Pediatric Surgery: Diagnosis and Management 3030814874, 9783030814878

Citation preview

Pediatric Surgery Diagnosis and Management Prem Puri Michael E. Höllwarth Editors Second Edition

123

Pediatric Surgery

Prem Puri  •  Michael E. Höllwarth Editors

Pediatric Surgery Diagnosis and Management Second Edition

Editors Prem Puri Beacon Hospital University College Dublin Dublin, Ireland

Michael E. Höllwarth University Clinic of Paediatric and Adolescent Surgery Medical University of Graz Graz, Austria

ISBN 978-3-030-81487-8    ISBN 978-3-030-81488-5 (eBook) https://doi.org/10.1007/978-3-030-81488-5 © Springer Nature Switzerland AG 2009, 2023 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

To Veena and Christa for their love, support, and inspiration

Preface to the Second Edition

It has been 14 years since the first edition of this book was published in 2009. In these intervening years, major advances in prenatal diagnosis, imaging, anesthesia, intensive care, and minimally invasive surgery including robotic technology have radically altered the management of infants and children with congenital and acquired surgical conditions. This second edition of Pediatric Surgery has been substantially revised and updated to reflect these advances in pediatric surgery. It contains 106 chapters from 234 contributors from five continents. Each chapter has been written by internationally renowned pediatric surgeons with significant experience in their respective field of interest. Many younger surgeons who will become the next generation of leaders in pediatric surgery were invited as authors or co-authors. This edition contains nine new chapters on important topics including respiratory management of the surgical patient, access to enteral nutrition, surgical safety, surgical problems in children with disabilities, surgical implications of HIV infection in children, esophageal replacement, variant Hirschsprung’s disease, and long-term outcomes in pediatric surgery. We have maintained the previous format, dividing the book into 13 sections. However, as the page extent of the second edition exceeds 1500 pages, we have split the book into two volumes for practical reasons. The new edition of this book provides an authoritative, comprehensive, and complete account of the pathophysiology and management of surgical disorders in infants and children. The book is mainly intended for pediatric surgical trainees and young pediatric surgeons, providing a comprehensive description of various surgical conditions in children with a major emphasis on diagnosis and management. The first edition was recognized worldwide as an important textbook dealing with surgical conditions in children. We hope that the thoroughly revised and updated second edition of the book will continue to act as a reference book for pediatric surgeons worldwide. We wish to thank most sincerely all the contributors for their outstanding work in producing this innovative textbook. We also wish to express our gratitude to the editorial staff of Springer, particularly Ms Melissa Morton and Mr Rakesh Kumar Jotheeswaran for all their help during the preparation and publication of the second edition of this book. Dublin, Ireland Graz, Austria

Prem Puri Michael E. Höllwarth

vii

Contents

Part I General Principles 1 The  Epidemiology of Birth Defects����������������������������������������������    3 Florian Friedmacher and Edwin C. Jesudason 2 Fetal  Counselling for Surgical Congenital Malformations������������������������������������������������������������   13 Kokila Lakhoo and Rebecca Black 3 Transport  of the Surgical Neonate ����������������������������������������������   25 Udo Rolle and Prem Puri 4 Pre-operative Management and Vascular Access������������������������   33 Ancuta Muntean, Ionica Stoica, John Gillick, and Prem Puri 5 Anaesthesia and Analgesia������������������������������������������������������������   55 Coilin Collins Smyth and Suzanne Crowe 6 Respiratory  Management of the Surgical Patient����������������������   71 Gregory Nolan and Suzanne Crowe 7 Fluid Management ������������������������������������������������������������������������   79 S. O’Sullivan and Suzanne Crowe 8 Sepsis ����������������������������������������������������������������������������������������������   85 Lexie H. Vaughn and Jeffrey S. Upperman 9 Nutrition������������������������������������������������������������������������������������������   97 Agostino Pierro and Simon Eaton 10 Access  for Enteral Nutrition ��������������������������������������������������������  109 Julia Brendel and Michael W. L. Gauderer 11 Hematological  Problems in Pediatric Surgery����������������������������  119 Peter McCarthy and Owen Patrick Smith 12 Genetics������������������������������������������������������������������������������������������  145 James J. O’Byrne and Andrew J. Green 13 Ethical  Considerations in Pediatric Surgery ������������������������������  155 Rita D. Shelby, Donna A. Caniano, and Benedict C. Nwomeh

ix

x

14 Minimal  Access Surgery in Infants and Children����������������������  163 Amulya K. Saxena, Roberta V. Iacona, and Keith Georgeson 15 Surgical  Safety in Children ����������������������������������������������������������  177 Iain Yardley and Holbrook Charlotte 16 Surgical  Problems of Children with Physical Disabilities����������  185 Casey M. Calkins 17 Surgical  Aspects of HIV Infection in Children����������������������������  203 Alastair J. W. Millar, Brian Eley, and Sharon Cox Part II Trauma 18 Birth Trauma����������������������������������������������������������������������������������  219 Thambipillai Sri Paran and Prem Puri 19 Pediatric Thoracic Trauma ����������������������������������������������������������  229 David E. Sawaya, Michael W. Morris, and Paul M. Colombani 20 Abdominal and Genitourinary Trauma��������������������������������������  239 Claire D. Gerall, Vincent P. Duron, and Steven Stylianos 21 Surgical  Treatment of Severe Head Trauma ������������������������������  261 Hans G. Eder 22 Pediatric Orthopedic Trauma������������������������������������������������������  273 Zacharias Zachariou, Eva E. Fischerauer, and Annelie M. Weinberg 23 Injuries  to the Tendons of the Hand��������������������������������������������  295 Georg Singer and Heidi Friedrich 24 Burns ����������������������������������������������������������������������������������������������  309 Alan David Rogers and Heinz Rode 25 Foreign Bodies��������������������������������������������������������������������������������  325 S. Shah, L. Nguyen, and R. Sun 26 Physical  and Sexual Child Abuse��������������������������������������������������  337 Michael E. Höllwarth Part III Head and Neck 27 Pierre Robin Sequence������������������������������������������������������������������  349 Udo Rolle, Aranka Ifert, and Robert Sader 28 Choanal Atresia������������������������������������������������������������������������������  359 R. Ben Speaker, Michael Harney, and John Russell 29 Thyroglossal  and Branchial Cysts, Sinuses, and Fistulas����������  365 Michael E. Höllwarth 30 Tracheostomy ��������������������������������������������������������������������������������  373 Lina Woods, Thom E. Lobe, and John Russell

Contents

Contents

xi

Part IV Chest 31 Chest Wall Deformities������������������������������������������������������������������  387 Robert E. Kelly and Donald Nuss 32 Breast  Disorders in Children and Adolescents����������������������������  405 Steffi Mayer, Jan-Hendrik Gosemann, Benno M. Ure, and Martin L. Metzelder 33 Congenital Airway Malformations ����������������������������������������������  413 Patricio Varela and Richard Azizkhan 34 Mediastinal  Masses in Children ��������������������������������������������������  429 Maria Molina and Israel Fernandez-Pineda 35 Pleural  Effusion and Empyema����������������������������������������������������  437 Michael Singh and Dakshesh Parikh 36 Congenital  Malformations of the Lung����������������������������������������  447 Ali A. Mokdad, David M. Gourlay, and Keith T. Oldham 37 Congenital Diaphragmatic Hernia ����������������������������������������������  463 Prem Puri and Nana Nakazawa 38 Extracorporeal Membrane Oxygenation������������������������������������  475 Brian P. Fallon, Samir K. Gadepalli, Jason S. Frischer, Charles J. H. Stolar, and Ronald B. Hirschl Part V Spina Bifida and Hydrocephalus 39 Spina  Bifida and Encephalocoele�������������������������������������������������  487 Martin T. Corbally 40 Hydrocephalus��������������������������������������������������������������������������������  499 Geraint Sunderland, Jonathan Ellenbogen, and Conor Mallucci 41 Dermal  Sinus Tract and Tethered Cord Syndrome��������������������  527 Geraint Sunderland and Jonathan Ellenbogen Part VI Anterior Abdominal Wall Defects 42 Omphalomesenteric Duct Remnants��������������������������������������������  543 Ampaipan Boonthai, Dhanya Mullassery, and Paul D. Losty 43 Omphalocele and Gastroschisis����������������������������������������������������  551 Leah M. Sieren, Duane S. Duke, and Marshall Z. Schwartz 44 Conjoined Twins����������������������������������������������������������������������������  563 Juan A. Tovar and Leopoldo Martinez

xii

Part VII Tumors 45 Vascular Anomalies������������������������������������������������������������������������  577 Anna McGuire, Steven J. Fishman, and Belinda Hsi Dickie 46 Congenital Nevi������������������������������������������������������������������������������  593 Hannes Prescher, Julia F. Corcoran, and Bruce S. Bauer 47 Lymphatic Malformations������������������������������������������������������������  609 Jeremy A. Goss, Mohammed Zamakhshary, Jacob C. Langer, and Emily Christison-Lagay 48 Sacrococcygeal Teratoma��������������������������������������������������������������  625 Thambipillai Sri Paran and Prem Puri 49 Neuroblastoma ������������������������������������������������������������������������������  633 Edward Kiely 50 Soft Tissue Sarcomas ��������������������������������������������������������������������  643 Sandeep Agarwala and Robert Carachi 51 Lymphomas������������������������������������������������������������������������������������  661 Christian Urban 52 Wilms’ Tumor��������������������������������������������������������������������������������  673 Michael E. Höllwarth 53 Ovarian Tumors ����������������������������������������������������������������������������  685 Alicia G. Sykes, Mary E. Fallat, and Romeo C. Ignacio Jr. 54 Testicular Tumors��������������������������������������������������������������������������  699 Amanda F. Saltzman and Jonathan Ross Part VIII Gastrointestinal 55 Esophageal Atresia and Tracheoesophageal Fistula ������������������  711 Michael E. Höllwarth and Paola Zaupa 56 Achalasia����������������������������������������������������������������������������������������  729 Fanny Yeung, Kenneth Wong, and Paul Tam 57 Esophageal  Perforations and Caustic Injuries in Children ����������������������������������������������������������������������  743 Shilpa Sharma and Devendra K. Gupta 58 Gastroesophageal Reflux Disease ������������������������������������������������  753 Michael E. Höllwarth and Valeria Solari 59 Esophageal Replacement��������������������������������������������������������������  777 Shilpa Sharma and Devendra K. Gupta 60 Infantile  Hypertrophic Pyloric Stenosis��������������������������������������  799 Takashi Doi and Takao Fujimoto

Contents

Contents

xiii

61 Gastrostomy and Jejunostomy ����������������������������������������������������  807 Julia Brendel and Michael W. L. Gauderer 62 Duodenal Obstruction ������������������������������������������������������������������  829 Yechiel Sweed and Arcady Vachyan 63 Intestinal Malrotation��������������������������������������������������������������������  849 Mark D Stringer and Prabal R. Mishra 64 Jejuno-Ileal Atresia������������������������������������������������������������������������  869 A. J. W. Millar, S. Cox, and A. Numanoglu 65 Meconium Ileus������������������������������������������������������������������������������  881 Valeria Solari and Massimo Rivosecchi 66 Duplications of the Alimentary Tract������������������������������������������  893 K. Hughes, A. Mortell, and Prem Puri 67 Necrotizing Enterocolitis ��������������������������������������������������������������  907 Heather L. Liebe, Henri R. Ford, Victoria Camerini, and Catherine J. Hunter 68 Constipation�����������������������������������������������������������������������������������  919 Hannah M. E. Evans-Barns, Sebastian K. King, Bridget R. Southwell, and John M. Hutson 69 Hirschsprung’s Disease������������������������������������������������������������������  933 Prem Puri 70 Variant Hirschsprung’s Disease����������������������������������������������������  949 Florian Friedmacher and Prem Puri 71 Anorectal Anomalies����������������������������������������������������������������������  967 Alberto Peña, Andrea Bischoff, and Luis De la Torre 72 Appendicitis������������������������������������������������������������������������������������  985 Markus Almström and Tomas Wester 73 Intussusception������������������������������������������������������������������������������  993 Holger Till and Erich Sorantin 74 Hernias�������������������������������������������������������������������������������������������� 1001 Anna Svenningsson and Tomas Wester 75 Short Bowel Syndrome������������������������������������������������������������������ 1015 Michael E. Höllwarth 76 Inflammatory Bowel Disease�������������������������������������������������������� 1031 Risto J. Rintala and Mikko P. Pakarinen 77 Paediatric Small Bowel Transplantation ������������������������������������ 1051 G. L. Gupte, K. Sharif, and A. J. W. Millar 78 Long-Term  Outcomes in Pediatric Surgery�������������������������������� 1061 Risto J. Rintala and Mikko P. Pakarinen

