015). Role of magnetic resonance imaging in the management of patients with multiple myeloma: a consensus statement. J. Clin. Oncol. 33: 657–664. 73 Shah, S.N. and Oldan, J.D. (2017). PET/MR imaging of multiple myeloma. Magn. Reson. Imaging Clin. N. Am. 25: 351–365. 74 Hall, M.N., Jagannathan, J.P., Ramaiya, N.H. et al. (2010). Imaging of extraosseous myeloma: CT, PET/CT, and MRI features. AJR Am. J. Roentgenol. 195: 1057–1065. 75 Shortt, C.P., Gleeson, T.G., Breen, K.A. et al. (2009). Whole-body MRI versus PET in assessment of multiple myeloma disease activity. AJR Am. J. Roentgenol. 192 (4): 980–986. 76 Mai, E.K., Hielscher, T., Kloth, J.K. et al. (2016). Association between magnetic resonance imaging patterns and baseline disease features in multiple myeloma: analyzing surrogates of tumour mass and biology. Eur. Radiol. 26: 3939–3948. 77 Derlin, T. and Bannas, P. (2014). Imaging of multiple myeloma: current concepts. World J. Orthop. 5: 272–282. 78 Fonti, R., Pace, L., Cerchione, C. et al. (2015). 18F-FDG PET/CT, 99mTc-MIBI, and MRI in the prediction of outcome of patients with multiple myeloma: a comparative study. Clin. Nucl. Med. 40: 303–308. 79 Zamagni, E., Patriarca, F., Nanni, C. et al. (2011). Prognostic relevance of 18-F FDG PET/CT in newly diagnosed multiple myeloma patients treated with up-front autologous transplantation. Blood 118: 5989–5995. 80 Fonti, R., Larobina, M., Del Vecchio, S. et al. (2012). Metabolic tumor volume assessed by 18F-FDG PET/CT for the prediction of outcome in patients with multiple myeloma. J. Nucl. Med. 53: 1829–1835. 81 Tirumani, S.H., Sakellis, C., Jacene, H. et al. (2016). Role of FDG-PET/CT in extramedullary multiple myeloma: correlation of FDG-PET/CT findings with clinical outcome. Clin. Nucl. Med. 41: e7–e13. 82 Sachpekidis, C., Hillengass, J., Goldschmidt, H. et al. (2015). Comparison of (18)F-FDG PET/CT and PET/MRI in patients with multiple myeloma. Am. J. Nucl. Med. Mol. Imaging 5: 469–478.

83 Gillman, J., States, L.J., and Servaes, S. (2020). PET in pediatric lymphoma. PET Clin. 15: 299–307. 84 Kwatra, N.S., Lim, R., Gee, M.S. et al. (2019). PET/MR imaging current updates on pediatric applications. Magn. Reson. Imaging Clin. N. Am.: 387–407. 85 Saade-Lemus, S., Teixeira, S.R., Vossough, A. et al. (2020). Emerging roles of PET/MR in the pediatric hospital. PET Clin. 15: 253–269. 86 Tartaglia, M.C., Rosen, H.J., and Miller, B.L. (2011). Neuroimaging in dementia. Neurotherapeutics 8: 82–92. 87 Villemagne, V.L., Burnham, S., Bourgeat, P. et al. (2013). Australian Imaging Biomarkers and Lifestyle (AIBL) Research Group. Amyloid β deposition, neurodegeneration, and cognitive decline in sporadic Alzheimer’s disease: a prospective cohort study. Lancet Neurol. 12: 357–367. 88 Vercher-Conejero, J.L., Rubbert, C., Kohan, A.A. et al. (2014). Amyloid PET/MRI in the differential diagnosis of dementia. Clin. Nucl. Med. 39: e336. 89 Dukart, J., Mueller, K., Horstmann, A. et al. (2011). Combined evaluation of FDG-PET and MRI improves detection and differentiation of dementia. PLoS One 6: e18111. 90 Teipel, S.J., Kurth, J., Krause, B. et al. (2015). Alzheimer’s disease neuroimaging initiative. The relative importance of imaging markers for the prediction of Alzheimer’s disease dementia in mild cognitive impairment – beyond classical regression. Neuroimage Clin. 8: 583–593. 91 Yanase, D., Matsunari, I., Yajima, K. et al. (2005). Brain FDG PET study of normal aging in Japanese: effect of atrophy correction. Eur. J. Nucl. Med. Mol. Imaging 32: 794–805. 92 Vaglio, A., Salvarani, C., and Buzio, C. (2006). Retroperitoneal fibrosis. Lancet 367: 241–251. 93 Vivas, I., Nicolás, A.I., Velázquez, P. et al. (2000). Retroperitoneal fibrosis: typical and atypical manifestations. Br. J. Radiol. 73: 214–222. 94 Ruhlmann, V., Poeppel, T.D., Brandt, A.S. et al. (2016). 18 F-FDG PET/MRI evaluation of retroperitoneal fibrosis: a simultaneous multiparametric approach for diagnosing active disease. Eur. J. Nucl. Med. Mol. Imaging 43: 1646–1652. 95 Washino, S., Hirai, M., Matsuzaki, A. et al. (2010). (18) F-fluorodeoxyglucose positron emission tomography for diagnosis and monitoring of idiopathic retroperitoneal fibrosis associated with mediastinal fibrosis. Ann. Nucl. Med. 24: 225–229. 96 Young, P.M., Peterson, J.J., and Calamia, K.T. (2008). Hypermetabolic activity in patients with active retroperitoneal fibrosis on F-18FDG PET: report of three cases. Ann. Nucl. Med. 22: 87–92. 97 Jansen, I., Hendriksz, T.R., Han, S.H. et al. (2010). (18) F-fluorodeoxyglucose position emission tomography

811

812

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

98

99

100

101

102

103

104

105

106

107

108

(FDGPET) for monitoring disease activity and treatment response in idiopathic retroperitoneal fibrosis. Eur. J. Intern. Med. 21: 216–221. Magrey, M.N., Husni, M.E., Kushner, I. et al. (2009). Do acute-phase reactants predict response to glucocorticoid therapy in retroperitoneal fibrosis? Arthritis Rheum. 61: 674–679. Laurent, C., Ricard, L., Fain, O. et al. (2019). PET/MRI in large-vessel vasculitis: clinical value for diagnosis and assessment of disease activity. Sci. Rep. 9: 12388. Jennette, J.C. et al. (2013). 2012 revised international Chapel Hill consensus conference nomenclature of vasculitides. Arthritis Rheum. 65: 1–11. Soussan, M., Nicolas, P., Schramm, C. et al. (2015). Management of large-vessel vasculitis with FDG-PET: a systematic literature review and meta-analysis. Medicine (Baltimore) 94: e622. Incerti, E., Tombetti, E., Fallanca, F. et al. (2017). (18) F-FDG PET reveals unique features of large vessel inflammation in patients with Takayasu’s arteritis. Eur. J. Nucl. Med. Mol. Imaging 44: 1109–1118. Andrews, J., Al-Nahhas, A., Pennell, D.J. et al. (2004). Non-invasive imaging in the diagnosis and management of Takayasu’s arteritis. Ann. Rheum. Dis. 63: 995–1000. Walter, M.A., Melzer, R.A., Schindler, C. et al. (2005). The value of [18F]FDG-PET in the diagnosis of large-vessel vasculitis and the assessment of activity and extent of disease. Eur. J. Nucl. Med. Mol. Imaging 65: 74–681. Grayson, P.C., Alehashemi, S., Bagheri, A.A. et al. (2018). (18)F-Fluorodeoxyglucose-positron emission tomography as an imaging biomarker in a prospective, longitudinal cohort of patients with large vessel vasculitis. Arthritis Rheumatol. 70: 439–449. Einspieler, I., Henninger, M., Mergen, V. et al. (2019). Three-dimensional fat-saturated T1-weighted Cartesian volumetric interpolated breath-hold examination (VIBE) for the diagnosis of aortitis in patients with suspected large vessel vasculitis: a comparative study with (18)F-FDG PET applying fully integrated PET/ MRI. Clin. Radiol. 74: e11–e731. Desai, M.Y., Stone, J.H., Foo, T.K. et al. (2005). Delayed contrast-enhanced MRI of the aortic wall in Takayasu’s arteritis: initial experience. Am. J. Roentgenol. 184: 1427–1431. Slart RHJA, Writing group, Reviewer group, Members of EANM Cardiovascular et al. (2018). FDG-PET/CT(A) imaging in large vessel vasculitis and polymyalgia rheumatica: joint procedural recommendation of the EANM, SNMMI, and the PET Interest Group (PIG), and endorsed by the ASNC. Eur. J. Nucl. Med. Mol. Imaging 45: 1250–1269.

109 Rudd, J.H., Narula, J., Strauss, H.W. et al. (2002). Imaging atherosclerotic plaque inflammation with [18F]-fluorodeoxyglucose positron emission tomography. Circulation 105: 2708–2711. 110 Baldursson, O., Steinsson, K., Bjornsson, J. et al. (1994). Giant cell arteritis in Iceland. An epidemiologic and histopathologic analysis. Arthritis Rheum. 37: 1007–1012. 111 Gonzalez-Gay, M.A., Garcia-Porrua, C., Llorca, J. et al. (2011). Biopsy-negative giant cell arteritis: clinical spectrum and predictive factors for positive temporal artery biopsy. Semin. Arthritis Rheum. 30: 249–256. 112 Hoffman, G.S., Cid, M.C., Hellmann, D.B. et al. (2002). A multicenter, randomized, double-blind, placebocontrolled trial of adjuvant methotrexate treatment for giant cell arteritis. Arthritis Rheum. 46: 1309–1318. 113 Salvarani, C., Crowson, C.S., O’Fallon, W.M. et al. (2004). Reappraisal of the epidemiology of giant cell arteritis in Olmsted County, Minnesota, over a fifty-year period. Arthritis Rheum. 51: 264–268. 114 Mahr, A., Saba, M., Kambouchner, M. et al. (2006). Temporal artery biopsy for diagnosing giant cell arteritis: the longer, the better? Ann. Rheum. Dis. 65: 826–828. 115 Vergani, V., Busnardo, E., Mapelli, P. et al. (2019). Hybrid cardiac PET/MR: the value of multiparametric assessment in cardiac sarcoidosis. Clin. Transl. Imaging 7: 317–326. 116 Iannuzzi, M.C., Rybicki, B.A., and Teirstein, A.S. (2007). Sarcoidosis. N. Engl. J. Med. 357: 2153–2165. 117 Syed, J. and Myers, R. (2004). Sarcoid heart disease. Can. J. Cardiol. 20: 89–93. 118 Birnie, D.H., Nery, P.B., Ha, A.C. et al. (2016). Cardiac sarcoidosis. J. Am. Coll. Cardiol. 68: 411–421. 119 Terasaki, F. and Yoshinaga, K. (2017). New guidelines for diagnosis of cardiac sarcoidosis in Japan. Ann. Nucl. Cardiol. 3: 42–45. 120 Birnie, D.H., Sauer, W.H., Bogun, F. et al. (2004). HRS expert consensus statement on the diagnosis and management of arrhythmias associated with cardiac sarcoidosis. Heart Rhythm 11: 1305–1323. 121 Ishimaru, S., Tsujino, I., Takei, T. et al. (2005). Focal up-take on 18F-fluoro-2-deoxyglucose positron emission tomography images indicates cardiac involvement of sarcoidosis. Eur. Heart J. 26: 1538–1543. 122 Blankstein, R., Osborne, M., Naya, M. et al. (2014). Cardiac positron emission tomography enhances prognostic assessments of patients with suspected cardiac sarcoidosis. J. Am. Coll. Cardiol. 63: 329–336. 123 Wicks, E., Menezes, L.J., Barnes, A. et al. (2018). Diagnostic accuracy and prognostic value of simultaneous hybrid 18F-fluorodeoxyglucose positron emission tomography/magnetic resonance imaging in cardiac sarcoidosis. Eur. Heart J. Cardiovasc. Imaging 19: 757–767.

Clinical Application of PET/MRI

124 Dweck, M.R., Abgral, R., Trivieri, M.G. et al. (2018). Hybrid magnetic resonance imaging and positron emission tomography with fluorodeoxyglucose to diagnose active cardiac sarcoidosis. JACC Cardiovasc. Imaging 11: 94–107. 125 Ohira, H., Tsujino, I., Ishimaru, S. et al. (2008). Myocardial imaging with 18F-fluoro-2-deoxyglucose positron emission tomography and magnetic resonance imaging in sarcoidosis. Eur. J. Nucl. Med. Mol. Imaging 35: 933–941. 126 Patel, M.R., Cawley, P.J., Heitner, J.F. et al. (2009). Detection of myocardial damage in patients with sarcoidosis. Circulation 120: 1969–1977. 127 Coleman, G.C., Shaw, P.W., Balfour, P.C. et al. (2017). Prognostic value of myocardial scarring on CMR in patients with cardiac sarcoidosis: a systematic review and meta-analysis. J. Am. Colloids Cardiol. Imaging 10: 411–420. 128 Kawakatsu, N., Suzuki, A., Serizawa, N. et al. (2016). Isolated cardiac sarcoidosis diagnosed by electroanatomic voltage mapping guided endomyocardial biopsy combined with magnetic resonance imaging and positron emission tomography. J. Cardiol. Cases 14: 107–110. 129 Bravo, P.E., Raghu, G., Rosenthal, D.G. et al. (2017). Risk assessment of patients with clinical manifestations of cardiac sarcoidosis with positron emission

130

131

132

133

134

135

tomography and magnetic resonance imaging. Int. J. Cardiol. 241: 457–462. Vita, T., Okad, D., Veillet-Chowdhury, M. et al. (2018). Complementary value of cardiac magnetic resonance imaging and positron emission tomography/computed tomography in the assessment of cardiac sarcoidosis. Circ. Cardiovasc. Imaging 11: e007030. Wisenberg, G., Thiessen, J.D., Pavlovsky, W. et al. (2020). Same day comparison of PET/CT and PET/MR in patients with cardiac sarcoidosis. J. Nucl. Cardiol. 27: 2118–2129. (Epub ahead of print). Hanneman, K., Kadoch, M., Guo, H. et al. (2017). Initial experience with simultaneous 18F-FDG PET/MRI in the evaluation of cardiac sarcoidosis and myocarditis. Clin. Nucl. Med. 42: e328–e334. Greulich, S., Deluigi, C.C., Gloekler, S. et al. (2013). CMR imaging predicts death and other adverse events in suspected cardiac sarcoidosis. JACC Cardiovasc. Imaging 6: 501–511. Nensa, F., Tezgah, E., Poeppel, T. et al. (2015). Diagnosis and treatment response evaluation of cardiac sarcoidosis using positron emission tomography/magnetic resonance imaging. Eur. Heart J. 36: 550. Kiko, T., Masuda, A., Nemoto, A. et al. (2018). Cardiac sarcoidosis after glucocorticoid therapy evaluated by 18F-fluorodeoxyglucose PET/MRI. J. Nucl. Cardiol. 25: 685–687.

813

814

30 68

Ga-FAPI, a Twin Tracer for 18F-FDG in the Era of Evolving PET Imaging

Reyhaneh Manafi-Farid1, GhasemAli Divband2,3, HamidReza Amini3, Thomas G. Clifford4, Ali Gholamrezanezhad4, Mykol Larvie5, and Majid Assadi6 1

Research Center for Nuclear Medicine, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran Nuclear Medicine Center, Jam Hospital, Tehran, Iran 3 Khatam PET-CT Center, Khatam Hospital, Tehran, Iran 4 Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA 5 Department of Radiology, Cleveland Clinic, Cleveland, OH, USA 6 Department of Radiology, School of Medicine, Nuclear Medicine and Molecular Imaging Research Center, Bushehr University of Medical Sciences, Bushehr, Iran 2

Introduction The rising incidence and prevalence of cancer has led to a search within the field of nuclear medicine for new ways to control and treat the diseases. Various positron emission tomography (PET)-tracers have been employed for functional imaging  [1]. Currently, 2-18F-fluoro-2-deoxy-d-glucose positron emission tomography/computed tomography (18F-FDG PET/CT) is the standard molecular imaging modality for many cancers. Nevertheless, FDG is a nonspecific metabolic marker with some limitations in certain malignancies, such as welldifferentiated neuroendocrine tumors, prostate adenocarcinoma, and mucin-producing tumors. Therefore, targeted molecular imaging is pursuing new tracers for imaging and therapy that may overcome the drawbacks of earlier agents [2]. In nuclear oncology, the theragnostic concept is an established method for specific molecular targeting, in both diagnosis and therapy. The scintigraphic detection of targets can be considered as an indirect sign that the patient may benefit from specific treatments [3]. Modern precision oncology is based on tumor biology and thus relies on accurate analysis of the relationship between tumor cells and their microenvironment to understand the underlying tumor pathophysiology [4]. In addition to cancer cells, the tumor microenvironment consists of multiple different cell types, including inflammatory cells, lymphocytes, blood vessels, fibroblasts, and the extracellular matrix  [5]. Therefore, this tumor microenvironment offers numerous potential targets for novel cancer drugs.

One such target is cancer-associated fibroblasts (CAFs), which are different from ordinary fibroblasts and abundant in the tumor microenvironment with potent effects on tumor development [6]. CAFs play a role in tumor growth and progression. A distinguishing feature of CAFs is their expression of fibroblast activation protein (FAP), a type II membranebound glycoprotein [7]. FAP expression in healthy adult tissues is negligible but is highly upregulated with tissue remodeling, fibrosis, wound healing, inflammation, and cancers [8, 9]. FAP can be considered as a key co-conspirator of cancer cells in tumor growth and metastasis [10]. Expression of FAP and its role in the majority of epithelial tumors encouraged scientists to employ it in oncologic imaging. However, early imaging with nonselective FAP antibodies was suboptimal, mainly due to their large size and long circulation time [11]. In recent years, several small-molecule inhibitors of FAP (FAPIs) have been investigated for the diagnosis and therapy of various malignancies. A radiolabeled FAPI, recently introduced by Loktev et al. revealed acceptable characteristics for imaging, including rapid, specific, and stable uptake [12]. Also, imaging with radiolabeled FAPIs reportedly provides high contrast in tumors with rich stromal content, and they may be considered highly tumor-specific [7]. FAP has been a target for imaging and treatment with different cold and radioactive FAP antibodies or inhibitors [13]. Here, we briefly describe the different subtypes of FAP inhibitors and their biodistribution. Moreover, we review the application of the novel promising imaging

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach, First Edition. Edited by Ali Gholamrezanezhad, Majid Assadi, and Hossein Jadvar. © 2023 John Wiley & Sons Ltd. Published 2023 by John Wiley & Sons Ltd.

68

Ga-FAPI, a Twin Tracer for 18F-FDG in the Era of Evolving PET Imaging

modality, radiolabeled-FAPI PET, in nuclear oncology. Finally, we also discuss its potential role in nonmalignant conditions.

FAPI Subtypes, Biodistribution, and Dosimetry Various radiolabeled FAPI subtypes have been used for PET imaging in preclinical and clinical settings, including, but not limited to, 68Ga-labeled FAPIs. FAPI-01 and FAPI-02  have shown specific binding to human FAP with a rapid and almost complete internalization [7]. FAPI-01 proved to be suboptimal due to enzymatic deiodination with efflux of the free iodine and consequently lower intracellular radioactivity after longer incubation intervals. FAPI-02, a DOTA-linked compound, demonstrated better pharmacokinetic and biochemical properties. FAPI-02 eliminated more slowly than FAPI-01 with approximately 10-fold higher retention after 24 hours. Furthermore, a rapid internalization into FAP-expressing cells has been reported with a high tumor uptake in patients with metastatic epithelial carcinomas [7, 14]. For optimization of tumor radiotracer uptake and retention, several FAPI-02-based compounds were investigated. Of synthesized FAPIs (FAPI-02 to FAPI-15), FAPI-04 was the most promising radiotracer as a potential theragnostic agent [15]. Similar to FAPI-02, FAPI-04 showed rapid internalization into FAP-positive tumors and fast clearance from the body, resulting in fast retention in lesions (in about 10 minutes after radiotracer injection). Furthermore, when compared to FAPI-02, FAPI-04 revealed 100% higher effective tumor uptake after 24 hours, which is more desirable for theragnostic application [15]. Several new compounds have been developed based on the lead compound FAPI-04, which is characterized by rapid radiotracer uptake in FAP-positive tumors followed by considerable background washout. Of these compounds, 68Ga-FAPI-46 displayed the highest tumor-toblood, tumor-to-muscle, and tumor-to-liver ratios [16]. Biodistribution has been studied on the following 177 lutetium-labeled radiotracers: FAPI-21, -35, -46, and -55. All of these compounds showed robust tumor accumulation with low radiotracer uptake by normal tissues. FAPI21 and -46 demonstrated higher tumor accumulation at 1 and 4 hours after injection in comparison to FAPI-04 [16]. Depending on the tumor type, tumor accumulation can be significantly prolonged using FAPI-46 [16]. Whereas earlier DOTA-based tracers (FAPI-02, -04, and -46) were developed with a focus on potential therapeutic applications, the NOTA-derivate FAPI-74 was developed as an exclusively diagnostic ligand, with slightly shorter

tumor retention  [17]. Recent analysis of biodistribution and dosimetry of 18F-FAPI-74 PET/CT and 68Ga-FAPI-74 PET/CT (by NOTA chelator ligand for 18F and 68Ga) in 10 lung cancer patients revealed high image contrast and low radiation burden. All currently available 68Ga-labeled FAPI subtypes have renal excretion with activity in the urinary bladder and otherwise minimal background activity in normal tissues. Usually, the radiotracer is seen in the renal collecting system and urinary bladder without accumulation in the renal parenchyma [14]. Some researchers have reported dosimetry results of FAPI PET tracers. Giesel et al. performed the preliminary dosimetry for 68Ga-FAPI-02 and 68Ga-FAPI-04 in two metastasized breast cancer patients. PET/CT acquisition was performed at 10 minutes, 1 hour, and 3 hours after radiotracer injection. The effective total body dose was 1.80E-22 mSv/MBq for 68 Ga-FAPI-02 and 1.64E-22 mSv/MBq for 68Ga-FAPI-04. The authors predicted that by injection of a dose of about 200 MBq, the effective dose from 68Ga-FAPI PET/CT would be 3–4 mSv. The urinary bladder received the highest effective dose for both 68Ga-FAPI-02 and -04 [14]. The dosimetry of 68Ga-FAPI-46 was investigated by Meyer et al. in six different cancer patients. Serial imaging was performed at 10minutes, 1hour, and 3hours after injection of 214–246MBq of 68Ga-FAPI-46. The average effective total body dose was 7.80E-03mSv/MBq. Organs with the highest effective doses were the bladder wall (2.41E-03mSv/MBq), ovaries (1.15E-03mSv/MBq), and red marrow (8.49E-04mSv/ MBq). The effective total body dose from 200MBq of 68GaFAPI-46 was 1.56±0.26mSv. This, along with 3.7mSv from low-dose CT, leads to approximately 5.3mSv total effective dose. They concluded that 68Ga-FAPI-46 PET/CT has a favorable dosimetry profile for imaging [18]. In summary, of the various FAPI tracers that have been developed, many of them have shown optimal imaging characteristics with favorable dosimetry. In particular, FAPI-21 and FAPI-46 seem to be the most appropriate agents for theragnostics.

Imaging Protocol At present there are no special patient preparation recommendations prior to performing 68Ga-FAPI PET/CT, such as fasting or discontinuing medications. However, adequate hydration is necessary for all agents with urinary excretion. Due to high and rapid renal excretion of most FAPI-based PET radiotracers, intravenous hydration with 500 mL of normal saline and injection of 20 mg of furosemide from 15 minutes before to 30 minutes after radiotracer administration have been implemented [19]. 68Ga-FAPI is

815

816

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

injected intravenously but a definite injection activity has not been established. Authors have used the range of 100– 370 MBq [14, 18, 19]. The appropriate injection activity is yet to be established for different scanners. In several clinical studies, especially for dosimetry purposes, different acquisition times for whole-body 68GaFAPI PET/CT scans were evaluated, mainly at three time points after radiotracer injection: 10 minutes, 1 hour, and 3 hours  [14, 15, 18]. Early time-point imaging at 10minutes with 68Ga-FAPI is possible, but it should be noted that the tumor-to-background contrast ratio improves with time [18]. Based on our experience, we orally hydrate patients and begin acquisition 30minutes after injection of 200MBq 68Ga-FAPI.

Oncologic Application The overexpression of FAP in the microenvironment of many epithelial tumors is associated with a worse prognosis. Therefore, targeting this enzyme for molecular imaging and endoradiotherapy may be a novel strategy in cancer management  [7]. 68Ga-FAPI PET/CT with various FAPI ligands (which will all be referred to as FAPI below) has been used for targeting FAP in tumor stroma [20]. The first human studies were reported in 2018  [12, 15]. To date, multiple clinical studies have been performed to evaluate the potential efficiency of FAPI PET in clinical oncology.

Imaging 68

Ga-FAPI PET/CT has been used to image various cancer types [14, 18], mostly in case series or case reports. In an early multicenter study, 68Ga-FAPI uptake was evaluated in 28 different cancer types in patients (n =80) [19], with suboptimal results from other imaging modalities. The result was promising, revealing remarkably high uptake and image contrast in several prevalent cancers such as breast, esophagus, lung, pancreatic, head–neck, and colorectal [19]. In another study, 75 patients with 12 different cancer types underwent imaging with 68Ga-FAPI and 18F-FDG PET/CT. Interestingly, the study showed that the detection rates of the primary tumor, lymph node, bone, and visceral metastases were significantly higher with 68Ga-FAPI PET/CT  [21]. Sarcoma, esophageal, breast, lung, pancreas, and primary liver cancers were among those with the highest uptake [19, 21]. In another recent study on 68 patients with inconclusive 18F-FDG PET/CT, 68Ga-FAPI PET/CT discriminated malignant lesions with an accuracy of 66.7% (12/18), localized primary cancer in 66.7% (4/12), detected tumor recurrence in 87.0% (20/23), and upstaged 33.3% (7/21) of patients [22].

When compared to 18F-FDG, 68Ga-FAPI showed higher intensity of uptake in most of the malignancies, providing higher contrast and tumor visualization [21–25], especially in gastrointestinal tract tumors such as primary esophageal cancer  [23, 26]. In 35 patients with gastric, duodenal or colorectal cancer, Pang et al. demonstrated that the sensitivity and intensity of uptake were higher with 68Ga-FAPI PET/CT for the primary lesions as well as lymph node, visceral, and bone metastases [24]. In the evaluation of hepatocellular carcinoma, 68Ga-FAPI PET/CT detected seven out of eight, 15 out of 16, and 20 out of 20 tumors in three different studies  [27–29]. Increased liver uptake was reported in cirrhosis [27, 28], although this apparently did not impact tumor visualization. The sensitivity of 68GaFAPI PET/CT was reportedly higher than 18F-FDG PET/CT (93 vs. 68%) and comparable with magnetic resonance imaging (93 vs. 100%) for the detection of primary liver malignancies  [28]. Also, 68Ga-FAPI PET/CT detected all (7/7) cholangiocarcinomas in contrast to 18F-FDG PET/CT (4/7)  [28]. Intense uptake in cholangiocarcinomas has already been reported in other studies [19, 21, 29]. When compared to 18F-FDG PET/CT, Zhao et al. reported higher sensitivity (97.67 vs. 72.09%) and intensity of uptake (median standardized uptake value [SUV] 9.82 vs. 3.48) for 68 Ga-FAPI PET/CT in peritoneal carcinomatosis in patients with various cancers, particularly gastric (20/35) [30]. On the contrary, Ballal et al. demonstrated that the intensity of uptake and the detection rate of 68Ga-FAPI and 18FFDG PET/CT were comparable in various malignancies, particularly breast and lung cancers [2]. They also reported lower intensity of 68Ga-FAPI uptake in lymph nodes  [2]. Differences in inclusion criteria may explain this discrepancy as most of the patients included in the aforementioned studies had malignancies with a known limited role of 18F-FDG PET/CT. Finally, in nasopharyngeal cancer, results have been somewhat inconclusive. Compared to 18F-FDG PET/CT, Zhao et al. (n =45) reported higher intensity of uptake in primary tumors, regional lymph nodes, and bone and visceral metastases  [31], whereas a study by Qin et  al. (n  =15) showed lower SUVmax of primary tumors and metastatic lymph nodes [32]. The number of detected lymph nodes was also noted to be lower with 68Ga-FAPI PET/MRI by Qin et al. [32]. It is clear that cancer staging and treatment algorithms hinge on imaging. 68Ga-FAPI PET/CT has altered stage and management in some cancers. Detecting further metastases as new findings in 47% (10/21) of patients, Koerber et  al. demonstrated a change in the management in 19% (4/21) of patients compared to standard imaging in lower gastrointestinal cancer [33]. The TNM stage was altered in three out of six treatment-naïve patients  [33]. When compared with CT in pancreatic cancer (n = 19), Röhrich

68

Ga-FAPI, a Twin Tracer for 18F-FDG in the Era of Evolving PET Imaging

et  al. showed that 68Ga-FAPI PET/CT changed stage in nine out of 12 and one out of seven with recurrent/progressive and newly diagnosed disease, respectively, while management was altered in seven out of 19 patients [34]. In nasopharyngeal cancer (n  = 39 treatment naïve), 68 Ga-FAPI PET/CT upstaged 10 patients in 26% and altered the management in seven (18%) when compared with 18FFDG  [31]. Also, when compared with MRI, T-stage was upgraded in four (10%) and underestimated in two (5%) patients  [31]. In another study, 68Ga-FAPI PET changed stage in six out of 15 patients (three upstaged and three downstaged) [32]. 68 Ga-FAPI PET/CT was also used for the assessment of treatment response in a case of esophageal cancer. It showed reduced uptake in the tumor 2weeks after chemoradiotherapy, while new regions of uptake were noted in the lungs as the result of radiation pneumonitis  [35]. It was also used before and after cytoreductive surgery of peritoneal carcinomatosis to evaluate the successfulness of the procedure [36]. Additionally, 68Ga-FAPI PET/CT has successfully detected disease recurrence in multiple studies with a reported sensitivity of 87.0% in one study  [22] and more true positive lesions compared to 18F-FDG PET/CT in another [31].

Finally, there are case reports of 68Ga-FAPI uptake in primary extranodal (gastric, brain) lymphoma  [37, 38], extramammary Paget disease  [39], neuroendocrine tumor  [40], cardiac angiosarcoma  [41], chromophobe renal cell carcinoma  [42], and a ground-glass lung adenocarcinoma [43].

Advantages FAP is expressed minimally or not at all in normal cells [12]. Therefore, imaging with FAPIs provides absent or low background activity. When compared with 18F-FDG, this feature is particularly useful for detection of malignant lesions in the brain and liver due to its higher contrast [2, 12, 44]. FAP is also an encouraging tracer in the evaluation of small lesions such as peritoneal carcinomatosis [21, 30, 45] and leptomeninges  [46]. Since even tiny tumors require supporting stromal tissue, it is hypothesized that tumors as small as 3–5 mm could be detected with FAPI PET [12]. Moreover, the uptake of 68Ga-FAPI is independent from the blood glucose level, which has important implications for patient preparation and particular utility in diabetic patients (Figure 30.1) [47].

Figure 30.1  A 69-year-old woman with recurrent cholangiocarcinoma underwent 68Ga-FAPI PET/CT. Because of uncontrollable diabetic hyperglycemia, 18F-FDG PET/CT could not be performed. The images revealed an intensely FAPI-avid (SUVmax, 24.5) soft tissue mass in the Whipple surgical bed along with several mildly FAPI-avid small peri-pancreatic lymph nodes and adjacent peritoneal nodules. A 68Ga-­FAPI–avid (SUVmax = 9.3) myoma was also detected in the uterus.

817

818

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

Figure 30.2  A 73-­year-­old woman with gallbladder adenocarcinoma who underwent cholecystectomy and evidence of progression after chemotherapy. (a) Maximum intensity projection image. 18F-FDG PET/CT scan reveals a mildly FDG-avid subcarinal metastatic lymph node (arrow) (SUVmax = 5.3) and numerous FDG-­avid liver metastatic lesions (SUVmax up to 17.6). (b) Maximum intensity projection image. 68 Ga-­FAPI-­46 PET/CT scan reveals a FAPI-­avid subcarinal metastatic lymph node (arrow) (SUVmax = 13.3), numerous FAPI-­avid metastatic liver lesions (SUVmax up to 21.1) as well as few FAPI-avid peritoneal seeding (arrow) without uptake on 18F-FDG PET/CT. Also a focal FAPI uptake in the pelvic region (above the bladder) is noted, corresponding to calcified uterine myoma (arrow). Caution must be taken for uterine pathology evaluation due to physiological FAPI uptake in the uterus and intense FAPI uptake in the uterine benign myoma.

Finally, FAP’s high target-to-background ratio is ideal for visualization of small lesions, target volume delineation, and targeted radionuclide therapy (Figures 30.2 and 30.3).

Disadvantages 68

Ga-FAPI PET/CT has been shown to be highly sensitive in detecting malignant lesions. Perhaps not surprisingly, this high sensitivity is at the expense of low specificity. Lower specificity has been reported for 68Ga-FAPI PET/ CT of 18F-FDG PET/CT, though this did not reach statistical significance  [21, 24]. The false-positive rate of 68Ga-FAPI PET/CT was approximately 20% (23/121) higher when compared to 18F-FDG PET/CT [21, 24] in one lesion-based analysis [22]. Many conditions that involve active fibrotic reaction can cause false-positive FAPI uptake [12, 21, 22, 34, 47],

including wound healing, arthritis, granulomatous diseases, myelofibrosis, and so on. FAPI reportedly does not accumulate in peritumoral or postsurgical/biopsy inflammation [48]. The ability to evaluate response to therapy earlier has been therefore been presumed to be another advantage. However, false-positive tracer uptake has been reported due to fibrosis after radiation therapy or surgery [22, 35, 49]. Also, Egger et al. showed FAP overexpression in bleomycin-induced pneumonitis, which may increase 68 Ga-FAPI uptake [50]. False-positive 68Ga-FAPI uptake has been reported in cases of recurrent angiomyolipoma [51], idiopathic retroperitoneal fibrosis  [52], thyroiditis  [53], benign breast lymphoid tissue  [54], benign pancreatic lesions  [55], splenic hemangioma  [56], tuberculosis  [57], postpartum breast and endometrium  [58], bone fracture  [59], chronic cholecystitis and osteophytes [60], solitary fibrous tumors [61],

68

Ga-FAPI, a Twin Tracer for 18F-FDG in the Era of Evolving PET Imaging

(a)

(b)

Figure 30.3  A 61-year-old male with biopsy proven anaplastic thyroid carcinoma. (a) Maximum intensity projection image. 68 Ga-FAPI-46 PET/CT scan reveals a huge cervical mass with intense 68Ga-FAPI-46 uptake, involving the left side of the neck with extension to the anterior mediastinum and invasion to the sternum. Also, multiple 68Ga-FAPI-46-avid metastatic cervical lymph nodes are depicted, bilaterally. (b) Fused transverse image. 68Ga-­FAPI-­46 PET/CT detects additional solitary bone metastasis in the right scapula (arrow).

and elastofibroma dorsi  [62]. Thus, false-positive uptake in regions of primary or secondary fibrosis seem inevitable with 68 Ga-FAPI PET/CT. In light of this, it may have potential roles in the evaluation of fibrotic conditions and their treatment response, which is elaborated on in the nononcologic section of this chapter. Finally, accurately recognizing false-negative findings is of crucial importance. Uptake intensity has been reported to be low in renal cell, differentiated thyroid, adenoid cystic cancer, and pheochromocytoma  [19] as well as prostate cancer metastases, which also show low-grade metabolism on 18F-FDG PET/CT [63].

Target Volume Delineation Target volume delineation is usually performed using conventional imaging. However, there is interobserver variability

in contouring the lesions, and normal-sized metastatic lymph nodes are usually not included [64]. Hence, functional imaging, predominantly 2-18F-FDG PET/CT, is used in some malignancies, such as head and neck cancer [65]. The era of conformal radiotherapy with advanced equipment necessitates more accurate tumor delineation. Due to high contrast and apparently high sensitivity, 68Ga-FAPI PET/CT is also being studied in functional target volume delineation. In one preclinical study, Röhrich et al. showed promising results when using 68Ga-FAPI PET/CT in glioblastoma target volume delieation  [66]. Windisch et  al. showed that 68 Ga-FAPI PET/CT-derived tumor volume (in thresholds of five-, seven-, and 10-fold) was incongruent with MRI [67] in glioblastoma, and that functional imaging with 68Garadiolabeled FAP-specific PET for target volume delineation resulted in gross tumor volumes (GTVs) containing tumor not covered by MRI-GTVs [67].

819

820

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

In esophageal cancer, Ristau et al. compared the target volume derived from 68Ga-FAPI PET/CT with CT and endoscopic clipping (n = 7) and reported changes in six out of seven patients [47]. Zhao et al. compared the target volume derived from 68Gaa-FAPI PET/CT with CT and 2-18F-FDG PET/CT (n = 21) [23]. They showed that 20, 30, and 40% of the standardized uptake value thresholds of 68 Ga-FAPI and 40% of 2-18F-FDG had similar lengths of tumor when compared to endoscopy [23]. 68Ga-FAPI PET/ CT increased target volume in up to 28.5% (6/21) of patients [23]. In head and neck cancer, Syed et  al. compared 68GaFAPI PET/CT with CT (n  = 14)  [48]. They concluded that the target volume was significantly larger using 68 Ga-FAPI PET/CT with the threefold threshold  [48]. Also, some of the high 68Ga-FAPI-avid regions, even in higher thresholds, were not included in CT-based target volume [48]. In pancreatic cancer (n  = 7), Liermann et  al. reported that target volumes derived from twofold 68Ga-FAPI PET/ CT are comparable with the volumes defined by most of the radiation oncologists using CT [64]. Moreover, Giesel et al. showed promising performance of fluorine-18-based FAPI in target volume delineation in lung cancer (n = 10) using a threefold threshold [17]. Thanks to its intense uptake in malignant lesions and high-contrast resolution, 68Ga-FAPI PET/CT may increase the target volume in some malignancies, thereby helping to improve conformational radiotherapy  [68]. However, its role in different cancer types and any potential survival benefit of increasing target volume based on 68Ga-FAPI PET/CT remains to be determined [67].

Radionuclide Treatment Molecular biological investigations have demonstrated an association between tumor invasiveness and resistance to therapy with higher density of CAFs and FAP expression  [48]. Targeting these molecules may therefore be therapeutic, particularly given its ability to depict tumoral lesions on imaging. Furthermore, its minimal uptake in normal tissues is an encouraging characteristic for radionuclide therapy, theoretically reducing side effects. In one preclinical study, copper-64 and actinium225 were successfully labeled with FAPI and shown to have specific uptake in tumoral lesions  [69]. Unfortunately, retention in tumors was shorter with 64Cu-FAPI compared to 68Ga-FAPI  [69], while retention in the liver was higher  [69]. Interestingly, 225Ac-FAPI resulted in tumor growth suppression  [69]. One of the concerns in FAPIbased radionuclide therapy is its efflux. However, newer

generations such as FAPI-21 and FAPI-46  have higher tumor retention  [69]. Also, instead of long half-life isotopes, isotypes like generator-based rhenium-188  with its 17 hour half-life may be used [15]. FAPI labeled with yttrium-90 or lutetium-177  has been administered in a limited number of end-stage ovarian, pancreas, and breast cancer patients  [2, 15, 70, 71]. Recently, 177Lu-FAP-2286, 90Y-FAPI-46, and 177LuFAPI-46 peptide-targeted radionuclide therapy (PTRT) was relatively well-tolerated with acceptable side effects when applied to a broad spectrum of cancers (Figure 30.4) [72–74]. FAPI-based radionuclide treatment is an intriguing modality for therapy of treatment-refractory malignancies, but its safety and efficacy have yet to be fully established.

Nononcologic Application As previously alluded to, FAPI is taken up in chronic inflammation, a source of false-positives in oncologic use. However, this attribute may be exploitable in nononcologic conditions to determine disease extent or treatment response. Incidental 68Ga-FAPI uptake has been reported in IgG4related disease [21]. Subsequently, Luo et al. [75] compared the performance of 68Ga-FAPI PET/CT with 18F-FDG PET /CT in 26 patients with IgG4-related disease. In their study, additional 18-organ involvement was detected in 13 out of 26 patients with 68Ga-FAPI PET/CT [75]. With the exception of lymph nodes, the intensity of 68Ga-FAPI uptake was higher [75]. Schmidkonz et al. demonstrated that 18F-FDG is positive in lymph nodes with lymphoplasmacytic infiltration IgG4 + cells whereas 68Ga-FAPI is positive in those with dense activated fibroblasts [76]. FAP expression is also increased after myocardial infarction (MI) [77] due to the presence of activated fibroblasts in myocardial remodeling  [78]. The role of 68Ga-FAPI PET/ CT imaging in the prediction of remodeling was evaluated in a preclinical study showing that 68Ga-FAPI is taken up in the border zone of infarction [78]. Also, 68Ga-FAPI uptake has been documented in a case of chronic heart failure [79], prior MI  [80], and presumed chemotherapy-induced cardiotoxicity  [81]. In two retrospective studies, cardiac 68 Ga-FAPI uptake was evaluated in cancerous patients [82, 83]. Siebermair et  al. (n  = 32) found a relationship between SUVmean and coronary artery disease, age and left ventricular ejection fraction  [82]. Heckmann et  al. (n  = 229) showed a relationship between focal uptake and cardiovascular disease and some cardiovascular disease risk factors  [83]. In another study, Finke et  al. investigated the checkpoint inhibitor-induced cardiac adverse autoimmune

68

Ga-FAPI, a Twin Tracer for 18F-FDG in the Era of Evolving PET Imaging

(a)

(b)

(d)

(c)

(e)

(f)

Figure 30.4  A 45-year-old woman with advanced breast cancer. Her breast cancer subtype was defined as luminal A-like (ER+ PR+ HER2−). 18F-FDG (a–c) and 68Ga-FAPI (d–f) PET/CT imaging showed numeral axial and appendicular skeletal lesions as well as several lesions in bilateral liver lobes (more prominent in FAPI PET). Therefore, the patient was candidate for 177Lu-FAPI-46 and underwent four therapy cycles of 177Lu-FAPI-46 (29.6 GBq).

effects (n = 23) and demonstrated that those with signs of cardiac adverse effects have slightly higher 68Ga-FAPI uptake than those without [84]. Therefore, quantification of fibroblast activation using 68Ga-FAPI PET/CT may have a role in the noninvasive evaluation and management of  cardiovascular disease and therapy-induced cardiac adverse effects. Additionally, Zhou et al. revealed that 68Ga-FAPI uptake is increased in renal fibrosis, enabling noninvasive evalua-

tion of this condition [85]. Zhang et al. evaluated the role of 68Ga-FAPI PET/CT in the process of healing after rotator cuff surgery in a preclinical study [49] and showed a similar trend in 68Ga-FAPI uptake and FAP-positive cell density on immunohistochemical analysis  [49]. Furthermore, increased 68Ga-FAPI uptake has been reported in probable Sjogren’s [47] and Erdheim–Chester diseases [86]. The overexpression of FAP has also been reported in synovial cells of rheumatoid arthritis. It may warrant

821

822

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

future study to examine whether 68Ga-FAPI PET/CT has a role in the differentiation of rheumatoid arthritis from osteoarthritis in unclear cases [87]. In summary, although 68Ga-FAPI PET/CT was introduced as a cancer imaging tracer, it depicts fibroblast activation in other diseases and its potential role in nononcologic conditions, such as rheumatologic disorders and cardiac conditions, warrants further study.