xiv

Part IX Hepatobiliary 79 Biliary Atresia�������������������������������������������������������������������������������� 1091 Hannah Thompson and Mark Davenport 80 Choledochal Cyst �������������������������������������������������������������������������� 1101 Hiroyuki Koga and Atsuyuki Yamataka 81 Hepatic Cysts and Abscesses �������������������������������������������������������� 1117 Priya Ramachandran 82 Portal Hypertension���������������������������������������������������������������������� 1123 Mark D Stringer 83 Gallbladder Disease ���������������������������������������������������������������������� 1141 Charlene Dekonenko, Shawn D. St. Peter, and George W. Holcomb III 84 Pancreatic Disorders���������������������������������������������������������������������� 1155 Elke Zani-Ruttenstock and Augusto Zani 85 Splenic Disorders �������������������������������������������������������������������������� 1173 Takashi Doi and Thom E. Lobe 86 Hepatic Tumors in Childhood������������������������������������������������������ 1185 Thambipillai Sri Paran and Michael P. La Quaglia 87 Pediatric Liver Transplantation �������������������������������������������������� 1197 Khalid Sharif and Alastair J. W. Millar Part X Genitourinary Disorders 88 Urinary Tract Infection ���������������������������������������������������������������� 1215 Thomas de los Reyes and Martin A. Koyle 89 Imaging  of the Paediatric Urogenital Tract�������������������������������� 1227 Michael Riccabona 90 Management of Antenatal Hydronephrosis�������������������������������� 1249 Jack S. Elder 91 Upper Urinary Tract Obstructions���������������������������������������������� 1269 Leon Chertin and Boris Chertin 92 Ureteric Duplication Anomalies���������������������������������������������������� 1281 Thambipillai Sri Paran and Prem Puri 93 Vesicoureteral Reflux�������������������������������������������������������������������� 1291 Prem Puri and Balazs Kutasy 94 Posterior Urethral Valves�������������������������������������������������������������� 1307 Salvatore Cascio, David Coyle, Simona Nappo, and Paolo Caione

Contents

Contents

xv

95 Neuropathic Bladder��������������������������������������������������������������������� 1329 Alaa El Ghoneimi, Annabel Paye-Jaouen, Valeska Bidault, Pauline Lopez, and Matthieu Peycelon 96 Bladder Exstrophy ������������������������������������������������������������������������ 1347 Wayland J. Wu and John P. Gearhart 97 Cloacal Exstrophy�������������������������������������������������������������������������� 1359 Duncan Wilcox, Andrea Bischoff, and Moritz M. Ziegler 98 Prune Belly Syndrome ������������������������������������������������������������������ 1371 Thambipillai Sri Paran and Prem Puri 99 End-Stage  Renal Disease and Renal Transplantation���������������� 1377 Atif Awan and Michael Riordan 100 Different Sexual Development������������������������������������������������������ 1389 Maria Marcela Bailez, Gabriela Guercio, and Santiago Weller 101 Cryptorchidism������������������������������������������������������������������������������ 1415 John M. Hutson and Jaya Vikraman 102 Acute Scrotum�������������������������������������������������������������������������������� 1425 Amulya K. Saxena, Matthew Jobson, and Michael Höllwarth 103 Hypospadias����������������������������������������������������������������������������������� 1435 Agneta Nordenskjöld and Göran Läckgren 104 Circumcision and Buried Penis���������������������������������������������������� 1451 Martin Kaefer 105 Hydrometrocolpos�������������������������������������������������������������������������� 1469 Shilpa Sharma and Devendra K. Gupta 106 Gynaecologic Conditions of Childhood���������������������������������������� 1479 Hazel Isabella Learner and Sarah M. Creighton Index�������������������������������������������������������������������������������������������������������� 1489

Contributors

Sandeep Agarwala  Department of Pediatric Surgery, All India Institute of Medical Sciences, New Delhi, India Markus  Almström Department of Pediatric Surgery, Astrid Lindgren Children’s Hospital, Karolinska University Hospital, Stockholm, Sweden Atif Awan  Department of Nephrology & Transplantation, Children’s Health Ireland at Temple Street, Dublin, Ireland Richard  Azizkhan Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA Maria  Marcela  Bailez Department of Pediatric Surgery, Hospital de Pediatría “Prof. Dr. J.P. Garrahan”, Buenos Aires, Argentina Bruce S. Bauer  Section of Plastic and Reconstructive Surgery, University of Chicago, Pritzker School of Medicine, Chicago, IL, USA Valeska Bidault  Department of Pediatric Urology, Robert-Debré University Hospital, APHP, National Reference Center of Rare Urinary Tract Malformations (MARVU), Université Paris Cité, Paris, France Andrea Bischoff  International Center for Colorectal and Urogenital Care, Children’s Hospital Colorado, University of Colorado School of Medicine, Aurora, CO, USA Rebecca Black  Oxford University Hospitals NHS Foundation Trust, Oxford, UK Ampaipan  Boonthai Paediatric Surgery and Transplantation Consultant Surgeon, Division of Pediatric Surgery, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand Julia Brendel  Department of Pediatric Surgery, Hannover Medical School, Hannover, Germany Paolo Caione  Division of Paediatric Urology, “Bambino Gesù” Children’s Hospital, Rome, Italy Casey  M.  Calkins Division of Pediatric Surgery, Medical College of Wisconsin, Children’s Hospital of Wisconsin, Milwaukee, WI, USA Victoria  Camerini Children’s Hospital Los Angeles, Los Angeles, CA, USA xvii

xviii

Donna  A.  Caniano Department of Surgery and Pediatrics, Ohio State University College of Medicine, Nationwide Children’s Hospital, Columbus, OH, USA Robert Carachi  Department of Surgical Pediatrics, University of Glasgow, Queen Elizabeth University Hospital, Glasgow, UK Salvatore Cascio  University College Dublin and Children’s Health Ireland, Dublin, Ireland Holbrook Charlotte  Evelina London Children’s Hospital, London, UK Boris Chertin  Department of Urology, Hebrew University, Jerusalem, Israel Leon  Chertin Sakler School of Medicine, Tel Aviv University, Tel Aviv, Israel Emily  Christison-Lagay Division of Pediatric Surgery, Department of Surgery, Yale University, New Haven, CT, USA Paul  M.  Colombani The Johns Hopkins Hospital Professor Emeritus, Baltimore, MD, USA Martin T. Corbally  Royal College of Surgeons in Ireland, Dublin, Ireland Julia F. Corcoran  Section of Plastic and Reconstructive Surgery, University of Illinois Chicago, Chicago, IL, USA Sharon  Cox University of Cape Town and Red Cross War Memorial Children’s Hospital, Cape Town, South Africa David Coyle  Children’s Health Ireland, Dublin, Ireland Sarah  M.  Creighton Department of Women’s Health University College London Hospitals, London, UK Suzanne  Crowe  Children’s Health Ireland, Crumlin, Dublin, Republic of Ireland Mark Davenport  Department of Paediatric Surgery, Kings College Hospital, London, UK Luis De la Torre  International Center for Colorectal and Urogenital Care, Children’s Hospital Colorado, Aurora, CO, USA Thomas  de los Reyes The Hospital for Sick Children and University of Toronto, Toronto, ON, Canada Charlene Dekonenko  Children’s Mercy Hospital, Kansas City, MO, USA Belinda  Hsi  Dickie Vascular Anomalies Center, Department of Surgery, Boston Children’s Hospital, Boston, MA, USA Takashi Doi  Department of Pediatric Surgery, Kansai Medical University, Osaka, Japan Duane S. Duke  Children’s Hospital of the King’s Daughter, Norfolk, VA, USA

Contributors

Contributors

xix

Vincent  P.  Duron Division of Pediatric Surgery, Columbia University Vagelos College of Physicians & Surgeons, NYP-Morgan Stanley Children’s Hospital, New York, NY, USA Simon Eaton  Great Ormond Street Hospital and Institute of Child Health, London, UK Hans  G.  Eder Department of Neurosurgery Medical University of Graz, Graz, Austria Alaa  El Ghoneimi Department of Pediatric Urology, Robert-Debré University Hospital, APHP, National Reference Center of Rare Urinary Tract Malformations (MARVU), Université Paris Cité, Paris, France Jack  S.  Elder Division of Pediatric Urology, Massachusetts General Hospital, Boston, MA, USA Brian  Eley University of Cape Town and Red Cross War Memorial Children’s Hospital, Cape Town, South Africa Jonathan Ellenbogen  Department of Neurosurgery, Alder Hey Children’s NHS Foundation Trust, Liverpool, UK Hannah M.E. Evans-Barns  Department of Paediatric Surgery, The Royal Children’s Hospital and Surgical Research Group, Murdoch Children’s Research Institute, Melbourne, VIC, Australia Mary E. Fallat  The Hiram C. Polk, Jr., Department of Surgery/Division of Pediatric Surgery, University of Louisville School of Medicine, Louisville, KY, USA Brian P. Fallon  University of Michigan, Ann Arbor, MI, USA Israel  Fernandez-Pineda Department of Pediatric Surgery, Virgen del Rocio Children’s Hospital, Sevilla, Spain Eva  E.  Fischerauer University Clinic of Paediatric and Adolesecent Surgery, Medical University, Graz, Austria Steven  J.  Fishman Vascular Anomalies Center, Department of Surgery, Boston Children’s Hospital, Boston, MA, USA Henri R. Ford  University of Miami, Miami, CA, USA Florian Friedmacher  Department of Pediatric Surgery, University Hospital Frankfurt, Goethe University Frankfurt, Frankfurt (Main), Germany Heidi  Friedrich University Clinic of Pediatric and Adolescent Surgery, Medical University of Graz, Graz, Austria Jason  S.  Frischer Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA Takao Fujimoto  Fujimoto Children’s Clinic, Tokyo, Japan Samir K. Gadepalli  University of Michigan, Ann Arbor, MI, USA

xx

Michael  W.L.  Gauderer Surgery and Pediatrics, University of South Carolina, Greenville, SC, USA University of South Carolina, School of Medicine Greenville, Greenville, SC, USA John  P.  Gearhart James Buchannan Brady Urological Institute, Roberts Jeffs Division of Pediatric Urology, Charlotte Bloomberg Children’s Hospital, Johns Hopkins University School of Medicine, Baltimore, MD, USA Keith Georgeson  Department of Paediatric Surgery, University of Alabama Health Services, Birmingham, AL, USA Claire D. Gerall  Division of Pediatric Surgery, Columbia University Vagelos College of Physicians & Surgeons, NYP-Morgan Stanley Children’s Hospital, New York, NY, USA John Gillick  Children’s Health Ireland at Temple Street, Dublin, Ireland Jan-Hendrik  Gosemann Department of Pediatric Surgery, University of Leipzig, Leipzig, Germany Jeremy A. Goss  Division of Plastic and Reconstructive Surgery, Department of Surgery, Yale University School of Medicine, New Haven, CT, USA David  M.  Gourlay Division of Pediatric Surgery, Medical College of Wisconsin, Milwaukee, WI, USA Andrew  J.  Green School of Medicine and Medical Science, University College Dublin, National Centre for Medical Genetics Our Lady’s Children’s Hospital, Crumlin, Dublin, Ireland Gabriela  Guercio Department of Endocrinology, Hospital de Pediatría “Prof. Dr. J.P. Garrahan” y CONICET, Buenos Aires, Argentina Devendra K. Gupta  Super Speciality Pediatric Hospital and Post Graduate Teaching Institute, Noida, India G.L. Gupte  Birmingham Children’s Hospital, Birmingham, UK Michael Harney  Bon Secours Hospital, Cork, Ireland Ronald B. Hirschl  C.S. Mott Children’s Hospital, University of Michigan, Ann Arbor, MI, USA Michael  E.  Höllwarth University Clinic of Paediatric and Adolescent Surgery, Medical University of Graz, Graz, Austria George  W.  Holcomb III Children’s Mercy Hospital, Kansas City, MO, USA K. Hughes  Children’s Health Ireland (CHI), Dublin, Ireland Catherine  J.  Hunter Children’s Hospital at Oklahoma University, Oklahoma City, OK, USA

Contributors

Contributors

xxi

John  M.  Hutson  University Clinic of Paediatric and Adolescent Surgery, Medical University of Graz, Graz, Austria Roberta  V.  Iacona Department of Paediatric Surgery, Chelsea and Westminster Hospital NHS Foundation Trust, Imperial College London, London, UK Aranka Ifert  Carolinum, Institute of Dentistry, Frankfurt, Germany Romeo C. Ignacio Jr  Department of Surgery/Division of Pediatric Surgery, University of California San Diego School of Medicine, La Jolla, CA, USA Edwin C. Jesudason  Department of Rehabilitation Medicine, Astley Ainslie Hospital, Edinburgh, UK University of Edinburgh, Edinburgh, UK Matthew Jobson  NHS Trust: The Luton and Dunstable University Hospitals NHS Foundation Trust, London, UK Martin  Kaefer Department of Urology, Division of Pediatric Urology, Indiana University School of Medicine, Indianapolis, IN, USA Robert E. Kelly  Children’s Hospital of The King’s Daughters, Norfolk, VA, USA Eastern Virginia Medical School, Norfolk, VA, USA Edward Kiely  Great Ormond Street Hospital for Children NHS Foundation Trust, London, UK Sebastian K. King  Department of Paediatric Surgery, The Royal Children’s Hospital and Surgical Research Group, Murdoch Children’s Research Institute, Melbourne, VIC, Australia Surgical Research Group, Murdoch Children’s Research Institute, Melbourne, VIC, Australia Hiroyuki Koga  Department of Pediatric and Urogenital Surgery, Juntendo University School of Medicine, Bunkyo-ku Tokyo, Japan Martin A. Koyle  The Hospital for Sick Children and University of Toronto, Toronto, ON, Canada Balazs Kutasy  Astrid Lindgren’s Children’s Hospital, Karolinska University Hospital, Stockholm, Sweden Michael P. La Quaglia  Pediatric Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY, USA Göran  Läckgren Department of Pediatric Surgery, Uppsala Academic Hospital, Uppsala, Sweden Kokila  Lakhoo  Children’s Hospital Oxford, Oxford University Hospitals, University of Oxford, Oxford, UK

xxii

Jacob  C.  Langer  Division of General and Thoracic Surgery, Hospital for Sick Children and Department of Surgery, School of Medicine Surgery, University of Toronto, Toronto, ON, Canada Hazel Isabella Learner  Department of Women’s Health University College London Hospitals, London, UK Heather L. Liebe  Children’s Hospital at Oklahoma University, Oklahoma City, OK, USA Thom E. Lobe  University of Tennessee Health Science Center, Memphis, TN, USA Pediatric Surgery, University of Illinois, Chicago, IL, USA Pauline Lopez  Department of Pediatric Urology, Robert-Debré University Hospital, APHP, National Reference Center of Rare Urinary Tract Malformations (MARVU), Université Paris Cité, Paris, France Paul D. Losty  Alder Hey Children’s Hospital NHS Foundation Trust, School of Health and Life Science, University of Liverpool, Liverpool, UK Conor Mallucci  Department of Neurosurgery, Alder Hey Children’s NHS Foundation Trust, Liverpool, UK Leopoldo  Martinez Departamento de Cirugía Pediátrica, Hospital Universitario La Paz, Madrid, Spain Steffi  Mayer Department of Pediatric Surgery, University of Leipzig, Leipzig, Germany Peter  McCarthy National Children’s Cancer Service, Children’s Health Ireland at Crumlin, Dublin, Ireland Systems Biology Ireland, School of Medicine, University College Dublin, Dublin, Ireland Anna McGuire  Vascular Anomalies Center, Department of Surgery, Boston Children’s Hospital, Boston, MA, USA Martin L. Metzelder  Department of Pediatric Surgery, University Clinic of Surgery, Medical University of Vienna, Vienna, Austria Alastair J.W. Millar  University of Cape Town and Red Cross War Memorial Children’s Hospital, Cape Town, South Africa Prabal  R.  Mishra Departments of Paediatric Surgery & Child Health, Wellington Hospital, Wellington, New Zealand Ali A. Mokdad  Division of Pediatric Surgery, Medical College of Wisconsin, Milwaukee, WI, USA Maria Molina  Department of Pediatric Surgery, Virgen del Rocio Children’s Hospital, Sevilla, Spain Michael W. Morris  University of Mississippi Medical Center, Jackson, MS, USA

Contributors

Contributors

xxiii

A. Mortell  Children’s Health Ireland (CHI), Dublin, Ireland Dhanya  Mullassery  Consultant Paediatric Surgeon, Great Ormond Street Hospital for Sick Children, University College London, London, UK Ancuta  Muntean Children’s Health Ireland at Temple Street, Dublin, Ireland Nana  Nakazawa Department of Pediatric Surgery, Juntendo University Nerima Hospital, Tokyo, Japan Simona Nappo  Division of Paediatric Urology, “Bambino Gesù” Children’s Hospital, Rome, Italy L. Nguyen  Montreal Children’s Hospital, Montreal, QC, Canada Gregory  Nolan Children’s Health Ireland, Crumlin, Dublin, Republic of Ireland Agneta  Nordenskjöld Department of Pediatric Surgery, Karolinska University Hospital, Stockholm, Sweden Department of Women’s and Children’s Health, Karolinska Institutet, Stockholm, Sweden A.  Numanoglu University of Cape Town and Red Cross War Memorial Children’s Hospital, Cape Town, South Africa Donald  Nuss  Children’s Hospital of The King’s Daughters, Norfolk, VA, USA Eastern Virginia Medical School, Norfolk, VA, USA Benedict  C.  Nwomeh Department of Surgery and Pediatrics, Ohio State University College of Medicine, Nationwide Children’s Hospital, Columbus, OH, USA James J. O’Byrne  National Centre for Inherited Metabolic Disorders, Mater Misericordiae Hospital, Dublin, Ireland and School of Medicine, University College Dublin, Dublin, Ireland Keith  T.  Oldham Division of Pediatric Surgery, Medical College of Wisconsin, Milwaukee, WI, USA S. O’Sullivan  Paediatric Intensive Care Unit, Our Lady’s Children’s Hospital Crumlin, Dublin, Ireland Mikko P. Pakarinen  Department of Pediatric Surgery, Hospital for Children and Adolescents, University of Helsinki, Helsinki, Finland Thambipillai  Sri  Paran Children’s Health Ireland at Crumlin, Dublin, Ireland and Trinity College Dublin and University College Dublin, Dublin, Ireland Dakshesh Parikh  Department of Paediatric Surgery, Birmingham Children’s Hospital, Birmingham, UK

xxiv

Annabel  Paye-Jaouen Department of Pediatric Urology, Robert-Debré University Hospital, APHP, National Reference Center of Rare Urinary Tract Malformations (MARVU), Université Paris Cité, Paris, France Alberto  Peña International Center for Colorectal and Urogenital Care, Children’s Hospital Colorado, University of Colorado School of Medicine, Aurora, CO, USA Matthieu  Peycelon Department of Pediatric Urology, Robert-Debré University Hospital, APHP, National Reference Center of Rare Urinary Tract Malformations (MARVU), Université Paris Cité, Paris, France Agostino  Pierro  Division of General and Thoracic Surgery, Translational Medicine Program, The Hospital for Sick Children, University of Toronto, Toronto, ON, Canada Hannes Prescher  University of Chicago Medicine & Biological Sciences, Chicago, IL, USA Prem Puri  Department of Pediatric Surgery, Beacon Hospital, and University College Dublin, Dublin, Ireland Priya Ramachandran  Department of Pediatric Surgery, Kanchi Kamakoti CHILDS Trust Hospital, Chennai, Tamil Nadu, India Michael Riccabona  Department of Pediatric Radiology, University Clinic of Radiology, Medical University Graz, Graz, Austria Risto  J.  Rintala Department of Pediatric Surgery, Children’s Hospital, Helsinki University Central Hospital, Helsinki, Finland Michael Riordan  Department of Nephrology & Transplantation, Children’s Health Ireland at Temple Street, Dublin, Ireland Massimo  Rivosecchi Department of Pediatric Surgery, “Bambino Gesù“ Children’s Hospital, Palidoro, Rome, Italy Heinz  Rode Division of Paediatric Surgery, Department of Surgery, University of Cape Town, Cape Town, South Africa and Red Cross War Memorial Children’s Hospital, Cape Town, South Africa Alan David Rogers  Ross Tilley Burn Centre, Sunnybrook Health Sciences Centre, Toronto, Canada Division of Plastic and Reconstructive Surgery, Department of Surgery, University of Toronto, Toronto, ON, Canada Udo Rolle  Department of Pediatric Surgery, University Hospital Frankfurt, Frankfurt/Main, Germany Department of Pediatric Surgery, Goethe University Frankfurt, Frankfurt, Germany Jonathan Ross  Department of Urology, Rush University, Chicago, IL, USA

Contributors

Contributors

xxv

John  Russell Department of Pediatric Otorhinolaryngology, Children’s Health Ireland, Crumlin, Dublin, Ireland Robert Sader  Department of Oral, Maxillofacial, and Plastic Facial Surgery, Goethe University Frankfurt, Frankfurt, Germany Amanda  F.  Saltzman Department of Urology, University of Kentucky, Lexington, KY, USA David E. Sawaya  University of Mississippi Medical Center, Jackson, MS, USA Amulya  K.  Saxena Department of Paediatric Surgery, Chelsea and Westminster Hospital NHS Foundation Trust, Imperial College London, London, UK Marshall  Z.  Schwartz  Department of Surgery and Urology, Wake Forest University School of Medicine, Wake Forest Institute for Regenerative Medicine, Wake Forest Baptist Health, Winston-Salem, NC, USA S. Shah  Baylor College of Medicine, Houston, TX, USA Khalid Sharif  Birmingham Women’s and Children’s Hospital, Birmingham, UK Shilpa  Sharma Department of Pediatric Surgery, All India Institute of Medical Sciences, New Delhi, India Rita D. Shelby  Ohio State University Wexner Medical Center, Columbus, OH, USA Leah M. Sieren  Wake Forest Baptist Health, Winston-Salem, NC, USA Georg  Singer University Clinic of Pediatric and Adolescent Surgery, Medical University of Graz, Graz, Austria Michael  Singh  Department of Paediatric Surgery, Birmingham Children’s Hospital, Birmingham, UK Owen Patrick Smith  National Children’s Cancer Service, Children’s Health Ireland at Crumlin, Dublin, Ireland Systems Biology Ireland, School of Medicine, University College Dublin, Dublin, Ireland Coilin  Collins  Smyth Our Ladys Children’s Hospital, Crumlin, Dublin, Ireland Valeria Solari  Department of Pediatric Surgery, Klinik Donaustadt, Vienna, Austria Erich  Sorantin  Department of Paediatric Radiology, University Clinic of Radiology, Medical University of Graz, Graz, Austria

xxvi

Bridget  R.  Southwell Surgical Research Group, Murdoch Children’s Research Institute, Melbourne, VIC, Australia R. Ben Speaker  Royal College of Surgeons Ireland, Dublin, Ireland Shawn D. St. Peter  Children’s Mercy Hospital, Kansas City, MO, USA Ionica Stoica  Children’s Health Ireland at Temple Street, Dublin, Ireland Charles  J.H.  Stolar  Stanley Morgan Children’s Hospital, New York, NY, USA Mark D Stringer  Department of Paediatric Surgery, Wellington Children’s Hospital and Department of Paediatrics and Child Health, Wellington School of Medicine, University of Otago, Wellington, New Zealand Steven  Stylianos Division of Pediatric Surgery, Columbia University Vagelos College of Physicians & Surgeons, NYP-Morgan Stanley Children’s Hospital, New York, NY, USA R. Sun  Texas Children’s Hospital, Houston, TX, USA Geraint  Sunderland Department of Neurosurgery, Alder Hey Children’s NHS Foundation Trust, Liverpool, UK Anna Svenningsson  Department of Pediatric Surgery, Karolinska University Hospital, Stockholm, Sweden Yechiel  Sweed Department of Pediatric Surgery, Galilee Medical Center, Naharia, Israel Bar Ilan University, Safed, Israel Alicia G. Sykes  Department of General Surgery, Naval Medical Center, San Diego, CA, USA Paul  Tam Macau University of Science and Technology, Macau, SAR, China Hannah  Thompson Department of Paediatric Surgery, Kings College Hospital, London, UK Holger Till  University Clinic of Pediatric and Adolescent Surgery, Medical University of Graz, Graz, Austria Juan A. Tovar  Departamento de Cirugía Pediátrica, Hospital Universitario La Paz, Madrid, Spain Jeffrey S. Upperman  Department of Pediatric Surgery, Monroe Carell Jr. Children’s Hospital at Vanderbilt, Nashville, CA, USA Christian  Urban  Department of Pediatric Hematology/Oncology, University Clinic of Pediatrics and Adolescent Medicine, Medical University of Graz, Graz, Austria Benno M. Ure  Hannover Medical School, Hannover, Germany