Future Direction 68

Ga-FAPI is a novel agent whose role in the imaging of specific cancer types and nononcologic disorders has only been evaluated in a limited number of studies. Using 68GaFAPI to evaluate treatment response also deserves more investigation. Intensity of 68Ga-FAPI uptake should be studied as it may offer prognostic value. Additionally, for radionuclide therapy, new compounds of FAPI may be developed with more stable tumoral retention, enabling prolonged internal radiation.

Conclusion Radiolabeled-FAPI PET is a promising modality that may become vital to everyday clinical practice in the future.

Simply put, FAPI PET demonstrates high-intensity uptake in most cancers. Radiolabeled-FAPI has an optimal biodistribution given its rapid tumoral uptake and fast background washout. It provides high-contrast images due to its minimal or absent background activity, especially in the brain and liver. Radiolabeled-FAPI demonstrates a particularly high-uptake intensity in multiple malignancies known to have low metabolism on 18F-FDG PET, such as hepatocellular carcinoma. In part thanks to its highcontrast images, radiolabeled-FAPI PET can be used for more precise target volume delineation for radiation therapy. Additionally, radiolabeled-FAPI is not taken up in acute inflammation, which may enable its use in in early assessment of the response to therapy, except for conditions with significant fibrosis. We hypothesize that radiolabeled-FAPI may have a more prominent role in initial staging and early response assessment rather than late response assessment or local recurrence, which may be more often complicated by substantial fibrosis. Another area of intrigue is FAPI’s potential theragnostic application, but further evaluation to determine its safety and efficacy are needed. Finally, radiolabeled-FAPI uptake in nonmalignant conditions provides another potential avenue of clinical use as preliminary studies suggest a role in the evaluation of fibrotic conditions.

References 1 Beheshti, M., Manafi-Farid, R., Rezaee, A., and Langsteger, W. (2020). PET/CT and PET/MRI, Normal Variations, and Artifacts. Switzerland Springer Nature. 2 Ballal, S., Yadav, M.P., Moon, E.S. et al. (2020). Biodistribution, pharmacokinetics, dosimetry of [(68)Ga] Ga-DOTA.SA.FAPi, and the head-to-head comparison with [(18)F]F-FDG PET/CT in patients with various cancers. Eur. J. Nucl. Med. Mol. Imaging 48 (6): 1915–1931. 3 Yordanova, A., Eppard, E., Kürpig, S. et al. (2017). Theranostics in nuclear medicine practice. Onco. Targets. Ther. 10: 4821–4828. 4 Baghban, R., Roshangar, L., Jahanban-Esfahlan, R. et al. (2020). Tumor microenvironment complexity and therapeutic implications at a glance. Cell Commun. Signal. CCS 18 (1): 59. 5 Joyce, J.A. and Fearon, D.T. (2015). T cell exclusion, immune privilege, and the tumor microenvironment. Science (New York N.Y.) 348 (6230): 74–80. 6 Quail, D.F. and Joyce, J.A. (2013). Microenvironmental regulation of tumor progression and metastasis. Nat. Med. 19 (11): 1423–1437.

7 Loktev, A., Lindner, T., Mier, W. et al. (2018). A new method for tumor imaging by targeting cancer associated fibroblasts. J. Nucl. Med. 59 (9): 1423–1429. 8 Li, M., Cheng, X., Rong, R. et al. (2020). High expression of fibroblast activation protein (FAP) predicts poor outcome in high-grade serous ovarian cancer. BMC Cancer 20 (1): 1032. 9 Hamson, E.J., Keane, F.M., Tholen, S. et al. (2014). Understanding fibroblast activation protein (FAP): substrates, activities, expression and targeting for cancer therapy. PROTEOMICS Clin. Appl. 8 (5–6): 454–463. 10 Kalluri, R. (2016). The biology and function of fibroblasts in cancer. Nat. Rev. Cancer 16 (9): 582–598. 11 Hicks, R.J., Roselt, P.J., Kallur, K.G. et al. (2021). FAPI PET/CT: will it end the hegemony of (18)F-FDG in oncology? J. Nucl. Med. 62 (3): 296–302. 12 Loktev, A., Lindner, T., Mier, W. et al. (2018). A tumorimaging method targeting cancer-associated fibroblasts. J. Nucl. Med. 59 (9): 1423–1429. 13 Lindner, T., Loktev, A., Giesel, F. et al. (2019). Targeting of activated fibroblasts for imaging and therapy. EJNMMI Radiopharmacy Chem. 4 (1): 16.

68

Ga-FAPI, a Twin Tracer for 18F-FDG in the Era of Evolving PET Imaging

14 Giesel, F.L., Kratochwil, C., Lindner, T. et al. (2019). (68) Ga-FAPI PET/CT: biodistribution and preliminary dosimetry estimate of 2 DOTA-containing FAP-targeting agents in patients with various cancers. J. Nucl. Med. 60 (3): 386–392. 15 Lindner, T., Loktev, A., Altmann, A. et al. (2018). Development of quinoline-based theranostic ligands for the targeting of fibroblast activation protein. J. Nucl. Med. 59 (9): 1415–1422. 16 Loktev, A., Lindner, T., Burger, E.-M. et al. (2019). Development of fibroblast activation protein-targeted radiotracers with improved tumor retention. J. Nucl. Med. 60 (10): 1421–1429. 17 Giesel, F.L., Adeberg, S., Syed, M. et al. (2021). FAPI-74 PET/CT using either (18)F-AlF or cold-kit (68)Ga labeling: biodistribution, radiation dosimetry, and tumor delineation in lung cancer patients. J. Nucl. Med. 62 (2): 201–207. 18 Meyer, C., Dahlbom, M., Lindner, T. et al. (2020). Radiation dosimetry and biodistribution of (68)GaFAPI-46 PET imaging in cancer patients. J. Nucl. Med. 61 (8): 1171–1177. 19 Kratochwil, C., Flechsig, P., Lindner, T. et al. (2019). (68) Ga-FAPI PET/CT: tracer uptake in 28 different kinds of cancer. J. Nucl. Med. 60 (6): 801–805. 20 Pang, Y., Zhao, L., Luo, Z. et al. (2021). Comparison of 68 Ga-FAPI and 18F-FDG uptake in gastric, duodenal, and colorectal cancers. Radiology 298 (2): 393–402. 21 Chen, H., Pang, Y., Wu, J. et al. (2020). Comparison of [(68)Ga]Ga-DOTA-FAPI-04 and [(18)F] FDG PET/CT for the diagnosis of primary and metastatic lesions in patients with various types of cancer. Eur. J. Nucl. Med. Mol. Imaging 47 (8): 1820–1832. 22 Chen, H., Zhao, L., Ruan, D. et al. (2021). Usefulness of [(68)Ga]Ga-DOTA-FAPI-04 PET/CT in patients presenting with inconclusive [(18)F]FDG PET/CT findings. Eur. J. Nucl. Med. Mol. Imaging 48 (1): 73–86. 23 Zhao, L., Chen, S., Chen, S. et al. (2021). (68)Ga-fibroblast activation protein inhibitor PET/CT on gross tumour volume delineation for radiotherapy planning of oesophageal cancer. Radiother. Oncol. 158: 55–61. 24 Pang, Y., Zhao, L., Luo, Z. et al. (2021). Comparison of (68)Ga-FAPI and (18)F-FDG uptake in gastric, duodenal, and colorectal cancers. Radiology 298 (2): 393–402. 25 Deng, M., Chen, Y., and Cai, L. (2021). Comparison of 68 Ga-FAPI and 18F-FDG PET/CT in the imaging of pancreatic cancer with liver metastases. Clin. Nucl. Med. 46 (7): 589–591. 26 Liu, Q., Shi, S., Xu, X. et al. (2021). The superiority of [(68)Ga]-FAPI-04 over [(18)F]-FDG PET/CT in imaging metastatic esophageal squamous cell carcinoma. Eur. J. Nucl. Med. Mol. Imaging 48 (4): 1248–1249.

27 Shi, X., Xing, H., Yang, X. et al. (2021). Fibroblast imaging of hepatic carcinoma with (68)Ga-FAPI-04 PET/CT: a pilot study in patients with suspected hepatic nodules. Eur. J. Nucl. Med. Mol. Imaging 48 (1): 196–203. 28 Guo, W., Pang, Y., Yao, L. et al. (2021). Imaging fibroblast activation protein in liver cancer: a single-center post hoc retrospective analysis to compare [(68)Ga]Ga-FAPI-04 PET/CT versus MRI and [(18)F]-FDG PET/CT. Eur. J. Nucl. Med. Mol. Imaging 48 (5): 1604–1617. 29 Shi, X., Xing, H., Yang, X. et al. (2021). Comparison of PET imaging of activated fibroblasts and (18)F-FDG for diagnosis of primary hepatic tumours: a prospective pilot study. Eur. J. Nucl. Med. Mol. Imaging 48 (5): 1593–1603. 30 Zhao, L., Pang, Y., Luo, Z. et al. (2021). Role of [(68)Ga] Ga-DOTA-FAPI-04 PET/CT in the evaluation of peritoneal carcinomatosis and comparison with [(18) F]-FDG PET/CT. Eur. J. Nucl. Med. Mol. Imaging 48 (6): 1944–1955. 31 Zhao, L., Pang, Y., Zheng, H. et al. (2021). Clinical utility of [(68)Ga]Ga-labeled fibroblast activation protein inhibitor (FAPI) positron emission tomography/ computed tomography for primary staging and recurrence detection in nasopharyngeal carcinoma. Eur. J. Nucl. Med. Mol. Imaging 48 (11): 3606–3617. 32 Qin, C., Liu, F., Huang, J. et al. (2021). A head-to-head comparison of (68)Ga-DOTA-FAPI-04 and (18)F-FDG PET/MR in patients with nasopharyngeal carcinoma: a prospective study. Eur. J. Nucl. Med. Mol. Imaging 48 (10): 3228–3237. 33 Koerber, S.A., Staudinger, F., Kratochwil, C. et al. (2020). The role of (68)Ga-FAPI PET/CT for patients with malignancies of the lower gastrointestinal tract: first clinical experience. J. Nucl. Med. 61 (9): 1331–1336. 34 Röhrich, M., Naumann, P., Giesel, F.L. et al. (2021). Impact of (68)Ga-FAPI-PET/CT imaging on the therapeutic management of primary and recurrent pancreatic ductal adenocarcinomas. J. Nucl. Med. 62 (6): 779–786. 35 Zhao, L., Chen, S., Lin, L. et al. (2020). [(68)Ga]GaDOTA-FAPI-04 improves tumor staging and monitors early response to chemoradiotherapy in a patient with esophageal cancer. Eur. J. Nucl. Med. Mol. Imaging 47 (13): 3188–3189. 36 Zhao, L., Pang, Y., Wei, J. et al. (2021). Use of 68Ga-FAPI PET/CT for evaluation of peritoneal carcinomatosis before and after cytoreductive surgery. Clin. Nucl. Med. 46 (6): 491–493. 37 Wang, G., Jin, X., Zhu, H. et al. (2021). (68)Ga-NOTAFAPI-04 PET/CT in a patient with primary gastric diffuse large B cell lymphoma: comparisons with [(18)F] FDG PET/CT. Eur. J. Nucl. Med. Mol. Imaging 48 (2): 647–648.

823

824

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

38 Zhang, Y., Cai, J., Lin, Z. et al. (2021). Primary central nervous system lymphoma revealed by 68Ga-FAPI and 18 F-FDG PET/CT. Clin. Nucl. Med. 46 (8): e421–e423. 39 Jiang, C. and Song, S. (2021). 68Ga-FAPI and 18F-FDG PET/CT in perineum extramammary Paget disease. Clin. Nucl. Med. 46 (4): 342–344. 40 Kömek, H., Gündoğan, C., and Can, C. (2021). 68Ga-FAPI PET/CT versus 68Ga-DOTATATE PET/CT in the evaluation of a patient with neuroendocrine tumor. Clin. Nucl. Med. 46 (5): e290–e292. 41 Zhao, L., Pang, Y., Lin, Q., and Chen, H. (2021). Cardiac angiosarcoma detected using 68Ga-fibroblast activation protein inhibitor positron emission tomography/magnetic resonance. Eur. Heart J. 42 (13): 1276. 42 Pang, Y., Wei, J., Shang, Q. et al. (2021). 68Ga-fibroblast activation protein inhibitor, a promising radiopharmaceutical in PET/CT to detect the primary and metastatic lesions of chromophobe renal cell carcinoma. Clin. Nucl. Med. 46 (2): 177–179. 43 Chen, H., Pang, Y., Meng, T. et al. (2021). 18F-FDG and 68 Ga-FAPI PET/CT in the evaluation of ground-glass opacity nodule. Clin. Nucl. Med. 46 (5): 424–426. 44 Giesel, F.L., Heussel, C.P., Lindner, T. et al. (2019). FAPI-PET/CT improves staging in a lung cancer patient with cerebral metastasis. Eur. J. Nucl. Med. Mol. Imaging 46 (8): 1754–1755. 45 Pang, Y., Zhao, L., and Chen, H. (2020). 68Ga-FAPI outperforms 18F-FDG PET/CT in identifying bone metastasis and peritoneal carcinomatosis in a patient with metastatic breast cancer. Clin. Nucl. Med. 45 (11): 913–915. 46 Hao, B., Wu, J., Pang, Y. et al. (2020). 68Ga-FAPI PET/CT in assessment of leptomeningeal metastases in a patient with lung adenocarcinoma. Clin. Nucl. Med. 45 (10): 784–786. 47 Ristau, J., Giesel, F.L., Haefner, M.F. et al. (2020). Impact of primary staging with fibroblast activation protein specific enzyme inhibitor (FAPI)-PET/CT on radiooncologic treatment planning of patients with esophageal cancer. Mol. Imaging Biol. 22 (6): 1495–1500. 48 Syed, M., Flechsig, P., Liermann, J. et al. (2020). Fibroblast activation protein inhibitor (FAPI) PET for diagnostics and advanced targeted radiotherapy in head and neck cancers. Eur. J. Nucl. Med. Mol. Imaging 47 (12): 2836–2845. 49 Zhang, X. and Chen, D. (2021). in vivo imaging of fibroblast activity using a 68Ga-labeled fibroblast activation protein alpha (FAP-α) inhibitor: Study in a mouse rotator cuff repair model. J. Bone Joint Surg. Am. 103 (10): e40. 50 Egger, C., Cannet, C., Gérard, C. et al. (2017). Effects of the fibroblast activation protein inhibitor, PT100, in a

51

52

53

54

55

56

57

58

59

60

61

62

63

64

murine model of pulmonary fibrosis. Eur. J. Pharmacol. 809: 64–72. Qin, C., Gai, Y., Liu, Q. et al. (2020). Elevated 68Ga-FAPI accumulation in a recurrent angiomyolipoma. Clin. Nucl. Med. 45 (12): 1034–1035. Pan, Q., Luo, Y., and Zhang, W. (2021). Idiopathic retroperitoneal fibrosis with intense uptake of 68Gafibroblast activation protein inhibitor and 18F-FDG. Clin. Nucl. Med. 46 (2): 175–176. Liu, H., Liu, L., Chen, L. et al. (2021). [(68)Ga]Ga-DOTAFAPI-04 PET/CT imaging of benign pulmonary solitary fibrous tumour. Eur. J. Nucl. Med. Mol. Imaging 48 (6): 2053–2054. Gündoğan, C., Güzel, Y., Can, C. et al. (2021). Falsepositive 68Ga-fibroblast activation protein-specific inhibitor uptake of benign lymphoid tissue in a patient with breast cancer. Clin. Nucl. Med. 46 (8): e433–e435. Zhang, X., Song, W., Qin, C. et al. (2021). Non-malignant findings of focal (68)Ga-FAPI-04 uptake in pancreas. Eur. J. Nucl. Med. Mol. Imaging 48 (8): 2635–2641. Liu, H., Wang, Y., Zhang, W. et al. (2021). Elevated 68 Ga-FAPI activity in splenic hemangioma and pneumonia. Clin. Nucl. Med. 46 (8): 694–696. Gu, B., Luo, Z., He, X. et al. (2020). 68Ga-FAPI and 18 F-FDG PET/CT images in a patient with extrapulmonary tuberculosis mimicking malignant tumor. Clin. Nucl. Med. 45 (11): 865–867. Dendl, K., Koerber, S.A., Adeberg, S. et al. (2021). Physiological FAP-activation in a postpartum woman observed in oncological FAPI-PET/CT. Eur. J. Nucl. Med. Mol. Imaging 48 (6): 2059–2061. Wu, J., Liu, H., Ou, L. et al. (2021). FAPI uptake in a vertebral body fracture in a patient with lung cancer: a FAPI imaging pitfall. Clin. Nucl. Med. 46 (6): 520–522. Liu, H., Chen, Z., Yang, X. et al. (2021). Increased 68 Ga-FAPI uptake in chronic cholecystitis and degenerative osteophyte. Clin. Nucl. Med. 46 (7): 601–602. Zhang, A., Zhang, H., Zhou, X. et al. (2021). Solitary fibrous tumors of the pleura shown on 18F-FDG and 68 Ga-DOTA-FAPI-04 PET/CT. Clin. Nucl. Med. 46 (11): 1e534–e537. Hayrapetian, A., Girgis, M.D., Yanagawa, J. et al. (2021). Incidental detection of elastofibroma dorsi with 68GaFAPI-46 and 18F-FDG PET/CT in a patient with esophageal cancer. Clin. Nucl. Med. 46 (2): e86–e87. Xu, T., Zhao, Y., Ding, H. et al. (2021). [(68)Ga]Ga-DOTAFAPI-04 PET/CT imaging in a case of prostate cancer with shoulder arthritis. Eur. J. Nucl. Med. Mol. Imaging 48 (4): 1254–1255. Liermann, J., Syed, M., Ben-Josef, E. et al. (2021). Impact of FAPI-PET/CT on target volume definition in radiation therapy of locally recurrent pancreatic cancer. Cancers 13 (4): 796.

68

Ga-FAPI, a Twin Tracer for 18F-FDG in the Era of Evolving PET Imaging

65 Lee, A.W., Ng, W.T., Pan, J.J. et al. (2018). International guideline for the delineation of the clinical target volumes (CTV) for nasopharyngeal carcinoma. Radiother. Oncol. 126 (1): 25–36. 66 Röhrich, M., Loktev, A., Wefers, A.K. et al. (2019). IDH-wildtype glioblastomas and grade III/IV IDHmutant gliomas show elevated tracer uptake in fibroblast activation protein-specific PET/CT. Eur. J. Nucl. Med. Mol. Imaging 46 (12): 2569–2580. 67 Windisch, P., Röhrich, M., Regnery, S. et al. (2020). Fibroblast activation protein (FAP) specific PET for advanced target volume delineation in glioblastoma. Radiother. Oncol. 150: 159–163. 68 Windisch, P., Zwahlen, D.R., Koerber, S.A. et al. (2020). Clinical results of fibroblast activation protein (FAP) specific PET and implications for radiotherapy planning: systematic review. Cancer 12 (9): 2629. 69 Watabe, T., Liu, Y., Kaneda-Nakashima, K. et al. (2020). Theranostics targeting fibroblast activation protein in the tumor stroma: (64)Cu- and (225)Ac-labeled FAPI-04 in pancreatic cancer xenograft mouse models. J. Nucl. Med. 61 (4): 563–569. 70 Lindner, T., Altmann, A., Krämer, S. et al. (2020). Design and development of (99m)Tc-labeled FAPI tracers for SPECT imaging and (188)Re therapy. J. Nucl. Med. 61 (10): 1507–1513. 71 Ballal, S., Yadav, M.P., Kramer, V. et al. (2021). A theranostic approach of [(68)Ga]Ga-DOTA.SA.FAPi PET/ CT-guided [(177)Lu]Lu-DOTA.SA.FAPi radionuclide therapy in an end-stage breast cancer patient: new frontier in targeted radionuclide therapy. Eur. J. Nucl. Med. Mol. Imaging 48 (3): 942–944. 72 Assadi, M., Rekabpour, S.J., Jafari, E. et al. (2021). Feasibility and therapeutic potential of 177Lu-fibroblast activation protein inhibitor-46 for patients with relapsed or refractory cancers: a preliminary study. Clin. Nucl. Med. 46 (11): e523–e530. 73 Ferdinandus, J., Costa, P.F., Kessler, L. et al. (2021). Initial clinical experience with 90Y-FAPI-46 radioligand therapy for advanced stage solid tumors: a case series of nine patients. J. Nucl. Med. https://doi.org/10.2967/ jnumed.121.262468. 74 Baum, R.P., Schuchardt, C., Singh, A. et al. (2021). Feasibility, biodistribution and preliminary dosimetry in peptide-targeted radionuclide therapy (PTRT) of diverse adenocarcinomas using 177Lu-FAP-2286: first-in-human results. J. Nucl. Med. https://doi.org/10.2967/jnumed.120.259192. 75 Luo, Y., Pan, Q., Yang, H. et al. (2021). Fibroblast activation protein-targeted PET/CT with (68)Ga-FAPI for

76

77

78

79

80

81

82

83

84

85

86

87

imaging IgG4-related disease: comparison to (18)F-FDG PET/CT. J. Nucl. Med. 62 (2): 266–271. Schmidkonz, C., Rauber, S., Atzinger, A. et al. (2020). Disentangling inflammatory from fibrotic disease activity by fibroblast activation protein imaging. Ann. Rheum. Dis 79 (11): 1485–1491. Tillmanns, J., Hoffmann, D., Habbaba, Y. et al. (2015). Fibroblast activation protein alpha expression identifies activated fibroblasts after myocardial infarction. J. Mol. Cell. Cardiol. 87: 194–203. Varasteh, Z., Mohanta, S., Robu, S. et al. (2019). Molecular imaging of fibroblast activity after myocardial infarction using a (68)Ga-labeled fibroblast activation protein inhibitor, FAPI-04. J. Nucl. Med. 60 (12): 1743–1749. Shi, X., Lin, X., Huo, L., and Li, X. (2020). Cardiac fibroblast activation in dilated cardiomyopathy detected by positron emission tomography. J. Nucl. Cardiol. https://doi.org/10.1007/s12350-020-02315-w. Zhu, W., Guo, F., Wang, Y. et al. (2020). 68Ga-FAPI-04 accumulation in myocardial infarction in a patient with neuroendocrine carcinoma. Clin. Nucl. Med. 45 (12): 1020–1022. Totzeck, M., Siebermair, J., Rassaf, T., and Rischpler, C. (2020). Cardiac fibroblast activation detected by positron emission tomography/computed tomography as a possible sign of cardiotoxicity. Eur. Heart J. 41 (9): 1060. Siebermair, J., Köhler, M.I., Kupusovic, J. et al. (2020). Cardiac fibroblast activation detected by Ga-68 FAPI PET imaging as a potential novel biomarker of cardiac injury/ remodeling. J. Nucl. Cardiol. 28 (3): 812–821. Heckmann, M.B., Reinhardt, F., Finke, D. et al. (2020). Relationship between cardiac fibroblast activation protein activity by positron emission tomography and cardiovascular disease. Circ. Cardiovasc. Imaging 13 (9): e010628. Finke, D., Heckmann, M.B., Herpel, E. et al. (2021). Early detection of checkpoint inhibitor-associated myocarditis using (68)Ga-FAPI PET/CT. Front. Cardiovasc. Med. 8: 614997. Zhou, Y., Yang, X., Liu, H. et al. (2021). Value of [(68)Ga] Ga-FAPI-04 imaging in the diagnosis of renal fibrosis. Eur. J. Nucl. Med. Mol. Imaging 48 (11): 3493–3501. Wu, S., Pang, Y., Chen, Y. et al. (2021). 68Ga-DOTAFAPI-04 PET/CT in Erdheim-Chester Disease. Clin. Nucl. Med. 46 (3): 258–260. Bauer, S., Jendro, M.C., Wadle, A. et al. (2006). Fibroblast activation protein is expressed by rheumatoid myofibroblastlike synoviocytes. Arthritis Res. Ther. 8 (6): R171.

825

826

31 Artificial Intelligence in Diagnostic Imaging Martina Sollini1,3, Daniele Loiacono2, Daria Volpe1,3, Alessandro Giaj Levra3, Elettra Lomeo3, Edoardo Giacomello2, Margarita Kirienko4, Arturo Chiti1,3, and Pierluca Lanzi2 1

Department of Biomedical Sciences, Humanitas University, Via Rita Levi Montalcini 4, 20090 Pieve Emanuele – Milan, Italy Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milano, Italy 3 IRCCS Humanitas Research Hospital, Via Manzoni 56, 20089 Rozzano – Milan, Italy 4 Department of Nuclear Medicine, Istituto Nazionale per lo Studio e la Cura dei Tumori, Milano, Italy 2

Introduction Artificial intelligence (AI) research has developed machines able to perform many tasks that normally require human intelligence. Among all AI techniques and methods, machine learning (ML) includes all those approaches that allow computers to learn from data without being explicitly programmed. ML has been extensively applied to medical imaging [1], and the last few years have seen the rise of deep learning (DL), which is an area of ML focused on the application and training of artificial neural networks with a very large number of layers, also called deep neural networks. The major strengths of DL are the capability to extract relevant features from raw data and learn effective data representations, without the need for any human intervention, whereas most ML approaches rely on manually extracted features and human-designed data representations [2–4]. The introduction of AI algorithms to clinical practice would lead to important benefits, including less time to analyze images (i.e. faster diagnosis), higher expertise and standardization in image interpretation (i.e. accurate description of clinical findings), and a low rate of misdiagnosis (i.e. early and personalized therapeutic approaches). All these benefits would ultimately result in a significant improvement in the patient’s outcome. On the other hand, such algorithms are currently focused on specific tasks, without the possibility of approaching human or disease complexity in a holistic way. Moreover, the dataset used for the development of AI might affect its generalizability (i.e. target population, geographic distribution of endemic diseases). Other general concerns related to AI in the healthcare domain are the accuracy of data used for

algorithm development, the size of the training dataset, application to rare conditions or diseases, and legal, ethic, and privacy issues. This chapter aims to provide some insights into AI, including basic knowledge on ML and DL algorithms, challenges, and future directions. Finally, it provides some examples of AI applications in diagnostic imaging.

Deep learning DL emerged as a research area of ML, which is a vast field of AI that includes a variety of methods such as support vector machines (SVMs), decision trees, random forests, clustering algorithms, and neural networks. In particular, DL focuses on methods that stem from the field of neural networks. Indeed, the adjective “deep” refers to the enormous number of artificial neuron layers that are used in these approaches, in contrast with the “shallow” neural networks commonly used in ML. LeCun et al. [5] proposed LeNet as the first successful application of convolutional neural networks (CNNs) in handwritten document recognition, using a seven-layered network. While CNNs existed for decades, researchers regained interest in neural networks only at the beginning of the 2010s, mainly thanks to the technological advances that allowed graphics processor units (GPUs) to be used to speed up neural network computations and the increased availability of image data. In 2012, Krizhevsky et  al. designed AlexNet  [6], a CNN inspired by LeNet but much bigger and with many more layers. AlexNet won the ImageNet 2012 competition in which proposed algorithms are asked to classify each image

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach, First Edition. Edited by Ali Gholamrezanezhad, Majid Assadi, and Hossein Jadvar. © 2023 John Wiley & Sons Ltd. Published 2023 by John Wiley & Sons Ltd.

Artificial Intelligence in Diagnostic Imaging

in a dataset among 1000 classes. This result highlighted the effectiveness of CNN in solving image analysis and computer vision tasks and determine the success of DL [7]. The major advantage of DL over other ML approaches is that the deep structure allows more complex and higherlevel patterns to be captured directly from the data. As a notable example, in 2012 Le et al. [8] created a deep neural network capable of learning various concepts from images, such human or cat faces, without the need to know whether each image contained the concept or not. This capability of learning high-level concepts from the data itself is one of the strengths of DL: while other ML algorithms require an effective feature extraction phase to be successfully applied, DL algorithms can learn an effective data representation automatically. Furthermore, data representations learned from data are generally also useful for related tasks with limited effort  [9], whereas feature extraction is usually a problem-dependent and time-consuming task. In the last few years, the application of DL to medical imaging has attracted a lot of interest, as proved by the increasing number of papers published in the last 5 years on these topics (see Figure 31.1). On the other hand, despite being very promising, these techniques pose several challenges that are still subject to research, such as the availability of a large amount of labeled and annotated data as well as the limited number of public datasets and models shared by the research community. A key element that has allowed the successful application of DL to images and, more generally, to data with a meaningful spatial distribution, is the introduction of CNN. CNNs are neural networks that apply the convolution operation to extract features from overlapping image

1400 # of Publications

1200 1000 800 600 400 200 0 2015

2016

2017 Year

2018

2019

Figure 31.1 Number of publications on deep learning applied to medical imaging. Data retrieved from PubMed using the following search string: (“deep learning” OR “deep neural network” OR “convolution neural network”) AND (radiography OR x-ray OR mammography OR CT OR MRI OR PET OR ultrasound OR therapy OR radiology OR MR OR mammogram OR SPECT).

patches (see Figure 31.2). This operation in neural network layers relies on the assumption that pixels of an image are more tightly correlated with near pixels than distant ones. This allows the network to learn a high number of spatially small filters (typically 3 × 3 or 5 × 5 pixels) that represent the local features in the input image. In a deep neural network several convolutional layers are stacked one onto another. Accordingly, the first layers of the network can learn low-level features in the image, such edge orientations, while subsequent layers build their representation based on the previously learned features. An important property of convolutional layers is that they allow translational invariance: thanks to the convolution operation, which applies the same filter on overlapping patches of the input, each layer can detect the same feature independently from its spatial location. As an example, let look at training CNN to classify photos. The first layers of the network will learn the finest details in an image, such edges, the intermediate layers will combine the edges to recognize patterns as fur, grass or sand, and finally the deeper layers will combine these features to recognize shapes and high-level concepts like animals, trees, outdoor landscapes, people, etc. DL can be applied to a large variety of tasks. Depending on the data availability and the desired outcome, it is possible to identify different learning approaches: ●

Discriminative approaches: These include all the tasks in which the model learns to classify the input data. In the simplest case, a DL model can be a binary classifier, which assigns a label (e.g. positive or negative) to each input sample. This kind of approach can be extended to multiclass and multi-output problems to predict a different set of parameters at the same time, such as arrays of clinical parameters from a medical image. It is also possible to further extend the concept of image classification: instead of predicting a single label value for the entire image, it is possible to localize the region of interest that is relative to a certain label (detection), for example to draw a bounding box around an area labeled as pathological. An even more detailed approach is called segmentation, in which the network, instead of predicting a single label for the whole input image, provides a label probability distribution for each pixel. Thus, with a simple decision rule such thresholding, it is possible to generate an accurate mapping of a certain region of interest. An effective deep neural network architecture for segmentation that has been developed for a medical setting is U-Net [11], which has also become very popular in other, more general, computer vision fields. Figure 31.3 provides a visual representation of the four discriminative approaches mentioned above.

827

828

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

0

... 3

... 9

Flatten

Input Image

Convolutional Filters (Network Weights)

Activations (Layer Output)

Convolutional Filters (Network Weights)

Convolutional Filters (Network Weights)

Figure 31.2 Structure of a convolutional neural network for classifying handwritten numbers. Multiple layers of two-dimensional convolution operations are stacked to extract local features and reduce spatial dimensionality. At each convolutional layer, every image patch is multiplied by a set of filters, which are learned during the training process. In the last layer, a fully connected layer acts as a classifier to produce the final output. The image was generated with an interactive tool for the visualization of convolutional neural networks developed by Harley [10]. Source: Based on Harley [10].

Classification

Prediction Neural Network

Neural Network

Positive 0.98 Negative 0.02

T2 Weighted Brain MRI

Detection

Clinical Parameters T2 Weighted Brain MRI

Segmentation Neural Network

Neural Network

T2 Weighted Brain MRI

p = 0.98

T2 Weighted Brain MRI

Figure 31.3 An overview of the most common discriminative learning approaches used in deep learning.

●

Representation learning: Convolutional autoencoders is an interesting and novel idea to exploit DL for the processing of images. An autoencoder is a particular type of neural network that is trained to reproduce in the output the same sample that is provided as input (see Figure 31.4). The network is composed of two stages: an encoder reduces the dimensionality up to an intermediate numerical representation. The decoder stage, instead, learns to reproduce the input sample starting from the intermediate representation. Since all the information contained in the input sample is forced to be encoded in the intermediate vector, at the end of the training the

encoder network has learned to produce a compact representation of the input, possibly retaining all the relevant information in the data and compressing the less relevant ones. The learned representation can be further used in combination with a classical ML algorithm, e.g. for classification or clustering. Due to the compactness of the intermediate representation for the input size, this technique can be viewed as a form of dimensionality reduction. The decoder network, instead, could be used as a generative model by perturbing the learned representation of a specific input, which is the basic idea behind variational autoencoders (VAEs) [12].

Artificial Intelligence in Diagnostic Imaging Representation Learning

Training Input = Output Convolutional Autoencoder

T2 Weighted Brain MRI

T2 Weighted Brain MRI

Encoder

Decoder Learned Representation

Figure 31.4 An overview of how deep learning exploits representation learning.

Image-to-Image Translation Generator Network

T2 Weighted Brain MRI

T2-FLAIR Brain MRI

Encoder

Decoder

Figure 31.5 An overview of an image-to-image translation task using a generative model.

Image translation and generative approaches: Generative model-building approximates the distribution of the input data. Once such a model has been learned, it can be sampled to generate new unseen samples that follow the same distribution of the input and thus share the same characteristics. This kind of approach can be applied to perform image-to-image translation tasks, where the goal is to produce a target image from an input image, also called a source. Examples of applications include applying a style to an image, colorizing a blackand-white image, or, in medical imaging, generating an image with a different contrast modality. Figure  31.5 shows how DL is applied to image-toimage translation tasks: a generator network, composed

●

by an encoder stage and a decoder stage (as in the autoencoder model described before), is trained to generate images as close as possible to the examples available in the training data. Unfortunately, this training process is complex and often fails to provide the desired results. For this reason, in recent years a novel type of deep neural network has been proposed in the literature: generative adversarial networks (GANs) [13]. This kind of approach has proven to be successful in many computer vision tasks, in particular those related to image generation and translation. The basic idea of GANs (see Figure 31.6) is to train two deep neural networks that work in an adversarial setting. The first is the generator, whose objective is to solve the

829

830

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach Generative Adversarial Networks

Update Parameters Generator Network Discriminator Network

True

Random Noise

Fake

Decoder

Dataset

Figure 31.6 An overview of how generative adversarial networks work.

image-to-image translation task as seen before. The second network is the discriminator, or adversarial network, which is a network that takes in input that is either the ground truth samples, from the training dataset, or the samples generated by the generator. The discriminator is trained to classify whether a given sample comes from the dataset or it has been generated, assigning them the class “true” or “fake.” The mechanism is adversarial, so the two networks are optimized alternatively during the training process: the training process is thus designed in such a way that the generator learns both to produce “realistic” samples and to fool the discriminator in its task of detecting the fake samples. The discriminator, in turn, is trained to detect the fake samples and the real ones. GANs can be used for image-to-image translation [14] or for segmentation problems, in which a setting similar to the image-to-image translation has been used [15].

Hardware and Software To train a deep neural network, several parameters need to be updated at each training step. While this could be done on a desktop central processing unit (CPU), it would require a massive amount of time. Such a computationally expensive task previously would have required the use of an expensive cluster of CPUs to be run in reasonable amounts of time. As mentioned above, GPUs are commonly used for gaming became they are powerful and cost-effective, which also makes them suitable for other applications, including medical imaging. The architecture of a GPU is particularly

suitable for updating the parameters of a neural network, thanks to the possibility of executing hundreds or thousands of parallel computations, in contrast with the tens that are possible to run on a CPU: for this reason, the training time is reduced by several orders of magnitude. Thanks to the accessibility of powerful gaming GPUs and their relatively low cost, many DL frameworks have been developed, allowing users to train DL models directly using their desktop computers. The most popular software platforms are TensorFlow, Keras, PyTorch, Caffe, and Theano. The majority of frameworks are compatible with the most commonly used programming language for data analysis, such as Python or R, but some frameworks are specific for certain general-purpose languages such as Java and C++. Recently, several companies have started to offer cloud computing solutions to train DL models on an hourly cost. The most popular are Google Cloud, Amazon AWS, IBM Watson, Microsoft Azure, Paperspace, and Vast.Ai. There are also free services, such as Google Colab or Kaggle, which offer a limited amount of time and computational power to allow researchers and students to learn and prototype. Every cloud computing solution is different in price, ease of configuration, provided hardware and the number of supported DL frameworks.

Challenges and Open Directions Successful application of DL techniques in a healthcare setting poses a set of challenges that have to be carefully considered due to the particular context. The main

Artificial Intelligence in Diagnostic Imaging

challenges and the most promising open directions to deal with them are the following: ●

●

Labeling and annotations: In healthcare, more than in other fields that adopt DL, the correctness of the data is crucial for the successful application of these “intelligent” approaches. In particular, to obtain good performances, the algorithm must be trained on data labeled or annotated by domain experts (expert labeling problem). However, because of the complexity of DL models, large amounts of data are required for training and raising issues related to the big effort and the resources needed to label such large datasets. This issue maybe addressed by adopting a learning paradigm named active learning. Active learning is an approach in which a learner, such as a DL model, could actively ask a user simple questions or corrections regarding particular samples. Therefore, active learning establishes an interaction between the user and the algorithm, improving the model performance for the cases for which there is uncertainty or the model is having poor performance. Active learning offers the advantage that the data is labeled mostly for important scenarios, reducing in this way the data redundancy that could be introduced when preparing a large dataset a priori. Overfitting: A common problem in deep neural networks is related to overfitting. As deep neural networks could have millions of parameters, they can thus represent very complex functions. The major risk in training such powerful models is that they could fit almost perfectly the training data, but in contrast have low performances when the model is tested on unseen data in a real scenario. This problem, known as overfitting, can be mitigated by adopting specific architectural choices, although the primary cause of overfitting is usually the insufficiency of training data. Since labeled data, especially in healthcare, is a limited resource, it is not always possible to provide more training samples. The commonly adopted solution to this problem is to apply data augmentation techniques. These kinds of techniques allow the amount of data to be artificially increased by applying various transformation of the input data, where it makes sense to have such input. For example, a commonly adopted data augmentation technique for object recognition in natural images is to perform random cropping, random rotation and random scaling. Depending on the task, other techniques are also possible, for example contrast, brightness, saturation shifts, etc. A more recent and promising data augmentation technique is to use generative models, like GANs, to create a dataset of related data that can be used to pre-train the deep model before optimizing it with the real data.

●

●

Rare conditions and diseases: Rare conditions and diseases are the typical scenarios in which datasets are extremely limited and there is no way to further increase the amount of data. In this case, the most straightforward solution is to adopt an approach to transfer learning. The main idea underlying this approach is to transfer the knowledge already acquired on a closely related problem, for which there might be a greater amount of data. For example, let’s assume we want to train a deep neural network to diagnose a rare type of lung disease from X-rays, but we only have a very small dataset due to the rarity of the disease. A possible solution would be to consider an existing model that is trained either on a similar disease or a similar data domain (i.e. X-rays) and to fine-tune the model on our dataset for the rare disease. This approach generally works quite well because the network has to extract from the data every feature that is possibly useful to identify the disease, starting from learning to identify the edges and the color gradients. Indeed, a common approach for training DL in medical imaging deep neural networks is to perform transfer learning from a model previously trained on a natural image dataset, such as ImageNet. In this way, part of the low-level features that are necessary to understand the basic shapes in an image has been already learned on a much larger dataset [16]. Privacy and ethical issues: Privacy and ethics are crucial issues specifically related to particular areas of applications, including healthcare. As mentioned above, DL models rely on massive amounts of data to be trained. While an always increasing amount of data is being collected, privacy concerns limit the possibility of building centralized open datasets that can be used to train DL models. Moreover, legal issues and policy may differ among countries, preventing the possibility of collecting and analyzing data across centers. The commonly adopted solution is to anonymize sensitive data to make it difficult to trace back to the patient’s identity. However, this approach is timeconsuming, prone to error, and not always feasible or effective, especially in some areas (e.g. rare disease). Recently, new techniques have been proposed to face this issue. These techniques aim to change the perspective: instead of searching for new ways to safely exchange data addressing all legal and ethical issues, these new approaches study how to learn the locally available knowledge (i.e. in each institution) and then aggregate or exchange knowledge instead of data (i.e. raw data are not shared). Federated learning is one of these techniques which has been successfully applied to text prediction on smartphones  [16]. Some initial experiences on federated learning in healthcare, and particularly in imaging, have been recently reported with promising results [17–19].

831

832

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Clinical Examples of AI in Diagnostic Imaging AI has a wide spectrum of applications in medicine, although its use in daily practice is only at its dawn [20]. In the field of diagnostic imaging, this topic has become extremely interesting since it introduces tools that might facilitate workflow and might support imagers in diagnosis and interpretation. The areas in which such technology may be particularly impacting are related to those processes which require the immediate attention of the physician (e.g. emergency) or those in which the interpretation of specific imaging findings is subtle or difficult (e.g. screening). Additionally, intelligent computer systems are emerging as “guess machines” to provide additional information to the morphologic and/or molecular characteristics which are undetectable by imagers, such as defining the risk stratification or predicting outcome. As of March 2020, there is already some FDA approved and or/and CE Mark certified AI-based software that can be used to automatically notify the radiologist if imaging contains critical findings. ACCIPIO® is software used to aid the prioritization of the clinical evaluation of acute intracranial hemorrhage (aICH). The software analyzes baseline brain computed tomography (CT) images, identifies abnormal findings suggestive for aICH, and returns the case to the workstation or  picture archiving and communication system (PACS) associated with a case-level indicator, which helps the prioritization of the patient. Viz.AI ContaCT [21] is another example of clinical decision support software approved by the FDA. This computer-aided triage software recognizes findings compatible with large-vessel occlusion in CT angiograms of the brain and sends alerts directly to the mobile app or the PACS system. Both these devices are clinical decision support and are not intended for automatic diagnosis, the final responsibility rests with the physician. Other AI-based tools have not yet been certified – the majority – but may be used for different purposes. For example, Prevedello et  al.  [22] developed two serial algorithms using diverse window settings (requirement imposed by the heterogeneity of the examined conditions) to detect some brain injuries, including acute hemorrhage, mass effect, hydrocephalus, suspect acute infarction, encephalomalacia, and other nonurgent abnormalities or normal findings. Atici et al. [23] developed a CNN for the automatic detection of high-grade gliomas (HGGs) using T2-weighted MRI scans. Notably, this study demonstrated the ability of the algorithm to identify HGGs cases even when the lesion was not present in the given image due to

the unexpected capability of CNN to detect accompanying findings (e.g. edema) in addition to the tumor itself. As mentioned above, DL algorithms may be trained using any type of data. Ma et  al.  [24] proposed a system composed of two CNNs pretrained with the ImageNet dataset, which used as input 15 000 ultrasound images intended to classify thyroid nodules as benign or malignant. The same group proposed a modified DenseNet architecture network trained with about 2900 99mTc singlephoton emission computed tomography (SPECT) images to classify the different patterns of uptake typical for Grave’s disease, Hashimoto disease, subacute thyroiditis, and normal thyroid (Figure 31.7) [25]. Lung nodule detection has emerged as one of the most promising applications of DL. The main goal in this regard is to reach high sensitivity and a small number of false positives. Nam, et al. [26] developed a DL algorithm to detect malignant pulmonary nodules on chest radiographs. The algorithm performed better than radiologists, and when used as a second reader it improved physician performance. Similarly, Hwang et al. [27] developed an algorithm to detect active tuberculosis (TB) on chest X-rays with excellent results. Again, when compared with the work of radiologists it turned out to be better. However, it was not known whether the algorithm could identify other lung manifestations of TB or distinguish active TB from other lung diseases. The higher accuracy of the DL algorithm compared to radiologists has also been reported for the detection and diagnosis of breast cancer [28]. DL methods may be used also for highly specific timeconsuming tasks such as the quantification of the amount of coronary artery calcification or the automatic detection of new lesions in the assessment of follow-up examinations. Wolterink et al. [1] tested a new method to automatically identify and quantify coronary artery calcification with promising results. Interestingly, they used cardiac CT angiography without the need to acquire a dedicated cardiac calcium scoring CT, resulting in a significant reduction in radiation exposure (up to 50%). Vivanti et  al.  [29] developed a new DL-based method that enables a simplified radiologist-friendly workflow by automatically and accurately tracking lesion across studies and detecting new lesions. They used as input baseline segmented CT scans, and follow-up scans were analyzed and compared to baseline images. The algorithm provided lesion segmentation in the follow-up scan as output. This method benefited from baseline and follow-up scan information, outperforming existing stand-alone methods. Yang et al. [30] implemented weakly supervised CNNs, combined with magnetic resonance (MR) apparent diffusion coefficient (ADC) and T2 weighted sequences, which

Artificial Intelligence in Diagnostic Imaging

Grave’s disease

Hashimoto disease

Subacute thyroiditis

Normal

Figure 31.7 Representation of the steps followed for the validation of the algorithm used to classify Grave’s disease. The CNN is first pretrained with ImageNet dataset images, then fine-tuned on medical SPECT images representing Grave’s disease, Hashimoto disease, subacute thyroiditis, and normal subjects. Finally, the model is tested on the test data. Source: Reproduced by Ma et al. (Ma L, Ma C, Liu Y, Wang X. Thyroid Diagnosis from SPECT Images Using Convolutional Neural Network with Optimization. Comput Intell Neurosci. 2019 Jan 15;2019:1–11). Copyright © 2019 Liyong Ma et al. under the Creative Commons Attribution License.

learned pathological features from the whole prostate. The algorithm was trained to indicate if each MRI slice contained cancer or not and localized the cancerous tissues in positive slices. Automated lymph node involvement may improve cancer diagnosis and treatment. Currently, lymph nodal status based on radiological images is assessed on size. CNNs have shown to be extremely successful compared to stateof-the-art studies, increasing sensitivity and reducing the false-positive rate [31]. DL approaches may be used not only to detect an area or a lesion but also to segment a specific region (or more frequently a volume) of interest, a process that is extremely time-consuming and burdened by operator-dependency. Wu et al. [32] developed a deep morphology aided diagnosis network (DeepMAD) to segment carotid artery vessel walls and combine information resulting from the segmentation process to determine the presence of carotid atherosclerosis on black-blood vessel wall MRI. Avendi et al. [33] developed a DL algorithm that could represent the cornerstone for the future development of a fully automatic left ventricle segmentation tool. Issues related to segmentation are particularly relevant in radiation oncology where high accuracy is crucial not only for the target lesions but also for the organs at risk. CNN has been reported to be extremely accurate in automatically segmenting organs at risk [34] by determining an approximately 50% reduction of the mean segmentation time [35]. Notably, not even the application of AI methods was able to accurately segment the pancreas due to its high anatomical variability [36].