Contributors

Contributors

xxvii

Arcady  Vachyan Pediatric Advanced Laparoscopic Surgical Unit, Ruth Rappaport Children’s Hospital, Rambam Medical Center, Haifa, Israel Patricio Varela  Dr Luis Calvo Mackenna Children’s Hospital, Division of Pediatric Surgery, Pediatric Airway and Chest Wall Malformation Center, University of Chile, Santiago, Chile Division of Pediatric Surgery, Clinica Alemana Medical Center, Santiago, Chile Lexie  H.  Vaughn Department of General Surgery, Vanderbilt University Medical Center, Nashville, CA, USA Jaya  Vikraman Department of Paediatrics, University of Melbourne, Melbourne, VIC, Australia Annelie  M.  Weinberg Department of Pediatric Traumatology, Johannes Gutenberg University, Mainz, Germany Santiago  Weller Department of Pediatric Urology, Hospital de Pediatría “Prof. Dr. J.P. Garrahan”, Buenos Aires, Argentina Tomas Wester  Department of Pediatric Surgery, Astrid Lindgren Children’s Hospital, Karolinska University Hospital, Stockholm, Sweden Duncan  Wilcox Department of Pediatric Urology, Children’s Hospital Colorado, University of Colorado School of Medicine, Aurora, CO, USA Kenneth  Wong University of Hong Kong, Queen Mary Hospital, Hong Kong, SAR, China Lina  Woods Department of Pediatric Otorhinolaryngology, Children’s Health Ireland, Dublin, Ireland Wayland J. Wu  Division of Pediatric Urology, Cohen Children’s Medical Center of New York, Zucker School of Medicine at Hofstra/Northwell, New Hyde Park, NY, USA James Buchannan Brady Urological Institute, Roberts Jeffs Division of Pediatric Urology, Charlotte Bloomberg Children’s Hospital, Johns Hopkins University School of Medicine, Baltimore, MD, USA Atsuyuki  Yamataka Department of Pediatric and Urogenital Surgery, Juntendo University School of Medicine, Bunkyo-ku Tokyo, Japan Iain Yardley  Evelina London Children’s Hospital, London, UK Fanny Yeung  Queen Mary Hospital, Hong Kong, SAR, China Zacharias Zachariou  Department of Pediatric Surgery, Makarios Children’s Hospital, Nicosia, Cyprus Medical School, University of Cyprus, Nicosia, Cyprus Mohammed  Zamakhshary Department of Pediatric Surgery, Alfaisal University, Riyadh, Saudi Arabia

xxviii

Augusto Zani  Division of General and Thoracic Surgery, The Hospital for Sick Children, Toronto, Canada Department of Surgery, University of Toronto, Toronto, Canada Elke  Zani-Ruttenstock Division of General and Thoracic Surgery, The Hospital for Sick Children, Toronto, Canada Department of Surgery, University of Toronto, Toronto, Canada Paola Zaupa  University Clinic of Pediatric and Adolescent Surgery, Medical University of Graz, Graz, Austria Moritz  M.  Ziegler  Children’s Hospital Colorado, University of Colorado School of Medicine, Aurora, CO, USA

Contributors

Part I General Principles

1

The Epidemiology of Birth Defects Florian Friedmacher and Edwin C. Jesudason

1.1 Introduction Surgical correction of birth defects helped to establish the specialty of pediatric surgery during the mid-twentieth century. Around this time, pioneering neonatal operations were successfully performed for the first time, enabling survival of newborns with life-threatening conditions like congenital diaphragmatic hernia (CDH) or esophageal atresia. Indeed, high survival rates can now be achieved for many previously fatal anomalies, due to innovations like parenteral nutrition as well as the concentration of surgical, anesthetic and intensive care expertise. For certain birth defects that still show high rates of mortality and morbidity (e.g. CDH or myelomeningocele), fetal surgery represents a promising approach to harm reduction. For many pediatric surgeons, birth defects represent only a small part of their overall work,

F. Friedmacher (*) Department of Pediatric Surgery, University Hospital Frankfurt, Goethe University Frankfurt, Frankfurt (Main), Germany e-mail: [email protected] E. C. Jesudason Department of Rehabilitation Medicine, Astley Ainslie Hospital, Edinburgh, UK University of Edinburgh, Edinburgh, UK

because rates are relatively low or afflicted children do not survive to reach them. However, if neonatal surgeons are properly able to judge their outcomes and to advise expectant parents, they need to have good understanding of the epidemiology of particular birth defects in their catchment area and to have a feel for which of the above best explains their local rates.

1.1.1 Birth Defects Are Leading Causes of Infant Mortality and Long-term Morbidity Worldwide Due to the advances made in the treatment of infectious diseases, birth defects have emerged as one of the leading causes of infant death. This pertains not only to countries with well-funded healthcare systems but in fact anywhere that infant mortality has fallen below 50 per 1000 births (Carmona 2005). Prevention is possible for some birth defects. For example, congenital rubella syndrome may be virtually eradicated by adoption of an effective maternal vaccination program (Banatvala and Brown 2004). Nevertheless, a subset of neural tube defects continues to occur perhaps due to inadequate peri-­ conceptional folic acid supplementation (Khoshnood et al. 2015) and lack of fortification policies (Arth et al. 2016). For many other birth defects, prevention does not seem possible, fea-

© Springer Nature Switzerland AG 2023 P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_1

3

4

sible or effective. In these circumstances, the epidemiological challenge extends beyond the known, preventable defects to newer issues and their potential consequences (e.g., COVID-19 pandemic or Zika virus epidemic) (Dang et  al. 2020; Pierson and Diamond 2018). Hence, birth defects will continue to cause not only infant deaths but also premature birth and long-term morbidity. Initially, pediatric services tended to focus on perinatal management of birth defects, but the focus has now widened to include improving quality of life and the transition toward ­independent adulthood. This is not helped by the separate development of pediatric and adult specialists, which means there can often be a lack of meaningful engagement and communication between the two. After all, it is likely that pediatric surgeons can do more to address transition and mitigate long-term sequelae (Rothstein and Dasgupta 2016).

1.1.2 Birth Defects Epidemiology and Teratology Have Emerged from Outbreak Investigations Modern birth defects epidemiology and teratology arose from two hazards that became apparent during the course of the last century: the recognition of congenital rubella syndrome (noted by clinical ophthalmological examination) (Cooper 1985) and the thalidomide disaster (phocomelia and other defects associated with maternal thalidomide administration for morning sickness) (Botting 2002). These unfortunate events vividly illustrated the devastating consequences of prenatal infection and drug exposure, respectively. They highlighted the urgent need to formalize reliable birth defect surveillance. Today, this serves a wide range of purposes, including early warning of outbreaks, identification of potential environmental or genetic causes, rational planning of neonatal services, facilitation of prenatal counseling and the comparison of outcomes (as a guide toward best practice) (Khoury 1989).

F. Friedmacher and E. C. Jesudason

1.1.3 Causation of Birth Defects Remains Often Complex and Poorly Understood Before considering the methods of birth defect surveillance, it is worth outlining the developmental biology and embryology that underlie birth defects (Donnai and Read 2003). Causes of birth defects can be considered as parental, fetal and environmental. In reality, they will often overlap. For instance, grandparental behaviors (or exposures) may produce epigenetic modifications that only manifest themselves in the developing fetus. A familiar example of a parental factor is the impact of maternal age on the prevalence of Down syndrome (Loane et al. 2013). Moreover, maternal conditions such as diabetes (Bell et al. 2012) and overweight/obesity (Stothard et  al. 2009) are well-described risk factors for the formation of birth defects. Unfortunately, the role of paternal age and/or exposures is more difficult to quantify (Yang et  al. 2007). Fetal causes may include genetically determined inborn errors of metabolism such as those causing intersex anomalies in congenital adrenal hyperplasia, chromosomal lesions such as Down or Edwards syndrome and twinning with its increased risk of birth anomalies. Environmental causes include those related to prenatal addiction and drug exposure (e.g. alcohol, smoking, illicit drugs, thalidomide, valproate, phenytoin, warfarin) as well as the impact of intrauterine infections (e.g. toxoplasmosis, rubella, cytomegalovirus) (Blotière et al. 2019; Hackshaw et al. 2011; Jentink et al. 2010; Ernhart et al. 1987). In turn, the effect of assisted reproductive technologies such as in vitro fertilization and intracytoplasmic sperm injection on the prevalence of birth defects is quite difficult to assess (Tararbit et al. 2013). The suggestion that anomaly rates are higher in assisted pregnancies needs to take into account several confounding factors, including the increased rates of multiple pregnancy, parental abnormalities and predispositions that may have resulted in the need for assisted reproduction.

1  The Epidemiology of Birth Defects

Other environmental contributors to birth defects may include “endocrine disrupting chemicals” (McLachlan 2001). These estrogenic compounds are conjectured to contribute to anomalies of sexual development in fetal males (e.g. hypospadias) as well as putative impairment of adult male sperm quality (Colborn et  al. 1993). In light of such difficulties in attributing causes, it is simpler to admit that only the minority of birth defects are known to arise from a simple environmental or genetic cause. At present, the majority of birth defects appear to have multifactorial origins. In such circumstances, it is helpful to consider birth defect causation as the result of complex interactions between genes and environment. Hence, some cases of spina bifida may result from micronutrient deficiency in the context of predisposing enzyme polymorphisms (Brody et  al. 2002). Similarly, teratogenic drugs may interact with pharmacogenomic predispositions to help explain why only certain pregnancies are affected (Leeder and Mitchell 2007). Beyond consideration of complex causation, it remains likely that simple chance has a major role to play, similar to the stochastic effects noted in radiation biology (Whitaker et al. 2003).

1.1.4 Birth Defects Appear to Arise Typically (But Not Exclusively) in the First Trimester Developmental biologists refer to “competence windows” to describe periods in development when particular cells and tissues are capable of responding appropriately to certain growth and transcription factors (Johansson et al. 2007; Kim et  al. 2005). In a similar manner, developing organs are assumed to have particular temporal windows when an otherwise non-specific teratogenic stimulus will impact disproportionately on formation of that organ system. During the first trimester, organ morphogenesis predominates, while later trimesters are devoted to organ growth and maturation. Therefore, sensitivity to teratogens is held to

5

peak during the first trimester. Hence, pregnant women are advised to avoid medications during this part of gestation in particular. Teleologically, morning sickness, which peaks during the first trimester, is postulated to help reduce ingestion of potential teratogens during this period of maximum vulnerability. While the model of first-­ trimester teratogenesis appears appropriate for many birth defects, it is now clear that certain anomalies appear to arise later as a result of fetal events (e.g. amniotic band formation, intussusception or vascular insult). Gastroschisis and intestinal atresia may be considered in this latter category (Feldkamp et al. 2007). Indeed, the difference between gastroschisis and omphalocele in terms of associated anomalies (and hence prognosis) may be explained by the different times they are held to originate during development. Omphalocele is considered an embryonic lesion that is accompanied by contemporaneous lesions of organogenesis in other systems such as the heart. In contrast, gastroschisis is thought to result from a discrete fetal vascular accident (like the associated intestinal atresia) and thus lacks extraintestinal manifestations. An alternative view, however, is that intestinal atresiae result only rarely from fetal accidents such as intussusception and are in fact better understood as failures of mesenteric vascular development (Shorter et al. 2006.). Similarly, a contrast between duodenal atresia and small bowel atresia may likewise be understood as the result of their differing onsets and etiologies. Duodenal atresia was historically explained as an embryonic failure of luminal recanalization, although this “solid core” theory has been contradicted by more recent animal studies. The strong association between duodenal atresia and other defects (e.g. cardiac defects, esophageal atresia and Down syndrome) supports an embryonic origin of this malformation (Meio et al. 2008). In contrast, small bowel atresiae are thought to follow mesenteric vascular occlusion usually in fetal life (Nichol et al. 2011). Hence, aside from gastroschisis, intestinal atresiae are unlikely to be associated with other structural lesions. Between these two extremes

F. Friedmacher and E. C. Jesudason

6

are birth defects where an embryonic lesion has deleterious knock-on effects later in fetal development. Based on experimental models, the neurological sequelae of spina bifida are postulated to result not only from the primary failure of neural tube closure but also from consequent exposure of the neural placode to amniotic fluid (Stiefel and Meuli 2007). Similarly, pulmonary hypoplasia in CDH may be explained by two insults, one affecting both lungs before diaphragm development and one affecting the ipsilateral lung after defective diaphragm development (Keijzer et  al. 2000). In circumstances such as these, where the pathology is thought to progress during fetal life, prenatal surgical correction has been a logical proposal to meet the challenge of refractory mortality and morbidity (Jancelewicz and Harrison 2009).

Table 1.1 Prevalence rates of congenital anomalies according to EUROCAT subgroups including genetic anomalies (2011–2018) (Available at: https://eu-­rd-­ platform.jrc.ec.europa.eu/eurocat/eurocat-­d ata/ prevalence_en) EUROCAT subgroups All anomaliesb – Congenital heart defects – Chromosomal – Limb – Urinary – Nervous system – Genital – Digestive system – Oro-facial clefts – Abdominal wall defects – Respiratory – Eye – Ear, face and neck

Prevalence per 10,000 births (95% CI)a 255.28 (253.97–256.60) 79.76 (79.02–80.50) 44.61 (44.06–45.16) 38.09 (37.58–38.60) 35.12 (34.63–35.61) 26.12 (25.70–26.54) 20.78 (20.41–21.16) 18.65 (18.30–19.01) 14.40 (14.09–14.71) 6.63 (6.42–6.85) 4.03 (3.87–4.20) 3.98 (3.82–4.15) 1.77 (1.66–1.88)

  Including live births, fetal deaths/stillbirths (from 20 weeks of gestation) and termination of pregnancy for congenital anomaly b  Excluding cases with only minor anomalies. Cases with more than one anomaly are only counted once a

1.1.5 Classification of Birth Defects for Epidemiological Purposes In general, the epidemiology of birth defects derives from the registration of anomalies by type. For instance, a European network of population-­based registries for the epidemiologic surveillance of congenital anomalies (EUROCAT) surveys over 1.7 million births (29% of European birth population) per year from multiple sources in 21 countries in Europe (Morris et al. 2018). All EUROCAT registries use a classification scheme based around organ systems (Table 1.1), specific diagnoses and International Classification of Diseases codes (Table  1.2). The “International Clearinghouse for Birth Defects” (http://www. icbdsr.org) is another good starting resource for pediatric surgeons wishing to know more about the epidemiology of birth defects. Cooperation between registries helps by pooling data and also by building consensus on certain issues like exclusion of minor anomalies without major and/or long-term sequelae (e.g. cryptorchidism or congenital hydrocele) or how abnormalities of gut fixation in CDH may be recorded. Although birth defects are currently classified by their structural anomaly (e.g. CDH, esophageal atresia) or defined diagnosis (e.g.

Down syndrome), it is likely that in the future, anomalies may be classified or at least subgrouped by genotypic differences rather than anatomic details alone. In certain cases, such distinctions may be prognostically and therapeutically important: e.g. omphalocele in BeckwithWiedemann syndrome is associated with the additional hazards of hypoglycemia, macrosomia and increased tumor risk due to disordered gene imprinting (Piedrahita 2011). Hence, the actual anatomical defect (i.e. omphalocele) becomes less important than the genetics and its multisystem sequelae. Similarly, it is postulated that subgroups of spina bifida may be folate-resistant due to underlying genetic/enzymatic variation (Pitkin 2007; Brody et al. 2002). The design of preconceptional prophylaxis for birth defects may need to acknowledge pharmacogenomically distinct subgroups to avoid benefits within one subgroup being overlooked due to a surrounding nonresponder cohort. Having a system of classification is, however, only one part of the task. Notification and classi-

1  The Epidemiology of Birth Defects Table 1.2  EUROCAT prevalence rates of congenital anomalies according to disease codes including genetic anomalies (2011–2018) (Available at: https://eu-­rd-­ platform.jrc.ec.europa.eu/eurocat/eurocat-­d ata/ prevalence_en) Prevalence per 10,000 births (95% CI)a Anomalies Down syndrome 24.43 (24.02–24.84) Hypospadias 17.77 (17.42–18.12) Congenital hydronephosis 12.82 (12.53–13.12) Edward syndrome 5.99 (5.79–6.19) Spina bifida 4.97 (4.79–5.16) Multicystic renal dysplasia 4.36 (4.19–4.53) Omphalocele 3.63 (3.47–3.79) Anorectal atresia and stenosis 3.37 (3.22–3.52) Congenital diaphragmatic hernia 2.93 (2.79–3.08) Esophageal atresia with/without 2.68 (2.55–2.82) tracheoesophageal fistula Gastroschisis 2.56 (2.43–2.69) Duodenal atresia or stenosis 1.45 (1.35–1.55) Hirschsprung disease 1.44 (1.34–1.54) Posterior urethral valve and/or 1.27 (1.18–1.37) prune belly 1.17 (1.08–1.26) Congenital pulmonary airway malformations Small bowel atresia or stenosis 0.94 (0.87–1.03) Situs inversus 0.81 (0.74–0.89) 0.62 (0.55–0.68) Bladder extrophy and/or epispadia Congenital construction bands/ 0.56 (0.50–0.62) amniotic bands Indeterminate sex 0.52 (0.47–0.59) Biliary atresia 0.33 (0.28–0.38) Conjoined twins 0.16 (0.13–0.19)   Including live births, fetal deaths/stillbirths (from 20 weeks of gestation) and termination of pregnancy for congenital anomaly a

fication are subject to local variations in practice. When resources exist for expert-led classification of birth defects by diagnosis, this approach to birth defect epidemiology appears the best currently available (Lin et  al. 2006). Nevertheless, even some North American registries lack clinician input in the classification and assignment of observed birth defects. The consequences of this omission for data quality remains to be determined. In the contrasting circumstances of rural China, expert-led assignment of cases has been substituted with simple photographic recording of malformations. This system not only allows the registry to function but also allows difficult

7

cases to be assigned later after remote assessment of images by experts (Li et al. 2003). In addition, the photographs potentially allow the classifiers to calibrate their judgments against those from other registries.

1.1.6 Counting of Birth Defects Is Affected by the Definition of Stillbirth Birth defect epidemiology becomes difficult whenever the classification of defects is not uniform or straightforward. However, an equal challenge remains the counting of birth defects. This task is complicated by practical barriers to case ascertainment (e.g. inadequate resources), the definition of stillbirth, and the effects of prenatal diagnosis and termination of pregnancy. Recording of anomaly prevalence lies at the core of birth defect epidemiology. To account for the unknowable incidence of a defect amongst vast numbers of naturally miscarried pregnancies, epidemiologists measure the prevalences of defects within a defined birth cohort: i.e. the number of live and stillborn cases of the defect, as a proportion of all births. This definition depends on the artificial distinction between miscarriage and stillbirth: EUROCAT’s recommendation is that spontaneous pregnancy losses prior to 20 weeks of gestation are counted as miscarriages (and therefore do not contribute to anomaly prevalence), while similar losses at 20 weeks of gestation and beyond are counted as stillbirths (and thus are included in prevalence statistics). Despite these guidelines, several countries have established different demarcations (e.g. 24 or 28 weeks or even 500 g birth weight). Clearly, some estimate of prenatal birth defects is required to avoid serious underestimation of overall prevalence (Duke et al. 2009). However, the demarcation of stillbirths begins to complicate matters. Countries using later gestational cut-offs may underestimate birth defect prevalence compared to registries where 20 weeks of gestation is used. Hence, minor changes in convention can lead to large but artificial differences in anomaly prevalence. While a definition of stillbirths is needed

8

for data collection, the sharp demarcation (whether 20  weeks or later) also appears arbitrary from a biological perspective. Consider a hypothetical prenatal medical therapy that reduces the prevalence of a specific birth defect. When the anomaly is rare (as most are), it may be difficult to determine whether an observed reduction in prevalence is truly due to fewer malformations or instead due to the promotion of earlier loss of affected pregnancies (i.e. prior to the 20th week or other agreed-upon margin). This latter phenomenon, termed “terathanasia”, has been invoked to explain how folate supplementation might influence the prevalence of neural tube defects (Godwin et al. 2008; Hook and Czeizel 1997).

1.1.7 Prenatal Diagnosis: The Greatest Challenge to Birth Defect Epidemiology? The classification of birth defects and the definition of stillbirth make anomaly surveillance complex. However, the impact of prenatal diagnosis is arguably still more important. Prenatal diagnosis (in particular, non-specific ultrasound screening) confounds birth defects surveillance in a number of ways: • Prenatal diagnosis increases identification of birth defects within the cohort of assessment by diagnosing those who may otherwise have perished prenatally (and were uncounted) or those who may have presented beyond the neonatal period (if at all). For instance, prenatal identification of cystic lung lesions. Some would never have been diagnosed (e.g. either regressing spontaneously or persisting asymptomatically). Others would have presented later (beyond the scope of the birth defects registry). • Prenatal diagnosis alters antenatal management and results in abortions or fetal interventions, which in turn affect the numbers of birth

F. Friedmacher and E. C. Jesudason

defects being counted. Therefore, most registries attempt to keep separate data on terminations for congenital anomalies, but these data are hard to find when abortions are prohibited by law. • Prenatal diagnosis may be inaccurate and unchecked. Pathological verification after termination of pregnancy may be incomplete or absent, yet the presumptive diagnosis is included in the birth defect list. • The resources and expertise to perform prenatal sonography vary with location, thereby hampering international comparison of birth defect prevalences. In summary, the apparently simple task of counting live and stillborn cases for birth defect surveillance is fraught with difficulty once the arbitrary definition of stillbirth is imposed and ubiquitous prenatal imaging prompts both terminations and identification of previously occult cases. Given these challenges in data collection, epidemiologists are aided by being able to compare a variety of surveillance databases. Many European registries are incorporated into the EUROCAT initiative. Similarly, several other registries feed into birth defect surveillance data furnished by the World Health Organization (WHO). Their “Atlas of selected congenital anomalies” is an interesting publication available in the public domain (https://apps.who.int/nutrition/publications/birthdefects_atlas/en/index. html). Most importantly, it is instructive to read and consider the caveats that EUROCAT and WHO place upon their data. These interpretational issues not only highlight the problems discussed in the previous sections but also allude to the ongoing challenge of inadequate resources and expertise for birth defect reporting. This in turn impairs the data accuracy and may help explain insufficient action upon findings. Major studies reinforce the logistical shortcomings of birth defect reporting in the United Kingdom (Boyd et al. 2005).

1  The Epidemiology of Birth Defects

1.1.8 Pediatric Surgeons Often Focus on Their Institutional Series of Birth Defects Small institutional series are the staple of pediatric surgeons’ reporting. However, several studies have shown how institutional series are vulnerable to bias and confounding (Mah et  al. 2009). Indeed, these studies face broader problems than population-based registries. Ascertainment remains a particular issue. For example, prenatal diagnosis, terminations or fetal deaths/stillbirths prior to transfer can each give the misleading impression that the institution is improving outcomes when, in fact, extramural changes are responsible. Moreover, pediatric surgeons try to stratify for disease severity to show that their (good) results are not simply the product of a low-risk caseload. But in such circumstances, it can be misleading to use the frequency of interventions to stratify for severity in a birth defect cohort. For instance, in CDH, the decision to patch and/or use extracorporeal membrane oxygenation and/or nitric oxide may owe in reality more to institutional protocols than differences in pathophysiology between cases. Ultimately, institutional series are often subject to substantial bias with apparently poorer results perhaps not even being submitted for publication.

1.1.9 A “Life-Course” Approach to Birth Defects Given the aforementioned precautions, where can pediatric surgery make progress? Advances in prenatal imaging may improve prenatal prognostication and case selection for fetal therapies (Jancelewicz and Harrison 2009). However, better imaging may also identify more defects of questionable significance. To help families balance these developments, pediatric surgeons will need to keep abreast of birth defect epidemiology and to collaborate with other specialties. A larger prize may be realized if pediatric surgery follows epidemiology in widely adopting “life-course” studies to assess the impact over time of correct-

9

ing birth defects surgically. Embracing a life-­ course approach means that, like pediatric oncology, further thought and training can be focused on the needs of teenage and young adult survivors. Pediatric surgery can also adopt the biopsychosocial model and the creative, problem-­ solving approach exemplified by, e.g., trauma rehabilitation. These efforts may help the team broach issues well before the hurdles of transitional care. It may also allow such issues to be addressed using the full breadth of non-surgical, non-pharmacological approaches.