Segmentation algorithms can also be integrated into multimodel CNNs (e.g. detection and segmentation, segmentation, and classification) [37, 38]. Ding et al. [39] developed and trained a CNN to distinguish Alzheimer disease (AD), mild cognitive impairment (MCI), or non-AD/MCI using baseline 18-fluoro-deoxyglucose positron emission tomography (18-FDG-PET) images and compared the algorithm’s output to that of experienced radiologists. Data from 1000 patients were prospectively collected within the Alzheimer Disease Neuroimaging Initiative (ADNI) database (2005–2017). In recognizing non-AD and non-MCI, the model demonstrated higher specificity and lower sensitivity when compared to radiologists. The model was able to anticipate the final diagnosis with an average time of 75.8 months. Hwang et al. [40] developed a DL-based algorithm to distinguish normal from abnormal chest radiographs and to classify common diseases including lung cancer, active pulmonary tuberculosis, pneumonia, and pneumothorax. The algorithm, as expected, outperformed nonradiology physicians, but surprisingly also boardcertified radiologists and thoracic radiologists in both image classification and lesion detection. The performance of physicians and radiologists was improved when they were assisted by the algorithm (Figures 31.8 and 31.9). Ciompi et al. [41] developed a ML approach to identify pulmonary peri-fissural nodules. These are benign lesions that represent the 30% of nodules detected during the screening program. They applied a transfer learning approach from a pretrained CNN able to recognize visual

833

834

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

Figure 31.8 Comparing the radiologist-only reading with the deep learning-based automatic detection (DLAD)-assisted doctor. (a) In the chest X-ray, opacity on the left lung was not seen by any radiologist. (b) CT showing ground-glass opacity. (c) The localization of the pneumonia lesion is evident: this is the result of the automatic learning algorithm based on deep learning. Source: Reprinted from Hwang EJ, Park S, Jin K-N, Kim JI, Choi SY, Lee JH, et al. Development and Validation of a Deep Learning–Based Automated Detection Algorithm for Major Thoracic Diseases on Chest Radiographs. JAMA Netw Open. 2019 Mar 22;2(3):e191095. under the terms of the CC-BY license, which permits unrestricted use, distribution, and reproduction in any medium.

(a)

(b)

(c)

Figure 31.9 Compare the radiologist-only reading with the DLAD-assisted doctor: (a) In the chest X-ray, only two out of 15 doctors found the opacity of the right lung. (b) CT scan of the lung nodule. (c) The localization of the pulmonary nodule is evident: this is the result of the automatic learning algorithm based on deep learning. Source: Reprinted from Hwang EJ, Park S, Jin K-N, Kim JI, Choi SY, Lee JH, et al. Development and Validation of a Deep Learning–Based Automated Detection Algorithm for Major Thoracic Diseases on Chest Radiographs. JAMA Netw Open. 2019 Mar 22;2(3):e191095. under the terms of the CC-BY license, which permits unrestricted use, distribution, and reproduction in any medium.

features from 2D images and used them to recognize these lesions, which are characterized by a peculiar shape. The diagnostic accuracy of the CNN was close to that of expert radiologists. Similarly, Walsh et al. [42] trained a DL algorithm to classify idiopathic pulmonary fibrosis. The algorithm achieved higher accuracy than radiologists. McKinney et  al.  [43] developed computer-aided detection software to improve the accuracy of mammography

screening on a very large cohort from the UK and the USA. The algorithm was first trained to recognize suspicious lesions by cross-referencing data from mammograms with those from biopsy, and then its effectiveness was assessed. False-positive and -negative rates were reduced (about 1% in the UK and 6% USA, and 3% in the UK and 9% in the USA, respectively). The algorithm exceeded the average performance of radiologists, supporting its use as a second opinion in mammographic screening.

Artificial Intelligence in Diagnostic Imaging

Yasaka et al. [44] explored the diagnostic performance of CNNs to classify liver masses using dynamic contrast-enhanced CT images. This retrospective analysis used more than 50 000 CT images of liver masses (obtained in 2013) to train the CNN, and then tested the algorithm performance using imaging acquired in 2016. CNN had a high diagnostic potential and was able to classify the liver masses into five categories: classic hepatocellular carcinoma (HCC), other malignant tumors other than classic and early HCC, undetermined masses (including early HCC and dysplastic nodules), hemangiomas, and cysts. Ben-Cohen et  al.  [45] proposed a learning-based approach using CT data as a noninvasive automatic methodology to categorize metastatic liver lesions according to their unknown primary tumor. Each secondary lesion was segmented by a radiologist in portal phase noncontrast CT images. The experiment was then conducted on a separate set of lesions characterized by different primary sites. Overall, the proposed systems showed promising capabilities even if the accuracy was low. The same group explored the use of DL to stage liver fibrosis (confirmed by biopsy or surgery) using gadoxetic acid–enhanced hepatobiliary phase MR images. The model proved to have high diagnostic performance in the staging of liver fibrosis, and the fibrosis score provided by the algorithm was significantly correlated with fibrosis stage [46].

Conclusions AI has become a very hot topic in the field of diagnostic imaging. Many studies have been published on AI algorithms with one goal: technology will support human intelligence in the interpretation of images. Such software capable of identifying critical findings in the acute setting would be extremely useful, especially in the triage system. The support of a medical device which notifies the most urgent cases or promptly identifies image abnormalities would doubtless result in a shorter time for treatment decision, thus potentially impacting on patient outcomes, especially in time-sensitive diseases. Organ segmentation is a fundamental step, especially in treatment planning. Manual delineation is time-consuming and poorly reproducible. Many studies have proposed a fully automatic method to optimize and standardized the procedure. Finally, a possible application of AI in the interpretation of images is the prediction of an outcome. Such a process may be useful for the early identification of patients who may benefit from more aggressive treatment or for predicting the outcome. Indeed, in many cases the benefit that a patient can have from a certain approach is not clear and state-of-the-art prognostic scores or nomograms are not accurate enough to predict an outcome. In this regard, it is interesting how, with these new technologies, it is possible to integrate clinical and radiological data to achieve a more accurate prognosis.

References 1 Wolterink, J.M., Leiner, T., de Vos, B.D. et al. (2016). Automatic coronary artery calcium scoring in cardiac CT angiography using paired convolutional neural networks. Med. Image Anal. 34: 123–136. 2 Sollini M, Cozzi L, Ninatti G, Antunovic L, Cavinato L, Chiti A, Kirienko M. PET/CT radiomics in breast cancer: Mind the step. Methods. 2020 Jan 21: S1046– 2023(19)30263-4. DOI: 10.1016/j.ymeth.2020.01.007. Epub ahead of print. PMID: 31978538. 3 Avanzo, M., Stancanello, J., and El Naqa, I. (2017). Beyond imaging: the promise of radiomics. Phys Med. 38: 122–139. 4 Forghani, R., Savadjiev, P., Chatterjee, A. et al. (2019). Radiomics and artificial intelligence for biomarker and prediction model development in oncology. Comput. Struct. Biotechnol. J. 17: 995–1008. 5 Lecun, Y., Bottou, L., Bengio, Y., and Haffner, P. (1998). Gradient-based learning applied to document recognition. Proc. IEEE 86 (11): 2278–2324.

6 Krizhevsky, A., Sutskever, I., and Hinton, G.E. (2012). ImageNet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems 25 (eds. F. Pereira, C.J.C. Burges, L. Bottou and K.Q. Weinberger), 1097–1105. Curran Associates, Inc. 7 Russakovsky, O., Deng, J., Su, H. et al. (2015). ImageNet large scale visual recognition challenge. Int. J. Comput. Vis. 115 (3): 211–252. 8 Le Q V, Ranzato M, Monga R, Devin M, Chen K, Corrado GS, et al. Building high-level features using large scale unsupervised learning. In: Proceedings of the 29th International Conference on International Conference on Machine Learning (ICML’12). Madison, WI, USA. Omnipress, 2012. p. 507–514. 9 Shin, H.-C., Roth, H.R., Gao, M. et al. (2016). Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning. IEEE Trans. Med. Imaging 35 (5): 1285–1298.

835

836

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

10 Harley, A.W. (2015). An interactive node-link visualization of convolutional neural networks. In: Advances in Visual Computing. ISVC 2015. Lecture Notes in Computer Science, vol. 9474 (eds. G. Bebis et al.), 867–877. Cham. https://doi.org/10.1007/978-3-319-278575_77: Springer. 11 Ronneberger, O., Fischer, P., and U-Net, B.T. (2015). Convolutional networks for biomedical image segmentation. In: Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015 (eds. N. Navab, J. Hornegger, W.M. Wells and A.F. Frangi), 234–241. Cham: Springer International Publishing. 12 Kingma DP, Welling M. Auto-Encoding Variational Bayes. 2013 Dec 20. arXiv:1312.6114. 13 Goodfellow IJ, Pouget-Abadie J, Mirza M, Xu B, WardeFarley D, Ozair S, et al. Generative adversarial nets. In: Proceedings of the 27th International Conference on Neural Information Processing Systems, Volume 2 (NIPS’14). Cambridge, MA, USA: MIT Press; 2014. p. 2672–2680. 14 Isola P, Zhu J-Y, Zhou T, Efros AA. Image-to-image translation with conditional adversarial networks. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE; 2017. p. 5967–5976. 15 Luc P, Couprie C, Chintala S, Verbeek J. Semantic Segmentation using Adversarial Networks. 2016 Nov 25. arXiv:1611.08408. 16 Sahiner, B., Pezeshk, A., Hadjiiski, L.M. et al. (2019). Deep learning in medical imaging and radiation therapy. Med. Phys. 46 (1): e1–e36. 17 Remedios, S.W., Roy, S., Bermudez, C. et al. (2020). Distributed deep learning across multisite datasets for generalized CT hemorrhage segmentation. Med. Phys. 47 (1): 89–98. 18 Remedios, S., Roy, S., Blaber, J. et al. (2019). Distributed deep learning for robust multi-site segmentation of CT imaging after traumatic brain injury. In: Medical Imaging 2019: Image Processing (eds. E.D. Angelini and B.A. Landman), 9. SPIE. 19 Chang, K., Balachandar, N., Lam, C. et al. (2018). Distributed deep learning networks among institutions for medical imaging. J. Am. Med. Inform. Assoc. 25 (8): 945–954. 20 Sollini, M., Antunovic, L., Chiti, A., and Kirienko, M. (2019). Towards clinical application of image mining: a systematic review on artificial intelligence and radiomics. Eur. J. Nucl. Med. Mol. Imaging 46: 2656–2672. 21 US Food and Drug Administration. DEN170073 Viz.AI. Vol. 90. 2018. p. 1–22. https://www.accessdata.fda.gov/ cdrh_docs/reviews/DEN170073.pdf. 22 Prevedello, L.M., Little, K.J., Qian, S., and White, R.D. (2017). Neuroradiology: artificial intelligence for

23

24

25

26

27

28

29

30

31

32

detection of neurologic disease. Radiology 285 (3): 923–931. Atici, M.A., Sagiroglu, S., Celtikci, P. et al. (2020). A novel deep learning algorithm for the automatic detection of high-grade gliomas on t2-weighted magnetic resonance images. A preliminary machine learning study. Turk Neurosurg. 30 (2): 199–205. Ma, J., Wu, F., Zhu, J. et al. (2017). A pre-trained convolutional neural network based method for thyroid nodule diagnosis. Ultrasonics 73: 221–230. Ma, L., Ma, C., Liu, Y., and Wang, X. (2019). Thyroid diagnosis from SPECT images using convolutional neural network with optimization. Comput. Intell. Neurosci. 2019: 1–11. Nam, J.G., Park, S., Hwang, E.J. et al. (2019). Development and validation of deep learning-based automatic detection algorithm for malignant pulmonary nodules on chest radiographs. Radiology 290: 218–228. Hwang, E.J., Park, S., Jin, K.-N. et al. (2019). Development and validation of a deep learning–based automatic detection algorithm for active pulmonary tuberculosis on chest radiographs. Clin. Infect. Dis. 69 (5): 739–747. Becker, A.S., Marcon, M., Ghafoor, S. et al. (2017). Deep learning in mammography diagnostic accuracy of a multipurpose image analysis software in the detection of breast cancer. Investig. Radiol. 52 (7): 434–440. Vivanti, R., Szeskin, A., Lev-Cohain, N. et al. (2017). Automatic detection of new tumors and tumor burden evaluation in longitudinal liver CT scan studies. Int. J. Comput. Assist. Radiol. Surg. 12 (11): 1945–1957. Yang, X., Liu, C., Wang, Z. et al. (2017). Co-trained convolutional neural networks for automated detection of prostate cancer in multi-parametric MRI. Med. Image Anal. 42: 212–227. Roth HR, Lu L, Seff A, Cherry KM, Hoffman J, Wang S, et al. A new 2.5D representation for lymph node detection using random sets of deep convolutional neural network observations. In: Golland P., Hata N., Barillot C., Hornegger J., Howe R. (eds) Medical Image Computing and Computer-Assisted Intervention – MICCAI 2014. MICCAI 2014. Lecture Notes in Computer Science, vol. 8673 (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 2014. Springer, Cham. https://doi. org/10.1007/978-3-319-10404-1_65. Wu, J., Xin, J., Yang, X. et al. (2019). Deep morphology aided diagnosis network for segmentation of carotid artery vessel wall and diagnosis of carotid atherosclerosis on black-blood vessel wall MRI. Med. Phys. 46 (12): 5544–5561.

Artificial Intelligence in Diagnostic Imaging

33 Avendi, M.R., Kheradvar, A., and Jafarkhani, H. (2016). A combined deep-learning and deformable-model approach to fully automatic segmentation of the left ventricle in cardiac MRI. Med. Image Anal. 30: 108–119. 34 Ibragimov, B., Toesca, D., Chang, D. et al. (2017). Combining deep learning with anatomical analysis for segmentation of the portal vein for liver SBRT planning. Phys. Med. Biol. 62 (23): 8943–8958. 35 Men, K., Geng, H., Cheng, C. et al. (2019). Technical note: more accurate and efficient segmentation of organs-atrisk in radiotherapy with convolutional neural networks cascades. Med. Phys. 46 (1): 286–292. 36 Farag, A., Lu, L., Roth, H.R. et al. (2017). A bottom-up approach for pancreas segmentation using cascaded Superpixels and (deep) image patch Labeling. IEEE Trans. Image Process. 26 (1): 386–399. 37 Liu, M., Li, F., Yan, H. et al. (2020). A multi-model deep convolutional neural network for automatic hippocampus segmentation and classification in Alzheimer’s disease. NeuroImage 208 (December 2019): 116459. 38 Trebeschi, S., Van Griethuysen, J.J.M., Lambregts, D.M.J. et al. (2017). Deep learning for fully-automated localization and segmentation of rectal cancer on multiparametric MR. Sci. Rep. 5301 https://doi. org/10.1038/s41598-017-05728-9. 39 Ding, Y., Sohn, J.H., Kawczynski, M.G. et al. (2019). A deep learning model to predict a diagnosis of Alzheimer disease by using 18 F-FDG PET of the brain. Radiology 290 (3): 456–464.

40 Hwang, E.J., Park, S., Jin, K.-N. et al. (2019). Development and validation of a deep learning–based automated detection algorithm for major thoracic diseases on chest radiographs. JAMA Netw. Open 2 (3): e191095. 41 Ciompi, F., Chung, K., van Riel, S.J. et al. (2017 Apr 19). Towards automatic pulmonary nodule management in lung cancer screening with deep learning. Sci Rep. 7: 46479. https://doi.org/10.1038/srep46479. Erratum in: Sci Rep. 2017 Sep 07;7:46878. PMID: 28422152; PMCID: PMC5395959. 42 Walsh, S.L.F., Calandriello, L., Silva, M., and Sverzellati, N. (2018). Deep learning for classifying fibrotic lung disease on high-resolution computed tomography: a case-cohort study. Lancet Respir. Med. 6 (11): 837–845. 43 McKinney, S.M., Sieniek, M., Godbole, V. et al. (2020). International evaluation of an AI system for breast cancer screening. Nature 577: 89–94. 44 Yasaka, K., Akai, H., Abe, O., and Kiryu, S. (2018). Deep learning with convolutional neural network for differentiation of liver masses at dynamic contrast-enhanced CT: a preliminary study. Radiology 286 (3): 887–896. 45 Ben-Cohen, A., Klang, E., Diamant, I. et al. (2017). CT image-based decision support system for categorization of liver metastases into primary cancer sites: initial results. Acad. Radiol. 24 (12): 1501–1509. 46 Yasaka, K., Akai, H., Kunimatsu, A. et al. (2018). Liver fibrosis: deep convolutional neural network for staging by using gadoxetic acid–enhanced hepatobiliary phase MR images. Radiology 287 (1): 146–155.

837

838

32 Radionuclide Therapies and Correlative Imaging Ashwin Singh Parihar1,2 and Erik Mittra3 1

Department of Nuclear Medicine, Postgraduate Institute of Medical Education and Research, Chandigarh, India Mallinckrodt Institute of Radiology, Washington University School of Medicine, St Louis, MO, USA 3 Department of Diagnostic Radiology, Division of Nuclear Medicine & Molecular Imaging, Oregon Health & Science University, Portland, OR, USA 2

Introduction Radionuclide therapies are currently expanding rapidly due to a variety of factors, including novel targets, their proven efficacy and safety, and the concept of theranostics. The latter is a portmanteau of diagnostics and therapeutics. Accurate noninvasive visualization, using diagnostic molecular targets for functional imaging, is a critical tool for staging, restaging, and response assessment for a large number of cancers. Using those same molecular targets to deliver precision radionuclide therapy is the underlying principle of theranostics in nuclear medicine. Initially used for the treatment of hyperthyroidism and differentiated thyroid malignancies using 131I, the concept of theranostics has since grown to cover a diverse range of malignancies, such as those of prostate, liver, neuroendocrine neoplasms (NENs), painful osseous metastases, and hematologic malignancies, among others. In this chapter, we review the principal radionuclide therapies with regards to their clinical applications, correlative functional imaging, pre- and post-therapeutic considerations, their efficacy, and safety.

­ heranostics in Neuroendocrine T Neoplasms Introduction NENs, originating from the secretory cells of the neuroendocrine system, have an endodermal lineage. These heterogeneous tumors, with widely varying grades and degrees of differentiation and ultimately prognosis, are classified according to their site of origin (e.g. bronchopulmonary,

gastro-entero-pancreatic (GEP), breast, thyroid, uterus, and others)  [1]. The most common among these are the GEP NENs. Based on the embryonic origin, the gastrointestinal (GI) NENs are further classified into foregut (stomach, duodenum), midgut (jejunum, ileum, cecum, appendix, proximal colon), and hindgut (distal colon, rectum) tumors  [2]. Midgut NENs are the most common and, by secretion of vasoactive hormones, most frequently lead to the functional tumors or the characteristic “carcinoid syndrome” constituted by flushing, diarrhea, and, less frequently, right-sided valvular heart disease [3]. NENs are known to abundantly express somatostatin receptors (SSTRs), which have a role in modulating the proliferation and secretory activity of the tumor cells. This has therapeutic implications, as the somatostatin analogues such as octreotide and lanreotide are used for achieving symptomatic control in patients with functional tumors and have antiproliferative properties  [4, 5]. SSTR expression has also been utilized for functional imaging, initially for gamma scintigraphy with diethylenetriamine pentaacetic acid (DTPA)-octreotide (pentetreotide) then onto positron emission tomography (PET) with the 1,4,7,1 0-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) peptides (DOTA-[1-Nal3]octreotide, DOTANOC; DOTA[Tyr3]octreotide, DOTATOC; DOTA-[Tyr3]octreotate, DOTATATE) [6]. Out of the five types of SSTRs (SSTR1–5), SSTR2 forms the predominant receptor in the GEP NENs and is bound by all these receptor agonists. DOTANOC has  demonstrated high affinity and a broader spectrum of  binding to SSTR2, -3, and -5  in comparison to its counterparts [7]. Development of peptide receptor radionuclide therapy (PRRT) is also based on targeting the same SSTR. 111Inpentetreotide was used in initial trials, albeit with limited

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach, First Edition. Edited by Ali Gholamrezanezhad, Majid Assadi, and Hossein Jadvar. © 2023 John Wiley & Sons Ltd. Published 2023 by John Wiley & Sons Ltd.

Radionuclide Therapies and Correlative Imaging

success due to the physical characteristics of the auger electron emissions of this primary gamma-emitter [8]. As newer analogues were developed that could be attached to beta-emitting radionuclides such as 90Y and 177Lu, the efficacy of PRRT also improved significantly [9]. Based on the results of the NETTER-1 trial  [10] and a study from Netherlands [11], Lutathera (177Lu-DOTATATE, Advanced Accelerator Applications Inc., USA) was the first (and so far only) approved PRRT radiopharmaceutical for GEP NENs by the United States Food and Drug Administration (US FDA) in 2018. Therefore, the remainder of this section will focus on 177Lu-DOTATATE-based PRRT. There are, however, ongoing studies exploring the use of PRRT using other agents such as receptor antagonists with higher binding activity to the SSTR or labeled with alpha-emitters such as 225Ac and 213Bi, with initial reports showing promising results [12, 13].

Objectives of Functional Imaging

Histologic Features • Confirmation of NEN • Tumor Grade: Grades 1-3 (Based on Ki-67 index/Mitotic activity) • Tumor Differentiation: Well/Moderately/Poorly Differentiated

Functional Imaging • SSTR PET imaging (68Ga-DOTANOC/DOTATATE PET/CT) - Disease extent, Site, number and size of lesions, lesional radiotracer avidity • 18F-FDG PET/CT - Lesional radiotracer avidity,differential radiotracer avidity among 18F-FDG PET/CT and SSTR PET/CT

Patient Characteristics • Karnofsky Performance Status, ECOG Score • General well being, functional symptoms - frequency, severity • Prior/ Ongoing treatment with cold somatostatin analogues • Improvement/ Deterioration (Follow-up)

Laboratory Parameters

Functional imaging with radiolabeled somatostatin analogues is useful both from a purely diagnostic perspective and as an adjunct to PRRT (selecting patients and assessing response) (Figures 32.1 and 32.2) https://doi.org/10.2967/ (a)

jnumed.120.251512. Presently, the DOTA-peptides mentioned above are typically labeled with 68Ga for PET-based SSTR imaging of NENs in diagnostic and theranostic settings [14]. DOTATATE is also available labeled with 64Cu, which has the advantages of not requiring a generator, a

(b)

• Complete Blood Count • Liver, Renal Function Tests • Serum Chromogranin levels • 99mTc-DTPA for GFR estimation

Figure 32.1

Checklist for PRRT planning.

(e)

(c)

(d)

Figure 32.2 63-year-old male with a pancreatic neuroendocrine neoplasm with liver metastases previously treated with a pancreaticoduodenectomy, adjuvant carboplatin, and liver-directed therapy. Due to continued progression of disease, he was referred for PRRT. 68Ga-DOTATATE PET/CT [maximum intensity projection image (a), axial CT (b), PET (c), and fusion PET/CT (d)] shows extensive SSTR-expressing metastatic disease, with maximal avidity higher than that of spleen. Based on this, he started PRRT. Planar posttherapy scintigraphy image (e) shows 177Lu-DOTATATE accumulation in the same distribution as seen on the pre-therapy PET images, albeit with lower resolution due to gamma-imaging.

839

840

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Table 32.1 Present and evolving indications of functional imaging with somatostatin receptor analogues. Clinical setting

Indication

Suspected NEN

Clinical or biochemical suspicion of NEN with inconclusive findings on conventional imaging and nonavailability of a histologic diagnosis Raising the diagnostic confidence in lesions not amenable to biopsy, but otherwise showing typical characteristics of NEN on conventional imaging Selection of biopsy site in patients with suspected NEN and previous inconclusive conventional image-guided biopsy

Histologically confirmed NEN, baseline evaluation

Initial staging of histopathologically confirmed NEN

Histologically confirmed NEN, re-evaluation

Assessment of residual disease after surgery

Detection of primary tumor in patients with confirmed metastasis Assessment for suitability of PRRT in surgically unfit candidates

Table 32.2

Goals of PRRT in different clinical settings.

Clinical setting

Goal of therapy

Neo-adjuvant

To achieve surgical eligibility in presently inoperable NENs

Adjuvant

Disease control in cases with incomplete resection of the primary NEN Disease control at the time of recurrence after initial surgery

Unresectable, Metastatic NEN

Disease control

Salvage PRRT

Disease control in patients previously treated with PRRT, presenting with recurrence

Symptom control in functional tumors

Table 32.3 Tumor grade assessed by mitotic activity and Ki-67 labeling index (in gastro-entero-pancreatic neuroendocrine tumors).

Tumor grade

Mitotic activity (per 2 mm2)a

Ki-­67 labeling index

Restaging in patients with clinical or biochemical suspicion of disease recurrence

G1 (low)

20%

Assessment for suitability of PRRT as an adjuvant therapy

a 2 mm2 denotes 10 high power fields at 40× magnification with a field diameter of 0.5 mm.Source: Based on Klimstra et al. [15].

Assessment of response to PRRT

much longer half-life, and a shorter positron range than 68 Ga, but the disadvantage of a lower positron branching ratio (0.17 versus 0.89 with 68Ga). A list of practical indications of SSTR imaging is provided in Table 32.1.

Radionuclide Therapy: Indications and Procedure PRRT with 177Lu-DOTATATE requires careful patient screening with the assessment of appropriateness by a dedicated multidisciplinary team, including medical and surgical oncology, nuclear medicine, diagnostic and interventional radiology, endocrinology, and pathology. The individualized goals of therapy (Table 32.2) have to be clearly identified during the patient screening procedure. The initial screening of a patient for PRRT includes review of the tumor histopathology. Tumor grade (assessed by mitotic activity or Ki-67 labeling index; Table 32.3) and degree of differentiation are two key histologic parameters

affecting choice of management and disease prognosis. While tumor grade denotes the proliferative activity of the tumor cells, degree of differentiation assesses their resemblance to the normal counterparts [1]. Patients with G1 or G2 tumors that are well-differentiated are typically suitable candidates for PRRT. 111 In-pentreotide and 99mTc-hydrazinonicotinamide (HYNIC)-TOC based single-photon emission computed tomography (SPECT) imaging or 68Ga- or 64Cu-labeled somatostatin-analogue based PET imaging is performed to noninvasively document SSTR expression on tumor cells  [16]. PET is favored over SPECT due to its higher sensitivity, better radiotracer kinetics, resolution, and dosimetry  [17]. Deoxy-2-[18F]fluoroglucose (18F-FDG) PET/CT, when performed pretherapy, can provide additional useful prognostic information, especially for those with higher-grade or more poorly differentiated tumors [18] (Figure 32.3). The laboratory investigations that have to be performed for patient selection are provided in Table 32.4 with the reference cutoff values [19]. Figure 32.1 provides a checklist for planning PRRT with 177Lu-DOTATATE.

Radionuclide Therapies and Correlative Imaging

Figure 32.3 63-year-old female with biopsy proven metastatic right axillary neuroendocrine tumor underwent 68GaDOTATATE PET/CT which showed somatostatin receptor expressing (SUVmax 5.9) left axillary lesion (a, thick arrow) and a faintly somatostatin receptor expressing (SUVmax 2.6) soft tissue lesion in the posterior basal segment of the left lung lower lobe (not seen on the maximum intensity projection image). Both lesions showed higher avidity on 18F-FDG PET/CT (b, axillary lesion [SUVmax 7.4], thick arrow; lung lesion [SUVmax 4.1], thin arrow), as expected of a higher-grade tumor.

(a)

Table 32.4 Laboratory investigations prior to first cycle of PRRT with 177Lu-DOTATATE. Laboratory test

(b)

Table 32.5 Requirements of amino acid formulation for 177 Lu-DOTATATE-based PRRT.

Acceptable value at baseline

Parameter

Requirement

Hemoglobin

>8 g/dL >2000/mm3

Lysine content

18–24 g

White blood count

>70 000/mm3

Arginine content

18–24 g

Platelet count Total bilirubin

3 g/dL

Osmolarity

50 mL/min

Source: Modified from Hope et al. [19].

Source: Modified from Hope et al. [19].

Therapy Administration

Pretreatment

177

●

●

Anti-emetics such as ondansetron (4 or 8 mg intravenous) or palonosetron (0.25 mg intravenous over 30 seconds) given 30 minutes prior to amino acid infusion reduce the risk of nausea/vomiting associated with the amino acid infusion. Amino acid infusion: Infusion of positively charged amino acids (lysine, arginine) is recommended to reduce the risk of nephrotoxicity. The amino acid infusion should start at least 30 minutes prior to the radiopharmaceutical administration, be continued concurrently with the radiopharmaceutical injection, and last for a total period of 4 hours[19, 20]. Various amino-acid preparations are available and a generic composition required to prevent nephrotoxicity is provided in Table  32.5. The emetogenicity will vary with the specific formulation.

Lu-DOTATATE, diluted in normal saline with an activity of 7.4GBq (200mCi), is administered intravenously over 30minutes during each cycle of therapy. A running intravenous line is preferred for infusing the radiopharmaceutical as this minimizes the risk of extravasation. Appropriate radiation shielding has to be kept in place to avoid extra radiation burden to staff. Care should also be taken to avoid spillage and contamination. The patient should be monitored for any adverse events during the therapy and availability of octreotide, and corticosteroids and intravenous fluids should be ensured to manage carcinoid crisis in the rare event that it occurs.

Post-­therapy: Discharge and Follow-­Up The patient can be discharged after clinical assessment and in line with local radiation safety protocols. A post-therapy whole-body scintigraphy (with or without SPECT), performed

841

842

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

immediately after or within 48hours after the therapy, helps in confirmation of radiotracer uptake at the involved sites. Since significant radionuclide excretion occurs via urine, and precautions to avoid urinary contamination have to be explained to the patients and their care-givers. Post-therapy, patients require laboratory investigations (complete blood count, liver and renal function tests) at 4 (±2) and 8 weeks. Typically, the interval between each cycle of therapy is 8 weeks with a total of four cycles. Diagnostic imaging for response assessment should be done at 1–3 months (may be performed at the end of two cycles if clinically indicated), 6 months, and 12 months after completion of all cycles. National Comprehensive Cancer Network (NCCN) guidelines recommend contrastenhanced abdominal and pelvic CT or contrast-enhanced magnetic resonance imaging (MRI) with chest CT for diagnostic imaging to assess treatment response [21]. The role of SSTR-based functional imaging for response assessment is still being evaluated but can be done as an adjunct to conventional imaging, especially when that is ambiguous.

Efficacy PRRT with 177Lu-DOTATATE has been proven to have good efficacy in NENs in multiple studies across various clinically diverse settings. The NETTER-1 phase 3 trial in patients with midgut NEN compared 177Lu-DOTATATE with 30mg octreotide long-acting repeatable (LAR) in the test group versus octreotide LAR 60mg alone in the control group. The test group had a significantly higher progression-free survival rate of 65.2% (95% confidence interval [CI] 50.0–76.8) at 20months versus 10.8% (95% CI 3.5–23.0) in the control group. The test group also had a significantly higher treatment response rate of 18% versus 3% in the control group. The median progressionfree survival was not reached in the test group and was 8.4months (95% CI 5.8–9.1) in the control group [10]. Similar results were shown by a retrospective study including the Dutch population, in which an overall best objective response rate (including complete and partial response [PR], as per RECIST 1.1) of 39% was seen with 177 Lu-DOTATATE therapy in a total of 443 patients with different NENs. The median overall and progression-free survivals were 63 and 29 months, respectively [11]. Based on the findings of a systematic review and network meta-analysis of multiple randomized control trials (RCTs), the combination therapy of 177Lu-DOTATATE with a somatostatin analogue showed the second highest treatment efficacy in GI NENs, exceeded only by combination therapy with bevacizumab and a somatostatin analogue. Furthermore, combination therapy of 177Lu-DOTATATE with a somatostatin analogue had the lowest hazard for disease progression [22].

177

Lu-DOTATATE-based PRRT has also been shown to improve the quality of life and significant reduction of symptoms in up to 40–70% patients. Additionally, improvement in symptoms, global health status, and Karnofsky performance status is shown to occur irrespective of the treatment outcome [23]. This is important, especially in the setting of functional tumors, where the symptoms lead to significant distress and reduction in the quality of life. The combined benefit of increased survival with symptomatic improvement leads to increased overall patient satisfaction with the therapy. Combination of PRRT with chemotherapy, such as capecitabine, or targeted therapy, such as sunitinib or everolimus, has shown to have better efficacy than monotherapies, although proof from more large-scale studies is needed before definite recommendations can be made [24–27].

Safety Multiple studies have shown the overall safety of 177LuDOTATATE-based PRRT. An overview of the adverse effects associated with PRRT is provided in Table 32.6. The acute side effects, such as nausea and vomiting, are related to the amino acid infusion not the radionuclide therapy, and are well managed with antiemetics. Moreover, the newer amino acid formulations are minimally emetogenic. Hormonal crises after PRRT in functional NENs is a rare adverse effect, with reported occurrence in 75 000/mm3

Liver transaminases

50% decline in PSA levels has been seen in 34–51% of patients [51–53]. In contrast, docetaxel-based chemotherapy, which forms the first line of treatment in metastatic CRPC (mCRPC), has shown a >50% decline in PSA levels in ~48% of patients [54]. Cabazitaxel in the Phase 3 TROPIC trial showed >50% decline in PSA level in ~39.2% of patients. The FIRSTANA trial comparing docetaxel and cabazitaxel as first-line therapy in mCRPC showed >50% decline in PSA levels in ~68.4% patients treated with docetaxel versus ~68.7% in the cabazitaxel group. However, patients currently undergoing 177LuPSMA RLT belong to a heavily pretreated population, having exhausted standard of care therapies, and thus a direct comparison of these parameters should be made with caution. Pooled results from eight studies on 175 patients have shown PR in 37.2%, stable disease (SD) in 38.3%, and progressive disease (PD) in 24.5% of patients [52]. In comparison, PR was noted in 40.5%, SD in 42.8%, and PD in 2.3% of patients receiving cabazitaxel as first-line therapy in mCRPC [55]. The median overall survival with 177Lu-PSMA RLT is reported at ~8.3 months in patients with mCRPC, compared to ~18.7 months with docetaxel therapy [52, 54]. However, again, considerable heterogeneity exists in the patient population of the two groups with regards to the temporal stage of the disease and prior treatment. 177 Lu-PSMA RLT leads to significant symptomatic improvement in patients, with pain due to osseous metastases at baseline. A Phase 2, single-arm trial showed that 177Lu-PSMA RLT led to improvement in pain in all time points in all patients with pain at baseline. The improvement in pain was both in terms of severity and the duration. Additionally, improvement was noted in cognitive functioning and insomnia during the treatment course [56]. The recently concluded phase 3 prospective randomized registry trial of 177Lu-PSMA-617 in patients with mCRPC has shown promising results (Endocyte’s VISION trial). Patients treated with 177Lu-PSMA-617 plus standard of care had a significantly higher overall survival (median 15.3 vs 11.3 months), and progression-free survival (median 8.7 vs 3.4 months) in comparison with those treated with standard of care alone. Although the incidence of significant adverse events (grade-3 or above) was higher with the 177Lu-PSMA-617 arm compared to only standard of care, the quality of life  was  not adversely affected (http://dx.doi. org/10.1056/NEJMoa2107322). These encouraging results combined with the existing data on 177Lu-PSMA RLT are likely to result in FDA approval in the coming year(s).

Table 32.9 Adverse effects with 177Lu-PSMA radioligand therapy in prostate cancer. Adverse effects

Incidence

Salivary gland toxicity [56, 61]

0–87%

Hematologic toxicity [57]

Leukopenia

All grades ~40%, grade 3–4 ~3%

Anemia

All grades ~34%, grade 3–4 ~10%

Thrombocytopenia

All grades ~31%, grade 3–4 ~4%

Nephrotoxicity [61, 62]

All grades 4.5–93%

Overall, 177Lu-PSMA RLT has been shown to be an effective therapy in terms of achieving symptomatic improvement as well as a reduction in tumor burden in patients with mCRPC.

Safety 177

Lu-PSMA-based RLT has been primarily explored in heavily pretreated patients with mCRPC. These patients are already exposed to multiple adverse effects resulting from chemotherapeutic drugs, such as docetaxel and cabazitaxel, including but not limited to hematologic toxicity, cardiotoxicity, paraesthesias, alopecia, and GI toxicity  [54]. In contrast, 177Lu-PSMA RLT is well tolerated with fewer side effects and a low incidence of grade 3–4 toxicities [56, 57]. Xerostomia is a commonly reported adverse effect, resulting from physiologic PSMA expression in the salivary glands. However, most patients experience grade I xerostomia, which is not dose limiting. The incidence of grade 3 hematotoxicity and nephrotoxicity is low, and grade 4 toxicities are extremely rare [58–60]. Table 32.9 provides a list of commonly associated adverse effects with 177Lu-PSMA RLT.

­ IBG-­based Theranostics in Neuroblastoma m and Pheochromocytoma/Paraganglioma Introduction meta-Iodobenzylguanidine (mIBG), a guanethidine analogue of norepinephrine, targets the norepinephrine transporter (NET) which is required for the reuptake of norepinephrine at the presynaptic terminals  [63]. This reuptake process is primarily driven by a saturable, active adenosine triphosphate-dependent system, and to a lesser degree by a passive diffusion pathway  [64]. Once inside the  tumor cells, mIBG is stored in neurosecretory granules  (pheochromocytoma, paraganglioma [PPGL]) via the vesicular mono-amine transporter or stored in the

845

846

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

cytoplasm/mitochondria (neuroblastoma). These tumors derived from the primordial neural crest (e.g. neuroblastoma) or sympathetic nervous system (e.g. PPGLs) show increased expression of NET and thus are potential targets for mIBG-based imaging and therapy. Neuroblastoma is the most common extracranial solid malignancy of childhood, with most cases occurring in infants and children  [65]. The majority of cases present with an abdominal tumor, with around 35% having regional nodal involvement at diagnsosis [66]. Distant metastases are present in 60–70% of patients, requiring aggressive, multimodality treatment (chemoradiation, autologous stem cell transplant). Disease staging is one of the important prognostic factors in neuroblastoma, affecting treatment and survival outcomes  [67]. 123I/131I-mIBG-based imaging is used for the detection of distant metastases prior to considering the patient for surgery and is the gold standard for staging and follow-up of neuroblastoma patients [67, 68]. In patients with high-risk metastatic disease, mIBG-based imaging is also a pre-requisite for evaluation toward high-dose 131I-mIBG therapy. PPGLs originate from the chromaffin cells in the adrenal medulla or the extra-adrenal sympathetic and parasympathetic paraganglia [69]. The functional tumors, accounting for over 50% of all PPGLs, present with adrenergic symptoms (hypertension, deadache, palpitations, arrhythmias), which contribute to significant morbidity and mortality [70, 71]. Surgery is the first-line and potentially curative treatment for localized, primary PPGLs. However, around 10% of patients have unresectable primary tumor at diagnosis, while ~35% present with distant metastases, bearing poor prognoses [72]. These patients are treated with cyclophosphamide-, vincristine-, and dacarbazine-based chemotherapies, which have limited efficacy in terms of objective tumor response rates (complete response [CR] ~4%, PR ~37%) [73]. 131I-mIBG therapy is of value in these patients, having the potential of reducing the morbidity and mortality, and achieving sustained objective tumor response.

Objectives of Functional Imaging 123

I-mIBG is preferred over 131I-mIBG for diagnostic imaging because of its favorable emissions (159 vs. 364 keV) and shorter half-life (13.2 hours vs. 8 days), allowing earlier imaging with a lower radiation burden and permitting higher injectable doses, thereby improving sensitivity. Addition of SPECT/CT to planar imaging improves diagnostic accuracy and anatomic localization of the tumors. 131I-mIBG imaging is then primarily reserved for pre-therapy dosimetry. Oncologic indications of 123I-mIBG imaging are provided in Table  32.10 (see also Figure 32.5). It should be noted that SSTR PET has an evolving role for PPGLs since many of these tumors, especially those with succinate dehydroge-

Table 32.10 Oncologic indications of 123I-mIBG imaging. Clinical setting

Indication

Suspected neuroendocrine origin tumors

Imaging-based confirmation of mIBG avid tumors: neuroblastoma, ganglioneuroblastoma, ganglioneuroma, PPGL, carcinoid, Merkel cell tumors, and medullary thyroid cancer

Confirmed NENs

Initial staging and restaging

Unresectable/ metastatic tumors

Evaluation for high-dose 131 I-mIBG-based therapy: eligibility and dosimetry Evaluation of response to therapy

Source: Based on Bombardieri et al. [74].

nase complex (SDHx) gene mutations, express SSTRs and given the higher resolution of PET over SPECT. However, SSTR PET would not, of course, inform about eligibility for mIBG therapy so it can only be used for diagnostic imaging purposes, or for planning PRRT in mIBG negative tumors.