1.2 Conclusion and Future Directions It has been argued that birth defects are best interrogated with a systems approach rather than via molecular biology. A similar shift away from linear drug-receptor paradigms in non-communicable adult diseases may alleviate the current stagnation within the blockbuster drug pipeline (DiMasi et al. 2004). Computing is more powerful than ever and data storage is becoming relatively cheap. Therefore, intriguing possibilities exist these days to use bigger data to do in-depth research into birth defects epidemiology. For example, scientists have been able to use social media like Facebook and Twitter to track the impacts of adverse weather, earthquakes or even impeding flu epidemics (Jesudason 2016). Store purchases may also give an early hint of such events. Given what store chains can divine about individual lifestyles, it remains to be seen whether their techniques, such as collaborative filtering, can be used to identify obscure risk factors for birth defects (Jesudason 2016). Certainly, the current generation of children are perhaps the first in which a store card records most purchases to which they have ever been exposed. At present, this seems like an area ripe for exploration. Another advance that one may see is the introduction of “near patient interfaces” that allow the parent to enter more of their information and that of their child in cases of birth defects. Shifting

10

data ownership may be an effective way to increase participation, particularly when resources for dedicated data gatherers are scarce. The question then arises whether nations, rich and poor, are building healthcare teams equipped to do such work. Medicine is still taught around the great empires of biomedicine. To help the shift from biological reductionism, these new challenges will require a reintroduction of engineering know-how into medicine: to improve quantitative modeling, to manipulate complex systems, and to program computers to explore these data-rich opportunities (Jesudason 2016). Looking at wider policy, birth registries are too important to neglect. Births show demographers how the world’s population will look in the future. By this standard, birth defects monitoring provides an early warning system for humanity as a whole. Therefore, if climate change exerts subtle effects, it may be that these will be detected first in the birth prevalences of key defects (Van Zutphen et al. 2012). Similarly, increased use of genetically modified crops seems likely to fuel public demand for good data on birth defects to ensure that risk to humans is minimal (Maghari and Ardekani 2011). Finally, areas of Iraq exposed to depleted uranium shells report increased birth defect rates (Hindin et al. 2005). Disturbingly, the WHO is alleged to have been complicit in efforts to suppress this “bad news” (Ahmed 2013). Birth defects epidemiology is often difficult in peacetime, so it is understandable that there would be controversy about conflict-­related birth defects. However, healthcare professionals have a responsibility to speak truth to power even on these uncomfortable matters.

References Ahmed N (2013) How the World Health Organisation covered up Iraq’s nuclear nightmare. The Guardian; [internet]. Available from: https://www.theguardian.com/environment/earth-­i nsight/2013/oct/13/ world-­health-­organisation-­iraq-­war-­depleted-­uranium Arth A, Kancherla V, Pachón H, Zimmerman S, Johnson Q, Oakley GP Jr (2016) A 2015 global update on folic acid-preventable spina bifida and anencephaly. Birth Defects Res A Clin Mol Teratol 106:520–529 Banatvala JE, Brown DWG (2004) Rubella. Lancet 363:1127–1137

F. Friedmacher and E. C. Jesudason Bell R, Glinianaia SV, Tennant PW, Bilous RW, Rankin J (2012) Peri-conception hyperglycaemia and nephropathy are associated with risk of congenital anomaly in women with pre-existing diabetes: a population-based cohort study. Diabetologia 55:936–947 Blotière PO, Raguideau F, Weill A, Elefant E, Perthus I, Goulet V et al (2019) Risks of 23 specific malformations associated with prenatal exposure to 10 antiepileptic drugs. Neurology 93:e167–e180 Botting J (2002) The History of Thalidomide. Drug News Perspect 15:604–611 Boyd PA, Armstrong B, Dolk H, Botting B, Pattenden S, Abramsky L et al (2005) Congenital anomaly surveillance in England  – ascertainment deficiencies in the national system. BMJ 330:27 Brody LC, Conley M, Cox C, Kirke PN, McKeever MP, Mills JL et  al (2002) A polymorphism, R653Q, in the trifunctional enzyme methylenetetrahydrofolate dehydrogenase/methenyltetrahydrofolate cyclohydrolase/formyltetrahydrofolate synthetase is a maternal genetic risk factor for neural tube defects: report of the Birth Defects Research Group. Am J Hum Genet 71:1207–1215 Carmona RH (2005) The global challenges of birth defects and disabilities. Lancet 366:1142–1144 Colborn T, vom Saal FS, Soto AM (1993) Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ Health Perspect 101:378–384 Cooper LZ (1985) The history and medical consequences of rubella. Rev Infect Dis 7:S2–S10 Dang D, Wang L, Zhang C, Li Z, Wu H (2020) Potential effects of SARS-CoV-2 infection during pregnancy on fetuses and newborns are worthy of attention. J Obstet Gynaecol Res 46:1951–1957 DiMasi JA, Grabowski HG, Vernon J (2004) R&D costs and returns by therapeutic category. Ther Innov Regul Sci 38:211–223 Donnai D, Read AP (2003) How clinicians add to knowledge of development. Lancet 362:477–484 Duke CW, Correa A, Romitti PA, Martin J, Kirby RS (2009) Challenges and priorities for surveillance of stillbirths: a report on two workshops. Public Health Rep 124:652–659 Ernhart CB, Sokol RJ, Martier S, Moron P, Nadler D, Ager JW et al (1987) Alcohol teratogenicity in the human: a detailed assessment of specificity, critical period, and threshold. Am J Obstet Gynecol 156:33–39 Feldkamp ML, Carey JC, Sadler TW (2007) Development of gastroschisis: review of hypotheses, a novel hypothesis, and implications for research. Am J Med Genet A 143A:639–652 Godwin KA, Sibbald B, Bedard T, Kuzeljevic B, Lowry RB, Arbour L (2008) Changes in frequencies of select congenital anomalies since the onset of folic acid fortification in a Canadian birth defect registry. Can J Public Health 99:271–275 Hackshaw A, Rodeck C, Boniface S (2011) Maternal smoking in pregnancy and birth defects: a systematic review based on 173 687 malformed cases and 11.7 million controls. Hum Reprod Update 17:589–604

1  The Epidemiology of Birth Defects Hindin R, Brugge D, Panikkar B (2005) Teratogenicity of depleted uranium aerosols: a review from an epidemiological perspective. Environ Health 4:17 Hook EB, Czeizel AE (1997) Can terathanasia explain the protective effect of folic-acid supplementation on birth defects? Lancet 350:513–515 Jancelewicz T, Harrison MR (2009) A history of fetal surgery. Clin Perinatol 36:227 Jentink J, Loane MA, Dolk H, Barisic I, Garne E, Morris JK et  al (2010) Valproic acid monotherapy in pregnancy and major congenital malformations. N Engl J Med 362:2185–2193 Jesudason EC (2016) The epidemiology of birth defects. In: Puri P (ed) Pediatric surgery. Springer, Berlin, Heidelberg Johansson KA, Dursun U, Jordan N, Gu G, Beermann F, Gradwohl G et al (2007) Temporal control of neurogenin3 activity in pancreas progenitors reveals competence windows for the generation of different endocrine cell types. Dev Cell 12:457–465 Keijzer R, Liu J, Deimling J, Tibboel D, Post M (2000) Dual-hit hypothesis explains pulmonary hypoplasia in the nitrofen model of congenital diaphragmatic hernia. Am J Pathol 156:1299–1306 Khoshnood B, Loane M, de Walle H, Arriola L, Addor MC, Barisic I et al (2015) Long term trends in prevalence of neural tube defects in Europe: population based study. BMJ 351:h5949 Khoury MJ (1989) Epidemiology of birth defects. Epidemiol Rev 11:244–248 Kim J, Wu HH, Lander AD, Lyons KM, Matzuk MM, Calof AL (2005) GDF11 controls the timing of progenitor cell competence in developing retina. Science 308:1927–1930 Leeder JS, Mitchell AA (2007) Application of pharmacogenomics strategies to the study of drug-induced birth defects. Clin Pharmacol Ther 81:595–599 Li S, Moore CA, Li Z, Berry RJ, Gindler J, Hong SX et al (2003) A population-based birth defects surveillance system in the People’s Republic of China. Paediatr Perinat Epidemiol 17:287–293 Lin AE, Forrester MB, Cunniff C, Higgins CA, Anderka M (2006) Clinician reviewers in birth defects surveillance programs: survey of the National Birth Defects Prevention Network. Birth Defects Res A Clin Mol Teratol 76:781–786 Loane M, Morris JK, Addor MC, Arriola L, Budd J, Doray B et al (2013) Twenty-year trends in the prevalence of Down syndrome and other trisomies in Europe: impact of maternal age and prenatal screening. Eur J Hum Genet 21:27–33 Maghari BM, Ardekani AM (2011) Genetically modified foods and social concerns. Avicenna J Med Biotechnol 3:109–117 Mah VK, Zamakhshary M, Mah DY, Cameron B, Bass J, Bohn D et al (2009) Absolute vs relative improvements in congenital diaphragmatic hernia survival: what happened to “hidden mortality”. J Pediatr Surg 44:877–882

11 McLachlan JA (2001) Environmental signaling: what embryos and evolution teach us about endocrine disrupting chemicals. Endocr Rev 22:319–341 Meio IB, Siviero I, Ferrante SMR, Carvalho JJ (2008) Morphologic study of embryonic development of rat duodenum through a computerized three-dimensional reconstruction: critical analysis of solid core theory. Pediatr Surg Int 24:561–565 Morris JK, Springett AL, Greenlees R, Loane M, Addor MC, Arriola L et  al (2018) Trends in congenital anomalies in Europe from 1980 to 2012. PLoS One 13:e0194986 Nichol PF, Reeder A, Botham R (2011) Humans, mice, and mechanisms of intestinal atresias: a window into understanding early intestinal development. J Gastrointest Surg 15:694–700 Piedrahita JA (2011) The role of imprinted genes in fetal growth abnormalities. Birth Defects Res A Clin Mol Teratol 91:682–692 Pierson TC, Diamond MS (2018) The emergence of Zika virus and its new clinical syndromes. Nature 560:573–581 Pitkin RM (2007) Folate and neural tube defects. Am J Clin Nutr 85:285S–288S Rothstein DH, Dasgupta R (2016) Delivery of Surgical Care Committee of the American Academy of Pediatrics Section on Surgery. Transition of care from pediatric to adult surgery. Pediatrics 138:e20161303 Shorter NA, Georges A, Perenyi A, Garrow E (2006) A proposed classification system for familial intestinal atresia and its relevance to the understanding of the etiology of jejunoileal atresia. J Pediatr Surg 41:1822–1825 Stiefel D, Meuli M (2007) Scanning electron microscopy of fetal murine myelomeningocele reveals growth and development of the spinal cord in early gestation and neural tissue destruction around birth. J Pediatr Surg 42:1561–1565 Stothard KJ, Tennant PW, Bell R, Rankin J (2009) Maternal overweight and obesity and the risk of congenital anomalies: a systematic review and meta-­ analysis. JAMA 301:636–650 Tararbit K, Lelong N, Thieulin AC, Houyel L, Bonnet D, Goffinet F et al (2013) The risk for four specific congenital heart defects associated with assisted reproductive techniques: a population-based evaluation. Hum Reprod 28:367–374 Van Zutphen AR, Lin S, Fletcher BA, Hwang SA (2012) A population-based case-control study of extreme summer temperature and birth defects. Environ Health Perspect 120:1443–1449 Whitaker SY, Tran HT, Portier CJ (2003) Development of a biologically-based controlled growth and differentiation model for developmental toxicology. J Math Biol 46:1–16 Yang Q, Wen SW, Leader A, Chen XK, Lipson J, Walker M (2007) Paternal age and birth defects: How strong is the association? Hum Reprod 22:696–701

2

Fetal Counselling for Surgical Congenital Malformations Kokila Lakhoo and Rebecca Black

2.1 Introduction Paediatric surgeons are often called to counsel parents once a surgical abnormality is diagnosed on a prenatal scan. The referral base for a paediatric surgeon now includes the perinatal period. Favourable impact of prenatal counselling has been confirmed to influence the site of delivery in 37% of cases, change the mode of delivery in 6.8%, reverse the decision to terminate a pregnancy in 3.6% and influence the early delivery of babies in 4.5%. Counselling parents about prenatally suspected surgically correctable anomalies should not be solely performed by obstetricians or paediatricians. Similarly, the paediatric surgeon performing these prenatal consultations must be aware of differences between the prenatal and postnatal natural history of the anomaly. There is often a lack of understanding regarding the natural history and prognosis of a condition presenting in the newborn and the same condition diagnosed prenatally.

K. Lakhoo (*) Children’s Hospital Oxford, Oxford University Hospitals, University of Oxford, Oxford, UK e-mail: [email protected] R. Black Oxford University Hospitals NHS Foundation Trust, Oxford, UK e-mail: [email protected]

The diagnosis and management of complex fetal anomalies require a team effort by obstetricians, neo-natologists, geneticists, paediatricians and paediatric surgeons to deal with all the maternal and fetal complexities of a diagnosis of a structural defect. This team should be able to provide information to prospective parents on fetal outcomes, possible interventions, appropriate setting, time and route of delivery and expected postnatal outcomes. The role of the surgical consultant in this team is to present information regarding the prenatal and postnatal natural history of an anomaly, its surgical management and the long-term outcome (Lakhoo 2007).

2.2 Historical Overview Prenatal diagnosis has remarkably improved our understanding of surgically correctable congenital malformations. It has allowed us to influence the delivery of the baby, offer prenatal surgical management and discuss the options for termination of pregnancy in the case of seriously handicapping or lethal conditions. Antenatal diagnosis has also defined an in utero mortality for some lesions, such as diaphragmatic hernia and sacro-­ coccygeal teratoma, so that true outcomes can be measured. Prenatal ultrasound scanning has improved since its first use 50  years ago, thus providing better screening programmes and more accurate assessment of fetal anomalies. Screening

© Springer Nature Switzerland AG 2023 P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_2

13

14

for Down’s syndrome may now be offered in the first trimester e.g. the combined test (using a combination of nuchal translucency measurement and maternal blood markers) or second trimester tests e.g. quadruple blood test. Better ultrasound resolution has led to the recognition of ultrasound soft markers that have increased the detection rate of fetal anomalies, but at the expense of higher false positive rates. Routine ultrasound screening identifies anomalies and places these pregnancies into a higher risk ­category. Such pregnancies may be referred to Fetal Medicine Units for further scanning and other investigations. Parents may be offered further invasive diagnostic investigations, such as amniocentesis or chorionic villous sampling. Some structural abnormalities which are difficult to define on ultrasound, such as hindbrain lesions or in the presence of oligohydramnios, may be better imaged with magnetic resonance imaging. With the increasing range of options and sophistication of diagnostic methods, parents today are faced with more information, choice and decisions than ever before, which can create as well as help to solve dilemmas. The different tests and screening procedures commonly in use are outlined below under diagnosis.

2.3 Incidence Congenital malformations account for one of the major causes of perinatal mortality and morbidity. Single major birth defects affect 3% of newborns and multiple defects affect 0.7% of babies. The prenatal hidden mortality is higher since the majority abort spontaneously. Despite improvements in perinatal care, serious birth defects still account for 20% of all deaths in the newborn period and an even greater percentage of serious morbidity later in infancy and childhood. The major causes of congenital malformation are chromosomal abnormalities, mutant genes, multifactorial disorders and teratogenic agents (Lakhoo 2007).

K. Lakhoo and R. Black

2.4 Prenatal Diagnosis 2.4.1 Screening for Fetal Anomalies The NHS fetal anomaly screening programme (FASP) offers screening to all pregnant women in England (NHS Fetal Anomaly Screening Programme Handbook 2018). The first scan is performed at 10 to 14 weeks of gestation. It can: • Confirm viability. • Accurately date the pregnancy. • Diagnose multiple pregnancy and chorionicity. • Detect major structural anomalies, such as anencephaly. The combined test can be used to assess the chance of the baby being born with Down’s syndrome (trisomy 21), Edward’s syndrome (trisomy 18) or Patau’s syndrome (trisomy 13). It combines maternal age, gestational age, ultrasound measurement of the nuchal translucency (Fig. 2.1) at between 11 and 14 weeks’ gestation with two biochemical markers—PAPPA and free beta hCG—to calculate the risk of the pregnancy being affected by T21, 13 or 18. If the nuchal translucency cannot be measured, a quadruple test can be offered. This measures four biochemical markers—AFP, hCG, uE3 and inhibin-A.  This test can be performed between 14 + 2 weeks and 20 + 0 weeks.

Fig. 2.1  Nuchal translucancy scan

2  Fetal Counselling for Surgical Congenital Malformations

The combined test has a detection rate of 80% for a screen positive rate of 2.5%. The quadruple test has a detection rate of 80% for a screen positive rate of up to 3.5%. A second scan (often referred to as ‘the anomaly scan’ is offered at 18 + 0 to 20 + 6 weeks of pregnancy. This scan is designed to identify anomalies which indicate: • Conditions that may benefit from treatment before or after birth. • Conditions whose outcome may be improved by planning an appropriate place, mode and timing of birth, along with optimal postnatal management. • That the baby may not survive the neonatal period. As a minimum, the conditions screened for at the anomaly scan, along with their detection rates, are: Anencephaly Open spina bifida Cleft lip Diaphragmatic hernia Gastroschisis Exomphalos Cardiac anomalies Bilateral renal agenesis Lethal skeletal dysplasia

98% detection rate 90% 75% 60% 98% 80% 50% 84% 60%

The quality of the images obtained is dependent on many factors, including the skill of the operator, maternal habitus and the position of the fetus. Scan technology is improving images, including 3-D and 4-D options. Scanning for fetal cardiac defects remains challenging. There is an association between an increased nuchal translucency measurement and the risk of a cardiac defect, or a wide range of other syndromes, even in the context of a normal karyotype. For this reason, those fetuses with a raised NT and normal karyotype are offered fetal echocardiography. Some anomalies may not be visible at the time of the anomaly scan, but only present later in the third trimester. Examples include duodenal atresia and other forms of bowel obstruction, and

15

non-lethal skeletal dysplasias; in achondroplasia, the femur is of normal length at the time of the anomaly scan.

2.4.2 Invasive Diagnostic Tests Amniocentesis and chorionic villous sampling (CVS) are the two most commonly performed invasive diagnostic tests (Fetal Medicine Foundation 2019). Since the introduction of the national screening programme for common trisomies, the second trimester scan is interpreted differently. Certain ultrasound findings, previously referred to as ‘soft markers’ should not be used to recalculate a risk for these trisomies. These findings include choroid plexus cysts, a dilated cisterna magna, echogenic foci in the heart and a 2-vessel umbilical cord. However, other findings, termed ‘markers’ are of significance and should be referred for further assessment in a Fetal Medicine Unit. These include an increased nuchal fold, cerebral ventriculomegaly, echogenic bowel, renal pelvic dilatation and a small fetus (95%) are unilateral, involving one lobe or segment of the lung. Less common lung anomalies include bronchogenic cysts, congenital lobar emphysema and bronchial atresia. CPAMs are usually isolated. The risk of chromosomal or genetic disorders are not increased, so invasive testing is not usually offered. Fetal echocardiography is performed in cases where there is mediastinal shift, as this can make cardiac assessment more challenging. Many CPAMs will run a benign course. Some, however, can be associated with hydrops (with a

K. Lakhoo and R. Black

18

poor prognosis) or polyhydramnios. If these complications do develop and there is a large cyst, aspiration or shunting (insertion of a pig tail catheter connecting the cyst with the amniotic fluid through the fetal chest) can improve outcome. In the case of type 3 CPAM, a course of antenatal steroids given to the mother can sometimes improve prognosis. Open fetal surgery with excision of the CPAM has been performed in a few cases. CPAMs, particularly microcystic (type 3) ones, can be difficult to see on ultrasound in the third trimester as the fetus grows and the surrounding lung becomes more echogenic. In 80%, however, the lesion is still present and needs postnatal follow-up. A small CPAM with minimal medistainal shift should not affect the timing, place or mode of birth. A large lesion, where postnatal surgery is contemplated, should deliver in a tertiary unit. Accurate antenatal multiprofessional planning is essential. Postnatal manage-

ment is dictated by clinical status at birth. Symptomatic lesions require urgent radiological evaluation with chest radiograph and ideally a CT scan (Fig. 2.2) followed by surgical excision. In asymptomatic cases, postnatal investigation consists of chest CT scan within 1 month of birth, even if regression or resolution is noted on prenatal scanning. Plain radiography should not be relied on, because it will miss and underestimate many lesions. Surgical excision of postnatal asymptomatic lesions remains controversial, with some centres opting for conservative management. The approach to treating this asymptomatic group has evolved in some centres, whereby a CT scan is performed within 1 month post birth, followed by surgery before 6 months of age due to the inherent risk of infection and malignant transformation (Annunziata et al. 2019). Small lesions less than 1 cm may be managed expectantly, as these may not represent CCAM but artifact or end on

Fig. 2.2  Prenatal scan and postnatal radiological image of CPAM

2  Fetal Counselling for Surgical Congenital Malformations

view of a vessel. True resolution of these lesions is exceptional. Successful outcome of greater than 90% have been reported for these surgically managed asymptomatic lung lesions.

19

mortality in isolated exomphalos; however, the majority of these children survive to live normal lives.

2.5.3.2 Gastroschisis Gastroschisis is an isolated lesion that usually 2.5.3 Abdominal Wall Defects occurs on the right side of the umbilical defect with evisceration of the abdominal contents Exomphalos and gastroschisis are both common directly into the amniotic cavity. The incidence is but distinct abdominal wall defects with an increasing from 1.66 per 10,000 births to 4.6 per unclear aetiology and a controversial prognosis. 10,000 births affecting mainly young mothers Antenatal detection rates for both conditions are typically less than 20  years old. Associated high. Most will be found by the time of the anom- anomalies are noted in only 5–24% of cases with aly scan; an increasing number is being found in bowel atresia the most common co-existing the first trimester. abnormality. The incidence of chromosomal and genetic syndromes is not increased. On prenatal scan, with a detection rate approaching 100%, 2.5.3.1 Exomphalos Exomphalos is characteristically a midline the bowel appears to be free floating, and the defect, at the insertion point of the umbilical loops may appear to be thickened due to damage cord, with a viable sac composed of amnion and by amniotic fluid exposure causing a “peel” forperitoneum containing herniated abdominal con- mation. Dilated loops of bowel (Fig. 2.3) may be tents. Incidence is known to be 1  in 4000 live seen from obstruction secondary to protrusion births. Associated major abnormalities that from a defect or atresia due to intestinal include trisomy 13, 18 and 21, Beckwith-­ ischaemia. Predicting outcome in fetuses with gastroschiWiedemann syndrome (macroglossia, gigantism, exomphalos), Pentology of Cantrell sis based on prenatal ultrasound finding remains (sternal, pericar-dial, cardiac, abdominal wall a challenge. There is some evidence that internal and diaphragmatic defect), cardiac, gastrointes- bowel dilatation may be predictive; however, tinal and renal abnormalities are noted in thickened matted bowel and Doppler measure60–70% of cases; thus, karyotyping, in addition ments of the superior mesenteric artery are not to detailed sonographic review and fetal echo- accurate predictors of outcome. Fetal growth cardiogram, is essential for complete prenatal restriction is common (30–60% of cases) and screening. Fetal intervention is unlikely in this more difficult to monitor because the abdominal condition. If termination is not considered, normal vaginal delivery at a centre with neonatal surgical expertise is recommended and delivery by caesarean section only is reserved for large exomphalos with exteriorised liver to prevent damage. Surgical repair includes primary closure or a staged repair with a silo for giant defects. Occasionally, in vulnerable infants with severe pulmonary hypoplasia or complex cardiac abnormalities the exomphalos may be left intact and allowed to slowly granulate and epithelialise by application of antiseptic solution. Postnatal morbidity occurs in 5–10% of cases. Malrotation and Fig. 2.3  Prenatal ultrasound of dilated bowel in gastrosadhesive bowel obstruction does contribute to chisis

20

circumference is more difficult to measure and interpret. Some centres will opt for elective caesarean section for all, but most will offer induction of labour by around 37 weeks of gestation. Delivery needs to be at a centre with paediatric surgical facilities. Various methods of postnatal surgical repair include the traditional primary closure, reduction of bowel without anaesthesia, reduction by preformed silo, or by means of a traditional silo. Co-existing intestinal atresia could be repaired by primary anastamosis or staged with stoma formation. Variation in achieving full enteral feeding due to prolonged gut dysmotility is expected in all cases. The long-term outcome in gastroschisis is dependent on the condition of the bowel. In uncomplicated cases, the outcome is excellent in more than 90% of cases. The mortality of live born infants is 5%, with further 5% suffering short bowel syndrome and 10% requiring surgery for adhesive bowel obstruction. Late third trimester fetal loss should always be mentioned during fetal counselling (Gamba and Midrio 2014).

2.5.4 Tracheo-Oesophageal Fistula (TOF) and Oesophageal Atresia (OA) Repair of TOF/OA is a condition that measures the skill of paediatric surgeons from trainees to independent surgeons. The incidence is estimated at 1 in 3000 births. Prenatally, the condition may be suspected from maternal polyhydramnios and absence of a fetal stomach bubble at any time from the 20-week anomaly scan. However, if there is an associated tracheoesophageal fistula, the stomach may appear normal on ultrasound scan. It is therefore estimated that oesophageal atresia is suspected prenatally in only about 40% of cases (Bradshaw et al. 2016). Additional diagnostic clues are provided by associated anomalies, such as trisomy (13, 18, 21), VACTERL sequence (vertebral, anorectal, cardiac, tracheo-oesophageal, renal, limbs) and CHARGE association (coloboma, heart defects,

K. Lakhoo and R. Black

atresia choanae, retarded development, genital hypoplasia, ear abnormality). Associated anomalies, mainly cardiac, are present in more than 50% of cases and worsen the prognosis; fetal echocardiography and invasive testing are therefore usually offered. Duodenal atresia may co-­exist with TOF/OA. Amnioreduction (draining of the amniotic fluid) can be offered, particularly for symptomatic relief for the mother, but carries a risk of preterm birth and is only a temporary measure as the fluid will reaccumulate. The risk of recurrence in subsequent pregnancies for isolated TOF/OA is less than 1%. Delivery is advised at a specialised centre with neonatal surgical input. Postnatal surgical management is dependent on the size and condition of the baby, length of the oesophageal gap and associated anomalies. Primary repair of the oesophagus is the treatment of choice; however, if not achieved, staged repair with upper oesophageal pouch care and gastrostomy or organ replacement with stomach or large bowel are other options. Associated anomalies require evaluation and treatment. Advanced paediatric endosurgical centres may offer minimally invasive thoracoscopic approach to the repair of TOF.  Early outcome of a high leak rate and oesophageal stricture requiring dilatation in 50% of cases are expected where the anastamosis of the oesophagus is created under tension. Improved perinatal management and inherent structural and functional defects in the trachea and oesophagus indicate long-term outcome. In early life, growth of the child is reported to be below the 25th centile in 50% of cases, respiratory symptoms in two-thirds of TOF/OA and gastro-oesophageal reflux recorded in 50% of patients. Quality of life is better in the isolated group with successful primary repair compared to those with associated anomalies and delayed repair.