Radionuclide Therapy: Indications and Procedure 131

I-mIBG therapy is indicated in inoperable or metastatic tumors with sufficient mIBG avidity, as documented on pretherapy 123I/131I-mIBG imaging. Baseline results from conventional imaging (ultrasound, CT, MRI) and biochemical tumor markers should be available for response assessment post-therapy. The primary objectives of 131I-mIBG therapy vary per patient, but commonly include complete or partial disease remission, preventing tumor progression, and relief of functional symptoms, including the reduction of hypertensive medications. It is important to be aware of two different 131I-mIBG preparations: low and high specific activity (LSA and HSA). HSA 131I-mIBG (specific activity ~92.5 MBq/μg) should be preferred wherever available over LSA (specific activity ~1.59 MBq/μg) as HSA has higher efficacy (increased radiation dose delivered to the tumor per unit injected activity) and lower incidence of cardiovascular adverse effects (low levels of unlabeled mIBG) [75]. The laboratory investigations required prior to therapy are provided in Table 32.11 with the reference values.

Pretreatment ●

Interfering medications: Several drug classes are reported to interfere with the uptake and storage of mIBG, including calcium channel blockers, sympathomimetics,

Radionuclide Therapies and Correlative Imaging

(a)

(R) ANTERIOR (L)

(b)

(c)

(L) POSTERIOR (R)

(d)

(R) ANTERIOR (L)

(L) POSTERIOR (R)

Figure 32.5 Eight-year-old female with metastatic left retroperitoneal neuroblastoma treated with multiple therapies, including resection, chemotherapy, external beam radiation, stem cell transplant, and immunotherapy, presented with a left posterior skull mass concerning for recurrent disease. Planar whole-body 123I-mIBG scintigraphy images in anterior (a) and posterior (b) projections show widespread osseous metastases with prominent tracer uptake in the occiput (thick arrow) and right pelvis (thin arrow). She was treated with 131I-mIBG therapy and follow-up anterior (c) and posterior (d) 123I-mIBG scans obtained 1-month after therapy show a marked improvement. Table 32.11 Laboratory investigations prior to 131I-mIBG therapy. Laboratory test

Acceptable value at baseline

White blood count

>3000/mm3

Absolute neutrophil count

≥1000/mm3

Platelet count

>80 000/mm3

Liver transaminases

2.5 times the upper limit of normal

Serum creatinine

2.5 times the upper limit of normal

Estimated glomerular filtration rate (eGFR)

>30 mL/min

Source: Based on Gonias et al. [76], Pryma et al. [77], and Giammarile et al. [78].

●

anti-arrhythmics, anti-psychotics, anti-depressants, and central nervous system stimulants [78]. If clinically feasible, these drugs should be withdrawn before therapy for an appropriate amount of time relative to their halflife and replaced with alternative medications. Thyroid blockade: Adequate thyroid blockade is required to prevent unnecessary irradiation of the thyroid tissue by unlabelled or free radioiodine. This is commonly

●

performed with potassium iodide capsules (adult dose 130 mg/day) or supersaturated potassium iodide (SSKI) drops (adult dose 2–3 drops three times/day) beginning 1–2 days before and continuing for 2–3 weeks after therapy. Potassium perchlorate may be additionally given to facilitate radioiodine washout from thyroid (adult dose 400 mg/day), starting 12 hours prior to and continuing for 1 week after therapy [78, 79]. Adequate hydration: More than 50% of the injected activity of 131I-mIBG is excreted via urine in the first 48 hours  [79]. Adequate hydration is thus required prior to therapy and up to 1 week later to reduce the radiation dose to nontarget organs  [80]. Patients should also be instructed to void frequently after therapy administration. Incontinent patients require prior urinary bladder catheterization.

Therapy Administration Therapeutic dose of 131I-mIBG varies across institutions. A weight-based activity of 450–1850 MBq/kg or a standard activity of 7 GBq is used in children with neuroblastoma, with the availability of autologous hematopoietic stem cell rescue (AHSCR) for myelosuppression  [79]. Adult patients undergoing treatment for PPGLs typically

847

848

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

receive 400–500 mCi (high-dose regimens >600 mCi with AHSCR) 131I-mIBG per treatment cycle  [77, 79]. Bone marrow is the dose-limiting organ in 131I-mIBG therapy, thus the injected activity requires reduction in patients with myelosuppression or impaired renal function and is based on patient-specific pretherapy dosimetry using serial whole-body scintigraphy. Higher activities may be injected when AHSCR support is available. The vital signs (blood pressure [BP], pulse rate) of the patient must be measured immediately prior to, during, and at frequent intervals after therapy. This is especially important for PPGLs, where mIBG administration can lead to unstable BP. Sudden elevation of BP during therapy can be commonly managed by temporarily halting the infusion. Alpha- or beta-blockers may be required in refractory cases. 131 I-mIBG diluted in normal saline is infused slowly (over 45 minutes to 4 hours) via an indwelling intravenous cannula under a lead-shielded infusion system. The infusion line is flushed with normal saline after completion of 131 I-mIBG infusion at the same initial rate [78].

Post-­therapy: Discharge and Follow-­Up Post-therapy whole-body scintigraphy (with or without SPECT/CT) is typically done to confirm tracer uptake in the lesions (delayed imaging at/ after ~4 days) and can also be done for post-therapy dosimetry (serial imaging) [78, 79]. The patients must be monitored regularly for hematologic toxicities, especially during the first 6 weeks after therapy (weekly or higher frequency, depending on administered dose). Temporary myelosuppression is expected in the majority of the patients, but frequent monitoring can help in early institution of supportive therapies such as granulocytecolony stimulating factors (G-CSF), platelets, whole-blood transfusion, and erythropoietin. Patients receiving highdose therapy are more likely to develop prolonged myelosuppression and require AHSCR, the harvesting of which has to be planned and executed prior to therapy [76]. Response evaluation using conventional imaging (CT, MRI) and biochemical tumor markers is typically performed at 3–6 months after treatment. 123I-mIBG imaging can also be done to assess functional response. Retreatment can be planned as early as 3–6 months, provided the blood parameters have recovered and meet the baseline requirement for therapy [80, 81].

Efficacy 131

I-mIBG therapy is instituted mostly in pretreated patients with stage III–IV neuroblastoma, where the prognosis is already poor and the patients are at high-risk of developing adverse effects such as myelosuppression.

In a phase II study of 163 patients treated with relapsed/ refractory neuroblastoma, treated with LSA 131 I-mIBG, 8% had CR, 28% had PR (overall objective response rate 36%) and 27% had PD. The overall survival at 1 and 2 years was 49% and 29%, respectively  [82]. A phase II study using 131I-mIBG therapy in combination with topotecan in 16 treatment-naïve patients with highgrade neuroblastoma showed an overall objective response rate of 57% and primary tumor response rate of 94% after two treatment courses  [83]. A systematic review of 131I-mIBG therapy in relapsed/refractory neuroblastoma showed an overall mean tumor response rate of 32% with an overall survival at 1 year ranging from 38% to 100% [84]. The currently available data suggests that high-dose 131I-mIBG therapy (18 mCi/kg body weight) combined with AHSCR can improve survival and objective tumor response rates in children with grade III/IV neuroblastoma [85, 86]. A phase II study of LSA 131I-mIBG in 50 patients with metastatic PPGLs showed 8% CR and 14% PR. An overall response rate, including patients with some degree of tumor regression was 57%  [76]. HSA 131I-mIBG therapy has been explored in unresectable/metastatic/recurrent PPGLs. A phase II clinical trial showed that 92.2% of patients had PR or stable disease as the best tumor response, with no complete responses. Twenty-five percent of patients could achieve BP control, with the majority receiving two treatment doses [77]. These results were superior to those observed from cyclophosphamide-, vincristine-, and dacarbazine-based chemotherapy in metastatic PPGLs (4% CR, 37% PR)  [73]. The results of this study paved the way for FDA approval of HSA 131 I-mIBG in 2018 for metastatic mIBG-avid PPGLs.

Safety Myelosuppression is the most concerning toxicity associated with 131I-mIBG therapy. Early, temporary myelosuppression is typically observed at 2–4 weeks of therapy, with the nadir values occurring at 4–6 weeks [79]. G-CSF support, wholeblood or platelet transfusions may be required to manage early grade 3–5 toxicities. Around 6–33% of patients may require AHSCR due to persistent myelosuppression [76, 82, 87]. The late adverse effects are similar to those observed with radioiodine (131I-NaI) therapy for thyroid cancer. Despite adequate thyroid blockade, abnormalities in thyroid function tests are observed in up to 22% of patients undergoing 131I-mIBG therapy [88]. Several secondary hematologic and solid tumors have been described in patients on longterm follow-up after 131I-mIBG therapy. However, establishing a definite causality is difficult in these heavily pretreated patients [79–90]. Table 32.12 provides a list of the early and late adverse effects reported with 131I-mIBG therapy.

Radionuclide Therapies and Correlative Imaging

Table 32.12 Early and late adverse effects with 131I-mIBG therapy. Early

Late

Nausea, vomiting (~21–78%) [77, 91]

Hypothyroidism (~5% requiring thyroxine supplementation) [88]

Temporary myelosuppression (~60–87%) [76, 78, 92]

Persistent myelosuppression (~6–33%) [76, 82]

Transient sialadenitis (~41–90%) [77, 93]

Secondary hematologic/solid tumors (~4–14%) [79, 87, 89, 90]

Hypertensive crisis (~0–15%) [76, 77, 94] Othersa [76, 77, 87, 91] Values in parentheses represent incidence rates. a Other toxicities: acute respiratory distress syndrome, bronchiolitis obliterans organizing pneumonia, pulmonary embolism, hypogonadism, GI adverse effects, hypotension, hepatotoxicity, infections, fever.

­ adiopharmaceutical Therapies for Pain R Palliation from Osseous Metastases Introduction Pain resulting from osseous metastases is frequently debilitating and a major source of morbidity in patients with various malignancies. The pathophysiologic basis of metastatic osseous pain in the absence of a fracture is multifactorial and likely the result of an interplay of several factors such as periosteal dilatation, nerve entrapment/compression, mechanical stress on the bone, and release of cytokines and other algesic chemicals in the marrow [95]. This partly explains why the occurrence and severity of pain is not correlated with the size and number of metastases, type of primary tumor or its location [96].

The palliation of osseous pain involves multimodality management involving pharmacologic anti-inflammatory and analgesic agents, local and systemic anticancer medical therapies, invasive procedures, radiation therapy, and radionuclide therapies  [97]. Pharmacologic therapy, including nonsteroidal anti-inflammatory drugs and various opioids, is typically the initial treatment of choice. Surgery and locally directed therapies are useful in cases of solitary painful metastasis or limited oligo-metastatic disease. Radiation therapy provides immediate relief and is useful in the setting of osseous pain with imminent risk of fracture or spinal cord/neural compression. The adverse effects with radiation therapy, especially with total body or hemi-body irradiation, include severe myelo-suppression, and GI and local toxic effects [95]. However, for disseminated skeletal metastases, systemic therapies such as palliative chemotherapy or radionuclide therapy are best suited. Radionuclide therapy has the advantage of lower incidence of adverse effects, ease of therapy administration, prolonged pain relief, and the ability to target all (including subclinical) sites of osseous metastases [95, 98]. The radiopharmaceuticals used in pain palliation include two major categories: inherently bone-seeking radionuclides such as 32P, 89Sr, and 223Ra, or labeled radiopharmaceuticals such as 153Sm-ethylenediamine tetra(methylene phosphonic acid) (EDTMP) and 90Y-citrate. For the second group, the pharmaceutical acts as the bone-seeking agent. The vast majority of these isotopes are beta-emitting and an overview of their physical characteristics is presented in Table 32.13. Some also have gamma-emissions, which provide an advantage for imaging (post-therapy or for dosimetry), albeit at the cost of added radiation dose from the added penetrance of those gamma-photons. 223 Ra (radium dichloride) is currently the only FDAapproved alpha-emitter. As a result of their high linear energy

Table 32.13 Properties of common beta-emitting therapeutic radiopharmaceuticals in pain palliation radionuclide therapy.

Radiopharmaceutical

Production

Half-­life (days)

Energy (Eβmax, MeV)

89

Reactor

50.5

1.46

7

–b

4

103 (28%)

Sr-chloridea

153

a

Sm-EDTMP

Soft tissue penetration (mmmax)

Gamma emission keV (frequency)

Reactor

1.9

0.81

186

Reactor

3.7

1.1

5

137 (9%)

188

Reactor/generator

0.7

2.1

10

155 (15%)

90

Reactor

2.7

2.3

12

–b

Reactor

6.7

0.497

2

208 (11%)

Reactor

14.3

1.7

8

–b

Re-HEDP Re-HEDP

Y-citrate

177 32

Lu-DOTMP/EDTMP a

P-orthophosphate

a

FDA-approved radiopharmaceuticals for metastatic osseous pain palliation therapy. 32P-orthophosphate has been largely replaced by Sr-chloride and 153Sm-EDTMP. b Pure beta-emitters. 89

849

850

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

transfer (~100keV/μm), alpha-emitters deposit a high amount of energy over a much shorter distance, with less potential for marrow suppression and have shown significant benefit in terms of pain relief as well as prolonged survival.

Pretreatment ●

●

Objectives of Functional Imaging Skeletal scintigraphy, most commonly with 99mTc-methylene diphosphonate (MDP), is necessary for determining eligibility for radiopharmaceutical therapy for metastatic osseous pain palliation. The objectives of performing a 99mTc-MDP skeletal scintigraphy include: i) Confirming the presence of osteoblastic metastases or osteolytic metastases with significant sclerotic reaction as evidenced by sites of increased radiotracer uptake. ii) Documenting the degree of tracer avidity of the lesions which reflects the osteoblastic activity and is predictive of response to radiopharmaceutical therapy. iii) Confirming the presence of poly-metastatic disease as local therapies may be better suited for solitary or oligo-metastatic disease. iv) Verifying that the sites of increased radiotracer uptake correspond with the clinical localization of pain.

Radionuclide Therapy: Indications and Procedure

● ●

Therapy Administration After intravenous access, and with care to avoid tracer extravasation, the radiopharmaceuticals (89Sr, 153Sm, 32P, and 223 Ra) are administered as a slow injection over 1–2 minutes, followed by a flush of 10–20 mL of normal saline. 32P can also be given orally. The recommended doses, per cycle of therapy, for the common radiopharmaceuticals are [98, 101]: ●

89

●

153

● ●

The principal indication for radionuclide therapy is the presence of multiple painful osseous metastases that show increased radiotracer uptake on skeletal scintigraphy [99]. The mechanism of pain palliation is the sterilization of inflammatory cells in the metastatic tumor foci, which are responsible for secretion of chemokines and algesic substances  [100]. Tumoricidal activity may be seen with some radiopharmaceuticals, leading to some degree of objective response but it is not the primary objective of therapy. The laboratory cutoff values for this therapy are shown in Table 32.14. Table 32.14 Laboratory investigations prior to radionuclide therapy for metastatic osseous pain palliation.

Laboratory test

Acceptable value at baseline

Hemoglobin

>9 g/dL

White blood count

>3500/mm3

Absolute neutrophil count

≥1500/mm3

Platelet count

>60 000/mm3

Estimated glomerular filtration rate (eGFR)

>30 mL/min

Source: Based on Pandit-Taskar et al. [99].

●

Skeletal scintigraphy: ideally within 8 weeks of therapy, documenting sites of increased abnormal radiotracer uptake, corresponding to clinical localization of pain. Anatomic imaging (CT/MRI): to rule out significant extra-osseous disease and risk of spinal cord compression from the osseous metastases. Refractory metastatic osseous pain: despite analgesic or other pharamacologic therapy. Life expectancy: should be more than 3 months. Prior external beam radiotherapy, chemotherapy, bisphosphonates: last dosage should be at least ≥6 weeks prior to therapy.

Sr: 148 MBq (1.5–2.2 MBq/kg) Sm: 37 MBq/kg 32 P: 185–370 MBq (intravenous), 370–444 MBq (oral) 223 Ra: 50 kBq/kg

Post-­Therapy: Discharge and Follow-­Up Optionally, post-therapy imaging may be performed at 24 hours to confirm the radiotracer biodistribution. This may be done by Bremsstrahlung imaging for pure beta emitters (e.g. 32P, 89Sr) and routine gamma camera imaging for radionuclides with gamma emissions (e.g. 153Sm, 177Lu). The patients have to be informed about the possible occurrence of a “flare phenomenon,” which is the paradoxical increase in osseous pain usually at 2–4 weeks following therapy, possibly due to radiation-induced inflammation. Therefore, the analgesics should be continued after therapy till the time that the osseous pain reduces. The duration of analgesic effect rendered by the therapy is typically 2–6 months  [98]. Patients with pain relief after radionuclide therapy may be planned for retreatment provided they still meet the initial prerequisites for the therapy.

Efficacy Pain relief (of any degree) is reported in ~50–90% of patients [98]. The majority of studies on radionuclide therapy in metastatic osseous pain palliation involve patients with breast or PC as the primary malignancy. A systematic review and meta-analysis reported a single-agent efficacy of radiopharmaceutical therapy

Radionuclide Therapies and Correlative Imaging

as 70% with CR in 27% of patients [102]. The clinical response varies slightly with the radiopharmaceutical used for therapy, although with no significant differences among the commonly used beta-emitting agents. 89Sr (SrCl2) therapy has response rates from 60% to 84%, with onset of pain relief by 1–3weeks and a maximum pain relief duration up to 14months (mean 6months)  [99, 103]. 153Sm therapy has response rates ranging from 40% to 85% [104–106]. A systematic review on metastatic osseous pain palliation with 89Sr, 153 Sm, and 188Re showed complete symptomatic response in 32% (range 8–77%) and PR in 44% of patients [107]. The use of 223Ra (radium dichloride) in a randomized, double-blinded trial (ALSYMPCA) in patients with castration resistant PC and osseous metastases showed a significantly improved overall survival (median 14.9 months vs. 11.3 months in the placebo arm, 95% CI 0.58–0.83), improved quality of life, and lower incidence of adverse effects [101]. The group treated with 223Ra had a 30% reduction in the risk of mortality (HR 0.7, 95% CI 0.58–0.83). The improvement in overall survival with 223Ra is not seen with other betaemitting radiopharmaceuticals although the earlier trials were not necessarily powered for those endpoints.

Safety Myelosuppression, including thrombocytopenia and leukopenia, is the most common adverse effect, with the cell counts approaching nadir at 3–5weeks (153Sm) or 12–16weeks (89Sr, 32P)98. The hematologic toxicity is temporary in most patients, but prior compromised marrow reserves, older age groups, prior myelosuppressive therapy (chemo-radiation), and extensive skeletal marrow involvement are high-risk factors for significant post-treatment myelosuppression. A phase 3 trial with 153Sm for osseous pain palliation in men with metastatic hormone-refractory PC showed mild, transient myelosuppression as the only adverse effect. Posttherapy, the leukocyte counts fell by ~45% and platelet counts by ~40%, returning to baseline values by 8weeks  [108]. Similarly, reduction in leukocyte counts by 11–65% occurs in 12–80% and reduction in platelet counts of ~29% in 29–80% of patients with 89Sr therapy  [107]. Overall, the clinically used beta-emitters have a similar toxicity profile.

­ heranostics in Hyperthyroidism T (Graves’ Disease, Toxic Adenoma, Toxic Multinodular Goiter) Introduction Hyperthyroidism is a syndrome caused by higher than normal levels of serum thyroid hormone (T4 and/or T3). Usually, this causes suppression of thyroid-stimulating

hormone (TSH), which then becomes a sensitive marker of the disease. There are many causes of hyperthyroidism, but broadly it is due to either overstimulation of the thyroid epithelium causing overproduction of hormones or destruction of the thyroid follicles and epithelium resulting in the release of stored thyroid hormones, termed thyroiditis. The most common causes of overproduction are diffuse toxic goiter (Grave’s disease) or nodular toxic goiter, and these tend to be a permanent issue in most patients. The broad divide between overproduction versus destruction as a cause of hyperthyroidism is important for imaging and therapy with radioiodine because overproduction results in increased uptake of radioiodine, which enables treatment, whereas destruction results in low uptake of radioiodine, which precludes its use in treatment [109]. Once hyperthyroidism is biochemically confirmed, the next step is typically a radioiodine uptake and scan [109]. A neck ultrasound (with or without fine-needle aspiration [FNA]) may also be done if there are palpable nodules or if the radioiodine scan demonstrates hyper- or especially hypofunctioning nodules. If the cause of the hyperthyroidism is due to thyroiditis, only symptomatic management is needed as the disease is typically self-limited. If the cause is due to overproduction of thyroid hormone, the first step is medical management with propylthiouracil (PTU) or methimazole (MMI). However, these medications can be difficult to manage and have significant potential side effects, including agranulocytosis, liver failure, and birth defects [110], so they cannot be used in the long term. If the hyperthyroidism doesn’t resolve, definitive treatment options are either surgical removal of part or all of the thyroid gland or radioiodine ablation of the thyroid. Both have various pros and cons [111] which should be carefully discussed with the patient prior to a decision for therapy.

Objectives of Functional Imaging As mentioned, the main indication for a radioiodine thyroid uptake and scan is for further evaluation of biochemically confirmed hyperthyroidism [112] (Figure 32.6). It can also be used to evaluate the functional status of a nodule seen on ultrasound even in the face of normal labs. The rationale is that hypofunctioning nodules are more likely to be cancerous than hyperfunctioning ones and so merit biopsy  [113]. Nowadays, ultrasound features rather than radioiodine uptake often makes this determination, although it remains a consideration. Preparation for an uptake and scan involves being off thyroid medications (PTU or MMI) for 5 days and ensuring that no exogenous iodine (most commonly from a diagnostic CT with intravenous iodinated contrast) has been administered in the

851

852

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

(c)

(d)

6 weeks prior to therapy  [114]. For the same reason, it is ideal if the patient does not have an iodine-rich meal (i.e. seafood or seaweed) in the 48 hours prior to imaging. A true low-iodine diet, however, is typically not necessary. The procedure is composed of two distinct parts: an uptake measurement (nonimaging using a thyroid probe) and a scan (imaging using a gamma camera). Both parts can be done using a single administration of 123I. The uptake can also be done on its own using a smaller (half) dose of 123I. Interpretation of the scan and uptake is relatively straightforward. Uptake values below the normal range (30% at 24 hours) suggest either Grave’s disease or toxic nodular goiter as the gland is active despite a low TSH. In this case, the scan will show diffuse uptake for Grave’s, often with visualization of the thyroid isthmus and pyramidal lobes. For toxic nodule(s), the scan will typically show high uptake in the nodule(s) with variable suppression of the surrounding normal thyroid gland. Because of the latter issue, the uptake value may be in the normal range for

Figure 32.6 Composite image of 123I (NaI) scans of the neck using a pinhole collimator. (a) Diffuse increased uptake with visualization of the isthmus and pyramidal lobe (arrow) in a patient with Graves’ disease. (b) Solitary hyperfunctioning nodule (arrow) with some suppression of the remainder of the normal gland. (c) Multinodular goiter with several hyper-functioning nodules and suppression of the remainder of the normal gland. (d) Thyrotoxicosis with minimal tracer uptake (near background) in the thyroid gland, suggestive of subacute thyroiditis.

nodular goiter if the uptake in the nodule is balanced by the low uptake in the suppressed gland. If the scan and uptake results are clear, then obtaining thyroid function tests at the time of the scan has limited added value. If, however, the uptake and scan are normal, then obtaining concurrent thyroid function tests is useful to know if the thyroid labs are also normal or abnormal at the time of the study [112].

Radionuclide Therapy: Indications and Procedure A hyperthyroid patient would be amenable to therapy with 131 I if the following conditions are met: i) The radioiodine uptake at 24 hours is greater than 30%. ii) The scan shows either diffusely increased uptake or solitary or multiple hyperfunctioning nodules. iii) No hypo-functioning nodules seen which have not already been evaluated by ultrasound and/or biopsy to confirm they are not malignant. The contraindications for radioiodine therapy include pregnant and breast-feeding status. Moderate to severe or sight-threatening orbitopathy is also a contraindication for radioiodine therapy.

Radionuclide Therapies and Correlative Imaging

Pretreatment Prior to therapy, it has to be ensured that the patient has had a thorough discussion with their endocrinologist regarding alternative treatment options, a negative urine or serum pregnancy test within 48 hours of the planned therapy, and appropriate counseling about radiation precautions, including avoiding pregnancy for at least 6 months, ideally 1 year. If the uptake is high normal, but the patient is otherwise a good candidate for radioiodine therapy and had proper preparation, then it may be beneficial to have the patient undergo a low-iodine diet for 1–2 weeks and return for an uptake measurement. The idea is to further stimulate the thyroid gland and increase the uptake value above normal so as to enhance the efficacy of the therapy.

Therapy Administration The dose of 131I for therapy of hyperthyroidism can be chosen either empirically or by calculation [115]. For Graves’ disease, empiric doses are typically in the range of 10–15 mCi and for hyperfunctioning nodule(s) in the range of 20–25 mCi as the latter can be more radioresistant. Alternatively, the dose can be calculated to deliver a specific radiation dose to the gland or nodule(s). For Graves’ disease, the formula is 50–200 μCi per gram of tissue, divided by the radioiodine uptake. For hyperfunctioning nodules, the formula is 150–200 μCi per gram of tissue, divided by the radioiodine uptake. That is, the dose is directly related to the gland size and indirectly related to the uptake. To use the formula then the gland size must be known. Here again there are two methods. The most accurate is to use the measurements from a recent thyroid ultrasound. In this case, the formula for gland size is that of an oval (length × width × height/2). When an ultrasound is not available, a physical examination can be used to estimate the gland size. The goal of the therapy is not necessarily to achieve a euthyroid state, but rather to destroy the gland completely and render the patient hypothyroid. When that happens, they will require thyroid hormone replacement for the remainder of their life. Even with this approach, up to 20% of patients may require a second treatment to be fully effective [111]. Lower doses are correlated to greater treatment failure  [115], but this should be balanced by not giving high radiation doses to everyone, which may increase the risk of side effects (see below) as well as their lifetime radiation burden.

Post-­Therapy: Discharge and Follow-­Up The patient is discharged with a retained activity of less than ~33 mCi corresponding to a dose rate of ~7 mrem/hour at 1m  [115]. The timeline of therapy should be clarified with

patients as 131I therapy requires time for rendering the patient hypothyroid. The patient needs to be followed up with free T4, total T3, and TSH levels at 4–6-weekly intervals for the first 6 months after 131I therapy or until the patient becomes hypothyroid. During this time, symptomatic patients should continue their PTU/MMI and other medications such as beta-blockers (propranolol). Hypothyroidism may occur from 4 weeks onwards, with 40% of patients becoming hypothyroid by 8 weeks and >80% by 16 weeks. Retreatment may be considered if the hyperthyroid state persists beyond 6 months of therapy [115]. Once the patient is rendered hypothyroid, lifelong thyroid hormone supplementation is needed.

Efficacy A prospective study of 131I therapy versus antithyroid drugs showed a 100% success rate (denoted by permanent hypothyroidism) with 131I at 100 months. In addition, 24.4% patients were hypothyroid at 6 months and 62.2% achieved permanent hypothyroidism by 2 years. Overall, 131I had better cure rates and lower relapse rates when compared to antithyroid drugs [116]. The cure rate with 131I in Graves’ disease has a strong correlation with the administered activity: ~61% cure rate with 200 MBq 131I and ~86% cure rate with 580 MBq [115]. A retrospective review of 398 patients with Graves’ disease comparing the outcomes of surgery (n = 103) versus radioiodine therapy (n  =  295) showed an 81.4% success rate following the first dose and an overall 90.1% success rate after additional doses of 131I (NaI). The study also reported a significantly higher incidence of complications (although transient) following surgery (36.9%) than with radioiodine therapy (2.7%)  [117]. A study of 720 patients comparing patients with Graves’ disease managed by radioiodine, antithyroid drugs or surgery showed a similar success rate of radioiodine: 92% versus 52% for antithyroid drugs and 100% for surgery  [118]. Although most of the studies are retrospective in design and have different biases with regards to patient selection and distribution among different treatment groups, the success rate with radioiodine in an appropriately selected patient population has been shown to consistently outperform that with antithyroid drugs and is nearly on a par with surgery.

Safety Table  32.15 lists the adverse effects of radioiodine treatment, which can be described as short- or long-term effects  [119]. Short-term side effects are typically benign and also resolve within the first several days after the therapy. Ensuring adequate hydration and controlling

853

854

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Table 32.15 Adverse effects of radioiodine therapy (131I) for hyperthyroidism. Short-­term adverse effects

Long-­term adverse effects

Nausea

Persistent dry mouth

Sialadenitis

Development of malignancy (rare)

Swelling of the thyroid gland Dysgeusia Generalized malaise Thyroid storm (rare)

symptoms with nonsteroidal anti-inflammatory medications and/or antiemetics can mitigate these effects. Sialagogues such as sour candies or lemon have been advocated to reduce the risk of sialadenitis, but the timing of their use is not well understood [120]. Several studies show that starting sialagogues immediately after 131I administration (and continuing for several days) can reduce the uptake of 131I in the salivary gland [121, 122]. However, one study showed that starting the sialagogues within 24 hours of the 131I administration can actually increase uptake in the salivary gland (the so-called “rebound effect”), resulting in greater side effects [123]. Long-term side effects of radioiodine therapy include prolonged or permanent damage to the salivary glands causing persistent dry mouth. Infrequently, this can present itself several months after the therapy even in patients who did not initially experience sialadenitis. Again, the use of sialagogues has been shown to reduce these effects. Also, rarely, alteration in taste can be more prolonged or rarely permanent. There is a theoretical risk of developing a malignancy from the radiation, but this is more of a concern at the higher doses used for thyroid cancer not with the low doses used for hyperthyroidism.

­ heranostics in Differentiated T Thyroid Cancer Introduction The incidence of thyroid cancer has been increasing over time, although the mortality has been relatively stable except for advanced-stage disease [124, 125]. There are several pathologic subtypes of thyroid cancer, although the most common are the differentiated thyroid cancers, including papillary and follicular, together accounting for over 90% of thyroid cancers. This is fortuitous because these malignancies have a favorable prognosis and are amenable to radioiodine therapy. Other subtypes, such as medullary

thyroid cancer, anaplastic thyroid cancer, and other poorly differentiated thyroid cancers, have a worse prognosis and do not take up radioiodine. The remainder of this section pertains only to differentiated thyroid cancers. The most commonly used staging classification for differentiated thyroid cancer is the American Joint Committee on Cancer Tumor Node Metastasis (AJCC/TNM) staging system, which was revised in October 2016 as the 8th edition. The primary change in the new edition was the downstaging of a significant number of patients into lower stages, reflecting their low risk of thyroid cancer-related death [126, 127]. Thyroid cancers do not typically cause symptoms and are often found incidentally during a physical examination or by imaging of the neck done for other reasons. However a suspicious nodule is first identified, the next step is ultrasound-guided FNA followed by a core biopsy. Once differentiated thyroid cancer is confirmed by pathology, the next step is typically total thyroidectomy, although this varies based on size of the primary tumor [128]. It is not uncommon for some normal thyroid tissue to be left behind either inadvertently or on purpose after surgery. The next step in treatment is potential radioiodine treatment with high-dose 131I.

Objectives of Functional Imaging In general, 123I (or low-dose 131I if 123I is not available or for dosimetry purposes) whole-body scans are done in conjunction with a possible 131I therapy either after initial thyroidectomy for the purpose of remnant ablation or adjuvant therapy, or subsequently to treat residual/recurrent disease (Figure 32.7). Routine surveillance 123I whole-body scans in patients without rising thyroglobulin levels is generally not recommended [129]. The use of an 123I pretherapy scan at the time of the initial 131I therapy is one of several controversies in the imaging and therapy of differentiated thyroid cancer  [130]. Some centers will routinely treat all thyroid cancer patients with a 100–150 mCi dose of 131I after thyroidectomy, in which case doing a pretherapy 123I scan is arguably of limited utility. However, other centers base the dose of the initial therapy on the extent of disease seen on the pretherapy scan  [131]. There are other advantages as well. Sometimes the amount of residual thyroid tissue is high enough to warrant a repeat surgery to ensure the effectiveness of the therapy. Also sometimes (usually young) female patients will have prominent physiologic uptake in breast tissue, during which time it is not advisable to do the therapy as the radiation exposure to the breast tissue may be too high. These are examples of issues which would not be known if the pretherapy 123I scan was not done. Regardless of the specific indication for an 123I whole-body scan, the

Radionuclide Therapies and Correlative Imaging

Figure 32.7 28-year-old female post thyroidectomy for papillary thyroid carcinoma. She received 100 mCi of oral 131I (NaI) for ablation of residual thyroid tissue and treatment of cervical nodal metastases. The planar whole-body scans obtained in the anterior (a) and posterior (b) projections, taken 1 week after the therapy, show uptake of radioiodine in the residual thyroid tissue and regional nodal metastases (arrow). No distant metastases are seen. Low-dose 131 I (NaI) whole-body planar images (anterior, c; posterior, d) taken 1 year after therapy show physiologic tracer distribution with no residual uptake in the neck.

(a)

preparation and technique are generally the same. The patient preparation involves ensuring sufficiently raised TSH levels (at least >30 mIU/L, preferably >50 mIU/L), low iodide pool in the body (low-iodine diet for 1–2 weeks, no recent iodinated contrast-based imaging).

Radionuclide Therapy: Indications and Procedure The treatment decision for high-dose radioiodine (131I) therapy is based on the surgical pathology, molecular features, other surgical findings, stage, and other medical conditions and comorbidities. If done, concurrent laboratory results (serum thyroglobulin, antithyroglobulin levels with serum TSH), 123I pretherapy whole-body scan findings, and other imaging (i.e. CT chest) are also incorporated. Patients with very low-risk disease may defer treatment until their thyroglobulin/antithyroglobulin levels start rising. For patients who will be treated, the next step is to decide on a dose. The goals of radioiodine treatment are given in Table 32.16.

(b)

(c)

(d)

Table 32.16 Goals of 131I treatment in patients with differentiated thyroid carcinoma. Clinical setting

Remnant ablation

Goal

Improving the specificity of serum thyroglobulin/antithyroglobulin as tumor markers Improving the specificity of 123I scanning for detection of residual/recurrent metastatic disease by sterilization of the normal residual thyroid tissue

Adjuvant treatment

Eradication of any subclinical tumor deposits that might persist after surgical resection of the known tumor and its metastases

Treatment of known disease

Treatment of known malignant disease either visualized on imaging (conventional or 123I scanning) or based on elevated thyroglobulin levels, especially when not surgically amenable

Pretreatment Considerations prior to treatment with high-dose 131I are as follows: 1) Initial risk-stratification per the American Thyroid Association (ATA) guidelines. 2) Laboratory values: stimulated TSH, thyroglobulin, and anti-thyroglobulin.

3) Anatomic imaging: neck ultrasound, chest CT. 4) Functional imaging: pretherapy 123I whole-body scan (±SPECT/CT). 5) Ensuring no iodine interference: history, laboratory testing as needed for serum and urinary iodide levels.

855

856

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Therapy Administration 131

The dose of I can be empirical or based on the extent of disease [132]. Empiric dosing is typically done in the initial post-thyroidectomy setting for remnant ablation or adjuvant therapy in the range of 100–150 mCi. If based on the extent of disease, it is helpful to incorporate other imaging such as the pretherapy 123I whole-body scan and other conventional imaging. A higher dose will be more efficacious for patients with higher-risk disease. This is especially of concern in that repeated therapies may become less effective as each treatment selects for the radioresistant clones of cancer. Using this rational, the typical range for doses is 30–250 mCi based on the extent of disease: i) 30–50mCi for ablation of residual normal thyroid tissue ii) 75–150 mCi if there is the possibility of residual cancer in the neck (extracapsular extension, lymphovascular invasion, or lymph node metastases) iii) 150–200 mCi if there are lung metastases iv) 200–250 mCi for bone metastases. With respect to high doses for patients with extensive metastases, ultimately a limit must be reached so radiation to normal surrounding tissues is not harmful. The two primary areas of concern are bone marrow suppression and radiation pneumonitis resulting in pulmonary fibrosis. For these patients, pretherapeutic dosimetry should be used to calculate an individual patient’s maximum tolerated dose (MTD) or maximum tolerated activity (MTA). Based on each patient’s biology, renal function, and distribution and burden of metastases, the same dose of 131I will have different residence times in that individual and therefore different radiation exposure to normal organs. It should be noted, however, that this is another area of controversy as some studies show that pretherapy dosimetry is helpful [133], while others do not [134].

Post-­therapy: Discharge and Follow-­Up The discharge limits are similar to those in hyperthyroid patients treated with 131I. Thyroid cancer patients treated with high-dose 131I should undergo a post-therapy scan 5–8 days after the treatment is given. The expectation is that the post-therapy scan will show uptake in the neck in the same distribution as the pretherapy 123I scan (if done), although the uptake in the residual thyroid tissue or cancer will be much more intense. The larger goal is to look for distant metastases, if any, that were previously unknown. The detection of additional metastases helps in tailoring future management and affects prognosis. Addition of SPECT/CT can further aid in accurate delineation of the disease and aid in distinguishing sites of pathologic uptake

from physiological ones. Follow-up of patients with thyroid cancer is typically life-long, with the follow-up interval being ~6 months initially and gradually increased to ~1- or 2-yearly follow-up [120]. The modalities used for follow-up commonly include serum thyroglobulin and antithyroglobulin levels with corresponding TSH values and ultrasound of the neck. Other imaging modalities may be utilized in case-appropriate scenarios. The patients are kept on suppressive doses of levothyroxine based on the initial risk and dynamic risk stratification guiding the TSH levels [120].

Efficacy Radioiodine (131I) therapy has been shown to improve overall survival in patients with intermediate or high-risk differentiated thyroid cancer. In a study of 21 870 patients categorized as ATA intermediate risk papillary thyroid cancer, the use of 131I was associated with improved overall survival (P < 0.001) and a 29% reduction in risk of mortality (hazard risk 0.71, 95% CI 0.62–0.82, P < 0.001)  [135]. Among high-risk patients, the benefit of 131I therapy is more evident. The National Thyroid Cancer Treatment Cooperative Study Group showed an improved overall survival in stage III patients treated with 131I (relative risk 0.66, 95% CI 0.46–0.98)  [136]. In a retrospective study of 444 patients with metastatic thyroid cancer, 43% of 295 patients with radioiodine uptake achieved resolution of radioiodineavid disease. The 10-year survival rate in this group with resolution of radioiodine-avid lesions was 92% versus 19% in those with persistent disease [137].

Safety The adverse effects from 131I therapy of thyroid cancer are generally the same as for 131I therapy of hyperthyroidism. As noted before, some studies show a dose relationship to side effects such that these side effects may be more prevalent in patients treated for thyroid cancer [120]. The primary concern for long-term side effects of radioiodine therapy for differentiated thyroid cancer is the development of second malignancies due to the radiation. Overall, there is increased risk of secondary cancer in thyroid cancer survivors and the type of cancer depends on whether patients receive radioiodine therapy, age of diagnosis and treatment, and latency since treatment  [138]. Leukemia is the best-established link for those who have received radioiodine treatment  [139, 140]. Most leukemia cases occur within 10 years of initial diagnosis. The link to other cancers is less strong, but some that have been reported include renal and colorectal cancer since radioiodine is secreted through the kidney and GI tracts. Additionally, breast and PC have been suggested, although there are conflicting reports since

Radionuclide Therapies and Correlative Imaging

the risk is not statistically different in thyroid cancer patients who received or did not receive radioiodine treatment. Some of the increased detection of secondary cancer can be attributed to genetic and environmental factors as well as sampling bias due to increased surveillance.

­ elective Internal Radiation Therapy S for Liver Tumors

decompensated disease, and in the setting of portal vein thrombosis. The commonly used radionuclides used in SIRT are listed in Table 32.17. SIRT also has a role in the treatment of liver metastases, commonly from primary colorectal carcinoma (CRC). Liver metastases are the most common cause of mortality in patients with CRC and ~85% of patients with CRC and liver metastases have unresectable liver lesions [142]. SIRT thus offers an alternate therapeutic modality, prolonging survival and improving quality of life in these patients.

Introduction

Objectives of Functional Imaging

Primary hepatocellular carcinoma (HCC) is the sixth most commonly diagnosed malignancy, with an increasing incidence of ~2–3% annually. Primary HCC also has one of the lowest survival rates (for all stages combined), with a 5-year relative survival rate of 18%  [65]. Surgical resection of the affected segment/lobe is the ideal treatment with the goal of achieving adequate resection margins and preservation of sufficient remnant normal parenchyma to sustain physiologic functions and avoid post-operative liver failure. However, factors such as large tumors (>5cm in size), multifocal involvement, vascular invasion, and metastatic spread often lead to unresectable tumors, portending a poor prognosis [141]. Liver transplantation is the only potential curative option for patients with unresectable disease limited to the liver. However, decompensated cirrhosis, advanced disease, and clinical worsening often lead to ineligibility for liver transplant. Selective internal radiation therapy (SIRT) is a form of trans-arterial locoregional radiation therapy that utilizes selective delivery of radioactive microspheres to the tumor vasculature (radioembolization). The normal liver parenchyma receives ~80% of its total blood supply by the portal vein, whereas hepatic malignancies receive 80–100% of their blood supply from the hepatic arteries. Because of the selective (and at times supra-selective) trans-arterial delivery of the radioactive microspheres to the tumor vasculature, SIRT can  be performed in patients with multifocal disease and

99m

Tc-macroaggregated albumin (MAA) scintigraphy is necessary prior to treatment with 90Y-microspheres  [143] (Figures 32.8 and 32.9). The objectives of performing a pretherapy 99mTc-MAA scan include: i) Evaluation and quantification of hepato-pulmonary vascular shunt fraction. ii) Identification of potential extra-hepatic sites where radiation may be deposited, such as the upper abdominal organs. iii) Optimization of activity to be delivered by calculating the tumor-to-background absorbed dose ratio.

Radionuclide Therapy: Indications and Procedure The current clinical and experimental indications of SIRT are provided in Table 32.18. SIRT has been used in the palliative setting for treatment of primary HCC. It is also used in the treatment of hepatic metastases from primary CRC. Its use in the neo-adjuvant setting prior to resection or liver transplantation, for maintaining the disease stability or for downstaging has shown encouraging results. However, this indication has not been approved for routine clinical use and continues to be used in the setting of research trials.

Table 32.17 Properties of common radionuclides used in SIRT for liver malignancies.

Soft tissue penetration (mmmax)

Gamma emission keV (frequency)

Radionuclide

Compounds

90

Microspheres (resin/glass based)

2.7

2.3

12

–a

131

Lipiodol

8.0

0.61

2.4

364 (81%)

188

Lipiodol

0.7

2.1

10

155 (15%)

Y

I Re

a

Half-­life (days)

Energy (Eβmax, MeV)

Pure beta-emitter.

857

858

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

(a)

(b)

Liver Art Liver Post

(c)

(d)

Lungs Post

Lungs Art

(a)

(b)

Figure 32.8 80-year-old male with choroidal melanoma metastatic to the liver presenting for pre-radio embolization imaging. This study was done to assess hepatic arterial perfusion prior to intra-arterial delivery of 90 Y-microspheres. The planar images show expected nonuniform perfusion of the right and left lobes of the liver (a, b), without extrahepatic uptake. The anterior and posterior planar images are used to calculate relative lung perfusion (calculated as the ratio of lung activity to lung + liver activity using the regions of interest shown, c and d); the relative lung perfusion is 5.7% (normal 20%).

(c)

Figure 32.9 88-year-old male with metastatic melanoma to the liver. This examination was done after therapy to assess the distribution of 90Y-microspheres administered intra-arterially. The trans-axial CT (a), SPECT (b) and fused SPECT/CT (c) imaging of the Bremsstrahlung radiation confirms localization of 90Y-microspheres to the liver, predominantly to hepatic segments IVA and IVB (arrow). There is also heterogeneous scattered tracer activity within the peripheral medial aspects of segments II and III.

Pretreatment The tests/procedures required prior to SIRT are outlined in Table 32.19.

Therapy Administration The desired activity of the radionuclide is administered via a selective (or supra-selective) intra-arterial catheter as a continuous infusion to ensure that the radiation dose is deposited to the tumor, sparing the normal hepatic parenchyma.

Post-­therapy: Discharge and Follow-­Up ●

●

●

Prophylactic medications: anti-ulcer medication (for 2 weeks following treatment) and steroids (for 1 week following treatment). Post-therapy whole-body scintigraphy (optional): to confirm the desired distribution of the radionuclide. It may be performed at 1 week after 131I-Lipiodol therapy or a few hours after 90Y-microsphere therapy. Clinical, biochemical, imaging follow-up: for prompt identification of any adverse events (at 1 month) and response assessment (2- to 3-month intervals).

Radionuclide Therapies and Correlative Imaging

Table 32.18 Indications of SIRT. Status

Indication

Clinical

Palliative treatment of histopathologically confirmed, inoperable primary hepatocellular carcinoma Treatment of hepatic metastases from primary colorectal carcinoma

Experimental

Bridging patients awaiting liver transplantation Downstaging patients for eligibility toward liver transplantation Neo-adjuvant therapy for eligibility toward hepatic resection

Source: Based on Giammarile et al. [144].

Table 32.19 Procedures and tests required prior to SIRT. Procedure

Description

Laboratory tests

Serum liver enzymes, coagulation parameters, total blood count, renal function tests, and baseline tumor markers

18

F-FDG PET/CT

Exclusion of extrahepatic metastases and assessment of hepatic tumor burden

CT angiography

For identification and selection of vessel for access and SIRT activity administration, identify anatomic vascular variants, intervention planning if required (e.g. coiling), checking for portal venous thrombosis

99m Tc-MAA scintigraphy

Functional imaging for dosimetry and shunt assessment

Source: Based on Giammarile et al. [144].

SIRT used as salvage therapy in patients with chemotherapy refractory liver metastases from CRC has shown objective response rates of 24–58.6% and median overall survival (OS) of 8.3–12.6 months [148, 149]. In a study of 58 patients, those given SIRT with best supportive treatment had a median OS of 8.3 months vs. 3.5 months in patients only on the best supportive treatment (P < 0.001)  [148]. SIRT as first-line therapy, with chemotherapy (5-fluorouracil, leucovorin, oxaliplatin) has been shown to have significantly improved survival and objective response rates in patients with hepatic metastases from CRC in comparison to chemotherapy alone [150, 151].

Safety SIRT is usually well tolerated, except for common adverse effects such as nausea, abdominal pain, and fever. The serious adverse effects result from the off-target delivery of the radioactive microspheres to the adjacent organs, such as gallbladder, stomach, duodenum, lungs, and especially the normal liver parenchyma (Table  32.20). The pretherapy scan helps limit off-target delivery by indicating if any aberrant arteries need to be ligated prior to the therapy.

­ adioimmunotherapy in Hematologic R Malignancies Introduction Radioimmunotherapy (RIT) involves a similar approach to the radionuclide therapies discussed thus far but attaches the therapeutic radioisotope to an antibody which delivers the radiation to tumor cells expressing a particular antigen. Antibody-dependent cellular-mediated cytotoxicity (ADCC)

Efficacy SIRT when compared with trans-arterial chemoembolization (TACE) in intermediate-stage patients with unresectable primary HCC showed a best overall response rate of 30.8 vs. 13.3% and disease control rates of 76.9 vs. 73.3%, respectively  [145]. Another study including Barcelona Clinic Liver Cancer (BCLC) stage A or B patients showed a significantly longer time to disease progression (>26 months) in patients who underwent SIRT versus those who underwent conventional TACE (6.8 months, P  =  0.0012)  [146]. In a study of 86 patients treated with either TACE or SIRT (90Y) for downstaging (liver transplantation eligibility), SIRT downstaged 58% patients vs. 31% for TACE. Event-free survival was also significantly higher for SIRT (17.7 months) than for TACE (7.1 months, P = 0.0017) [147].

Table 32.20 Adverse effects of SIRT. Short-­term adverse effects

Long-­term adverse effects

Acute pain, nausea

Gastro-duodenal ulceration

Anorexia, malaise, fever

Cholecystitis

Transient elevation of transaminases

Pancreatitis

Radioembolizationinduced liver disease Radiation pneumonitis Source: Based on Stubbs et al. [152].

859

860

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

already affects those cells to which the antibody directly binds. However, cells having a reduced expression of antigen or internally located cells in a poorly vascularized, bulky tumor may escape from these cytotoxic effects. However, RIT induces cellular killing, not only in cells that directly bind with the antibody, but adjacent cells as well due to the bystander effect of radiation. Thus, RIT combines the efficacy of ADCC and radiation-induced cytotoxicity. The clinical applications of RIT have been most successful in hematologic malignancies such as leukemias and lymphomas. RIT has been tried in several solid tumors as well, but the results have not been as encouraging. The reason for this discrepancy is likely because hematologic malignancies are more radiosensitive than solid tumors and the radio-resistant effects of hypoxia (due to increased tumor bulk) are more frequent in the latter. As such, to achieve a similar objective response, solid tumors require up to 5–10 times higher deposited radiation dose in comparison to that in hematologic neoplasms [153]. The following considerations are important for RIT [153]. 1) Radionuclide: The commonly used beta-emitting radionuclides are preferred where deeper tissue penetration is beneficial (e.g. bulky disease). Thus far, 90Y and 131I have been used in over 95% of the clinical trials with RIT. Table 32.21 lists the common beta-emitting radionuclides used in RIT. They have favorable emission characteristics and radiolabelling properties, abundant prior clinical safety data, and are readily available. The alpha-emitters (e.g. 225Ac, 213Bi, 211At) have higher relative biologic effectiveness and deposit a very high radiation dose over a short distance (high linear energy transfer). The cytotoxic potential of the alpha-emitters is also unaffected by tumor hypoxia, in contrast to that of the beta-emitters. They are thus helpful in malignancies with single cell tumors or small cell clusters, such as leukemias. 2) Antigen: The antigen of choice should have a high (>100 000 sites per cell) uniform expression on the

Table 32.21

The first RIT agents approved for treatment of tumors were 131I-tositumomab (Bexxar®) and 90Y-ibritumomab tiuxetan (Zevalin®). Both these agents target the CD20 antigen present on the mature B cells and were approved for the treatment of relapsed/refractory low-grade follicular B-cell lymphomas. 177Lu-lilotomab satetraxetan (Betalutin®) was granted fast-track designation by the FDA in 2020 for the treatment of adult patients with relapsed/refractory marginal zone lymphoma.

Objectives of Functional Imaging Functional imaging with lower activities of the radionuclide-labeled antibodies are required for pretherapy dosimetry and verifying biodistribution (Figure 32.10). Marrow toxicity is dose-limiting in patients where a nonmyeloablative treatment approach is being considered. Bone marrow biopsy and 18F-FDG PET/CT are utilized in a pretherapy setting to estimate the marrow disease burden as >25% marrow involvement is a relative contraindication for RIT [153]. 18F-FDG PET/CT is also utilized for assessing response to treatment in comparison to the baseline study, as it is for any therapy for lymphoma.

Properties of common beta-emitting radionuclides used in RIT.

Radionuclide

Antibodies

Half-­life (days)

90

Energy (Eβmax, MeV)

Soft tissue penetration (mmmax)

Gamma emission keV (frequency)

Ibritumomab

2.7

2.3

12

–a

131

Tositumomab

8.0

0.61

2.4

364 (81%)

177

Lilotomab

6.7

0.497

2

208 (11%)

Y I Lu

a

tumor cells and low to no expression on normal cells. The antigens should not be released into the circulation and the metabolic fate of the antigen–antibody complex should be well established. 3) Antibody: The antibody (or antibody fragment) of choice should have a high binding affinity for the antigen pair, permitting high selectivity and high immunoreactivity (>90%). The antibody binding with the normal cells should be negligible. A desirable therapeutic index should be >10 (kidneys) and >50 (red marrow). 4) Targeted malignancy: As previously discussed, RIT has higher efficacy in hematologic malignancies in comparison to solid tumors. Further, intracompartmental approaches to RIT have better response rates in solid malignancies in comparison to intravenous administration.

Pure beta-emitter.

Radionuclide Therapies and Correlative Imaging

(a)

(b)

(c)

(d)

Figure 32.10 66-year-old female with follicular lymphoma. This study was done using 111In-ibritumomab tiuxetan to determine the expected biodistribution of 131I-ibritumomab tiuxetan. Planar images of the whole body are obtained in the anterior (a, c) and posterior projections (b, d) 48 hours post intravenous injection of 5.3 mCi 111In-ibritumomab tiuxetan. Apart from the normal biodistribution in the blood pool and liver, there is increased activity along the bilateral iliac and inguinal (dotted arrows), supraclavicular, axillary (solid arrows), and mediastinal nodal regions represents known sites of lymphomatous involvement.

Radionuclide Therapy: Indications and Procedures RIT is typically used in the setting of relapsed/refractory B-cell lymphomas. However, with increasing evidence of its efficacy, RIT is also being integrated along with chemotherapy as first-line therapy in lymphomas, where it has shown objective response rates of 90–100% and CRs of 60–100%  [154–157]. The efficacy of RIT can be further improved by using myeloablative doses with marrow rescue, using pretargeted RIT (PRIT) and targeting multiple antigens at a single time-point. PRIT involves multistep targeting, where the cold antibody is administered first and allowed to bind to the tumor antigens throughout the body. Next, a radionuclide bound with a highly selective moiety is administered that binds to the antibody, thus localizing to the antigen–antibody complex. A cold antibody preparation is administered prior to RIT to saturate the circulating cellular antigens, thereby minimizing the deleterious background radiation. Administering cold antibodies prior to RIT does not diminish the efficacy of therapy as no saturation of tumor antigens has been observed.

Efficacy The efficacy of RIT in B-cell lymphomas in some key clinical studies is outlined in Table 32.22.

Safety RIT is well-tolerated with reversible hematotoxicity being dose-limiting in the setting of non-myeloablative radiation doses. Delivery of myeloablative doses with the availability of AHSCR improves the efficacy of RIT and avoids the commonly encountered hematologic toxicities. The early and late adverse events observed with RIT are outline in Table 32.23.

­Future Directions The scope of radionuclide therapy has been expanding rapidly, with the underlying foundation of theranostics, the amalgamation of diagnostics, and therapeutics using complementary agents. While a comprehensive list of current research targets is beyond the scope of this chapter, some of the most promising theranostic targets are discussed. The

861

862

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

Table 32.22

Key results of clinical studies evaluating efficacy of RIT in B-cell lymphomas.

RIT agent

Treatment approach

Target antigen

Results

131

Nonmyeloablative

CD20

60–80% ORR;15–40% CR

90

Nonmyeloablative

CD20

60–89% ORR;15–40% CR

90

Nonmyeloablative

CD22

61% ORR;48% CR

177

Nonmyeloablative

CD37

61% ORR;30% CR

131

Myeloablative

CD20

2 years PFS 68%;OS after 2 years 83%

90

Myeloablative

CD20

2 years PFS 78;%2 years OS 92%

I-tositumomab [158–160]

Y-ibritumomab tiuxetan [161, 162] Y-epratuzumab [163] Lu-lilotomab satetraxetan [164] I-tositumomab [165]

Y-ibritumomab tiuxetan [166]

CR, complete response rate; PFS, progression-free survival; ORR, objective response rate; OS, overall survival.

Table 32.23

Early and late adverse effects observed with RIT.

Short-­term adverse effects

Long-­term adverse effects

Hematologic toxicity (dose-limitinga)

Elevation of TSH, hypothyroidismb

Fever, asthenia, nausea, chills

Myelodysplastic syndromes

Infusion reactions

Secondary solid malignancies Human anti-mouse antibody reaction

a

In patients with nonmyeloablative treatment approach. I-labeled agents only.

b 131

chemokine receptor subtype-4 (CXCR4) has been studied and found to be overexpressed in several solid and hematologic malignancies  [167]. Additionally, it represents an alternative target in patients with NENs that do not have high SSTR expression. 68Ga-pentixafor has been studied for CXCR-4 targeted PET imaging with 177Lu-pentixather its therapeutic counterpart  [168]. 177Lu/90Y-pentixather showed significant favorable response in heavily pretreated patients of multiple myeloma in a first-in-human study [169], and the agent is presently under investigation in several clinical trials. The αvβ3 integrin is overexpressed on the surface of the endothelial cells in the newly developing vessels and is vital for neo-angiogenesis, a key process in tumor spread. 68 Ga-RGD (tripeptide sequence of arginine-glycineaspartate), targeting the αvβ3 integrin, has been used for PET imaging of several solid tumors, such as lung, breast, head and neck, brain, and thyroid malignancies, among others  [170, 171]. Its therapeutic counterpart 177Lu-RGD has been shown to have favorable pharmacokinetic and dosimetry characteristics in preclinical models  [172] and demonstrated favorable therapeutic response in its first-inhuman study [173].

Melanocortin-1 receptor (MC1-R), a G-protein coupled receptor, is abundantly expressed in both melanotic and amelanotic melanomas, making it a suitable target for both imaging and therapy [174]. The radiolabeled α-melanocyte stimulating hormone (α-MSH) is utilized for targeting MC1-R. 188Re/177Lu/212Pb-labeled MC1-R targeting peptides have shown selective uptake in the melanoma cells with rapid whole-body clearance and no significant toxicities in preclinical models [174]. Another preclinical study with 225Ac-MC1R ligand in metastatic uveal melanoma showed high biostability, low toxicity, rapid blood clearance, and selective uptake at sites expressing MC1R [175]. Further clinical studies are required to explore this novel target presenting an alternate therapeutic modality in patients with advanced metastatic melanoma. Fibroblast activation protein (FAP) is a surface protein on cancer-associated fibroblasts and was recently shown to be overexpressed in 28 malignancies with high selectivity and tumor-to-background ratios [176]. 68Ga-FAP-inhibitor (FAPI) apart from a superior diagnostic modality in several malignancies represents a potential theranostic pair with 90 177 Y/ Lu-FAPI in patients with advanced metastatic and refractory malignancies. In a patient with advanced breast carcinoma, 90Y-FAPI-4 administration led to symptomatic improvement without any significant toxicity  [177]. However, the rapid clearance of FAPI-4 from tumor tissue limited the maximum radiation dose delivery due to low retention time. Modifications to the FAPI compounds and several clinical trials are underway.

­Conclusion Therapeutic nuclear medicine (or nuclear oncology) will continue to have growing importance in cancer care. With all the biologic targets discussed in this chapter that have already been identified and FDA approved, and with the ability to both image and treat those targets (via theranostics), nuclear

Radionuclide Therapies and Correlative Imaging

medicine will be at the forefront of precision/personalized medicine in oncology for many years to come. As novel targets and their imaging and therapeutic radiopharmaceutical surrogates continue to be developed, it is critically important to obtain robust, quality data to determine the

most appropriate clinical settings where these therapies could be of most benefit to the patients. Also, it is important to work closely with colleagues in medical, surgical, and radiation oncology to deliver the best care to patients within a multidisciplinary setting.

­References 1 Rindi, G., Klimstra, D.S., Abedi-Ardekani, B. et al. (2018). A common classification framework for neuroendocrine neoplasms: an International Agency for Research on Cancer (IARC) and World Health Organization (WHO) expert consensus proposal. Mod. Pathol. 31: 1770–1786. 2 Williams, E.D. and Sandler, M. (1963). The classification of carcinoid tumours. Lancet 1: 238–239. 3 Soga, J., Yakuwa, Y., and Osaka, M. (1999). Carcinoid syndrome: a statistical evaluation of 748 reported cases. J. Exp. Clin. Cancer Res. 18: 133–141. 4 Rinke, A., Müller, H.H., Schade-Brittinger, C. et al. (2009). Placebo-controlled, double-blind, prospective, randomized study on the effect of octreotide LAR in the control of tumor growth in patients with metastatic neuroendocrine midgut tumors: a report from the PROMID study group. J. Clin. Oncol. 27: 4656–4663. 5 Caplin, M.E., Pavel, M., Ćwikła, J.B. et al. (2014). Lanreotide in metastatic enteropancreatic neuroendocrine tumors. N. Engl. J. Med. 371: 224–233. 6 Pettinato, C., Sarnelli, A., Di Donna, M. et al. (2008). 68 Ga-DOTANOC: biodistribution and dosimetry in patients affected by neuroendocrine tumors. Eur. J. Nucl. Med. Mol. Imaging 35: 72–79. 7 Wild, D., Bomanji, J.B., Benkert, P. et al. (2013). Comparison of 68Ga-DOTANOC and 68Ga-DOTATATE PET/CT within patients with gastroenteropancreatic neuroendocrine tumors. J. Nucl. Med. 54: 364–372. 8 Kwekkeboom, D., Krenning, E.P., and De Jong, M. (2000). Peptide receptor imaging and therapy. J. Nucl. Med. 41: 1704–1713. 9 Kunikowska, J., Królicki, L., Hubalewska-Dydejczyk, A. et al. (2011). Clinical results of radionuclide therapy of neuroendocrine tumours with 90Y-DOTATATE and tandem 90 177 Y/ Lu-DOTATATE: which is a better therapy option? Eur. J. Nucl. Med. Mol. Imaging 38: 1788–1797. 10 Strosberg, J., El-Haddad, G., Wolin, E. et al. (2017). Phase 3 trial of 177Lu-Dotatate for midgut neuroendocrine tumors. N. Engl. J. Med. 376: 125–135. 11 Brabander, T., Van Der Zwan, W.A., Teunissen, J.J.M. et al. (2017). Long-term efficacy, survival, and safety of [177Lu-DOTA0,Tyr3]octreotate in patients with gastroenteropancreatic and bronchial neuroendocrine tumors. Clin. Cancer Res. 23: 4617–4624.

12 Morgenstern, A., Apostolidis, C., Kratochwil, C. et al. (2018). An overview of targeted alpha therapy with 225 actinium and 213 bismuth. Curr. Radiopharm. 11: 200–208. 13 Navalkissoor, S. and Grossman, A. (2019). Targeted alpha particle therapy for neuroendocrine tumours: the next generation of peptide receptor radionuclide therapy. Neuroendocrinology 108: 256–264. 14 Hope, T.A., Bergsland, E.K., Bozkurt, M.F. et al. (2018). Appropriate use criteria for somatostatin receptor PET imaging in neuroendocrine tumors. J. Nucl. Med. 59: 66–74. 15 Klimstra DS, Kloppel G, La Rosa S, et al. Classification of neuroendocrine neoplasms of the digestive system. In: WHO Classification of Tumours: Digestive System Tumours, 5th edn, WHO Classification of Tumours Editorial Board. 2019. p. 16. 16 Virgolini, I., Ambrosini, V., Bomanji, J.B. et al. (2010). Procedure guidelines for PET/CT tumour imaging with 68 Ga-DOTA-conjugated peptides: 68Ga-DOTA-TOC, 68 Ga-DOTA-NOC, 68Ga-DOTA-TATE. Eur. J. Nucl. Med. Mol. Imaging 37: 2004–2010. 17 Deppen, S.A., Blume, J., Bobbey, A.J. et al. (2016). 68 Ga-DOTATATE compared with 111In-DTPA-octreotide and conventional imaging for pulmonary and gastroenteropancreatic neuroendocrine tumors: a systematic review and meta-analysis. J. Nucl. Med. 57: 872–878. 18 Binderup, T., Knigge, U., Loft, A. et al. (2010). 18 F-fluorodeoxyglucose positron emission tomography predicts survival of patients with neuroendocrine tumors. Clin. Cancer Res. 16: 978–985. 19 Hope, T.A., Abbott, A., Colucci, K. et al. (2019). NANETS/SNMMI procedure standard for somatostatin receptor-based peptide receptor radionuclide therapy with 177Lu-DOTATATE. J. Nucl. Med. 60: 937–943. 20 Zaknun, J.J., Bodei, L., Mueller-Brand, J. et al. (2013). The joint IAEA, EANM, and SNMMI practical guidance on peptide receptor radionuclide therapy (PRRNT) in neuroendocrine tumours. Eur. J. Nucl. Med. Mol. Imaging 40: 800–816. 21 Kulke, M.H., Shah, M.H., Benson, A.B. et al. (2015). Neuroendocrine tumors, version 1.2015. JNCCN J. Natl. Compr. Cancer Netw. 13: 78–108.

863

864

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

22 Kaderli, R.M., Spanjol, M., Kollár, A. et al. (2019). Therapeutic options for neuroendocrine tumors: a systematic review and network meta-analysis. JAMA Oncol. 5: 480–489. 23 Khan, S., Krenning, E.P., Van Essen, M. et al. (2011). Quality of life in 265 patients with gastroenteropancreatic or bronchial neuroendocrine tumors treated with [177Lu-DOTA 0,Tyr 3]octreotate. J. Nucl. Med. 52: 1361–1368. 24 Ashwathanarayana, A.G., Biswal, C.K., Sood, A. et al. (2017). Imaging-guided use of combined 177LuDOTATATE and capecitabine therapy in metastatic mediastinal paraganglioma. J. Nucl. Med. Technol. 45: 314–316. 25 Claringbold, P.G. and Turner, J.H. (2015). Neuroendocrine tumor therapy with Lutetium-177octreotate and Everolimus (NETTLE): a phase I study. Cancer Biother. Radiopharm. 30: 261–269. 26 Cives, M. and Strosberg, J.R. (2018). Gastroenteropancreatic neuroendocrine tumors. CA Cancer J. Clin. 68: 471–487. 27 Kwekkeboom, D.J. and Krenning, E.P. (2016). Peptide receptor radionuclide therapy in the treatment of neuroendocrine tumors. Hematol. Oncol. Clin. North Am. 30: 179–191. 28 Bergsma, H., Van Lom, K., Raaijmakers, M.H.G.P. et al. (2018). Persistent hematologic dysfunction after peptide receptor radionuclide therapy with 177Lu-DOTATATE: incidence, course, and predicting factors in patients with gastroenteropancreatic neuroendocrine tumors. J. Nucl. Med. 59: 452–458. 29 De Keizer, B., Van Aken, M.O., Feelders, R.A. et al. (2008). Hormonal crises following receptor radionuclide therapy with the radiolabeled somatostatin analogue [177Lu-DOTA0,Tyr 3]octreotate. Eur. J. Nucl. Med. Mol. Imaging 35: 749–755. 30 Bodei, L., Cremonesi, M., Grana, C.M. et al. (2011). Peptide receptor radionuclide therapy with 177LuDOTATATE: the IEO phase I-II study. Eur. J. Nucl. Med. Mol. Imaging 38: 2125–2135. 31 Kesavan, M. and Turner, J.H. (2016). Myelotoxicity of peptide receptor radionuclide therapy of neuroendocrine tumors: a decade of experience. Cancer Biother. Radiopharm. 38: 189–198. 32 Bodei, L., Kidd, M., Paganelli, G. et al. (2015). Long-term tolerability of PRRT in 807 patients with neuroendocrine tumours: the value and limitations of clinical factors. Eur. J. Nucl. Med. Mol. Imaging 42: 5–19. 33 Zhai, Z., Zheng, Y., Li, N. et al. (2020). Incidence and disease burden of prostate cancer from 1990 to 2017: results from the Global Burden of Disease Study 2017. Cancer 126 (9): 1969–1978.

34 Bander, N.H., Trabulsi, E.J., Kostakoglu, L. et al. (2003). Targeting metastatic prostate cancer with radiolabeled monoclonal antibody J591 to the extracellular domain of prostate specific membrane antigen. J. Urol. 170: 1717–1721. 35 Eder, M., Schäfer, M., Bauder-Wüst, U. et al. (2012). 68 Ga-complex lipophilicity and the targeting property of a urea-based PSMA inhibitor for PET imaging. Bioconjug. Chem. 23: 688–697. 36 Afshar-Oromieh, A., Haberkorn, U., Eder, M. et al. (2012). [68Ga]Gallium-labelled PSMA ligand as superior PET tracer for the diagnosis of prostate cancer: comparison with 18F-FECH. Eur. J. Nucl. Med. Mol. Imaging 39: 1085–1086. 37 Kozikowski, A.P., Nan, F., Conti, P. et al. (2001). Design of remarkably simple, yet potent urea-based inhibitors of glutamate carboxypeptidase II (NAALADase) [1]. J. Med. Chem. 44: 298–301. 38 Singh, B., Sharma, S., Bansal, P. et al. (2021). Comparison of the diagnostic utility of 99mTc-PSMA scintigraphy versus 68Ga-PSMA-11 PET/CT in the detection of metastatic prostate cancer and dosimetry analysis. Nucl. Med. Commun. 42 (5): 482–489. 39 Virgolini, I., Decristoforo, C., Haug, A. et al. (2018). Current status of theranostics in prostate cancer. Eur. J. Nucl. Med. Mol. Imaging 45: 471–495. 40 Baum, R.P., Kulkarni, H.R., Schuchardt, C. et al. (2016). 177 Lu-labeled prostate-specific membrane antigen radioligand therapy of metastatic castration-resistant prostate cancer: safety and efficacy. J. Nucl. Med. 57: 1006–1013. 41 Ahmadzadehfar, H., Rahbar, K., Kürpig, S. et al. (2015). Early side effects and first results of radioligand therapy with 177Lu-DKFZ-617 PSMA of castrate-resistant metastatic prostate cancer: a two-centre study. EJNMMI Res. 5: 114. 42 Parihar, A.S., Chandekar, K., Singh, H. et al. (2021). Orbital and brain metastases on 68Ga-PSMA PET/CT in a patient with prostate carcinoma refractory to 177Lu-PSMA and 225Ac-PSMA therapy. Asia Ocean J. Nucl. Med. Biol. 9: 67–70. 43 Satapathy, S., Das, C.K., Parihar, A.S. et al. (2021). Response to concomitant enzalutamide and 177LuPSMA-617 radioligand therapy in ATM-mutated metastatic castration resistant prostate cancer. Clin. Nucl. Med. 46 (7): 582–583. 44 Perera, M., Papa, N., Christidis, D. et al. (2016). Sensitivity, specificity, and predictors of positive 68Ga– prostate-specific membrane antigen positron emission tomography in advanced prostate cancer: a systematic review and meta-analysis. Eur. Urol. 70: 926–937. 45 Fendler, W.P., Eiber, M., Beheshti, M. et al. (2017). 68 Ga-PSMA PET/CT: joint EANM and SNMMI procedure

Radionuclide Therapies and Correlative Imaging

46

47

48

49

50

51

52

53

54

55

56

57

guideline for prostate cancer imaging: version 1.0. Eur. J. Nucl. Med. Mol. Imaging 44: 1014–1024. Kratochwil, C., Fendler, W.P., Eiber, M. et al. (2019). EANM procedure guidelines for radionuclide therapy with 177Lu-labelled PSMA-ligands (177Lu-PSMA-RLT). Eur. J. Nucl. Med. Mol. Imaging 46: 2536–2544. Fendler, W.P., Kratochwil, C., Ahmadzadehfar, H. et al. (2016). 177Lu-PSMA-617 therapy, dosimetry and follow-up in patients with metastatic castration-resistant prostate cancer. Nuklearmedizin 55: 123–128. Langbein, T., Chaussé, G., and Baum, R.P. (2018). Salivary gland toxicity of PSMA radioligand therapy: relevance and preventive strategies. J. Nucl. Med. 59: 1172–1173. Taïeb, D., Foletti, J.M., Bardies, M. et al. (2018). PSMAtargeted radionuclide therapy and salivary gland toxicity: why does it matter? J. Nucl. Med. 59 (5): 747–748. Yordanova, A., Becker, A., Eppard, E. et al. (2017). The impact of repeated cycles of radioligand therapy using [177Lu]Lu-PSMA-617 on renal function in patients with hormone refractory metastatic prostate cancer. Eur. J. Nucl. Med. Mol. Imaging 44: 1473–1479. Kim, Y.J. and Kim, Y.I. (2018). Therapeutic responses and survival effects of 177Lu-PSMA-617 radioligand therapy in metastatic castrate-resistant prostate cancer: a metaanalysis. Clin. Nucl. Med. 43: 728–734. Yadav, M.P., Ballal, S., Sahoo, R.K. et al. (2019). Radioligand therapy with 177Lu-PSMA for metastatic castration-resistant prostate cancer: a systematic review and meta-analysis. Am. J. Roentgenol. 213: 275–285. Calopedos, R.J.S., Chalasani, V., Asher, R. et al. (2017). Lutetium-177-labelled anti-prostate-specific membrane antigen antibody and ligands for the treatment of metastatic castrate-resistant prostate cancer: a systematic review and meta-analysis. Prostate Cancer Prostatic Dis. 20: 352–360. Tannock, I.F., De Wit, R., Berry, W.R. et al. (2004). Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N. Engl. J. Med. 351: 1502–1512. Oudard, S., Fizazi, K., Sengeløv, L. et al. (2017). Cabazitaxel versus docetaxel as first-line therapy for patients with metastatic castration-resistant prostate cancer: a randomized phase III trial – FIRSTANA. J. Clin. Oncol. 35: 3189–3197. Hofman, M.S., Violet, J., Hicks, R.J. et al. (2018). [177Lu]PSMA-617 radionuclide treatment in patients with metastatic castration-resistant prostate cancer (LuPSMA trial): a single-centre, single-arm, phase 2 study. Lancet Oncol. 19: 825–833. Rahbar, K., Ahmadzadehfar, H., Kratochwil, C. et al. (2017). German multicenter study investigating 177LuPSMA-617 radioligand therapy in advanced prostate cancer patients. J. Nucl. Med. 58: 85–90.

58 Rasul, S., Hacker, M., Kretschmer-Chott, E. et al. (2020). Clinical outcome of standardized 177Lu-PSMA-617 therapy in metastatic prostate cancer patients receiving 7400 MBq every 4 weeks. Eur. J. Nucl. Med. Mol. Imaging 47: 713–720. 59 Heck, M.M., Tauber, R., Schwaiger, S. et al. (2019). Treatment outcome, toxicity, and predictive factors for radioligand therapy with 177Lu-PSMA-I&T in metastatic castration-resistant prostate cancer(figure presented.). Eur. Urol. 75: 920–926. 60 Rahbar, K., Afshar-Oromieh, A., Jadvar, H. et al. (2018). PSMA Theranostics: current status and future directions. Mol. Imaging 17. 61 Rahbar, K., Bode, A., Weckesser, M. et al. (2016). Radioligand therapy with 177Lu-PSMA-617 as a novel therapeutic option in patients with metastatic castration resistant prostate cancer. Clin. Nucl. Med. 41: 522–528. 62 Fendler, W.P., Reinhardt, S., Ilhan, H. et al. (2017). Preliminary experience with dosimetry, response and patient reported outcome after 177Lu-PSMA-617 therapy for metastatic castration-resistant prostate cancer. Oncotarget 8: 3581–3590. 63 Mandela, P. and Ordway, G.A. (2006). The norepinephrine transporter and its regulation. J. Neurochem. 97: 310–333. 64 Tellioglu, T. and Robertson, D. (2001). Genetic or acquired deficits in the norepinephrine transporter: current understanding of clinical implications. Expert Rev. Mol. Med. 3: 1–10. 65 Siegel, R.L., Miller, K.D., and Jemal, A. (2020). Cancer statistics, 2020. CA Cancer J. Clin. 70: 7–30. 66 Park, J.R., Eggert, A., and Caron, H. (2010). Neuroblastoma: biology, prognosis, and treatment. Hematol. Oncol. Clin. North Am. 24: 65–86. 67 Newman, E.A., Abdessalam, S., Aldrink, J.H. et al. (2019). Update on neuroblastoma. J. Pediatr. Surg. 54: 383–389. 68 Brisse, H.J., McCarville, M.B., Granata, C. et al. (2011). Guidelines for imaging and staging of neuroblastic tumors: consensus report from the international neuroblastoma risk group project. Radiology 261 (1): 243–257. 69 Tischler, A.S., Kimura, N., and Mcnicol, A.M. (2006). Pathology of pheochromocytoma and extra-adrenal paraganglioma. Ann. N. Y. Acad. Sci. 1073 (1): 557–570. 70 Goldstein, R.E., O’Neill, J.A., Holcomb, G.W. et al. (1999). Clinical experience over 48 years with pheochromocytoma. Ann. Surg. 229: 755. 71 Plouin, P.F., Chatellier, G., Fofol, I. et al. (1997). Tumor recurrence and hypertension persistence after successful pheochromocytoma operation. Hypertension 29: 1133–1139. 72 Hamidi, O., Young, W.F., Iñiguez-Ariza, N.M. et al. (2017). Malignant pheochromocytoma and

865

866

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

73

74

75

76

77

78

79

80

81

82

83

84

paraganglioma: 272 patients over 55 years. J. Clin. Endocrinol. Metab. 102: 3296–3305. Niemeijer, N.D., Alblas, G., Van Hulsteijn, L.T. et al. (2014). Chemotherapy with cyclophosphamide, vincristine and dacarbazine for malignant paraganglioma and pheochromocytoma: systematic review and metaanalysis. Clin. Endocrinol. (Oxf) 81: 642–651. Bombardieri, E., Giammarile, F., Aktolun, C. et al. (2010). 131 123 I/ I-Metaiodobenzylguanidine (mIBG) scintigraphy: procedure guidelines for tumour imaging. Eur. J. Nucl. Med. Mol. Imaging 37 (12): 2436–2446. Barrett, J.A., Joyal, J.L., Hillier, S.M. et al. (2010). Comparison of high-specific-activity ultratrace 123 131 / I-MIBG and carrier-added 123/131I-MIBG on efficacy, pharmacokinetics, and tissue distribution. Cancer Biother. Radiopharm. 25: 299–308. Gonias, S., Goldsby, R., Matthay, K.K. et al. (2009). Phase II study of high-dose [131I]metaiodobenzylguanidine therapy for patients with metastatic pheochromocytoma and paraganglioma. J. Clin. Oncol. 27: 4162–4168. Pryma, D.A., Chin, B.B., Noto, R.B. et al. (2019). Efficacy and safety of high-specific-activity 131I-MIBG therapy in patients with advanced pheochromocytoma or paraganglioma. J. Nucl. Med. 60: 623–630. Giammarile, F., Chiti, A., Lassmann, M. et al. (2008). EANM procedure guidelines for 131I-metaiodobenzylguanidine (131I-mIBG) therapy. Eur. J. Nucl. Med. Mol. Imaging 35: 1039–1047. Grünwald, F. and Ezziddin, S. (2010). 131I-Metaiodobenzy lguanidine therapy of neuroblastoma and other neuroendocrine tumors. Semin. Nucl. Med. 40: 153–163. Jimenez, C., Erwin, W., and Chasen, B. (2019). Targeted radionuclide therapy for patients with metastatic pheochromocytoma and paraganglioma: from lowspecific-activity to high-specific-activity Iodine-131  metaiodobenzylguanidine. Cancers (Basel) 11: 1018. Kayano, D. and Kinuya, S. (2018). Current consensus on I-131 MIBG therapy. Nucl. Med. Mol. Imaging (2010) 52: 254–265. Matthay, K.K., Yanik, G., Messina, J. et al. (2007). Phase II study on the effect of disease sites, age, and prior therapy on response to iodine-131-metaiodobenzylguanidine therapy in refractory neuroblastoma. J. Clin. Oncol. 25: 1054–1060. Kraal, K.C.J.M., van Dalen, E.C., Tytgat, G.A.M. et al. (2017). Iodine-131-meta-iodobenzylguanidine therapy for patients with newly diagnosed high-risk neuroblastoma. Cochrane Database Syst. Rev. 2017. Wilson, J.S., Gains, J.E., Moroz, V. et al. (2014). A systematic review of 131I-meta iodobenzylguanidine molecular radiotherapy for neuroblastoma. Eur. J. Cancer 50: 801–815.

85 de Kraker, J., Hoefnagel, K.A., Verschuur, A.C. et al. (2008). Iodine-131-metaiodobenzylguanidine as initial induction therapy in stage 4 neuroblastoma patients over 1 year of age. Eur. J. Cancer 44: 551–556. 86 Yanik, G.A., Villablanca, J.G., Maris, J.M. et al. (2015). 131I-Metaiodobenzylguanidine with intensive chemotherapy and autologous stem cell transplantation for high-risk neuroblastoma: a new approaches to neuroblastoma therapy (NANT) phase II study. Biol. Blood Marrow Transplant. 21: 673–681. 87 Polishchuk, A.L., Dubois, S.G., Haas-Kogan, D. et al. (2011). Response, survival, and toxicity after iodine-131metaiodobenzylguanidine therapy for neuroblastoma in preadolescents, adolescents, and adults. Cancer 117: 4286–4293. 88 Quach, A., Ji, L., Mishra, V. et al. (2011). Thyroid and hepatic function after high-dose 131 I-metaiodobenzylguanidine (131I-MIBG) therapy for neuroblastoma. Pediatr. Blood Cancer 56: 191–201. 89 Weiss, B., Vora, A., Huberty, J. et al. (2003). Secondary myelodysplastic syndrome and leukemia following 131 I-metaiodobenzylguanidine therapy for relapsed neuroblastoma. J. Pediatr. Hematol. Oncol. 25: 543–547. 90 Huibregtse, K.E., Vo, K.T., DuBois, S.G. et al. (2016). Incidence and risk factors for secondary malignancy in patients with neuroblastoma after treatment with 131I-me taiodobenzylguanidine. Eur. J. Cancer 66: 144–152. 91 Bleeker, G., Schoot, R.A., Caron, H.N. et al. (2013). Toxicity of upfront 131I-metaiodobenzylguanidine (131I-MIBG) therapy in newly diagnosed neuroblastoma patients: a retrospective analysis. Eur. J. Nucl. Med. Mol. Imaging 40: 1711–1717. 92 George, S.L., Falzone, N., Chittenden, S. et al. (2016). Individualized 131I-mIBG therapy in the management of refractory and relapsed neuroblastoma. Nucl. Med. Commun. 37: 466–472. 93 Modak, S., Pandit-Taskar, N., Kushner, B.H. et al. (2008). Transient sialoadenitis: a complication of 131Imetaiodobenzylguanidine therapy. Pediatr. Blood Cancer 50: 1271–1273. 94 Loh, K.C., Fitzgerald, P.A., Matthay, K.K. et al. (1997). The treatment of malignant pheochromocytoma with iodine-131 metaiodobenzylguanidine (131I-MIBG): a comprehensive review of 116 reported patients. J. Endocrinol. Invest. 20: 648–658. 95 Mercadante, S. (1997). Malignant bone pain: pathophysiology and treatment. Pain 69: 1–18. 96 Oster, M.W., Vizel, M., and Turgeon, L.R. (1978). Pain of terminal cancer patients. Arch. Intern. Med. 138: 1801–1802. 97 Levy, M.H. (1996). Pharmacologic treatment of cancer pain. N. Engl. J. Med. 335: 1124–1132.

Radionuclide Therapies and Correlative Imaging

98 Handkiewicz-Junak, D., Poeppel, T.D., Bodei, L. et al. (2018). EANM guidelines for radionuclide therapy of bone metastases with beta-emitting radionuclides. Eur. J. Nucl. Med. Mol. Imaging 45: 846–859. 99 Pandit-Taskar, N., Batraki, M., and Divgi, C.R. (2004). Radiopharmaceutical therapy for palliation of bone pain from osseous metastases. J. Nucl. Med. 45: 1358–1365. 100 Knapp, F.F., Dash, A., Knapp, F.F. et al. (2016). Therapeutic radiopharmaceuticals for bone pain palliation. In: Radiopharmaceutical Therapeutics, 225–252. Springer https://link.springer.com/ chapter/10.1007%2F978-81-322-2607-9_12. 101 Parker, C., Nilsson, S., Heinrich, D. et al. (2013). Alpha emitter Radium-223 and survival in metastatic prostate cancer. N. Engl. J. Med. 369: 213–223. 102 D’angelo, G., Sciuto, R., Salvatori, M. et al. (2012). Targeted “bone-seeking” radiopharmaceuticals for palliative treatment of bone metastases: a systematic review and meta-analysis – PubMed. Q. J. Nucl. Med. Mol. Imaging 56: 538–543. 103 Sciuto, R., Festa, A., Pasqualoni, R. et al. (2001). Metastatic bone pain palliation with 89-Sr and 186-ReHEDP in breast cancer patients. Breast Cancer Res. Treat. 66: 101–109. 104 Liepe, K. and Kotzerke, J. (2007). A comparative study of Re-HEDP, Re-HEDP, Sm-EDTMP and Sr in the treatment of painful skeletal metastases. Nucl. Med. Commun. 28: 623–630. 105 Coronado, M., Redondo, A., Coya, J. et al. (2006). Clinical role of Sm-153 EDTMP in the treatment of painful bone metastatic disease. Clin. Nucl. Med. 31: 605–610. 106 Bączyk, M., Czepczyński, R., Milecki, P. et al. (2007). Sr versus Sm-EDTMP: comparison of treatment efficacy of painful bone metastases in prostate and breast carcinoma. Nucl. Med. Commun. 28: 245–250. 107 Finlay, I.G., Mason, M.D., and Shelley, M. (2005). Radioisotopes for the palliation of metastatic bone cancer: a systematic review. Lancet Oncol. 6: 392–400. 108 Sartor, O., Reid, R.H., Hoskin, P.J. et al. (2004). Samarium-153-lexidronam complex for treatment of painful bone metastases in hormone-refractory prostate cancer. Urology 63: 940–945. 109 Sarkar, S.D. (2006). Benign thyroid disease: what is the role of nuclear medicine? Semin. Nucl. Med. 36: 185–193. 110 Andersen, S.L., Olsen, J., and Laurberg, P. (2016). Antithyroid drug side effects in the population and in pregnancy. J. Clin. Endocrinol. Metab. 101: 1606–1614. 111 Törring, O., Tallstedt, L., Wallin, G. et al. (1996). Graves’ hyperthyroidism: treatment with antithyroid drugs, surgery, or radioiodine – a prospective, randomized

112

113

114

115

116

117

118

119

120

121

122

123

study. Thyroid Study Group. Intergovernmental Panel on Climate Change, editor. J. Clin. Endocrinol. Metab. 81: 2986–2993. Meller, J. and Becker, W. (2002). The continuing importance of thyroid scintigraphy in the era of high-resolution ultrasound. Eur. J. Nucl. Med. Mol. Imaging Ger. 29 (S2): S425–S438. Carnell, N.E. and Valente, W.A. (1998). Thyroid nodules in Graves’ disease: classification, characterization, and response to treatment. Thyroid 8: 647–652. Scheff A. ACR-SNM-SPR Practice Guideline for the Performance of Thyroid Scintigraphy and Uptake Measurements. American College of Radiology in collaboration with other societies. https://s3.amazonaws. com/rdcms-snmmi/files/production/public/docs/ Thyroid_Scintigraphy_1382732120053_10.pdf 2009. Ross, D.S., Burch, H.B., Cooper, D.S. et al. (2016). 2016 American Thyroid Association guidelines for diagnosis and management of hyperthyroidism and other causes of thyrotoxicosis. Thyroid 26: 1343–1421. Chen, D.Y., Jing, J., Schneider, P.F. et al. (2009). Comparison of the long-term efficacy of low dose 131I versus antithyroid drugs in the treatment of hyperthyroidism. Nucl. Med. Commun. 30: 160–168. Wu, V.T., Lorenzen, A.W., Beck, A.C. et al. (2017). Comparative analysis of radioactive iodine versus thyroidectomy for definitive treatment of graves disease. Surgery 161: 147–155. Sundaresh, V., Brito, J.P., Thapa, P. et al. (2017). Comparative effectiveness of treatment choices for Graves’ hyperthyroidism: a historical cohort study. Thyroid 27: 497–505. Mumtaz, M., Lin, L.S., Hui, K.C. et al. (2009). Radioiodine I-131 for the therapy of graves’ disease. Malays. J. Med. Sci. Malaysia 16 (1): 25–33. Haugen, B.R., Alexander, E.K., Bible, K.C. et al. (2016). 2015 American Thyroid Association Management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid 26: 1–133. Kulkarni, K., Van Nostrand, D., Atkins, F. et al. (2014). Does lemon juice increase radioiodine reaccumulation within the parotid glands more than if lemon juice is not administered? Nucl. Med. Commun. 35: 210–216. Van Nostrand, D., Atkins, F., Bandaru, V.V. et al. (2009). Salivary gland protection with sialagogues: a case study. Thyroid 19: 1005–1008. Nakada, K., Ishibashi, T., Takei, T. et al. (2005). Does lemon candy decrease salivary gland damage after radioiodine therapy for thyroid cancer? J. Nucl. Med. 46: 261–266.

867

868

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

124 Milano, A.F. (2018). Thyroid cancer: 20-year comparative mortality and survival analysis of six thyroid cancer histologic subtypes by age, sex, race, stage, cohort entry time-period and disease duration (SEER*Stat 8.3.2) a systematic review of 145,457 cases for diagnosis year. J. Insur. Med. United States 47: 143–158. 125 Lim, H., Devesa, S.S., Sosa, J.A. et al. (1974–2013). Trends in thyroid cancer incidence and mortality in the United States. JAMA United States 317: 1338–1348. 126 Tam, S., Boonsripitayanon, M., Amit, M. et al. (2018). Survival in differentiated thyroid cancer: comparing the AJCC cancer staging seventh and eighth editions. Thyroid 28: 1301–1310. 127 Casella, C., Ministrini, S., Galani, A. et al. (2018). The new TNM staging system for thyroid cancer and the risk of disease downstaging. Front. Endocrinol. (Lausanne) 9: 541. 128 Owens, P.W., McVeigh, T.P., Fahey, E.J. et al. (2018). Differentiated thyroid cancer: how do current practice guidelines affect management? Eur. Thyroid J. 7: 319–326. 129 Lamartina, L., Grani, G., Durante, C. et al. Follow-up of differentiated thyroid cancer – what should (and what should not) be done. Nat. Rev. Endocrinol. Engl. 14: 538–551. 130 Van Nostrand, D. (2017). Selected controversies of radioiodine imaging and therapy in differentiated thyroid cancer. Endocrinol. Metab. Clin. North Am. 46: 783–793. 131 Song, H., Mosci, C., Akatsu, H. et al. (2018). Diagnostic 123 I whole body scan prior to ablation of thyroid remnant in patients with papillary thyroid cancer: implications for clinical management. Clin. Nucl. Med. 43: 705–709. 132 Bal, C.S., Kumar, A., and Pant, G.S. (2004). Radioiodine dose for remnant ablation in differentiated thyroid carcinoma: a randomized clinical trial in 509 patients. J. Clin. Endocrinol. Metab. 89: 1666–1673. 133 Klubo-Gwiezdzinska, J., Van Nostrand, D., Atkins, F. et al. (2011). Efficacy of dosimetric versus empiric prescribed activity of 131I for therapy of differentiated thyroid cancer. J. Clin. Endocrinol. Metab. 96: 3217–3225. 134 Deandreis, D., Rubino, C., Tala, H. et al. (2017). Comparison of empiric versus whole-body/-blood clearance dosimetry-based approach to radioactive iodine treatment in patients with metastases from differentiated thyroid cancer. J. Nucl. Med. 58: 717–722. 135 Ruel, E., Thomas, S., Dinan, M. et al. (2015). Adjuvant radioactive iodine therapy is associated with improved survival for patients with intermediate-risk papillary

136

137

138

139

140

141

142

143

144

145

146

147

thyroid cancer. J. Clin. Endocrinol. Metab. 100: 1529–1536. Carhill, A.A., Litofsky, D.R., Ross, D.S. et al. (2015). Long-term outcomes following therapy in differentiated thyroid carcinoma: NTCTCS registry analysis 1987–2012 [Internet]. J. Clin. Endocrinol. Metab. 100 (9): 3270–3279. Durante, C., Haddy, N., Baudin, E. et al. (2006). Long-term outcome of 444 patients with distant metastases from papillary and follicular thyroid carcinoma: benefits and limits of radioiodine therapy. J. Clin. Endocrinol. Metab. 91: 2892–2899. Brown, A.P., Chen, J., Hitchcock, Y.J. et al. (2008). The risk of second primary malignancies up to three decades after the treatment of differentiated thyroid cancer. J. Clin. Endocrinol. Metab. 93: 504–515. Molenaar, R.J., Sidana, S., Radivoyevitch, T. et al. (2018). Risk of hematologic malignancies after radioiodine treatment of well-differentiated thyroid cancer. J. Clin. Oncol. 36: 1831–1839. Molenaar, R.J., Pleyer, C., Radivoyevitch, T. et al. (2018). Risk of developing chronic myeloid neoplasms in well-differentiated thyroid cancer patients treated with radioactive iodine. Leukemia 32: 952–959. Gans, J.H., Lipman, J., Golowa, Y. et al. (2019). Hepatic cancers overview: surgical and chemotherapeutic options, how do Y-90 microspheres fit in? Semin. Nucl. Med. 49: 170–181. Alsultan, A.A., Barentsz, M.W., Smits, M.L.J. et al. (2019). Angiogenesis in 90 Y-radioembolization of colorectal liver metastases. Semin. Nucl. Med. 49: 204–210. Tafti, B.A. and Padia, S.A. (2019). Dosimetry of Y-90 microspheres utilizing Tc-99m SPECT and Y-90 PET. Semin. Nucl. Med. 49: 211–217. Giammarile, F., Bodei, L., Chiesa, C. et al. (2011). EANM procedure guideline for the treatment of liver cancer and liver metastases with intra-arterial radioactive compounds. Eur. J. Nucl. Med. Mol. Imaging 38: 1393–1406. Kolligs, F.T., Bilbao, J.I., Jakobs, T. et al. (2015). Pilot randomized trial of selective internal radiation therapy vs. chemoembolization in unresectable hepatocellular carcinoma. Liver Int. 35: 1715–1721. Salem, R., Gordon, A.C., Mouli, S. et al. (2016). Y90 radioembolization significantly prolongs time to progression compared with chemoembolization in patients with hepatocellular carcinoma. Gastroenterology 151: 1155–1163.e2. Lewandowski, R.J., Kulik, L.M., Riaz, A. et al. (2009). A comparative analysis of transarterial downstaging for hepatocellular carcinoma: chemoembolization versus radioembolization. Am. J. Transplant. 9: 1920–1928.

Radionuclide Therapies and Correlative Imaging

148 Seidensticker, R., Denecke, T., Kraus, P. et al. (2012). Matched-pair comparison of radioembolization plus best supportive care versus best supportive care alone for chemotherapy refractory liver-dominant colorectal metastases. Cardiovasc. Intervent. Radiol. 35: 1066–1073. 149 Cosimelli, M., Golfieri, R., Cagol, P.P. et al. (2010). Multi-centre phase II clinical trial of yttrium-90 resin microspheres alone in unresectable, chemotherapy refractory colorectal liver metastases. Br. J. Cancer 103: 324–331. 150 Wasan, H.S., Gibbs, P., Sharma, N. et al. (2017). First-line selective internal radiotherapy plus chemotherapy versus chemotherapy alone in patients with liver metastases from colorectal cancer (FOXFIRE, SIRFLOX, and FOXFIRE-Global): a combined analysis of three multicentre, randomised, phase 3 trials. Lancet Oncol. 18: 1159–1171. 151 Van Hazel, G., Blackwell, A., Anderson, J. et al. (2004). Randomised phase 2 trial of SIR-spheres® plus fluorouracil/leucovorin chemotherapy versus fluorouracil/leucovorin chemotherapy alone in advanced colorectal cancer. J. Surg. Oncol. 88: 78–85. 152 Stubbs, R.S. and Wickremesekera, S.K. (2004). Selective internal radiation therapy (SIRT): a new modality for treating patients with colorectal liver metastases. HPB 6: 133–139. 153 Larson, S.M., Carrasquillo, J.A., Cheung, N.K.V. et al. (2015). Radioimmunotherapy of human tumours. Nat. Rev. Cancer 15: 347–360. 154 Press, O.W., Unger, J.M., Braziel, R.M. et al. (2006). Phase II trial of CHOP chemotherapy followed by tositumomab/iodine I-131 tositumomab for previously untreated follicular non-Hodgkin’s lymphoma: five-year follow-up of Southwest Oncology Group protocol S9911. J. Clin. Oncol. 24: 4143–4149. 155 Leonard, J.P., Coleman, M., Kostakoglu, L. et al. (2005). Abbreviated chemotherapy with fludarabine followed by tositumomab and iodine I-131 tositumomab for untreated follicular lymphoma. J. Clin. Oncol. 23: 5696–5704. 156 Zinzani, P.L., Tani, M., Fanti, S. et al. (2008). A phase II trial of CHOP chemotherapy followed by yttrium 90 ibritumomab tiuxetan (Zevalin) for previously untreated elderly diffuse large B-cell lymphoma patients. Ann. Oncol. 19: 769–773. 157 Link, B.K., Martin, P., Kaminski, M.S. et al. (2010). Cyclophosphamide, vincristine, and prednisone followed by tositumomab and iodine-131-tositumomab in patients with untreated low-grade follicular lymphoma: eight-year follow-up of a multicenter phase II study. J. Clin. Oncol. 28: 3035–3041.

158 Kaminski, M.S., Zasadny, K.R., Francis, I.R. et al. (1993). Radioimmunotherapy of B-cell lymphoma with [131I] anti-B1 (anti-CD20) antibody. N. Engl. J. Med. 329: 459–465. 159 Kaminski, M.S., Zelenetz, A.D., Press, O.W. et al. (2001). Pivotal study of iodine I-131 tositumomab for chemotherapy-refractory low-grade or transformed low-grade B-cell non-Hodgkin’s lymphomas. J. Clin. Oncol. 19: 3918–3928. 160 Horning, S.J., Younes, A., Jain, V. et al. (2005). Efficacy and safety of tositumomab and iodine-131 tositumomab (Bexxar) in B-cell lymphoma, progressive after rituximab. J. Clin. Oncol. 23: 712–719. 161 Witzig, T.E., Gordon, L.I., Cabanillas, F. et al. (2002). Randomized controlled trial of yttrium-90-labeled ibritumomab tiuxetan radioimmunotherapy versus rituximab immunotherapy for patients with relapsed or refractory low-grade, follicular, or transformed B-cell non-Hodgkin’s lymphoma. J. Clin. Oncol. 20: 2453–2463. 162 Witzig, T.E., Flinn, I.W., Gordon, L.I. et al. (2002). Treatment with ibritumomab tiuxetan radioimmunotherapy in patients with rituximabrefractory follicular non-Hodgkin’s lymphoma. J. Clin. Oncol. 20: 3262–3269. 163 Morschhauser, F., Kraeber-Bodéré, F., Wegener, W.A. et al. (2010). High rates of durable responses with anti-CD22 fractionated radioimmunotherapy: results of a multicenter, phase I/II study in non-Hodgkin’s lymphoma. J. Clin. Oncol. 28: 3709–3716. 164 Kolstad, A., Illidge, T., Bolstad, N. et al. (2020). Phase 1/2a study of 177Lu-lilotomab satetraxetan in relapsed/ refractory indolent non-Hodgkin lymphoma. Blood Adv. 4: 4091–4101. 165 Press, O.W., Eary, J.F., Gooley, T. et al. (2000). A phase I/ II trial of iodine-131-tositumomab (anti-CD20), etoposide, cyclophosphamide, and autologous stem cell transplantation for relapsed B-cell lymphomas. Blood 96: 2934–2942. 166 Nademanee, A., Forman, S., Molina, A. et al. (2005). Aphase 1/2 trial of high-dose yttrium-90-ibritumomab tiuxetan in combination with high-dose etoposide and cyclophosphamide followed by autologous stem cell transplantation in patients with poor-risk or relapsed non-Hodgkin lymphoma. Blood 106: 2896–2902. 167 Malcolm, J., Falzone, N., Lee, B.Q. et al. (2019). Targeted radionuclide therapy: new advances for improvement of patient management and response. Cancers (Basel): 11. 168 Schottelius, M., Osl, T., Poschenrieder, A. et al. (2017). [177Lu]pentixather: comprehensive preclinical characterization of a first CXCR4-directed endoradiotherapeutic agent. Theranostics 7: 2350–2362.

869

870

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach

169 Herrmann, K., Schottelius, M., Lapa, C. et al. (2016). First-in-human experience of CXCR4-directed endoradiotherapy with 177Lu-and 90Y-labeled pentixather in advanced-stage multiple myeloma with extensive intra-and extramedullary disease. J. Nucl. Med. 57: 248–251. 170 Chen, H., Niu, G., Wu, H. et al. (2016). Clinical application of radiolabeled RGD peptides for PET imaging of integrin αvβ3 [Internet]. Theranostics: 78–92. 171 Parihar, A.S., Mittal, B.R., Kumar, R. et al. (2020). 68 Ga-DOTA-RGD2 positron emission tomography/ computed tomography in radioiodine refractory thyroid cancer: prospective comparison of diagnostic accuracy with 18F-FDG positron emission tomography/computed tomography and evaluation toward potential theranostics. Thyroid 30: 557–567. 172 Pirooznia, N., Abdi, K., Beiki, D. et al. (2020). 177Lulabeled cyclic RGD peptide as an imaging and targeted radionuclide therapeutic agent in non-small cell lung

173

174

175

176

177

cancer: biological evaluation and preclinical study. Bioorg. Chem. 102: 104100. Parihar, A.S., Sood, A., Kumar, R. et al. (2018). Novel use of 177Lu-DOTA-RGD2 in treatment of 68Ga-DOTARGD2-avid lesions in papillary thyroid cancer with TENIS. Eur. J. Nucl. Med. Mol. Imaging 45: 1836–1837. Miao, Y. and Quinn, T.P. (2008). Peptide-targeted radionuclide therapy for melanoma. Crit. Rev. Oncol. Hematol. 67: 213–228. Tafreshi, N.K., Tichacek, C.J., Pandya, D.N. et al. (2019). Melanocortin 1 receptor–targeted a-particle therapy for metastatic uveal melanoma. J. Nucl. Med. 60: 1124–1133. Kratochwil, C., Flechsig, P., Lindner, T. et al. (2019). 68 Ga-FAPI PET/CT: tracer uptake in 28 different kinds of cancer. J. Nucl. Med. 60: 801–805. Lindner, T., Loktev, A., Altmann, A. et al. (2018). Development of quinoline-based theranostic ligands for the targeting of fibroblast activation protein. J. Nucl. Med. 59: 1415–1422.

871

Index a

AAA. See Abdominal aortic aneurysm (AAA) ABC. See Aneurysmal bone cyst (ABC) Abdomen, anatomy of adrenal glands 105–106 biliary system 90–92 bowel, vascularization of 98–100 gallbladder 92–93, 94 gastrointestinal system 96–100 hepato‐biliary system and pancreas 86–93 inframesocolic compartment 104 ligaments 101 limbs and joints 123–128 liver 86–90 mesentery 101, 102 omentum 101, 103 pancreas 93–95 peritoneal compartments 101, 103 peritoneum 101 retroperitoneum 106–108 spleen 100–101 supramesocolic compartment 103–104 urinary system 104–105 Abdominal aortic aneurysm (AAA) 298–301, 304 Abnormal bronchus 76 Absorbed photon energy 23 ACC. See Adenoid cystic carcinoma (ACC) ACCIPIO® 832 ACCs. See Adrenocortical carcinomas (ACCs) Accurate attenuation correction 48 Acetabulum 123 Acetazolamide 168 Achalasia 385 ACM. See Alveolar‐capillary membrane (ACM) Acoustic impedance 43 Acquired immune deficiency syndrome (AIDS) 443 Acquisition system

CT 31, 32 PET 35–36 on SPECT‐CT 21 Acute aortic dissection 274, 301 Acute calculous cholecystitis 469 Acute cerebral infarction 166–167 Acute cholecystitis 690–691 Acute coronary syndromes 257 Acute pulmonary embolus 680–681 Acute rejection, renal transplant and 526, 527 Acute tubular necrosis (ATN) 526–528, 679 AD. See Alzheimer’s disease (AD); Aortic dissection (AD) ADC. See Apparent diffusion coefficient (ADC) Adenocarcinoma 407, 408 Adenoid cystic carcinoma (ACC) 224, 228 Adenomas 429 Adenomatous polyposis coli (APC) gene 441 Adrenal adenomas 498–499 Adrenal cysts 500 Adrenal glands 105–106 ACCs 501 adenomas 498–499 adrenal incidentaloma 497–498 anatomy of 497 correlative 498–499 cysts 500 hemorrhage 499–500 lymphomas 502 metastases 502, 503 myelolipomas 498 pheochromocytomas 500–503 Adrenal hemorrhage 499–500 Adrenal incidentaloma 497–498 Adrenal lymphoma 502 Adrenal metastases 502, 503 Adrenal myelolipomas 498 Adrenocortical carcinomas (ACCs) 498, 501

AEFs. See Aortoenteric fistulas (AEFs) Affine transformation 32 AI. See Artificial intelligence (AI) AIDS. See Acquired immune deficiency syndrome (AIDS) Alanine‐serine‐cysteine transporter 2 (ASCT2) 539 Alar fascia 61, 67 Alberta Stroke Program Early CT Score (ASPECTS) 166 ALND. See Axillary lymph node dissection (ALND) Alveolar‐capillary membrane (ACM) 315 Alzheimer’s disease (AD) 143, 164, 165 amyloid PET imaging 173 biomarker 172 diagnosis of 171 early‐onset patients 172, 173 late‐onset patients 173 medial temporal atrophy in 173 MRI, SPECT, and PET 172–174 parietotemporal cortex in 172–173 posterior cingulate gyrus and precuneus 172 preclinical 172 Amide proton transfer (APT) 198 Amiloyd imaging 143–144 Amino acid tracers 141, 199–207 FDOPA PET 202 FET PET 201–202, 206, 207 FLT PET 203 FMISO PET 203–204 MET PET 200 Amiodarone induced thyrotoxicosis (AIT) [Type I, II, and III] 487 Ammonia‐13 259, 265 Amplitude (A‐mode) 45 Amyloid b (Ab) accumulation 164, 165 Amyloidosis, cardiac 271–272 Amyloid PET imaging 173 Anaerobic glycolysis 2 Anal cancer 437–439 Analog SiPM (aSiPM) 39

Radiology-Nuclear Medicine Diagnostic Imaging: A Correlative Approach, First Edition. Edited by Ali Gholamrezanezhad, Majid Assadi, and Hossein Jadvar. © 2023 John Wiley & Sons Ltd. Published 2023 by John Wiley & Sons Ltd.

872

Index Analytical methods 38 Anaplastic gliomas 208 Anatomical imaging modalities 30 Anatomic staging, for breast cancer 365 Anatomy and variants, on cross‐sectional imaging 52 Anatomy of brain 53 basal ganglia 57–58 cerebral cortex 55–57 CNS anatomy of coverings of 54 compartments and structures of 54 cranial nerves 59–60 parenchyma 54–55 posterior fossa 61 semiology in neuroimaging 53–54 ventricular system 61 white matter 58–59 Aneurysmal bone cyst (ABC) 634 Anger camera 33 components of 34 principle 21 Angiodysplasia 688 Angiomyolipomas 528 Angiotensin‐converting enzyme inhibitors (ACEI) 526 Angular gyrus 164 Annihilation coincidence‐detection systems 355–356 Annihilation process 35 Anterior aortic arch 81 Anterior fibromuscular stroma (AFMS) 533 Anterior oral cavity 232 Anterior pararenal space 106, 107 Anterior segment bronchi 76, 79 Anterior subphrenic space 104 Anterior temporal pole 164 Antibody‐dependent cellular‐mediated cytotoxicity (ADCC) 859 Antineutrophil cytoplasmic antibody (ANCA) 285 Aorta diseases of 274 large vessel vasculitis of 288 Aortic arch and aorto‐pulmonary window 81–83 Aortic dissection (AD) 301–302 Aortic wall, changes in 286 Aortitis, imaging of 292 Aortoenteric fistulas (AEFs) 303–305 Apical segment bronchus 76, 79 Apparent diffusion coefficient (ADC) 138, 197, 334, 346, 412 Appendiceal malignancies 427–432 appendiceal NET imaging 428–429 mucinous neoplasms 429–431 Appendix adenocarcinoma 431 Area under the curve (AUC) 333

Arm, compartments of 126–127 Arrhythmia 271 Arrhythmogenic right ventricular cardiomyopathy (ARVC) 271 Arterial partial pressure of carbon dioxide (PaCO2) 168 Arterial spin labeling (ASL) 196, 197 Arterial spin labeling MRI (ASL‐MRI) 166, 169 Arthritis degenerative 590, 592 rheumatoid 585–589 18 F‐FDG PET/CT 586, 588–589 MRI 586 99m Tc‐MDP scan 586, 587 radiography 585–586 ultrasound 586 spondyloarthropathies 589–591 Artifacts 219–221 anatomical conditions of 219 dental metal 219, 220 reciprocal compensation of 222 swallowing 220, 223 in ultrasound images 44 Artificial intelligence (AI) 208 deep learning (DL) 826–830 deep learning‐based automatic detection (DLAD) 834 deep morphology aided diagnosis network (DeepMAD) 833 diagnostic imaging 832–835 DLAD‐assisted doctor 834 DL‐based algorithm 833 hardware 830 learning‐based approach 835 lung nodule detection 832 segmentation algorithms 833 software 830 ARVC. See Arrhythmogenic right ventricular cardiomyopathy (ARVC) ASL. See Arterial spin labeling (ASL) ASPECTS. See Alberta Stroke Program Early CT Score (ASPECTS) Aspiration area (RA) 386 Aspiration percentage (PA) 386 Atheromas. See Atherosclerotic plaques Atherosclerosis 286 imaging of 292–295 lesions, FDG uptake 293 Atherosclerotic plaques 292 Atlas‐based methods 48 ATN. See Acute tubular necrosis (ATN) ATN biomarker 172 Atrophic gastritis 444 Atropine 258 Attenuation 34, 43 Attenuation coefficients 43 Attenuation correction artifacts 12 ATTR. See Transthyretin amyloidosis (ATTR)

Autoimmune thyroiditis 700 Autologous hematopoietic stem cell rescue (AHSCR) 847 Autosomal dominant disorder 441 Avalanche photodiodes (APDs) 39 Axillary lymph node dissection (ALND) 361 Axillary lymph node metastasis, PET and 366 Axillary lymph nodes 351, 352 FDG avidity in 363 Azygo‐esophageal recess 83–84 Azygos lobe 81

b

Ball‐and‐socke joint 123 Bannayan–Ruvalcaba–Riley syndrome 441 Barium 397–398 Barium swallow videofluoroscopy esophageal transit, radiologic evaluation of 384–385 motility disorders investigated with 385 patient with esophageal candidiasis 390 patient with Zenker’s diverticulum 390 Barrett metaplasia 444 Basal ganglia (BG), anatomy of 57–58 Basal trunk 79 BBB. See Blood–brain barrier (BBB) Beam hardening artifact, in PET‐CT 12, 14 Behavioral variant of frontotemporal dementia (bvFTD) 175, 176 Benign breast conditions 363 Benign gastric diseases gastric emptying functional disorders of 390–391 radionuclide evaluation of 391 ulcers 413 Benign gastrointestinal (GI) disorders 383 Benign liver tumors 468, 470, 471 Benign prostatic hyperplasia (BPH) 534 Benign tumors 226 Benzodiazepine receptor imaging 177 Bile duct obstruction 473, 476 Biliary atresia 457 Biliary leaks 475, 477, 691 Biliary system, anatomy of 90–92 Bilinear segmentation‐scaling method 38 Biomarker subtypes of breast cancer 351 Bladder cancer 530–531 Bland thrombosis 295 Blood–brain barrier (BBB) 53, 194 Blood oxygen level‐dependent (BOLD) 163 Blood supply 310 BOLD. See Blood oxygen level‐dependent (BOLD) Bone anatomy of cranial fossa 54 Bone metastases in breast cancer 367, 369 Bone mineral density (BMD) 659. See also Osteoporosis

Index Bone scan bone and soft‐tissue tumors 605 multiple myeloma 640 spine metastases 642–643, 646 Bone scintigraphy 577 chronic nonbacterial osteomyelitis 696 Legg–Calvé–Perthes disease (LCPD) 696–697 neuroblastoma 697–698 occult fracture 578 periprosthetic infection 598 stress fractures 583 99m Tc‐labeled disphosphonates 696 Bone SPECT/CT (BSCT) imaging, spinal trauma 648–649 Bowel, vascularization of 98–100 Bowel wall, patterns of 400 Boyden classification, bronchopulmonary segmental anatomy and 310 Brachiocephalic artery, trunk of 80 Brain anatomy of 53 basal ganglia 57–58 bone anatomy of cranial fossa 54 of cerebral cortex 55–57 cranial nerves 59–60 lobar and cortical anatomy 55 parenchyma 54–55 posterior fossa 61 semiology in neuroimaging 53–54 ventricular system 61 white matter 58–59 CNS anatomy of coverings of 54 compartments and structures of 54 tumors classification 194 epidemiology and symptomatology 194 Brain death 683 Brain flow imaging in brain death 709 cisternography 709 ommaya reservoir 709 planar radionuclide angiogram 708 Brain ischemia 165 Brain perfusion, neuronuclear imaging of 163 Brain tumors advanced modalities 789 advantages 789 central nervous system (CNS) 788 computed tomography 788 18 F‐fluoromisonidazole (FMISO) 789 limitations 788 radiotracers 789 volumetric interpolated breath‐hold examination (VIBE) 789 Breast, anatomy of 351

Breast cancer biomarker subtypes 351 histopathological subtypes 351 lymphoscintigraphy role in staging 360–361 molecular subtypes 351 peripheral and nonperipheral 356 PET 362–363, 371 and axillary lymph node evaluation 366 future of 369–370 initial staging 365–366 suspected disease recurrence and restaging 369 radioguided surgery of 158 supplemental screening 356–357 systemic treatment of 368–369 treatment response, evaluation of 367–368 Breast imaging alternative to MRI 357 BI‐RADS 356 dedicated molecular breast‐imaging systems 354 diagnostic 357 early scintimammography 354 image‐guided biopsies 358 lymphoscintigraphic 362 MBI 354, 355 molecular techniques in 352–360 nonmolecular imaging modalities 353 nuclear medicine 371–373 single‐photon gamma imaging systems 354–355 ultrasound 353 Breast‐imaging reporting and data system (BI‐RADS) 356, 357 Breast lymphoscintigraphy imaging 362 injection technique 361–362, 363 role in cancer 360–361 Breast‐specific gamma imaging (BSGI) 354, 357 Brightness (B‐mode) 45 Bronchopulmonary segmental anatomy 308 and Boyden classification 310 fissures 308 nomenclature of 308 right and left lung 309 Bronchus, anatomy of 74–80 axial CT image 75 left bronchus 79 normal bronchial tree 75 right bronchus 74–78 Brown adipose tissue (BAT) 5, 6, 776 BSGI. See Breast‐specific gamma imaging (BSGI) Buccal space 66

Buccal squamous cell carcinoma, recurrent 299 Buccopharyngeal fascia 61, 62 “Butterfly glioblastoma” 195 bvFTD. See Behavioral variant of frontotemporal dementia (bvFTD)

c

CABG. See Coronary artery bypass surgery (CABG) CAC. See Coronary artery calcium (CAC) 11 C‐acetate PET/CT bladder cancer 531 prostate imaging 538–539 renal masses 530 CAD. See Coronary artery disease (CAD) Cadmium zinc telluride (CZT) 23, 39 CAD‐RADS. See Coronary Artery Disease Reporting and Data System (CAD‐RADS) CA‐125 levels, post‐ovarian cancer therapy assessment 569 Cancer‐associated fibroblasts (CAFs) 814 Capsule endoscopy 407 Captopril renography, renovascular hypertension 526 Carbonic anhydrase IX (CAIX) 530 Carcinoid syndrome 427, 838 Carcinoma of unknown primary (CUP) 137, 238 at lung 239 at tongue ground 238 Cardiac amyloidosis 271–272 Cardiac computed tomography angiography (CCTA) 274 diagnosis of aortic dissection 274 IHD, diagnosis and prognosis of 266–268 left anterior descending coronary artery 267 Cardiac death, risk of 263, 264 Cardiac iron overload 271 Cardiac magnetic resonance (CMR) cardiac sarcoidosis, diagnosis of 272–273 edema 270 role in IHD 268–269 Cardiac sarcoidosis 272–273 Cardiopulmonary diseases, pulmonary scintigraphy for 325–327 Cardiovascular infections cardiac implantable electronic device infections (CIEDs) 721–723 infective endocarditis (IE) 720–721 prosthetic vascular graft infections 723–725 cause of 723 diagnosis 723 foreign‐body inflammatory reaction 724

873

874

Index Cardiovascular infections (cont’d) infected aortic endovascular stent 724 test accuracy 723 treatment response 724–725 Carina, level of 79 Carney–Statakis syndrome 420 Carotid Occlusion Surgery Study 169 Carotid space 64 Cartesian plane 46 Catheter angiography, renal artery stenosis 526 CBF. See Cerebral blood flow (CBF) CBV. See Cerebral blood volume (CBV) 11 C‐choline PET/CT 139 prostate imaging 537, 538 CCTA‐derived fractional flow reserve (CT‐FFR) 266 CD. See Celiac disease (CD); Crohn’s disease (CD) CECT. See Contrast‐enhanced CT (CECT) Celiac disease (CD) 441 Centiloids 173 Central nervous system (CNS) 194 anatomy of coverings of 54 compartments and structures of 54 imaging 681–685 Central zone (CZ), prostate gland 533 Cerebral blood flow (CBF) 165, 169, 196 Cerebral blood volume (CBV) 165, 166, 196, 197 Cerebral commissures 59 Cerebral cortex, functional anatomy of 55–57 Cerebral hyperperfusion syndrome (CHS) 169–170 Cerebral metabolic rate of oxygen (CMRO2) 166, 167 Cerebral perfusion 681–683 Cerebral vasoreactivity (CVR) 168 Cerebrospinal fluid (CSF) 53, 61 Cerebrovascular disease 165–170 hemodynamic risk evaluation in 167–169, 170 ischemic severity evaluation in 166–167 postoperative CHS 169–170 Cervical cancer 445 Cervical lymph node metastasis 250 Cervical spine 117–119 kyphoscoliosis 223 lower 118–119 sequential interpretation of 119 upper 117–118 Cervix 110 CEST. See Chemical exchange saturation transfer (CEST) 11 C‐Flumazenil 178 CFR. See Coronary flow reserve (CFR) Chemical exchange saturation transfer (CEST) 198

Chemical shift MRI, adrenal adenomas 498 Chemotherapy 239, 240 Chest pain 671 Chest radiography (CRs) 311 mediastinal tumors 344–345 pleural diseases 343 SPN detection 333 teratomas 347 thymoma 345 Chest X‐ray for lung cancer 335 Cholangiocarcinoma 468, 469 Cholecystocolonic fistula 472, 475 Cholecystojejunostomy 475, 478 Choledochal cysts 457, 458 Choline 139–141 Chordoma, spinal 636–638 Chronic aortic dissection (AD) 302 Chronic cerebrovascular disease 167–169 Chronic gastroesophageal reflux 444 Chronic obstructive pulmonary disease (COPD) 324–325 Chronic rejection, renal transplant and 528 CHS. See Cerebral hyperperfusion syndrome (CHS) Classical lissencephaly 180 CMD. See Coronary microvascular dysfunction (CMD) 11 C‐methionine PET 181 CMR. See Cardiac magnetic resonance (CMR) CNS. See Central nervous system (CNS) Cobblestone lissencephaly 180 Coccyx 123 Cochlear nerve 60 Collecting system phase, in nephrography 523 Colon cancer multiple diagnoses of 436 recurrent 437 FDG uptake 5 transverse diameter of 98 Colonic‐type neoplasms 431 Colon transit scintigraphy 394–395, 396, 397 Color Doppler ultrasound prostate imaging 535 renal transplant evaluation 527 testicular torsion 675 Colorectal cancer 432–437 Commissural fibers 59 Common bile duct 97 Common hepatic artery (CHA) 89 Complete response (CR) 195 Computed tomography (CT) 1, 2, 30–31, 49, 133, 219 acquisition system 31, 32

acute cholecystitis 690 adrenal hemorrhage 499 myelolipoma 499 AEF 304 angiography 153 base ganglia, supratentorial axial of 57 bladder cancer 530 bone and soft‐tissue tumors 605 cerebral edema 681, 682 COVID‐19 327 dual‐energy 33 evaluation of GIST 423 fracture lines 583 gastric adenocarcinoma 413 gastrointestinal contrast agent and 10 GI tract bleeding 687 hepatic cholangiocarcinoma 468 hemangioma 468, 470 lymphoma 466 metastases 460, 462, 464 image of pelvic region 8, 9 intermediate bronchus 77 intravenous and oral contrast in 7–10 ionizing radiation 52 Krukenberg tumors 416 lower pelvis from cranial 125 mediastinal tumors 345 vs. MRI 53 multislice scanners 52, 53 nephroureteral obstruction 523 NETs 513 nuclear medicine physicians and 4 oral cavity 66 osteoporosis 662, 664 ovarian cancer 565 paranasal sinuses 71 physical principle on 31–32 pituitary imaging 496 pleural disease 343–344 post‐ovarian cancer therapy assessment 569 principle of 7, 10 prostate imaging 535 protocols in PET‐CT 11 Rathke’s cleft cysts 497 reconstruction methods 32–33 renal masses 528, 529 representation of images 32 right upper lobe bronchus 76 semiology of brain 53 small bowel 400–401 spinal chordoma 637, 638 EC 634 fibrous dysplasia 634 GCTs 634 hemangioma 635, 636

Index infection 626–627 lymphoma 637 metastases 642, 645 osteoblastoma 632–633 osteoid osteoma 631, 632 osteosarcoma 637, 639 SPN 333 teratomas 347 thoracic anatomy 73–74 thymoma 345 thyroid imaging 489 upper pelvis from cranial 124 Computed tomography angiography (CTA) 526 Computerized tomography pulmonary angiography (CTPA) 313 Conduction zone 308–310 Continuous wave (CW) Doppler 46 Contrast‐enhanced CT (CECT) 286, 295 spinal infection 627 Contrast‐enhanced magnetic resonance imaging (CE‐MRI) 353 Contrast‐enhanced US (CEUS) 768–769 hepatic hemangiomas 468 metastases 459–461 rheumatoid arthritis 586 Conventional angiography 286 Conventional gamma camera 23 Conventional mammography 353 Converse piezoelectric effect 44 Convolutional neural networks (CNNs) 826 Cornu ammonis 178 Coronary artery bypass surgery (CABG) 266 Coronary artery calcium (CAC) 266 Coronary artery disease (CAD) 261 Coronary Artery Disease Reporting and Data System (CAD‐RADS) 267 Coronary calcium scoring, in ischemic heart disease 266 Coronary flow reserve (CFR) 262, 264 Coronary microvascular dysfunction (CMD) 262 Coronavirus disease 2019 (COVID‐19) 327–328 Correlation, defined 1 Correlative assessment, across nuclear medicine and radiology 212 Correlative imaging 1–2 Corticobasal degeneration (CBD) 189 Corticobasal syndrome (CBS) 175 Coulomb interactions 35 COVID‐19 infection CT imaging 327 FDG PET/CT imaging 328 nuclear medicine 328 SPECT/CT 328

symptoms of 328 ventilation scintigraphy 327–328 Cowden disease 441 11 C‐PiB PET 165, 174 11 C‐Pittsburgh Compound‐B (PiB) 173 CPP. See Cerebral perfusion pressure (CPP) CR. See Complete response (CR) Cranial nerves anatomy of 60 overview of 59–60 Craniocaudal (CC) 353 Crohn’s disease (CD) 396–397, 442–443 Cross‐sectional imaging anatomy and its variants on 52 CT 52 CRs. See Chest radiography (CRs) CT. See Computed tomography (CT) CT angiography, GI tract bleeding 686–688 CT pulmonary angiogram (CTPA) 681 CUP. See Carcinoma of unknown primary (CUP) CVR. See Cerebral vasoreactivity (CVR) Cyclosporin nephrotoxicity 527 Cystic duct sign 472, 473 Cysts, adrenal 500 Cytoreductive surgery (CRS) 429

d

dbPET. See Dedicated breast positron emission tomography systems (dbPET) DBT. See Digital breast tomosynthesis (DBT) DCE‐MRI, bone and soft‐tissue tumor biopsy 606 DCF. See Deep cervical fascia (DCF) DECT. See Dual‐energy CT (DECT) Dedicated breast positron emission tomography systems (dbPET) 353, 355 Dedicated molecular breast‐imaging systems 354 Deep cervical fascia (DCF) 61 Deep layer of the deep cervical fascia (DLDCF) 61, 119 Deep learning (DL) advantage of 827 applications of 827 CNNs 826–827 discriminative approaches 827–828 generative approaches 829–830 graphics processor units (GPUs) 826 image translation 829–830 labeling and annotations 831 overfitting 831 privacy and ethical issues 831 rare conditions and diseases 831 representation learning 828–829 support vector machines (SVMs) 826

Deep learning‐based automatic detection (DLAD) 834 Deep morphology aided diagnosis network (DeepMAD) 833 Default mode network (DMN) 163 activation of 164 anatomy and function 164 Ab accumulation and 164, 165 concept 163–164 functional hubs 164 modulation 165 pathophysiology 164–165 task‐induced deactivation in 163 Degenerative arthritis 590, 592 Delayed second tumors, in head and neck 250 Delayed third tumors, in right piriform sinus 250 Dementia 170–177 Alzheimer’s disease 171–174 cause of 171 diagnosis of 136, 170–171 DLB 174 frontotemporal dementia 174–176 idiopathic NPH 176, 177 Dementia with Lewy bodies (DLB) 174 Dental metal artifacts 219, 219 effect depends on degree of head reclination 222 in MRI 219 in PET‐CT 220 Dental metal impact in MRI techniques 224 Dermal injection technique 361, 362 Diabetic foot infection 591, 593–595 diagnosis 731–732 18 F‐FDG PET and PET/CT 594–595 MRI 593 PET/MRI 594 plain radiography 593 SPECT/CT 594 three‐phase 99mTc‐MDP bone scan 593–594 WBC scintigraphy 731 Diabetic patients, with IHD 266 Diagnostic breast imaging 357 Diaphragm 86 Diaphragmatic apertures 86 Diethylenetriamine pentaacetic acid (DTPA) 838 Differential cortical retention 523 Differentiated thyroid cancer (DTC) 488–492 Differentiated thyroid carcinomas (DTC) 698 Diffuse esophageal spasm 385 Diffuse large B‐cell lymphoma (DLBCL) 439 Diffuse subtype gastric adenocarcinoma 412, 413

875

876

Index Diffusion kurtosis imaging (DKI) 197, 198 Diffusion tensor imaging (DTI) 58, 198 Diffusion weighted imaging (DWI) 166, 219, 334, 346 gliomas 197–198 ischemic core of 167 multiple myeloma 640 osteoporosis 666 Diffusion‐weighted imaging MRI (DW‐MRI) 566 Digital breast tomosynthesis (DBT) 353 Digital SiPM (dSiPM) 39 Direct piezoelectric effect 44 Disease‐based imaging findings cerebrovascular disease 165–170 corticobasal degeneration 189 dementia. See Dementia epilepsy. See Epilepsy movement disorder. See Movement disorders Display modes, in utrasound imaging 45 Disseminated peritoneal adenomucinosis 430 Distant metastases in breast cancer 366–369 in lung cancer 340, 433 DKI. See Diffusion kurtosis imaging (DKI) DLB. See Dementia with Lewy bodies (DLB) DLBCL. See Diffuse large B‐cell lymphoma (DLBCL) DLDCF. See Deep layer of the deep cervical fascia (DLDCF) DMN. See Default mode network (DMN) DNET. See Dysembryoplastic neuroepithelial tumor (DNET) Dobutamine 258 Dopamine transporter (DAT) 174 Doppler effect, in diagnostic ultrasound 45–46 Doppler shift. See Doppler effect Dorsal compartment, of forearm 127 Dorsal medial subsystems 164 Dorsal mPFC 164 Double‐contrast esophagography 390 Double cortex syndrome. See Subcortical band heterotopia Double‐layer detector technology 33 DSCT. See Dual‐source technology (DSCT) DTC. See Differentiated thyroid cancer (DTC) DTI. See Diffusion tensor imaging (DTI) Dual blue‐dye injection 362 Dual‐energy absorptiometry, osteoporosis 659–664 Dual‐energy CT (DECT) 33 Dual‐energy window (DEW) method 22, 37 Dual‐source technology (DSCT) 33 Duodenal adenocarcinoma 426

Duodenum 96–97 During‐reconstruction approach 37 DWI. See Diffusion‐weighted imaging (DWI) Dynamic contrast‐enhanced (DCE) MRI perfusion 196, 197 Dynamic (sequential) scintigraphy 522–523 renal transplant evaluation 527 Dynamic susceptibility contrast‐enhanced (DSC) MRI perfusion 196 Dysembryoplastic neuroepithelial tumor (DNET) 181, 182 Dysplasia 441

e

Early lymph node recurrence 249 Early scintimammography, limitations of 354 EATL. See Enteropathy‐associated T‐cell lymphoma (EATL) EC. See Eosinophilic granuloma (EC) ECE. See Extracapsular extension (ECE) ECG. See Electrocardiogram (ECG) Echo 43, 45 Echocardiography 269, 270 Edema 390 EEG. See Electroencephalography (EEG) EER. See Esophageal emptying rate (EER) EGD. See Esophagogastroduodenoscopy (EGD) E‐GISTS. See Extra‐gastrointestinal tract tumors (E‐GISTS) Elective lymph node dissection (ELND) 764 Electrocardiogram (ECG) 257 Electroencephalography (EEG) 163 Electromagnetic (EM) tracking 48–49 Emergency radiology 671–691 central nervous system 681–685 GI tract bleeding 685–688 hepatobiliary imaging 688–691 lung scintigraphy 679–682 myocardial perfusion imaging 671–674 renal scintigraphy 676–679 acute tubular necrosis 679 urinary tract obstruction 676–679 urine leak 679 vascular compromise 679 testicular torsion 673–676 Endocrine diseases, correlative imaging in 485–503 adrenal imaging 497–503 ACCs 501 adenomas 498–499 anatomy 497 correlative 498–499 cysts 500 hemorrhage 499–500

incidentaloma 497–498 lymphomas 502 metastases 502, 503 myelolipomas 498 pheochromocytomas 500–503 parathyroid imaging 492–495 anatomy and embryology 492 4D‐CT 492, 493 MRI 493 99m Tc‐sestamibi scintigraphy 493 overview of 492 PET and PET/CT 493 SPECT‐CT 493–495 ultrasound 492 pituitary imaging 494–497 anatomy 494–496 CT 496 18 F‐FDG PET/CT 496, 497 MRI 496 pituitary adenomas 496, 497 Rathke’s cleft cysts 497 thyroid imaging 485–492 anatomy 485 differentiated thyroid cancer 488–492 18 F‐FDG PET/CT 487–491 FNA biopsy 485–486 function 485 hybrid molecular imaging 488–492 123 I or 131I scan 489–491 124 I‐PET/US fusion imaging 489 99m Tc‐DMSA scintigraphy 486 99m Tc‐MIBI scintigraphy 486, 487 nodules 485 PET/MRI 492 SPECT/CT 489, 491, 492 SPECT/US 486 thyroid scintigraphy 486 ultrasound 485 Endoleaks 303 Endometrial cancer correlative imaging of 554–555 FDG PET/CT 556–561, 563 distant metastasis 557, 559 lymph node metastasis 557, 558 post‐therapy assessment 557–561 pre‐operative local staging 556–557 primary diagnosis 556 future directions for 563 MRI 555–556 abnormal findings 556 vs. other modalities 556 pre‐operative local staging 556 primary detection 556 recurrence 556 PET/MRI 561–562 PMB and 555 pre‐operative assessment 555 radiation therapy brachytherapy 562

Index CT simulation scan 562, 563 role of imaging in 562 risk factors for 554 staging of 555 TVS 555 type I 554 type II 554 Endometrium, FDG uptake 5 Endoscopic ultrasound (EUS), in gastric cancer 408, 413 Endoscopy 407–408 Endothoracic fascia 85 Endovascular aneurysm repair (EVAR) 301 Enteropathy‐associated T‐cell lymphoma (EATL) 441 Eosinophilic granuloma (EC) 634 Epididymitis 676 Epilepsy 177 FCD 178–179 FDG‐PET 177 hemimegalencephaly 179–180 hippocampal sclerosis 177–178 miscellaneous 181–183 neoplasms 181 other dysplastic diseases 180–181 Epipharynx. See Nasopharynx Epiploon. See Omentum Epithelioid hemangioendothelioma 468, 470 Epstein–Barr virus (EBV) 227 Erdheim–Chester diseases 821 Esophageal candidiasis 389–390 Esophageal diverticula 389 Esophageal dysmotility 390 Esophageal emptying rate (EER) 386 Esophageal malignancies 408 with bone metastases 409 imaging considerations 409–412 metastatic patterns 409 with peritoneal carcinomatosis 410 Esophageal squamous cell carcinoma 408 Esophageal transit, radiologic evaluation of 384 Esophageal transit time (ETT) 386 Esophagogastroduodenoscopy (EGD) 412, 444 Esophagus, structural abnormalities of 388–390 Ethmoid sinuses 71 EULAR. See European League Against Rheumatism (EULAR) European League Against Rheumatism (EULAR) 287 Eustachian tube 62, 64 EVAR. See Endovascular aneurysm repair (EVAR) Exercise stress test 257, 258 contraindications and indications for early termination 259

Extracapsular extension (ECE) 253–255 in large lymph node metastasis 254 in small lymph node metastasis 255 Extraconal orbital compartment 73 Extra‐gastrointestinal tract tumors (E‐GISTS) 420 Extrahepatic biliary tree 90–92

f

Falciform ligament 101 Fallopian tube 110–111 cancer, recurrent 445 False‐negative FDG‐avid lesions 363 False pelvis 109 False‐positive FDG‐avid lesions 363 Familial adenomatous polyposis (FAP) 441 FAP. See Familial adenomatous polyposis (FAP) Faraday’s law 41, 48 Fascia 219 Fast kV‐switching technology 33 FBP. See Filtered back projection (FBP) 18 F‐choline molecular structure of 139 PET/CT 139, 140 FDG‐avid breast lesions 363–364 FDG PET/CT endometrial cancer 556–561, 563 distant metastasis 557, 559 lymph node metastasis 557, 558 post‐therapy assessment 557–561 pre‐operative local staging 556–557 primary diagnosis 556 neuroendocrine tumors 518 ovarian cancer 566–568 post‐ovarian cancer therapy assessment 569–571 prostate imaging 537, 538 FDG‐6‐phosphate 3 FDOPA. See Fluoro‐dihydroxy‐ phenylalanine (FDOPA) 18 F‐DOPA and amino acid tracers 141 molecular structure of 141 NETs 518–519 PET‐CT, MIP images of 18 Female pelvis, anatomy of 109 cervix 110 ligaments of 112 ovaries 110–111 rectum 113–114 urethra 112 uterus 109–110 vagina 111 vascularization of 112–113 Female reproductive system, correlative imaging of 554–571. See also Endometrial cancer; Ovarian cancer

FET. See Fluoro‐ethyl‐thyrosine (FET) Fever of unknown origin 737–738 18 F‐FACBC (18F‐fluciclovine) PET/ CT 539–541 18 F‐FDG. See Fluorine‐18 (18F)‐ fluorodeoxyglucose (18F‐FDG) 18 F‐FDG PET adrenal adenomas 499 adrenal hemorrhages 500 adrenal metastases 502, 503 bone and soft‐tissue tumors 602, 605, 607 detection and diagnosis of 602, 605, 606 prognostic factors 607 response to therapy 607 bone metastases 610, 612–614 diabetic foot infection 594–595 hepatic HCC 467–468 hemangiomas 468, 471 metastases 461–463 hepatic melanomas 466 hyperparathyroidism and renal osteodystrophy 599, 601 musculoskeletal imaging physiologic uptake 577, 578 pitfalls 577–581 osteomyelitis 596 osteoporosis 666 periprosthetic infection 597–599 pheochromocytomas 501 pituitary adenomas 497 imaging 496 pituitary imaging 496, 497 rheumatoid arthritis 586, 588–589 spinal benign lesions 630 chordoma 637, 638 cord 647 EG 634 hemangiomas 635, 636 infections 627–628 malignant lesions 636 multiple myeloma 640 osteoblastoma 633 osteosarcoma 638, 639 spine metastases 644 spondyloarthropathies 590, 591 stress fractures 583 thyroid carcinoma 489–491 DTC 488 imaging 487–489 18 F‐FDG PET/CT adrenal adenomas 499 18 F‐FET 141, 142 18 F‐florbetaben 143

877

878

Index 18

F‐florbetapir 143 F‐FLT 141, 142 18 F‐fluciclovine 140, 411 18 F‐fluoride PET 531 18 F‐flutemetamol clinical indications 143 molecular structure of 143 PET 143, 144 FHWM. See Full‐width at half maximum (FHWM) Fibroblast activation protein (FAP) 814, 862 Fibromuscular stenosis 525 Fibrous dysplasia 634–635 FID. See Free induction decay (FID) Filtered back projection (FBP) 32 Fine needle aspiration (FNA) biopsy 485–486 Fissures 308 Fistula formation 445–447 Fluid Liquid Attenuation Inversion Recovery (FLAIR) 53, 178 Fluorinated PET radiopharmaceuticals 285 Fluorine‐18 (18F)‐fluorodeoxyglucose (18F‐FDG) 133, 136–139, 199, 286, 401 brain glucose metabolism using 164 molecular structure of 136 PET 264, 265 for aortitis 274 for cardiac sarcoidosis 273 for infective endocarditis 274 PET/CT 2, 16, 136–138, 285, 320, 402, 403, 408 AAA 301 anal cancer 439 angiography of heart 153 bilateral lung parenchyma 322 clinical examples 16–17 clinical indications of whole‐body 136 colon cancer examined by 139 COVID‐19 328 CUP 137 esophageal cancer 411 gastric cancer 414 GISTs 423 hibernating myocardium 138 image of pelvic region 9 lymph node metastases 418 lymphomas 440, 443 malignant pleural effusion 344 massive pulmonary fibrosis 323 mechanism of 3 in mediastinal tumors 345 normal biodistribution 3 peritoneal mucinous carcinomatosis 431 18

physiological causes and sites 3, 4, 5–6 preparation for 2–3 principle of 2–3 quantification of 3–4 sarcoidosis 320, 321 scan acquisition 3 in SPN 333–334 teratomas 347 thymoma 346–347 PET/MR 402–404 standardized uptake value 199 Fluoro‐dihydroxy‐phenylalanine (FDOPA) 199, 202 Fluoro‐ethyl‐thyrosine (FET) 199, 206, 207 Fluoromisonidazole (FMISO), PET 203–204 Fluoroscopy images 407 Fluoro‐thymidin (FLT) 203 18 F‐NaF. See 18F‐sodium fluoride (18F‐NaF) 18 F‐NaF PET/CT benign spinal lesions 630 bone metastases 609 hyperparathyroidism and renal osteodystrophy 599 multiple myeloma 641 osteoarthrosis 590, 592 osteoporosis 599, 666 Paget disease 599 periprosthetic infection 599 prostate cancer 547–548 spinal benign lesions 630 hemangiomas 635, 636 multiple myeloma 640 osteosarcoma 638, 639 spine metastases 644 stress fractures 583 traumatic fractures 582 Focal cortical dysplasia (FCD) 178–179 ILAE classification 179 MRI findings of 179 2‐year‐old boy with 180 Foot, compartments of 126 Forearm, compartment of 127, 128 Fracture risk assessment (FRAX) 659 Free induction decay (FID) 41, 42 Frequency, of sound waves 42, 43 Frontal lobes 55–56 Frontal sinus 71 Frontoparietal cortices, symmetric atrophy of 189 Frontotemporal dementia (FTD) 174–176 bvFTD 175, 176 progressive nonfluent aphasia 175 semantic dementia 175 18 F‐sodium fluoride (18F‐NaF) 148, 285 bone metastases 134 molecular structure 134

PET/CT with 134–136 uptake 294–295 FTD. See Frontotemporal dementia (FTD) Full‐width at half maximum (FHWM) 36 Functional gastric disorders 390–391 Functional hubs 164 Functional imaging modalities 30 Functional motility swallowing disorders 384 Fundus 109

g 67

Ga‐citrate scintigraphy, osteomyelitis 596, 597 Gadolinium‐based contrast agent (GBCA) 196 Gadolinium‐enhanced MRA renal artery stenosis 526 68 Ga‐DOTATOC 141–143 molecular structure of 142 PET/CT, neuroendocrine tumors 142, 143 68 Ga‐Exendin‐4 PET/CT 519 68 Ga‐FAPI PET/CT dosimetry results 815 future aspects 822 imaging protocol 815–816 nononcologic application 820–822 checkpoint inhibitor‐induced 820 Erdheim–Chester diseases 821 oncologic application advantages 817–818 disadvantages 818–819 imaging 816–817 radionuclide treatment 820 target volume delineation 819–820 tumor radiotracer uptake 815 Gallbladder 92–93, 94 Gallium‐67 (67Ga) citrate 320 Gallstone ileus 472, 474 Gamma camera 17, 19, 23, 33, 134, 144 components of 34 PET systems 35 Ganglioglioma 181, 182 68 Ga PET/CT osteomyelitis 596 in pulmonary nuclear medicine 320–324 68 Ga prostate‐specific membrane antigen (68Ga‐PSMA) 140, 204 Gas in aortic lumen 304 Gastric adenocarcinoma 412–419 colonic obstruction caused by recurrent 415 CT findings of 413 diffuse subtype 412 EUS 413 imaging in staging 417–418 M‐staging 418

Index N‐staging 418 T‐staging 418 Krukenberg tumors relating to 415–416 Lauren classification 412 metastatic patterns 414–415 direct invasion 414 hematogenous dissemination 415 lymphatic dissemination 414 subperitoneal dissemination 414 transperitoneal spread 414 MRI 413–414 PET/CT 414 recurrence evaluation 419 restaging and monitoring therapy 418–419 upper gastrointestinal imaging 412–413 WHO classification 412 Gastric emptying clinical indications 391 functional disorders of 390–391 radionuclide evaluation of 391 study in patient with diabetic gastroparesis 394 with dyspepsia 393 with gastroesophageal reflux disease 392 Gastric malignancies gastric adenocarcinoma 412–419 gastrointestinal stromal tumors 419–424 Gastrin‐releasing peptide receptor (GRPR) 546 Gastrocolic ligament (GCL) 414 Gastroesophageal reflux 385, 390 Gastrohepatic ligament (GHL) 414 Gastrointestinal contrast agent 10 Gastrointestinal (GI) lymphoma 407 Gastrointestinal scintigraphy gastroesophageal (GE) reflux 709–710 gastroparesis 710–711 Meckel’s diverticulum 711 Gastrointestinal stromal tumors (GISTs) 407, 419–424 clinical manifestations of 420 differential diagnosis 423 imaging features 422–423 management of patients with 420 multicentric small bowel 424 neurofibromatosis patient with 421 response to therapy in 423–424 treatment for 421–422 Gastrointestinal system 96–100 large intestine 97–98 small intestine 96–97, 98 spleen 100 stomach 96 Gastrointestinal (GI) tract 383 bleeding 685–688 direct invasion from non‐GI malignancy 445–447 metastatic disease to 444

Gaussian distribution 198 Gaussian kernel 22 GBCA. See Gadolinium‐based contrast agent (GBCA) GBM. See Glioblastoma (GBM) GCA. See Giant cell arteritis (GCA) GCT. See Giant cell tumor (GCT) Geiger mode 39 GE Millenium™ 21 Genetic disorders 440–441 Geometric center (GC) 394–395 calculating method 398 GI. See Gastrointestinal (GI) tract Giant cell arteritis (GCA) FDG PET/CT study in patient 289 US examination in 287 Giant cell tumor (GCT) 634 GISTs. See Gastrointestinal stromal tumors (GISTs) GI tract bleeding. See Gastrointestinal (GI) tract bleeding Glioblastoma (GBM) 195, 196 resection of right frontal 209 Gliomas anaplastic 208 arterial spin labeling 197 categories 194 histomolecular classification 195 intra‐axial tumor, patient with 210 left frontal 202 MRI for diagnosis 195–198, 201 CEST 198 DCE‐MRI perfusion 197 diffusion‐weighted imaging 197–198 DSC‐MRI perfusion 196–197 magnetic resonance spectroscopy 198 midline glioma 200 nuclear medicine imaging 198–199 amino acid tracers 199–207 18 F‐fluorodeoxyglucose 199 PET/MRI 207–212 recurrent IDH‐positive, patient with 211 response assessment in 195–198 WHO grade II and III 195 Glomerular filtration rate (GFR) 522, 526 Glottic larynx. See Glottis Glottis 68 Glucose transporters 2, 3 Gradients 40 Granulomatosis with polyangiitis (GPA) 285, 292 Granulomatous inflammations/ infections 320 Graves’ disease 486 Gray matter (GM) 53, 55 heterotopia 180–181 Greater omentum 101 Gross anatomy 308

GRPR. See gastrin‐releasing peptide receptor (GRPR) Gynecomastia in male patients 371, 372 in pediatric patients 373 Gyromagnetic ratio 40

h

Half‐time excretion (T1/2) 523 HAMN. See High‐grade appendiceal mucinous neoplasm (HAMN) HawkEye™ 21 Head–neck anatomy 219, 220 artifacts, affected by anatomical conditions of 219 dental metal artifacts 219 swallowing artifacts 220, 223 tumors chemotherapy 239–240 combined radio‐chemotherapy 246–249 CUP 238 delayed second 250 delayed third 250 extracapsular extension 253–255 nasal cavity and paranasal sinuses 224–225 post‐therapeutic findings 239–245 pre‐therapeutic findings 224 radiotherapy 243–245 surgical procedures 240–242 Heart failure 269 aorta, diseases of 274 left 326 with preserved ejection fraction 270–274 types of 269–274 valvular heart diseases in 274 Heart failure with preserved ejection fraction (HFpEF) 270–274 ARVC 271 cardiac amyloidosis 271–272 cardiac sarcoidosis 272–273 constrictive pericarditis, diagnosis of 270–271 HCM 271 Heart failure with reduced ejection fraction (HFrEF) 269–270 Heart mediastinum 84–85 Helical mode 31, 32 Helicobacter pylori infection 444 Hemangiomas hepatic 468, 470, 471 spinal 635, 636 Hematogenous dissemination, of gastric adenocarcinoma 415 Hematoma 527 Hemi‐liver 87 Hemimegalencephaly 179–180, 181

879

880

Index Hemodynamic parameters 196 Hemodynamic risk, in cerebrovascular disease 167–169, 170 Hemorrhage, adrenal gland 499–500 Hepatic IDA (HIDA) 689 Hepatic iminodiacetic acid (HIDA) scan 456 acute cholecystitis 471–472 bile duct obstruction 473, 476 biliary atresia 457 choledochal cyst 457, 458 Hepaticojejunostomy 475, 478 Hepatic vein, anatomical variants of 93 Hepatobiliary imaging 456–480 benign liver tumors 468, 470, 471 congenital hepatobiliary disorders 457–458 emergency radiology 688–691 hepatic metastases imaging of 459–464 origin of 458 malignant liver tumors 458–468 neoplasm 458–460 overview of 456 primary hepatic malignancy 467–470 Hepatobiliary scintigraphy 712–713 Hepato‐biliary system 86–95 Hepatocellular carcinoma (HCC) 467–468, 857 PET/CT images 300 Hepatoduodenal ligament (HDL) 414 Hereditary GI cancer 440–441 Heterotopic ossification (HO) 584–585 HGG. See High‐grade glioma (HGG) Hibernating myocardium 138 HIDA scanning. See Hepatic iminodiacetic acid (HIDA) scanning High‐grade appendiceal mucinous neoplasm (HAMN) 429 High‐grade glioma (HGG) 195 High probability V/Q scans 311 High resolution computed tomography (HRCT) 324 Hilar vessels, anatomy of 74–80 HIPEC. See Hyperthermic intraperitoneal chemotherapy (HIPEC) Hippocampal sclerosis 177–178 clinical characteristics 177 left 178 Hippocampus 164 Histopathological subtypes of breast cancer 351 HIV. See Human immunodeficiency virus (HIV) HO. See Heterotopic ossification (HO) Hounsfield units (HU) 7, 32, 52, 74 HPV. See Human papillomavirus (HPV) HRCT. See High resolution computed tomography (HRCT)

HU. See Hounsfield units (HU) Human immunodeficiency virus (HIV) 443–444 Human papillomavirus (HPV) 229 Hybrid imaging 1, 30, 46, 49, 133 advantages 47 in diagnosis of PE 312–315 PET/CT. See PET/CT PET/MR 47–48 SPECT‐CT. See SPECT‐CT triple modality scanner 48 Hybrid PET/MRI, detection of prostate cancer 549 Hydronephrosis 523–524, 676 Hyperglycemia 394 Hyperparathyroidism 599, 601 Hypertension, renovascular 524–526 Hyperthermic intraperitoneal chemotherapy (HIPEC) 429 Hyperthyroidism 144, 145 adverse effects 854 discharge and follow‐up 853 efficacy 853 functional imaging 851–852 pretreatment 853 safety 853–854 therapy administration 853 Hypertrophic cardiomyopathy (HCM) 271 Hypopharynx 68–69, 219 recurrence 249 tumors 236

i

IBD. See Inflammatory bowel disease (IBD) IBS. See Irritable bowel syndrome (IBS) ICA. See Internal carotid artery (ICA) ICD. See Implantable cardioverter defibrillator (ICD) Idiopathic normal pressure hydrocephalus (iNPH) 176, 177 IHD. See Ischemic heart disease (IHD) IHPBA. See International Hepatic Pancreatic Biliary Association (IHPBA) 131 I human serum albumin 310 123 I‐Ioflupane 154–155, 156 ILAE. See International League Against Epilepsy (ILAE) Ileocolic artery 100 Ileum 97 Image degrading factors, in SPECT 34–35 Image‐guided breast biopsies 358 comparison of 360 MBI‐guided biopsy 359 123 I‐metaiodobenzylguanidine (123I‐ mIBG) 156 NETs 519 pheochromocytomas 501, 502 131 I‐MIBG 519

123

I‐MIBG. See 123 I‐metaiodobenzylguanidine IMLN. See Internal mammary lymph nodes (IMLN) Immature teratoma 347 Immunotherapy 246–249, 320 Immunotherapy‐RANO (iRANO) 196 Implantable cardioverter defibrillator (ICD) 270 Indeterminate probability V/Q scans 311, 313 Indium‐111 (111In)‐labeled WBCs 292, 302, 303, 475, 477, 627 Infection/inflammation imaging cardiovascular infections cardiac implantable electronic device infections (CIEDs) 721–723 infective endocarditis (IE) 720–721 prosthetic vascular graft infections 723–725 fever of unknown origin 737–738 future aspects 738 intraabdominal infections 736–737 musculoskeletal infections acute osteomyelitis 725 diabetic foot infections 730–732 left tibial fracture nonunion 726 periprosthetic joint infections 726–728 septic arthritis 732 spondylodiscitis 728–730 pacemaker pocket infection 722, 723 pulmonary infections pneumonia 732 sarcoidosis 734–736 tuberculosis 732–734 radiopharmaceuticals 18 F‐fluorodeoxyglucose 720 67 Ga Citrate 717–718 indium‐111 labeled leukocyte scan 719 invitro labeled leukocyte scintigraphy 718–719 99m Tc Diphosphonates 717 technetium‐99m labeled leukocyte scan 719 Infective endocarditis 274 Inferior mesenteric artery 100 Inferior orbital fissure 71–72 Inferior vena cava (IVC), tumor thrombus in 295 Infero‐anterior nerve 60 Inflammation 390 Inflammatory and infectious disorders 441 Inflammatory arthritis 585–591 rheumatoid arthritis 585–589 18 F‐FDG PET/CT 586, 588–589 MRI 586 99m Tc‐MDP scan 586, 587

Index radiography 585–586 ultrasound 586 seronegative spondyloarthropathies 589–591 Inflammatory bowel disease (IBD) 383, 396–397, 442 Infra‐hyoid neck 61, 63. See also Supra‐hyoid neck posterior cervical space 67 visceral space 67–69 Infrahyoid spaces 220 Infrasound 43 Innermost intercostal muscle 85, 86 Innervation 310 131 I‐norcholesterol scans 498 111 In‐oxine‐WBC 401 Insufficiency fractures 582 Interfascial plane 108 Intermediate bronchus 76 axial CT image 77 level of 79 Internal carotid artery (ICA) 168 Internal mammary lymph nodes (IMLN) 361, 362 International Hepatic Pancreatic Biliary Association (IHPBA) 88 International League Against Epilepsy (ILAE) 179 International Myeloma Working Group (IMWG) 639, 798 Interventricular septum 85 Intraconal orbital compartment 73 Intra‐dermal injection 361 Intrahepatic biliary tree 90 Intravenous contrast, in CT scanning 7–10 Intravenous pyelography (IVP) 524 Intravenous urography (IVU). See Intravenous pyelography (IVP) Intravoxel incoherent motion (IVIM) 197 Invasive lobular carcinoma 363 Investing layer/fascia 61 111 In‐WBC scan 594, 595 131 Iodine 144–145 Iodine‐123 meta‐iodobenzylguanidine (123I‐MIBG) 258 124 I‐PET/US fusion imaging 489 Irritable bowel syndrome (IBS) 383 with diarrhea 392 MRI findings in 399 Ischemic heart disease (IHD) 257 diagnosis and prognosis CCTA 266–268 CMR imaging 268–269 coronary calcium scoring in 266 nuclear cardiology role in 257, 261–266 pharmacological stress test 258 Ischemic penumbra 166 Isocitrate dehydrogenase (IDH) 195 Iterative (IR) method 32–33, 38

Iterative reconstruction algorithm 22 IVIM. See Intravoxel incoherent motion (IVIM)

j

Jejunal arteries 100 Jejunum 97 Joints, anatomy of 123 Juvenile polyposis syndrome

441

k

Kaposi sarcoma (KS) 443 Kernels 32 Kidneys 104–105 anatomical variants 104–105 renal arteries 105 renal veins 105 Killian–Jamieson diverticulum 389 Krukenberg tumors 415–416

l

Labeled leukocyte imaging, osteomyelitis 596 Labeled leukocyte scan 154, 155 Lactating breasts, FDG uptake 5 Lactation 363, 364 LAD. See Left anterior descending coronary artery (LAD) LAMN. See Low‐grade appendiceal mucinous neoplasm (LAMN) Large intestine 97–98 Large lymph node metastasis 254 Large vessel vasculitides (LVV) 287–291 of aorta 288 PET/CT 287–289 PET/MRI 289 scoring methodology 289 US 287 Larmor equation 40, 41 Laryngeal cancer 234, 235, 255 Larynx 67–68, 219, 234–235 glottis 68 subglottis 68 supraglottis 67–68 tumor recurrence of 245 Lateral temporal cortex 164 Lauren classification, of gastric adenocarcinoma 412 LDCT. See Low dose‐computed tomography (LDCT) Left anterior descending coronary artery (LAD) 267 Left anterior perihepatic space 103 Left bronchus, anatomy of 79 left lower lobe bronchus 79 left upper lobe bronchus 79 lingular bronchus 79 Left common carotid artery 81

Left heart failure 326 Left hepatic vein (LHV) 88, 93 Left hilum, vascular anatomy of 79–80 Left intrahepatic bile duct, anatomical variants of 94 Left lower lobe (LLL) bronchus 79, 80 Left portal vein (LPV) 88, 89 Left posterior perihepatic space 104 Left posterior subhepatic space 104 Left subclavian artery 81 Left supramesocolic space 103–104 Left upper lobe (LUL) bronchus 74, 79–80 Left ventricular ejection fraction (LVEF) 262, 263, 269 Left vertical plane 88 Legg–Calvé–Perthes disease (LCPD) 696–697 Leptomeninges 54 LES. See Lower esophageal sphincter (LES) Lesser omentum 101 LGG. See Low‐grade glioma (LGG) Ligaments 101 Limbs, anatomy of 123 pelvic girdle and lower limbs 123–126 shoulder girdle and upper limbs 126–127 Linear attenuation coefficient (LAC) 31, 38, 48 Line of response (LOR) 35–36 Lingular bronchus 79, 80 Liver, anatomy of 86–90 classical 86 Couinaud classification 89, 90 fissures 86, 88 functional 87–90 ligaments 87 vascularization of 89 Liver Imaging Reporting and Data System (LI‐RADS) 467 Llymphoma, MIP image of FDG PET‐CT 13 Locoregional lymph nodes, anatomy of 351 Longitudinal magnetization recovery 42 Longitudinal relaxation 42 Low dose‐computed tomography (LDCT) 335 Low‐dose protocols 22 Lower cervical spine 118–119 Lower esophageal sphincter (LES) 383, 386 Lower leg, compartments of 124, 126 Lower limbs 123–126 Low‐grade appendiceal mucinous neoplasm (LAMN) 429 Low‐grade glioma (LGG) 195, 196 Low probability V/Q scans 311 177 Lu‐DOTATATE 142, 839 Lumbar spine 121–122

881

882

Index Luminal blood 286 Lung cancer causal risk factor 335 clinical symptoms and signs of 334–335 initial diagnosis and staging of M staging 336, 340 N staging 336–339 T staging 335–336 residual disease and recurrence 342 small cell lung cancer 335, 342, 343 treatment 340–342 Lung function, presurgical evaluation of 318–319 Lung perfusion imaging bronchopulmonary sequestration 705 cardiac shunts 704 congenital heart disease 704 99m Tc‐macroaggregated albumin (MAA) 703 neonatal patients 704 postsurgical evaluation 705–706 pulmonary embolism 706 vascular abnormalities 704 Lung perfusion scintigraphy 156, 157 Lung scintigraphy 679–682 Lung transplant rejection 682 177 Lu‐PSMA RLT adverse effects 845 efficacy 845 pre‐treatment 844 radionuclide therapy 844 safety 845 therapy administration 844 Lutetium‐yttrium oxyorthosilicate (LYSO) detector technology 21 LVEF. See Left ventricular ejection fraction (LVEF) LVV. See Large vessel vasculitides (LVV) Lymphatic dissemination, of gastric adenocarcinoma 414 Lymphatic drainage 310 patterns 432 Lymphatic system contrast‐enhanced ultrasound scan (CEUS) 768–769 ICG lymphography 762 intradermal lymphoscintigraphy 749 lymphatic imaging 751–752 lymphedema 747–748 lymphography 748 magnetic resonance imaging 769 non‐invasive 3.0 T MR lymphangiography 769 pattern visualization 752–762 peristaltic movement of 747 PET/CT lymphoscintigraphy 768 rest/ stress intradermal lymphoscintigraphy 749, 750, 752, 758

sentinel node 763 in breast cancer 766 in colorectal cancer 768 in head and neck cancer 766–767 in melanoma 764–765 in vulvar cancer 767–768 subcutaneous lymphoscintigraphy 748–749 superparamagnetic iron oxide (SPIO) nanoparticles 768 ultrahigh frequency ultrasound (UHFUS) 762–763 Lymph node metastases 249, 250 low conspicuity, unusual localization 252 recurrence with 253 remission after radio‐immunotherapy 248 of oropharyngeal SCC 246 Lymph nodes of neck 69–70 anatomical landmarks 69 levels 69–70 Lymphocele 527 Lymphoepithelial carcinoma 227 Lymphoma 439–440 brentuximab vedotin 773 contrast‐enhanced CT (CECT) 774 Deauville score 775, 776 diffusion‐weighted imaging (DWI) MRI 779 hepatic 466 Hodgkin lymphoma (HL) 772, 775 MALT 440 molecular imaging 773–774 non‐Hodgkin lymphoma (NHL) 772 PET/MRI 779 pitfalls false‐negative 779 false‐positive 776, 778–779 primary 439 programmed cell death‐ligand 1 (PD‐L1) 773 response assessment end‐of‐treatment response 775–776 immunotherapy response 776 interim response 774–775 staging 774 synopsis 779 TNM staging system 772 treatment in 773 WB‐MRI techniques aggressive lymphomas 780 apparent diffusion coefficient (ADC) 780 hodgkin lymphoma 780, 781 indolent lymphomas 780 synopsis 781 Lymphoma Response to Immunomodulatory Therapy Criteria (LYRIC) 776

Lymphomas adrenal 502 spinal 636, 637 Lymphoscintigraphy in breast cancer 363 injection techniques dermal injection technique 361, 362 dual blue‐dye and radiotracer SLN localization 362 peritumoral injections 361, 362 role of 360–361 Lynch syndrome 441

m

Macro aggregated albumin (MAA) 310 Macrophage‐laden atheromas 292 Macroscopic magnetization (MM) 41 Macroscopic magnetization longitudinal (MML) 41 Magnet, function of 40 Magnetic resonance cholangiopancreatography (MRCP) 690 Magnetic resonance imaging (MRI) 1, 2, 40, 49, 133, 163, 219 ACCs 501 acute cholecystitis 690 adrenal myelolipoma 499 basal ganglia, anatomy of 58 bladder cancer 530 components 40 CT vs. 52 dental metal artifacts in 219, 224 diabetic foot infection 593 endometrial cancer 555–556 abnormal findings 556 MRI vs. other modalities 556 pre‐operative local staging 556 primary detection 556 recurrence 556 evaluation of GIST 423 free induction decay and spatial encoding 41 gastric adenocarcinoma 413–414 gliomas, diagnosis of 195–198 hepatic cholangiocarcinoma 468 hemangioma 468, 470 lymphoma 466 metastases 460–462, 464 impact of dental metal 224 leg, compartments of 126 longitudinal magnetization recovery 42 malignant pleural disease 344 mediastinal tumors 345 musculoskeletal sarcomas grading 605 staging 601 nephroureteral obstruction 524 NETs 513

Index oral cavity 66 osteoarthrosis 590 osteoporosis 666–669 ovarian cancer 565–566 parathyroid imaging 493 periprosthetic infection 597 physical principles of 40 pituitary imaging 496 post‐ovarian cancer therapy assessment 569 prostate imaging 535–537 radiofrequency pulses 52 Rathke’s cleft cysts 497 renal masses 528, 529 semiology of brain 53 seronegative spondyloarthropathies 589 signal formation 40–41 small bowel 399–400 dynamic phase 399 extraintestinal disease 400 morphological phase 399 spinal ABCs 634 chordoma 637, 638 cord 646–647 EG 634 GCTs 634 hemangioma 635, 636 infection 627, 628 lymphoma 636, 637 osteoid osteoma 631, 632 trauma 648 spine metastases 643–646 osteosarcoma 637, 639 SPN 334 stress fractures 583 teratomas 347 thigh, compartments of 125 thoracic anatomy 74 thymoma 346 thyroid imaging 489 transverse magnetization decay 42 urinary tract obstruction 677 Magnetic resonance spectroscopy (MRS) 198 Magnetoencephalography. 163 Main pancreatic duct 97 Main photo‐peak energy window 22 Male patients, gynecomastia in 371, 372 Male pelvis, anatomy of 114–115 Malignant change, imaging patterns of 250 Malignant transformation of glial tumors 198 Malignant vascular thrombosis, imaging of 295–298 MALT. See Mucosa associated lymphoid tissue (MALT)

Mandibula 232, 233, 242 Mantle cell lymphoma 440 Masticator space 65–66 Mature teratoma 347 Maxillary sinus 71 Maximum intensity projection (MIP) image, of PET‐CT 7, 8, 10, 289 Maximum likelihood expectation maximization (MLEM) 38 MBI. See Molecular breast imaging (MBI) MCA. See Middle cerebral artery (MCA) Mean transit time (MTT) 196 Mechanical position encoding 49 Medial prefrontal cortex (mPFC) 164 Medial temporal atrophy, in AD 173 Medial temporal subsystems 164 Mediastinal tumors 344–345 Mediastinum 80 aortic arch and aorto‐pulmonary window 81–83 heart and paracardiac 84–85 normal cardiac anatomy 84 pulmonary arteries 83–84 supra‐aortic 80–81, 82 Mediolateral oblique (MLO) 353 Medium vessel vasculitides (MVV) 285, 291–292 Melanoma, hepatic 466 Mesentery 101, 102 Mesorectal fascia (MRF) 113 MET. See Metabolic equivalent (MET); Methionine (MET) Metabolic bone disease 599–603 hyperparathyroidism and renal osteodystrophy 599, 601 osteoporosis 599, 600 Paget disease 599, 602 Metabolic equivalent (MET) 258 Metabolic penumbra 167, 168 Meta‐Iodobenzylguanidine (mIBG) 156 adverse effects 849 discharge and follow‐up 848 efficacy 848 functional imaging 846 123 I‐mIBG 846 131 I‐mIBG 846 indications of 846 laboratory investigations 847 oncologic indications 846 pretreatment 846–847 safety 848 therapy administration 847–848 Metastases adrenal 502, 503 spine 641–644 bone scan 642–643, 646 combined imaging modalities 644–646 CT 642, 645 CT myelography 642

diagnosis of 642 18 F‐FDG PET 644 18 F‐NAF PET 644 MRI 643–646 SPECT 643 SPECT/CT 643 X‐ray 642 Metastasis‐directed therapy (MDT) 534 Metastatic breast cancer, systemic treatment of 368–369 Metastatic carcinoma rectum, FDG PET‐CTs of 17 Metastatic disease, to GI tract 444 Methionine (MET) 199, 200 MIBG. See Metaiodobenzylguanidine (MIBG) Middle cerebral artery (MCA) 168 Middle colic artery 100 Middle hepatic vein (MHV) 87, 93 Middle layer 61 Middle lobe (ML) bronchus 76 axial CT image 78 level of 79 Middle vertical plane 87 Mild cognitive impairment (MCI) 143 MIP. See Maximum intensity projection (MIP) Miscellaneous 181–183 MLO. See Mediolateral oblique (MLO) 99 Mo 145 Moderate probability V/Q scans 311 Molecular breast imaging (MBI) 354, 355, 358–360 clinical indications and practical uses of 356 image‐guided biopsies techniques 358, 359 Molecular imaging techniques 183 Molecular prostate imaging 537–545 11 C‐acetate PET/CT 538–539 11 C‐choline PET/CT 537, 538 FDG PET/CT 537, 538 18 F‐FACBC PET/CT 539–541 other radiotracers 545–546 Molecular subtypes of breast cancer 351 Molecular techniques, in breast imaging 352–360 Monte Carlo estimations 37 Motility disorders, investigated with barium swallow videofluoroscopy 385 Motion (M‐mode) 45 Motion artifacts, in PET‐CT 12 Mouth floor, recurrent tumor after reconstruction 242 Movement disorders 183 corticobasal degeneration 189 multiple system atrophy 185–188 Parkinson disease 183, 184–185 progressive supranuclear palsy 188–189

883

884

Index mPFC. See Medial prefrontal cortex (mPFC) MRA. See MRI angiography (MRA) MR‐based attenuation correction (MRAC) algorithms 48 MRI. See Magnetic resonance imaging (MRI) MRI angiography (MRA) 287 MRS. See Magnetic resonance spectroscopy (MRS) MSA. See Multiple system atrophy (MSA) M staging in lung cancer 336, 340 99m Tc‐diethylenetriamine penta‐acetic acid (99mTc‐DTPA) 522 99m Tc‐dimercaptosuccinic acid (99mTc‐DMSA) 522 99m Tc‐glucoheptonate 522 99m Tc‐labeled cholescintigraphy scintigraphy 689–690 99m Tc‐MDP bone scan osteoblastoma 633 osteoporosis 599, 600 rheumatoid arthritis 586, 587 99m Tc‐MDP SPECT osteoporosis 599 Paget disease 599 99m Tc‐mercaptoacetyltriglycine (99mTc‐MAG3) 522 99m Tc‐methylene diphosphonates (99mTc‐MDP) scan 609–611 Mucinous (colloid) adenocarcinoma 412 Mucinous neoplasms 427, 429–431 mucocele 429–430 pseudomyxoma peritonei 430–431 types 429 Mucocele 429–430 Mucosa associated lymphoid tissue (MALT) 439 Multidetector computed tomography (MDCT) 313, 335 Multiparametric prostate MRI (mpMRI) 535–536 Multiphase bone scintigraphy HO 585 stress fracture 583 Multiphase CT (4D‐CT) 492, 493 Multiple imaging protocols 258 Multiple myeloma (MM) 638–641, 781–784 bone scan 640 combined diagnostic modalities 641 CT 640 DWI 640 MRI 640 PET/CT 640–641 PET/MRI 641 skeletal survey 639 WBCT 640 Multiple system atrophy (MSA) 156, 185–188

Multislice CT scanners 52, 53 Musculoskeletal imaging 577–614 bone scintigraphy 577 degenerative arthritis 590, 592 18 F‐FDG PET physiologic uptake 577, 578 pitfalls 577–581 infection 591, 593–599 diabetic foot 591, 593–595 osteomyelitis 596–598 periprosthetic and other hardware infection 596–599 spondylodiscitis 596 inflammatory arthritis 585–591 metabolic bone disease 599–603 hyperparathyroidism and renal osteodystrophy 599, 601 osteoporosis 599, 600 Paget disease 599, 602 metastatic disease 608–614 18 F‐FDG PET 610, 612–614 18 F‐NaF PET 609 18 F‐NaF PET/CT 609 99m Tc‐methylene diphosphonates 609–611 normal variations 577–579 primary bone and soft‐tissue tumors 599, 601–608 biopsy 605–606 grading 605 initial detection/diagnosis and staging 601–605 other PET agents 607, 608 prognostic factors 607 response to therapy 607 restaging and detection of recurrence 607, 608 relevant anatomy 577 rheumatoid arthritis 585–589 18 F‐FDG PET/CT 586, 588–589 MRI 586 99m Tc‐MDP scan 586, 587 radiography 585–586 ultrasound 586 seronegative spondyloarthropathies 589–591 stress injuries 582–584 traumatic fractures 579–582 Musculoskeletal infections acute osteomyelitis 725 diabetic foot infections 730–732 left tibial fracture nonunion 726 periprosthetic joint infections 726–728 diagnosis 726 67 Ga imaging 727 molecular imaging 726 shoulder joint replacement surgery 728 symptoms of 726 treatment of 726

septic arthritis 732 spondylodiscitis 728–730 bone scintigraphy 729 diagnosis 729 infected spinal hardware 729 postoperative 729 postoperative changes 730 symptoms 728 MVV. See Medium vessel vasculitides (MVV) Myelodysplastic syndrome (MDS) 843 Myocardial delayed enhancement (MDE) 268 Myocardial perfusion imaging (MPI) 151–154, 257, 671–674 prognostic variables in cardiac 263 radiotracers and protocols 258 Myocardium, FDG uptake 5

n

N‐acetylaspartate 198 Na18F bone imaging 371 NaI (Tl) scintillation crystals 21, 23 Nasal cavity boundaries 70–71 malignant disease of 224–225 and paranasal sinuses 224–225 Nasopharyngeal carcinoma (NPC) 227, 228 Nasopharynx 62–64, 219, 227 National Comprehensive Cancer Network (NCCN) guidelines 547 Neck. See also Head–neck fascia 61 infra‐hyoid. See Infra‐hyoid neck lymph nodes of 69–70 supra‐hyoid. See Supra‐hyoid neck NECR. See Noise equivalent count rate (NECR) Negative predictive value (NPV) 262 Neoadjuvant chemotherapy (NAC) 368 Neoadjuvant therapy, for breast cancer 367–368 Neoplasms 181 Nephrography, phases of 523 Nephroureteral obstruction 523–525 Nephro‐urinary tract pathologies 521–531 anatomy 521–522 GFR 522 image acquisition 522–523 dynamic (sequential) scintigraphy 522–523 static scintigraphy 524 nuclear medicine bladder cancer 530–531 nephroureteral obstruction 524, 525 renal masses 528, 530 renal transplant evaluation 527–528 renovascular hypertension 526

Index radiology bladder cancer 530 nephroureteral obstruction 523–524 renal masses 528, 529 renal transplant evaluation 527 renovascular hypertension 526 tubular binding 522 tubular secretion 522 NETs. See Neuroendocrine tumors (NETs) NETTER‐1 trial 839 Neuroendocrine neoplasms (NEN) checklist 839 discharge and follow‐up 841–842 efficacy 842 functional imaging 839–840 goals of 840 indications of 840 overview of 838–839 pretreatment 841 safety 842–843 therapy administration 841 Neuroendocrine tumors (NETs) 407, 425–427, 462–465, 512–519 appendiceal 428–429 CT 513 functional imaging FDG PET 518 18 F‐DOPA PET 518–519 future perspectives 519 68 Ga‐Exendin‐4 PET/CT 519 123 I‐MIBG and 131I‐MIBG 519 indications for 515, 516 positron emission tomography 513 radiopharmaceuticals for 514, 515, 517 single‐photon emission computerized tomography 513 SSTR PET 516, 518 targets for 514, 515 functional NET 512 gastroenteropancreatic 512 grading systems 512 indications for radiological imaging 512 MIP image of 18 MRI 513 nonfunctional NET 512 PET/CT with 68Ga‐DOTATOC 142, 143 transabdominal USG 513 tumor heterogeneity 516 Neurogenic dysphagia 387 Neuroimaging 18 F‐sodium fluoride PET 694–695 modalities 171 PET‐CT 693–694 PET‐MRI 693 semiology of brain in 53–54 Neuromelanin‐sensitive MRI (NmMRI) 184, 185 Neuronuclear imaging of brain perfusion 163

Neuro‐oncology, technological insights and use 207–212 NmMRI. See Neuromelanin‐sensitive MRI (NmMRI) Noise 34 Noise equivalent count rate (NECR) 36 Nomenclature of bronchopulmonary segmental 308 Nonglial tumors 194 Non‐Hodgkin’s lymphoma, for staging FDG PET‐CT evaluation 15 Noninfectious vasculitides 285 Non‐invasive 3.0 T MR lymphangiography 769 Non‐Krukenberg metastatic tumors 416 Nonmolecular breast imaging modalities 353 Nonmucinous neoplasms 427–428, 431–432 Nonobstructive hydronephrosis 524 Nonradioactive iodine 144 Non‐small‐cell lung cancer (NSCLC) 335, 340 Nonsteroidal anti‐inflammatory drugs 328 Nontumoral thrombosis 295 Normal/near‐normal V/Q scans 311 Normal pressure hydrocephalus (NPH) 176 Novel PET agents 369–370 NPC. See Nasopharyngeal carcinoma (NPC) NPH. See Normal pressure hydrocephalus (NPH) NPV. See Negative predictive value (NPV) NSCLC. See Non‐small‐cell lung cancer (NSCLC) N staging in lung cancer 336–339 Nuclear cardiology guide therapy 264 and IHD diagnosis 261–262 management 257 prognosis 262–266 special patient populations 266 Nuclear magnetic resonance (NMR) 40 Nuclear medicine correlative assessment across 212 for COVID‐19 infection 328 68 Ga PET/CT imaging in 320–324 imaging 33, 133, 198–199 amino acid tracers. See Amino acid tracers 18 F‐fluorodeoxyglucose 199 PET 35–37 SPECT 33–35 incidental breast findings in 371–373 imaging male patients 371 male patients 371 Na18F bone imaging 371 pediatric patients 373 82 Rb cardiac imaging 371

metastable nuclide for 145 physicians, and CT 4 in pulmonary infection and inflammation 320

o

Obstructive hydronephrosis 524 Occipital lobes 56 Ocular globe 73 Odontoid process 118 OEF. See Oxygen extraction fraction (OEF) Olfactory fossa 71 Olfactory nerve 60 Oligometastatic prostate cancer 140 Omental bursa 103 Omentum 101, 103 One‐day protocol 258 OPES. See Oropharyngo‐esophageal scintigraphy (OPES) Optic nerve 60 Oral cavity 66 anterior 232 boundaries 67 posterior 234 structures 67 tumors of 231 Oral contrast, in CT scanning 7–10 Oral transit time (OTT) 386 Orbit bone structure 71 boundaries 71–72 ocular globe 73 preseptal space 71–72 retro‐ocular spaces 73 Ordered subsets expectation maximization (OSEM) algorithm 38 Oropharyingo‐esophageal tract 383 anatomy and physiology 383–384 functional motility swallowing disorders 384 radionuclide evaluation of 385–388 patient preparation and scintigraphic acquisition 386 qualitative and quantitative image processing 386–388 structural abnormalities of esophagus 388–390 Oropharyngeal carcinoma 229 chemotherapy of 240 combined radio‐chemotherapy 246 Oropharyngeal fascia 61 Oropharyngo‐esophageal scintigraphy (OPES) 385, 387, 388 with liquid bolus 389 patient with dermatomyositis and dysphagia 388 Oropharynx 64, 219, 229 Orthopantomography 233

885

886

Index Osseous metastases pain beta‐emitting therapeutic radiopharmaceuticals 849 discharge and follow‐up 850 efficacy 850–851 functional imaging 850 laboratory investigations 850 pretreatment 850 safety 851 therapy administration 850 Osseous metastatic disease 367, 369, 608–614 Osteoarthrosis (OA) 590, 592 Osteoblastoma 632–633 CT 632–633 18 F‐FDG PET/CT 633 MRI 633 99m Tc‐MDP bone scan 633 plain X‐ray 632 symptoms 632 Osteoid osteoma, spinal 630–632 clinical features 630 CT 631, 632 etiology of 630 MRI 631, 632 PET 631 PET/CT 632 plain X‐ray 630–631 scintigraphy 631 SPECT/CT 632 Osteomyelitis 596–598 Osteoporosis 599, 600, 659–669 clinical diagnosis of 659 fracture risk assessment 659 imaging CT 662, 664 dual‐energy absorptiometry 659–664 MRI 666–669 PET 665–666 quantitative computerized tomography 664–665 quantitative ultrasound 668 radiography 660, 661 pathophysiology 659 in United States 659 in women 659 Ostiomeatal complex 71 OTT. See Oral transit time (OTT) Ovarian cancer 563–571 CT scan 565 FDG PET/CT 566–568 future directions for 570, 571 MRI 565–566 PET/MRI 567, 569 post‐therapy assessment 569–570 CA‐125 levels 569 CT scan 569 FDG PET/CT 569–571 MRI 569

pre‐operative assessment 564 screening for 564 transvaginal sonography 564–565 Ovaries 5, 110–111 Oxygen extraction fraction (OEF) 165–166

p

PA. See Aspiration percentage (PA) Paget disease 599, 602 PAN. See Polyarteritis nodosa (PAN) Pancreas 93–95 parts 93 vascularization of 93–95 Papillary thyroid cancer (PTC) 237, 698 Paracardiac mediastinum 84–85 Parahippocampal gyrus 164 Parallax error 37 Paranasal sinuses 70 anatomical landmarks in CT 71 consist of 71 malignant disease of 224–225 SCC of 225 Parapharyngeal space 64, 65 Paraspinal muscles 119 Parathyroid anatomy 492 embryology 492 imaging 492–495 4D‐CT 492, 493 MRI 493 99m Tc‐sestamibi scintigraphy 493 overview of 492 PET and PET/CT 493 SPECT‐CT 493–495 ultrasound 492 Parenchymal peak‐time activity (Tmax) 523 Parietal lobes 55–56, 57 Parietal peritoneum 101 Parietotemporal cortex, in Alzheimer disease 172–173 Parkinson’s disease (PD) 141 DAT‐SPECT 184–185 diagnosis of 183, 184 neuromelanin‐sensitive MRI 184 other parkinsonisms 185 and secondary parkinsonian conditions 184 Parotid malignancy, SCC 227 Parotid space 66 Parotitis, post RT edema and 243 Partial response (PR) 195 Partial volume correction (PVC) methods 22, 37 Partial volume effects (PVEs) 34, 37 PCC. See Posterior cingulate cortex (PCC) PCP. See Pneumocytis carinii pneumonia (PCP) PD. See Parkinson’s disease (PD); Progressive disease (PD)

PE. See Pulmonary embolism (PE) Peak systolic velocity (PSV) 526, 527 Pediatric disease bone scintigraphy chronic nonbacterial osteomyelitis 696 Legg–Calvé–Perthes disease (LCPD) 696–697 99m Tc‐labeled disphosphonates 696 neuroblastoma 697–698 brain flow imaging in brain death 709 cisternography 709 ommaya reservoir 709 planar radionuclide angiogram 708 gastrointestinal scintigraphy gastroesophageal (GE) reflux 709–710 gastroparesis 710–711 Meckel’s diverticulum 711 hepatobiliary scintigraphy 712–713 lung perfusion imaging bronchopulmonary sequestration 705 cardiac shunts 704 congenital heart disease 704 99m Tc‐macroaggregated albumin (MAA) 703 neonatal patients 704 postsurgical evaluation 705–706 pulmonary embolism 706 vascular abnormalities 704 neuroimaging 18 F‐sodium fluoride PET 694–695 PET‐CT 693–694 PET‐MRI 693 nuclear voiding cystogram 708 PET‐CT imaging 693 renal imaging diuretic renography 706–707 radiopharmaceutical agents 706 renal cortical scintigraphy 707–708 SPECT myocardial perfusion imaging coronary artery anomalies 702–703 kawasaki disease 703 99m Tc‐tetrofosmin agents 700 thyroid scintigraphy pathophysiology 698 pediatric thyroid nodules 698 thyroiditis 700 thyroid malignancy 698, 700 Pediatric‐oncologic patients 371–373 Pediatric patients, gynecomastia in 373 Pelvic girdle 123–126 Pelvis, anatomy of 109–116 false pelvis 109 female 109–114 male 114–115 peritoneal and extraperitoneal spaces 115–116 true pelvis 109

Index Peptide receptor radionuclide therapy (PRRT) 838 adverse effects 842 amino acid formulation 841 checklist 839 goals of 839 laboratory investigations 841 177 Lu‐DOTATATE‐based 841 Perfusion defect 311 Perfusion imaging 196 DCE‐MRI 197 DSC‐MRI 196–197 Perfusion scintigraphy, ventilation scintigraphy and 324–325, 327–328 Perfusion‐weighted image (PWI) 166, 167 Perineural tumor, extension 227, 229 Peripheral zone (PZ), prostate gland 533 Periprosthetic infection 596–599 bone scintigraphy 598 18 F‐FDG PET 597–599 18 F‐NaF PET 599 MRI 597 SPECT/CT 598 Perirenal space 106, 108 Peritoneal dissemination, of gastric cancer 414, 417 Peritoneal ligaments 102 Peritoneal mucinous carcinomatosis 430 Peritoneum 101 Peritumoral radiotracer injection 361, 362 Perivertebral space (PVS) 67, 119 PET. See Positron emission tomography (PET) of spinal cord 647 spinal osteoid osteoma 631 PET/CT 47 artifacts in fusion 12–15 attenuation correction artifacts 12 beam hardening artifact 12, 14 motion artifacts 12 radiopharmaceutical related 15 truncation artifacts 14–15 basics 2 bone and soft‐tissue tumor biopsy 606–607 restaging and detection of recurrence 607, 608 bone and soft‐tissue tumors 605 carcinoma of left breast 12 11 C‐choline 139 CT protocols in 11 dental metal artifacts in 221 display of fused images 11 18 F‐choline 139, 140 FDG 2, 136–138 clinical examples 16–17 clinical indications of whole‐body 136 colon cancer examined by 139 non‐Hodgkin’s lymphoma for 15

normal biodistribution 3 physiological causes and sites 3, 4, 5–6 preparation for 3 principle 2–3 scan acquisition 3 and SUV 3–4 tracers beyond 16–17 18 F‐flutemetamol 143 with 18F‐NaF 134–136 with 68Ga‐DOTATOC 142 MIP image of 7, 8, 10, 13 multiple myeloma 640–641 pelvic region 9 prototype of scanner 11 PSMA 540–545 radiolabeled choline 139–140 spinal lymphoma 636, 637 spinal osteoid osteoma 631 spine metastases 644, 645 traumatic fractures 582 PET MPI 259, 261, 262 PET/MR 47–48 PET/MRI imaging systems 138, 369–370 advantages 208 bone and soft‐tissue tumors 608 development of 134 diabetic foot infection 594 for endometrial cancer 561–562 IDH polymorphism prediction 209 multiple myeloma 641 in neuro‐oncology 207–212 ovarian cancer 567, 569 of spinal cord 647 spine metastases 644 PETVAS score 289 Peutz–Jeghers syndrome 441 PFS. See Progression‐free survival (PFS) Pharmacological stress test 258–260 characteristics of vasodilator in 258 contraindications and indications for early termination 259 dobutamine 258 types 258 Pharyngeal‐basilar fascia 62, 63 Pharyngeal mucosal space 62–64 nasopharynx 62–64 oropharynx 64 Pharyngeal transit time (PTT) 386 Pharynx 220 Pheochromocytomas 500–503 Photomultiplier tubes (PMTs) 21, 23, 33–34, 47 Photo‐peak energy window 22 Phrenicocolic ligament 101 Physical principle on CT technique 31–32 Piezoelectricity 44 PIOPED. See Pulmonary embolism diagnosis (PIOPED)

Piriform recess, squamous cell carcinoma of 236 Pitch 31, 32 Pituitary gland 494–497 anatomy of 494–496 imaging CT 496 18 F‐FDG PET/CT 496, 497 MRI 496 pituitary adenomas 496, 497 Rathke’s cleft cysts 497 Plain radiography bone and soft‐tissue tumors 601 diabetic foot infection 593 Plain X‐rays, spinal infection 626 osteoid osteoma 630–631 osteosarcoma 637 trauma 647–648 Planar 99mTc‐MAA imaging “hot quadrate” sign on 476, 479 liver mapping process 480 Planar V/Q images 311–312, 315 perfusion defects of lung 313, 317, 318 pleural effusions 314 radiation exposure 321 Platelet derived growth factor alpha (PDGFRa) 419 Pleural diseases 343–344 Pleural surfaces 85–86 PMB. See Post‐menopausal bleeding (PMB) PMP. See Pseudomyxoma peritonei (PMP) PMR. See Polymyalgia rheumatica (PMR) PMTs. See Photomultiplier tubes (PMTs) Pneumocytis carinii pneumonia (PCP) 320 Pneumonia 325–326 Pneumonitis 320 PNFA. See Progressive nonfluent aphasia (PNFA) Polyarteritis nodosa (PAN) 291 Polymicrogyria 180 Polymyalgia rheumatica (PMR) 290 Portal vein, anatomical variants of 91 Positron–electron annihilation reaction 35 Positron emission tomography (PET) 1, 2, 35–37, 133, 163, 219, 838 and axillary lymph node evaluation 366 block detector arrangement. 36 in breast cancer 362–363, 370, 371 distant metastases 366–369 initial staging 365–366 primary 363–365 data acquisition 35–36 developments 38–39 FDOPA 202 FET 201–202, 206, 207 FLT 203 FMISO 203–204 image reconstruction 38

887

888

Index Positron emission tomography (PET) (cont’d) MET 200 NETs 513 and non‐PET radionuclides 285 novel agents 369–370 osteoporosis 665–666 physical principles of 35 quantitative data corrections in 37–38 radiopharmaceuticals 134–144 radiotracers, characteristics of 260 suspected disease recurrence and restaging 369 transaxial 38 Positron emission tomography‐computed tomography. See PET/CT Positron emission tomography/magnetic resonance imaging (PET/MRI) brain dementia frontotemporal dementia (FTD) 800 functional neuroimaging 800 morphological features 801 brain tumors advanced modalities 789 advantages 789 central nervous system (CNS) 788 computed tomography 788 18 F‐fluoromisonidazole (FMISO) 789 limitations 788 radiotracers 789 volumetric interpolated breath‐hold examination (VIBE) 789 cardiac sarcoidosis 804, 806–808 head and neck cancer challenges of 790 characterization 789 disadvantages 790 lymph node involvement 790 prognostication 789 hepatocellular carcinoma 791 multiple myeloma 796, 798 pancreas disease 18 F‐FDG PET/CT 792–794 intraductal papillary mucinous neoplasms (IPMNs) 792 pancreatic neuroendocrine tumors (PNETs) 795 pediatric lymphoma Ann Arbor classification 800 non‐Hodgkin lymphomas (NHLs) 798 prostate cancer diagnostic value 795 prostate‐specific antigen (PSA) 795 restaging settings 795 rectal cancer 796, 797 retroperitoneal fibrosis (RPF) baseline 802 evaluation of 802–803 imaging 802

medical treatment 802 post‐therapy 803 vasculitis 804, 807 Posterior aortic arch 81 Posterior cervical space 67 Posterior cingulate cortex (PCC) 164, 165, 172 Posterior cingulate gyrus 172 Posterior fossa 61 Posterior inferior parietal lobe 164 Posterior oral cavity 234 Posterior pararenal space 106, 108 Posterior segment bronchi 76 Posterior thoraco‐lumbar fascia (PTLF) 122 Post‐menopausal bleeding (PMB) 555 Postoperative cerebral hyperperfusion syndrome (CHS) 169–170 Post‐reconstruction approach 37 Post‐styloid parapharyngeal space 64 Post‐therapeutic findings, in head and neck diseases 239–245 Power Doppler 46 PR. See Partial response (PR) Precuneus 172 Pregnancy, pulmonary embolism and 680–681 Presacral fascia 113 Preseptal space 71–72 Pre‐styloid parapharyngeal space 64 Pre‐therapeutic findings, in head and neck diseases 224 Prevertebral layer/fascia 61 Pre‐vertebral space. See Perivertebral space Prevesical space 116 PRF. See Pulse repetition frequency (PRF) Primary breast cancer 363–365 Primary hepatic malignancy 467–470 Primary motor cortex 56 Primary spinal tumors 628, 629 Probe operation 45 Prognostic staging, for breast cancer 365 Progression‐free survival (PFS) 196 Progressive disease (PD) 195 Progressive nonfluent aphasia (PNFA) 175, 176 Progressive supranuclear palsy (PSP) 188–189 Propagation speed, of sound waves 42–43 ProstaScint 540 Prostate cancer (PC) 371, 373, 466, 534 bone‐specific imaging in 547–548 18 F‐sodium fluoride 547–548 99m Tc‐phosphonates 547 castrate‐resistant PC (CRPC) 843 discharge and follow‐up 844 efficacy 845 functional imaging 843 indications of 844

laboratory investigations 844 177 Lu‐PSMA RLT 843 PET/MRI 549–550 safety 845 theranostics 549 therapy administration 844 Prostate diseases 534 Prostate embryology and anatomy 533–534 Prostate gland, zonal anatomy 533 Prostate imaging 534–550 future prospects 550 molecular 537–545 11 C‐acetate PET/CT 538–539 11 C‐choline PET/CT 537, 538 FDG PET/CT 537, 538 18 F‐FACBC PET/CT 539–541 other radiotracers 545–546 non‐molecular 534–537 CT scan 535 MRI‐guided biopsy 536–537 multiparametric prostate MRI 535–536 transrectal ultrasound 534–535 ultrasound 534–535 prostate cancer 534 bone‐specific imaging in 547–548 18 F‐sodium fluoride 547–548 hybrid PET/MRI 549 99m Tc‐phosphonates 547 theranostics 549 Prostate‐specific membrane antigen (PSMA) 204, 411, 644, 843 ligands 540 miTNM staging 543, 545, 546 PET/CT 540–545 PSMA‐RADS score 545, 546 receptors 530 theranostics 549 Prostate vesicles 114–115 Prostatitis 534 Prosthetic graft material 302, 303 Pseudo‐diffusion coefficient 198 Pseudomyxoma peritonei (PMP) 427, 430–431 PSMA. See Prostate‐specific membrane antigen (PSMA) PSMA PET/CT 531 PSP. See Progressive supranuclear palsy (PSP) PSP‐Richardson syndrome (PSP‐RS) 188 Pterygopalatine fossa 72 PTLF. See Posterior thoraco‐lumbar fascia (PTLF) PTT. See Pharyngeal transit time (PTT) Pulmonary arteries 83–84, 679, 680 Pulmonary embolism (PE) 310–311, 680–681 chest radiograph 311

Index hybrid imaging in diagnosis 312–315 interpretation using planar V/Q images 311–312 using tomographic images 312 Pulmonary embolism diagnosis (PIOPED) 311 Pulmonary epithelial permeability studies 315–318 Pulmonary hilum, anatomy of 74 Pulmonary hypertension 326–327 Pulmonary infection and inflammation 320 Pulmonary infections pneumonia 732 sarcoidosis clinical presentation 734–735 treatment response 735 tuberculosis 732–734 common sites 733 diagnosis 733 latent tuberculosis infection 734 lymphatic pattern 733 standardized uptake value (SUV) 734 WBC imaging 733 Pulmonary scintigraphy for cardiopulmonary diseases 325–327 in COPD 324–325 Pulmonary system anatomy of bronchopulmonary segmental 308 gross anatomy 308 blood supply 310 conduction zone 308–310 COVID‐19 infection 327–328 lymphatic drainage 310 nuclear medicine 320 68 Ga PET/CT imaging in 320–324 in infection and inflammation 320 pulmonary embolism. See Pulmonary embolism (PE) pulmonary epithelial permeability studies 315–318 quantitative lung scintigraphy 318–319 respiratory zone 310 tracheobronchial tree 308 Pulmonary ventilation perfusion (V/Q) scans 680, 681 Pulsed wave (PW) Doppler 46 Pulse repetition frequency (PRF) 45 PWI. See Perfusion‐weighted image (PWI)

q

Quantitative data corrections 37–38 Quantitative lung scintigraphy 318–319 Quantitative method 293 Quantitative ultrasound (QUS) osteoporosis 668

r

RA. See Aspiration area (RA) Radiation therapy 320 Radio frequency (RF) system 40 Radiography HO 584 osteoarthrosis 590 Radioguided surgery 159 of breast cancer 158 Radioimmunotherapy (RIT) adverse effects 862 antibody 860 antigen 860 clinical applications 860 clinical studies 862 efficacy 861 functional imaging 860 properties of 860 radionuclide 860 safety 861 targeted malignancy 860 Radioiodine 144–145, 854 Radiolabeled autologous leukocytes, scintigraphy with 401–402 Radiolabeled choline 139–141 advantages 140 11 C‐choline 139 18 F‐choline 139 Radiology, correlative assessment across 212 Radionuclide angiography (RNA) 269 Radionuclide bone scan scintigraphy 367 Radionuclide therapy discharge and follow‐up 856 efficacy 856 future aspects 861–862 goals of 855 hyperthyroidism 852 meta‐Iodobenzylguanidine (mIBG) 846 neuroendocrine neoplasms (NEN) 840 osseous metastases pain 850 pretreatment 855 prostate cancer (PC) 844 radioimmunotherapy (RIT) 861 safety 856–857 selective internal radiation therapy (SIRT) 857 therapy administration 856 thyroid cancer 855 Radiopharmaceuticals 133 for NET 514, 515, 517 PET 134–144 amiloyd imaging 143–144 18 F‐DOPA 141 18 F‐FDG 136–139 18 F‐NaF 134–136 68 Ga‐DOTATOC 141–143 radioguided surgery 159 radiolabeled choline 139–141 small bowel and colon 393–394

SPECT 144–158 123 I‐Ioflupane 154–155 123 I‐metaiodobenzylguanidine 156 131 Iodine 144–145 labeled leukocyte 154, 155 myocardial perfusion imaging 151–154 radioguided surgery 159 renal scintigraphy 148–151 99m Tc‐macroaggregatealbumin 156– 158 99m Tc‐MDP 133, 134, 147–148 99m Technetium 145–146 Radiotherapy 243 and parotitis 243 of retro‐auricular SCC 244 tumor recurrence of larynx 245 Radiotracers 177 injection techniques 361–362, 363 myocardial perfusion imaging 258–260 SLN localization technique 362 uptake in breast 371–373 imaging male patients 371 imaging pediatric patients 373 Radon space 32, 33 Radon transform 31 RANO. See Response Assessment in Neuro‐ Oncology (RANO) Rasmussen’s encephalitis 181–182, 183 Rathke’s cleft cysts 497 82 Rb cardiac imaging 371 Real‐time ultrasound, image fusion of 48–49 RECIST. See Response Evaluation Criteria in Solid Tumors (RECIST) Reconstruction methods CT 32–33 PET 38 Rectal adenocarcinoma 434, 435, 437 Rectum 113–114 Red blood cells 304 Reflection 43, 44 Reflux esophagitis 444 Refraction 43 and artifacts 44 and reflection 44 Regional lymph node 366 Regional spine anatomy 117–123 cervical spine 117–119 lumbar spine 121–122 sacrum and coccyx 123 thoracic spine 119–121 Region of interest (ROI) 136 Relative renal uptake ratio 523 Relaxation time T1 42 Relaxation time T2 42 Remote effect 163 Renal arteries 105 Renal cell carcinoma (RCC) 528

889

890

Index Renal imaging diuretic renography 706–707 radiopharmaceutical agents 706 renal cortical scintigraphy 707–708 Renal masses 528–530 Renal osteodystrophy 599, 601 Renal parenchymal function phase 523 Renal perfusion phase 523 Renal scintigraphy 148–151, 676–679 acute tubular necrosis 679 renal transplant evaluation 527 urinary tract obstruction 676–679 urine leak 679 vascular compromise 679 Renal transplant evaluation 526–528 Renal vascular compromise 679 Renal veins 105 Renin‐angiotensin‐aldosterone system (RAAS) 525–526 Renovascular hypertension 524–526 Resolution correction 22 Resonance frequency 45 Respiratory zone 310 Response Assessment in Neuro‐Oncology (RANO) 196 Response Evaluation Criteria in Solid Tumors (RECIST) 408 Retention index (RI) 386 Retro‐ocular spaces 73 Retroperitoneum 106–108 central compartment 107 lateral compartments 106 posterior compartments 107–108 Retropharyngeal space 66–67 Retrosplenial cortex 164 Rheumatoid arthritis 585–589 18 F‐FDG PET/CT 586, 588–589 MRI 586 99m Tc‐MDP scan 586, 587 radiography 585–586 ultrasound 586 RI. See Retention index (RI) Right bronchus, anatomy of 74–78 intermediate bronchus 76 middle lobe bronchus 76 right lower lobe bronchi 76 RUL 76 Right colic artery 100 Right hepatic vein (RHV) 87, 93 Right hilum, vascular anatomy of 79 Right intrahepatic bile duct, anatomical variants of 94 Right lower lobe (RLL) bronchi 76 basal segment bronchus of 78, 79 Right lower quadrant (RLQ) mass 429 Right perihepatic space 103 Right portal vein (RPV) 88, 89 Right subhepatic space 103 Right subphrenic space 103

Right supramesocolic space 103 Right upper lobe (RUL) bronchus 74, 76 anterior and posterior segment bronchi 76 anterior segment bronchi 76 apical segment bronchus 76 level of 79 posterior segment bronchi 76 Right vertical plane 87 RNA. See Radionuclide angiography (RNA) ROI. See Region of interest (ROI) Rubidium‐82 259

s

Sacrum 123 SAEFs. See Secondary aortoenteric fistulas (SAEFs) Salivary gland tumors 226 Sarcoidosis 320 SBNET. See Small bowel NET (SBNET) SCA. See Spinocerebellar ataxia (SCA) Scatter correction, problem of 22 Scattered photons 34 SCC. See Squamous cell carcinoma (SCC) Schizencephaly 180 Scintigraphy myocardial perfusion 671 spinal osteoid osteoma 631 testicular torsion 675, 676 Scintillation camera system 33 components of 34 Scintimammography 354 SCLC. See Small‐cell lung cancer (SCLC) Scleroderma 385 Scoring methodology 289 Screening mammography 356–357, 358 SD. See Semantic dementia (SD); Stable disease (SD) SDH. See Succinate dehydrogenase (SDH) Secondary aortoenteric fistulas (SAEFs) 303–304 Secondary normal pressure hydrocephalus (NPH) 176 Segmentation‐based algorithm 48 Selective internal radiation therapy (SIRT) adverse effects 859 discharge and follow‐up 858 efficacy 859 functional imaging 857 indications 859 pretreatment 858 procedures and tests 859 properties of 857 safety 859 therapy administration 858 Semantic dementia (SD) 175 Seminal vesicles 114–115 Semiology of brain in neuroimaging 53–54 Semiquantitative method 293

Sentinel lymph node biopsy (SLNB) 360– 361, 363 Sentinel lymph node mapping 362 Sentinel node 763 in breast cancer 766 in colorectal cancer 768 in head and neck cancer 766–767 in melanoma 764–765 in vulvar cancer 767–768 Seronegative spondyloarthropathies 589–591 Severe acute respiratory syndrome virus‐2 (SARS‐CoV‐2) 327 Shoulder girdle 126–127 Shunt malfunctions, cerebrospinal 682–685 Sigmoid colon adenocarcinoma 433, 436 Signal formation, MRI 40–41 Signal processing 45 Signet ring cell carcinoma 412, 413 Silicon photomultipliers (SiPMs) 39 Single imaging modality 30 Single‐photon emission computed tomography (SPECT) 1, 2, 33–35, 133, 163 image degrading factors 34–35 lung perfusion 312 perfusion imaging of heart 152–153 physical principles of 33–34 protocols in 258, 260 quantitative data corrections in 37–38 radiopharmaceuticals 144–158 radiotracers, characteristics of 260 99m Tc‐HMPAO 166 Single‐photon emission computed tomography‐computed tomography (SPECT‐CT) imaging. See SPECT‐CT Single‐photon emission computerized tomography (SPECT) hepatic hemangiomas 468 NETs 513 Single‐photon gamma imaging systems 354–355 Sinogram 31 Sinonasal melanoma 226 Sinonasal region 70–71 Sinus of Morgagni 64 SISCOM. See Subtraction of ictal SPECT coregistered to MRI (SISCOM) Skeletal muscles, FDG uptake 6 SLDCF. See Superficial layer of the deep cervical fascia (SLDCF) SLNB. See Sentinel lymph node biopsy (SLNB) Small bowel and colon anatomy and physiology 391 CT imaging 400–401 functional disorders 392–393 magnetic resonance imaging 399–400

Index radionuclide evaluation 393–395 structural abnormalities of 395–398, 401–404 ultrasound imaging 398–399 Small bowel malignancies 424–427 Small bowel NET (SBNET) 425 Small bowel transit scintigraphy 394 Small‐cell lung cancer (SCLC) 335, 342, 343 Small intestine 98 duodenum 96–97 ileum 97 jejunum 97 Small lymph nodes 250, 252 unspecific vs. metastatic 251 Small vessel vasculitides (SVV) 285, 291–292 Smoking 316 and lung cancer 335 Smoldering myeloma, FDG PET/CT study for 294 Snell’s law 43 Sodium 18F‐fluorine (Na18F) 367 Software‐based hybrid imaging 313 Software fusion techniques 20 Solid‐state detectors 39, 47 SPECT‐CT scanners with 23 Solitary pulmonary nodules (SPNs) 333–334 chest radiography 333 CT scan 333 FDG PET/CT 333–334 MRI 334 Somatostatin receptor (SSTR)‐based PET imaging neuroendocrine tumor 462, 463 Somatostatin receptor PET (SSTR PET) NETs 516, 518 Somatostatin receptors (SSTRs) 142, 514, 515, 838 Somatostatin receptor subtype 2 (SSTR2) 207 Sonography. See Ultrasound imaging Sound waves, properties of 42 Spatial encoding 41 SPECT/CT 17, 47, 145 acquisition 21 advantages 19, 20 clinical examples 23–26 carcinoma of prostate 24 chronic kidney disease 25 left hip joint pain 24 pubic region pain, complaints of 26 recurrent pancreatitis 25 cost of 21 for COVID‐19 328 diabetic foot infection 594 HO 585 osteoarthrosis 590

periprosthetic infection 598 pleural effusions 314 protocols 21 reconstruction 21–23 with solid‐state detectors 23 spinal osteoid osteoma 632 spinal trauma 648–649 spine metastases 643 system information 19–20 SPECT myocardial perfusion imaging coronary artery anomalies 702–703 evaluation of chest pain 263 kawasaki disease 703 99m Tc‐tetrofosmin agents 700 radiotracers 258–260 Sphenoethmoidal recess 71 Sphenoid sinus 71 Spinal cord 644, 646–647 Spinal cord, FDG uptake 6 Spinal infection 625–628 bone scintigraphy 627 classification of 625 clinical features 626 CT 626–627 diagnosis of 626–629 18 F‐FDG PET/CT 627–628 frequency of 626 indications of 625 microbiology 626 MRI 627, 628 plain X‐ray 626 SPECT/CT 627 transmission mode 626 Spinal trauma 647–649 causes of 647 MDCTs 647–648 MRI 648 SPECT/CT 648–649 X‐rays 647–648 Spine anatomy 116–123 general 116–117 regional 117–123 cervical spine 117–119 lumbar spine 121–122 sacrum and coccyx 123 thoracic spine 119–121 Spine disorders 625–649 benign lesions 630–635 aneurysmal bone cyst 634 eosinophilic granuloma 634 fibrous dysplasia 634–635 giant cell tumor 634 hemangioma 635, 636 osteoblastoma 632–633 osteoid osteoma 630–632 infections 625–628 bone scintigraphy 627 classification of 625 clinical features 626

CT 626–627 diagnosis of 626–629 18 F‐FDG PET/CT 627–628 frequency of 626 indications of 625 microbiology 626 MRI 627, 628 plain X‐ray 626 SPECT/CT 627 transmission mode 626 malignant lesions 635–641 chordoma 636–638 correlative approach in diagnosis 635–636 Ewing sarcoma 637–639 lymphoma 636, 637 multiple myeloma 638–641 osteosarcoma 637–639 metastases. See Metastases, spine tumors 628, 629 Spin‐lattice relaxation 42 Spinocerebellar ataxia (SCA) 185, 187 Spin‐spin relaxation 42 Spiral mode 31 Spleen 100–101 Splenorenal ligament 101 Splenorenal ligament (SRL) 414 “Split atlas” 118 SPNs. See Solitary pulmonary nodules (SPNs) Spondyloarthropathies 589–591 Spondylodiscitis 596 Squamous cell carcinoma (SCC) 224, 407 anal canal 438 of frontal sinus 241 laryngeal malignant 234, 235, 255 oropharyngeal 229 of paranasal sinuses 225 parotid malignancy 227 persistence of maxillary 243 of piriform recess 236 radiotherapy of retro‐auricular 244 recurrence of 249, 299 tongue base 231, 232 tonsillar 230 of trigonum retromolare 234 Stable disease (SD) 195 Standardized uptake value (SUV) 136 FDG uptake 3–4, 227 Standardized uptake value ratio (SUVR) 173, 174 Standardized uptake values (SUVs) 220 Staphylococcus aureus 292 Static magnetic field 40, 41 Static scintigraphy 524 Stomach 96 Strength, defined 33 Stress injuries 582–584 Stress‐only protocol 258

891

892

Index Structural abnormalities of esophagus 388–390 Subarachnoid hemorrhage (SAH) 53 Subarachnoid space (SAS) 53, 61 Subcarinal space 83–84 Subcortical band heterotopia 180 Subglottic larynx. See Subglottis Subglottis 68 Subtraction of ictal SPECT coregistered to MRI (SISCOM) 177, 179 Succinate dehydrogenase (SDH) 420 Superficial cervical fascia 61 Superficial layer of the deep cervical fascia (SLDCF) 119 Superior mesenteric artery 98, 99, 100 Superior orbital fissure 71, 72 Superior vena cava 81, 85 Supero‐anterior nerve 60 Support vector machines (SVMs) 826 Supra‐aortic mediastinum 80–81, 82 Supra‐aortic trunks (SAT) 53 Supraglottic carcinoma, without infiltration of thyroid cartilage 235 Supraglottic larynx. See Supraglottis Supraglottis 67–68 Supra‐hyoid neck 61, 62. See also Infra‐hyoid neck buccal space 66 carotid space 64 compartments of 62–67 masticator space 65–66 oral cavity 67 parapharyngeal space 64, 65 parotid space 66 perivertebral space 67 pharyngeal mucosal space 62–64 retropharyngeal space 66–67 Suprahyoid spaces 219, 220 Supramesocolic compartment 103 Susceptibility weighted imaging (SWI) 53 Suspected disease recurrence and restaging 369 Suspensory ligaments 101 SUV. See Standardized uptake value (SUV) SUVR. See Standardized uptake value ratio (SUVR) SVV. See Small vessel vasculitides (SVV) Swallowing artifacts 220, 223

t

Takayasu arteritis (TAK) 287 FDG PET 290 lower sensitivity for 291 Target heart rate 258 Task‐negative network 163 99m Tc‐based radiotracers 259 99m Tc‐diethylenetriamine pentaacetic acid (99mTc‐DTPA) 133, 148, 315–316 molecular structure of 151

pulmonary capillary epithelial permeability 315–316 renal scintigraphy with 149, 152 99m Tc‐dimercaptosuccinic acid (99mTc‐DMSA) 149, 152 99m Tc‐DTPA. See 99mTc‐diethylenetriamine pentaacetic acid (99mTc‐DTPA) 99m Tc‐etyl cysteinate dimer (99mTc‐ECD) 177 brain perfusion SPECT using 164, 183 99m Tc‐hexamethylpropyleneamine oxime (99mTc‐HMAPO) 166, 167, 177, 183, 401, 402 99m Tc‐macroaggregatealbumin (99mTc‐MAA) 156–158 99m Tc‐mercapto acetyl tri glycine (MAG3) 149 99m Tc‐methyl‐diphosphonate (99mTc‐MDP) 133, 134, 147–148 diagnostic accuracy 148 molecular structure of 147 whole‐body bone scan bilateral knee prosthesis 151 breast cancer 150 prostate cancer 148 suspicion of prosthesis mobilization 149 Tc‐99m MDP scintigraphy 590 99m Tc. See 99mTechnetium (99mTc) 99m Tc‐pertechnetate 146 99m Tc‐pyrophosphate (PYP) 272 99m Tc‐radiocolloid 158, 159 99m Tc‐sestamibi 151, 153, 154 99m Tc‐World Health Organization (WHO) Tc‐sestamibi MBI‐guided breast biopsy 358, 359 99m Tc‐stannous fluoride (99mTc‐Sn‐F) 401 99m Tc‐tetrophosmin 151 TDLU. See Terminal ductal lobular unit (TDLU) 99m Technetium (99mTc) 145–146 Temporal lobes 55, 57 Temporoparietal junction 164 Teratoma 347 Terminal ductal lobular unit (TDLU) 351, 352 Testicular torsion 673–676 scintigraphy 675, 676 ultrasonography 674–675 Thallium‐201 (201Tl) 258, 259, 264 Theory of mind 164 Theranostics 134, 369–370 Thigh 124, 125 Thoracic anatomy 73 bronchus. See Bronchus CT 73–74 diaphragm 86 mediastinum. See Mediastinum MRI 74

pleural surfaces 85–86 pulmonary hilum 74 thoracic wall 85–86 Thoracic aorta, FDG uptake in 291 Thoracic cavity 85 Thoracic malignancies lung cancer. See Lung cancer mediastinal tumors 344–345 pleural diseases 343–344 solitary pulmonary nodules 333–334 teratoma 347 thymoma 345–347 Thoracic spine 119–121 Thoracic ultrasonography (TUS), pleural diseases 343 Thoracic wall 85–86 Thorax 219 Three‐phase 99mTc‐MDP bone scan 593–594 Thymic epithelial tumors 345 Thymoma 345–347 Thymus, FDG uptake 5, 7 Thyroglobulin 145 Thyroid anatomy 485 function 485 imaging 485–492 differentiated thyroid cancer 488–492 18 F‐FDG PET/CT 487–491 FNA biopsy 485–486 hybrid molecular imaging 488–492 123 I or 131I scan 489–491 124 I‐PET/US fusion imaging 489 99m Tc‐DMSA scintigraphy 486 99m Tc‐MIBI scintigraphy 486, 487 PET/MRI 492 SPECT/CT 489, 491, 492 SPECT/US 486 thyroid scintigraphy 486 ultrasound 485 nodules 485 Thyroid cancer 236 functional imaging 854–855 papillary 237 Thyroid cartilage, supraglottic carcinoma without infiltration of 235 Thyroid gland 81, 144, 296 Thyroid scintigraphy 146, 486 pathophysiology 698 pediatric thyroid nodules 698 thyroiditis 700 thyroid malignancy 698, 700 Thyroid‐stimulating hormone (TSH) 851 TID. See Transient ischemic dilatation (TID) Time–activity curve (TAC) 201 Time gain compensation (TGC) 45 Time needed to reach the peak filling (TTPF) 270 Time‐of‐flight (TOF) technology 36

Index Time to peak (TTP) 196 Tissue attenuation coefficients for 43 interaction of ultrasound with 43–44 TKIs. See Tyrosine kinase inhibitors (TKIs) TOF technology. See Time‐of‐flight (TOF) technology Tomographic imaging modalities multiple factors 1 overview of 2 pulmonary embolism 312 Tomographic theory 30–31 Tongue muscles 220 tumors of 231, 232 Tonsillar gland 238 Tonsils, tumors of 229 Trachea 308 Tracheal bronchus 76 Tracheobronchial tree 308 Tracking systems electromagnetic 48–49 mechanical position encoding 49 Traditional imaging techniques 30 Traditional scintigraphy 133 Transducers 44–45 components 44 piezoelectric principle 44 resonance frequency 45 Transglottic carcinoma, with cartilage infiltration 235 Transient ischemic dilatation (TID) 264 Transition zone (TZ), prostate gland 533 Transmission‐based methods 48 Transperitoneal spread, of gastric adenocarcinoma 414 Transrectal ultrasound (TRUS) 534–535 Transthyretin amyloidosis (ATTR) 271–272 Transvaginal sonography (TVS) endometrial cancer 555 ovarian cancer 564–565 Transverse magnetization decay, MRI 42 Transverse plane 87 Transversus thoracis muscle 85 Traumatic fractures 579–582 Trigeminal nerve 60 Trigonum retromolare 234 Triple‐energy window (TEW) technique 37 Triple modality scanner 48 True pelvis 109 Truncation artifacts, in PET‐CT 14–15 Truncation correction 48 T staging in lung cancer 335–336 Tuberculosis 626, 627 Tumor‐Node‐ Metastasis (TNM) system 409

Tumor thrombus 295 TUS. See Thoracic ultrasonography (TUS) Type I endometrial cancer 554–555 Type II endometrial cancer 554 Tyrosine kinase inhibitors (TKIs) 420, 422

Urinomas 527 UTE. See Ultrashort echo time sequences (UTE) Uterus 109–110 venous drainage of 113

u

v

UC. See Ulcerative colitis (UC) Ulcerative colitis (UC) 396–397 Ulcers 400 Ulegyria 182–183, 184 Ultrashort echo time sequences (UTE) 48 Ultrasonography 302 testicular torsion 674–675 Ultrasound (US) imaging 42, 236 acute cholecystitis 469, 690 adrenal hemorrhage 500 display modes 45 doppler 45–46 in GCA 287 hepatic metastases 459 hepatobiliary imaging, biliary atresia 457 HO 585 interaction with tissue 43–44 nephroureteral obstruction 523–524 parathyroid imaging 492 physical principles 42–43 probe operation 45 prostate imaging 535–536 seronegative spondyloarthropathies 589 signal processing 45 small bowel 398–399 thyroid imaging 485 transducers 44–45 Uncinate processes (UPs) 118 Unenhanced CT ACCs 501 adrenal adenomas 498 adrenal metastases 502 urinary tract obstruction 677 United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) 564 United States Food and Drug Administration (US FDA) 839 Upper cervical spine 117–118 Upper esophageal sphincter (UES) 383 Upper gastrointestinal (UGI) imaging, gastric adenocarcinoma 412–413 Upper limbs 126–127 Ureter 105 Urethra 112 Urinary bladder 105 Urinary system 104–105 FDG uptake 5 kidneys 104–105 ureter 105 urinary bladder 105

VAF. See Vertebral artery foramen (VAF) Vagina 111 Valvular heart diseases 274 Variceal bleeding 688 Vascular anatomy of left hilum 79–80 of right hilum 79 Vascular graft infections (VGIs), imaging of 302–303 Vascularization of bowel 98–100 of liver 89 of pancreas 93–95 of pelvis 112–113 of stomach 96 Vascular parkinsonism 185, 186 Vascular system, diagnostic imaging of 285 Vascular thrombosis, benign and malignant 295–298 Vasculitis, imaging of 285–292 CT 286 18 F‐FDG PET 286 LVV 287–291 MRI/MRA 286 MVV and SVV 291–292 ultrasound 286 Vasodilator stress test 258 Venous drainage of uterus 113 Venous thromboembolic disease 327 Venous thromboembolism (VTE) 295 Ventilation defect 311 Ventilation scintigraphy 319 and perfusion scintigraphy 324–325, 327–328 Ventricular system 61 Vertebral artery foramen (VAF) 119 Vesicoureteral reflux (VUR) 708 VGIs. See Vascular graft infections (VGIs) Visceral peritoneum 101 Visceral space 67–69 hypopharynz 68–69 larynx 67–68 Viz.AI ContaCT 832 Vocal cords FDG uptake 5 transaxial CT‐only image of 7 Volar compartment, of forearm 127 Voxel‐based morphometry (VBM) 171 Voxel‐based Specific Regional analysis system for AD (VSRAD®) 171 Voxels 31

893

894

Index V/Q scans and chest radiographs 311, 312 diagnostic probability 311–312 high probability 311 indeterminate probability 311, 313 low probability 311 moderate probability 311 normal/near‐normal probability 311 perfusion scan 314, 315 SPECT 312 VTE. See Venous thromboembolism (VTE)

w

Warburg effect

2

Wavelength, of sound waves 42 WBCs. See White blood cells (WBCs) Wegener’s. See Granulomatosis with polyangiitis (GPA) White blood cells (WBCs) 292, 401 White matter (WM) 53, 55, 58–59 WHO. See World Health Organization (WHO) Whole‐body computed tomography (WBCT) 640 World Health Organization (WHO) 194 classification of gastric adenocarcinoma 412 grade II and III gliomas 195 grade IV glioma 195

x

133

Xe inhalation method 168 X‐ray tubes 33 xSPECT technology 47

y

90

z

Y‐microspheres therapy

Zenker’s diverticulum

476

389, 390

WILEY END USER LICENSE AGREEMENT Go to www.wiley.com/go/eula to access Wiley’s ebook EULA.