2.5.5 Gastrointestinal Lesions The presence of dilated loops of bowel (>17 mm in length and 7 mm in diameter) on prenatal ultrasound scan is indicative of bowel obstruction.

2  Fetal Counselling for Surgical Congenital Malformations

Duodenal atresia has a characteristic ‘double bubble’ appearance on prenatal scan, resulting from the simultaneous dilatation of the stomach and proximal duodenum. This characteristic sign is, however, usually only present after 24 weeks’ gestation, so does not get picked up at the time of the routine anomaly scan. Associated anomalies are present in approximately 50% of cases, most notably trisomy 21  in 30% of cases, cardiac anomalies in 20% and the presence of VACTERL association (vertebral, anorectal, cardiac, tracheo-­ oesophageal, renal and limbs). The incidence of duodenal atresia is 1 in 5000 live births. The postnatal survival rate is >95% with associated anomalies, low birth weight and prematurity contributing to the 2.2 mMol/L, and a low total cell count. In contrast, the exudate contains higher protein content >30 g/L, low glucose   1.6 developed hydrops. Among those, 88% were intubated after delivery and only 53% survived, compared to 94% of those with a CVR  5  mm cysts) or microcystic ( 1.6 are at high risk for the development of hydrops. The develop-

ment of hydrops is associated with a high risk of fetal or neonatal demise due to the mass effect caused by the congenital pulmonary mass. Fetal intervention should be considered in fetuses who develop hydrops and the therapy tailored to the estimated gestational age and type of lesion

36  Congenital Malformations of the Lung

459

weekly, particularly if there is a high risk of developing hydrops. For fetuses less than 32 weeks of gestational age, with hydrops or at risk of hydrops (CVR  >  1.6), maternal steroids are advocated (Downard et al. 2017). Some series have demonstrated improvement of hydrops, regression in the size of the lesion, and/or a decrease in CVR with maternal betamethasone. A favorable response to antenatal steroids is often achieved with microcystic lesions, but these may be mixed with other morphologies. In case no improvement is noted with steroids, a thoracoamniotic shunt may be considered for macrocystic lesions. An ultrasound-guided percutaneously placed thoracoamniotic shunt has the best outcome with the lowest fetal-maternal risk for a CPAM with a large cyst. A systematic review of fetuses with a cystic adenomatoid malformation treated with thoracentesis or a thoracoamniotic shunt showed an 88% survival among those without hydrops and 65% with hydrops (Cavoretto et al. 2008). Open maternal-fetal surgery with pulmonary resection is an alternative for those with microcystic lesions that do not respond to maternal steroids. The procedure yields a 50% probability of survival to discharge from the NICU and imparts significant risk and potential morbidity to the mother. Given the technical complexity, this should only be performed in a center with experience in open maternal fetal surgery. For fetuses over 32  weeks estimated gestational age with hydrops, an ex utero intrapartum treatment (EXIT) procedure with thoracotomy and lobectomy using placental bypass can permit safe resection and avoid respiratory collapse. Several series reported an overall survival between 62 and 100% in fetuses in whom this approach was performed either because of fetal hydrops, extensive mediastinal shift, or persistently elevated CVR (David et al. 2016).

36.10.2 Bronchopulmonary Sequestration

36.10.1.2 Postnatal Therapy Prompt surgical resection should be performed in any symptomatic newborn. This is most often lobectomy of the affected area. In patients who remain asymptomatic after birth, an elective resec-

Prenatal management of bronchopulmonary sequestrations is similar to that for CPAM, as outlined above. Spontaneous involution of extralobar sequestrations is rare. Postnatal resection of a sequestration is indicated for multiple

tion is generally agreed upon among experts, but the timing of resection, surveillance strategy, and surgical approach are more contentious areas. A resection is advocated to preclude future respiratory compromise from recurrent infections or other respiratory sequalae and to prevent malignant degeneration. The likelihood of developing respiratory symptoms following the neonatal period varies between 3 and 86% in the literature. The risk of malignant degeneration in a CPAM to a pleuropulmonary blastoma or epithelial and mesenchymal malignancies has been estimated to be 9%. For patients with a positive DICER1 mutation or radiologic features that portend an increased likelihood for developing pleuropulmonary blastoma, early resection is warranted (Downard et al. 2017). Resection beyond the newborn period but during the first year of life is most common among pediatric surgeons and has been advocated by expert groups (Downard et  al. 2017). Delaying resection beyond the newborn period allows somatic growth that facilitates pulmonary resection and permits monitoring a CPAM as 4% resolve spontaneously. Operating around 6 months of age allows ample time for compensatory lung growth, avoids infectious complications, and precludes inflammation and adenopathy that increase the difficulty of the resection. The latter reason has prompted surgeons to recommend intervention as early as 3  months. Postnatal CT scan is recommended, the timing of which is dependent upon provider’s preference and operative plans. Resection most often takes the form of formal lobectomy; however, for small CPAMs, nonanatomical resection is reasonable. Traditionally, resection is done via open thoracotomy, but several contemporary reports have demonstrated excellent outcomes with thoracoscopic resection.

460

reasons. First, asymptomatic extralobar and intralobar sequestrations pose a risk of recurrent pulmonary infections or hemorrhage. Second, the abnormally developed lung tissue is ineffective in gas exchange and therefore is not of benefit to the patient. Third, in practice, sequestrations that present as a posterior mediastinal mass may be difficult to differentiate from solid tumors, CPAM, and possibly diaphragmatic hernia. Last, thoracoscopic resection particularly of extralobar sequestrations is often technically straightforward with low risk because there is no attachment to the normal lung. The goal of surgical resection is to remove only the abnormal portion of lung. For an extralobar sequestration, this is often easily achieved unless there is inflammation due to previous infection or hemorrhage. For intralobar sequestration, particularly if previously infected, this most often necessitates a lobar resection (Palla and Sockrider 2019). Endovascular embolization has been employed as well in the management of pulmonary sequestration. Transarterial, including transumbilical arterial, embolization of the anomalous systemic arteries feeding a sequestration resulted in regression of 80–90% of lesions treated (Chowdhury and Chakraborty 2015). Among those, complete regression varied between 7 and 53%. Leaving residual dysplastic tissue, however, is a major concern for this treatment modality and its adoption has been limited.

36.10.3 Congenital Lobar Emphysema Lobar resection is indicated in any symptomatic patient with CLE. This is commonly achieved by an open thoracotomy. A thoracoscopic approach is challenging, owing to a small working space, inability to deflate the affected lobe, and potential difficulty operating on the upper/middle lobes that are implicated most in CLE. In a recent large series of CLE, a quarter of patients initially approached thoracoscopically were converted to an open procedure. Of the overall study cohort, 6/53 (11.3%) patients had successful thoraco-

A. A. Mokdad et al.

scopic lobectomy (Kunisaki et  al. 2019). Emergent thoracotomy with delivery of the affected lobe can be lifesaving in a newborn with severe respiratory distress as a result of CLE. Although generally unnecessary in newborns, older children should generally undergo bronchoscopy preceding thoracotomy to evaluate for reversible endobronchial lesions not requiring pulmonary resection. Extrinsic bronchial compression is generally associated with a focal cartilaginous defect such that relief of the extrinsic compression alone is rarely adequate to relieve the bronchial obstruction. While bronchoplasty is theoretically attractive, the size of the bronchus in an infant or child poses an impractical obstacle to achieve an adequate functional result. For an asymptomatic or minimally symptomatic child with CLE, there are no long-term data to support a decision not to remove the lesion. A few studies with short-term follow-up on a small cohort of selected patients who have not undergone resection suggest that there is a low incidence of progression of disease. While the long-term outcome of these patients with untreated CLE remains unknown, the postoperative morbidity is minimal and the long-term outcome after surgical resection for CLE is excellent.

36.10.4 Bronchogenic Cyst Resection is typically indicated to alleviate symptoms; to prevent future infection, hemorrhage, airway obstruction, or malignant degeneration; and to provide pathologic identification. Generally, simple local resection with a minimally invasive approach can be accomplished with preservation of adjacent normal lung parenchyma. Resolution of active infection prior to attempted resection is helpful to minimize unnecessary pulmonary resection. Occasionally, formal pulmonary resection is required either due to the anatomic location of the cyst or inflammation from previous infections. Rarely, it is not possible to remove the cyst in its entirety without sacrificing vital structures; in such instances, partial cyst resection with ablation of the remaining cyst

36  Congenital Malformations of the Lung

wall can be performed. Long-term follow-up is necessary, however, as late recurrences have been reported for partially resected bronchogenic cysts.

36.11 Short- and Long-Term Postoperative Outcomes Pulmonary resection is generally very well tolerated, even in the newborn. Most contemporary pediatric surgical series of pulmonary resection for congenital pulmonary malformations in the absence of diffuse lung disease cite a mortality rate of  610 mmHg for more than 8 h. Older infants and children do not have as welldefined criteria for high mortality risk. A ratio of arterial PaO2 to FiO2 (P:F ratio) less than 60–100 in hypoxic respiratory failure or an arterial pH less than 7.0–7.25 due to hypercapnic respiratory failure is commonly used to identify candidates for ECMO.  The Paediatric-Pulmonary Rescue with Extracorporeal Membrane Oxygenation Prediction (P-PREP) score has been developed to predict in-hospital mortality for children over 7  days of age prior to the initiation of ECMO. It considers mode of support (V-A or V-V), duration of mechanical ventilation, P:F ratio, arterial pH, primary pulmonary diagnosis and comorbid conditions. Indications for support in patients with cardiac pathology are based on clinical signs of cardiovascular failure such as hypotension despite the administration of inotropes or volume resuscitation, metabolic acidosis, oliguria (urine output 100,000 mm3 and the ACT lowered to 180–200 s. Sometimes the temporary discontinuation of anticoagulation and normalization of the coagulation status are warranted to help stop the haemorrhage with a second circuit available in the event acute clotting of the circuit should occur. Aggressive surgical intervention is warranted if bleeding persists. Neurologic sequelae are a serious morbidity of the ECMO population and include learning

disorders, motor dysfunction and cerebral palsy. These outcomes appear to be as much due to hypoxia and acidosis prior to the ECMO course as the time on ECMO itself. ICH is the most devastating complication, occurring in 7% of newborn patients with an associated 72% mortality amongst newborns who have ICH on ECMO.  Frequent comprehensive neurologic exams should be performed and cranial ultrasounds obtained frequently while on ECMO based on local protocols. Blood pressure should be carefully monitored and maintained within normal parameters to help decrease the risk of ICH.  If necessary, electroencephalograms may be helpful in the neurologic evaluation. Acute tubular necrosis (ATN), marked by oliguria and increasing blood urea nitrogen and creatinine levels, is often seen in the ECMO patient during the initial 48 h, at which time renal function is expected to improve. If improvement does not occur, consideration must be towards poor tissue perfusion. This may be due to low cardiac output, insufficient intravascular volume or inadequate pump flow, all of which should be corrected. In the event of continued renal failure, haemofiltration or haemodialysis can be ­ performed to maintain proper fluid balance and electrolyte levels and are reported to be required in 26% of cases.

38.4 Conclusion As of January 2020, 32,385 neonates (87% survival) and 10,346 paediatric patients (72% survival) have been treated with ECMO for respiratory failure and 8830 neonatal (69% survival) and 12,538 paediatric (72% survival) patients for cardiac failure. Tables 38.1, 38.2 and 38.3 demonstrate the common neonatal and paediatric respiratory and cardiac diagnoses along with survival with ECMO support. In the neonatal period, the most common disorders treated with ECMO are CDH, MAS, PPHN, sepsis, RDS and cardiac support. For the paediatric population, viral and bacterial pneumonia, ARDS, acute respiratory failure (non-ARDS) and cardiac ­disease are the most common pathophysiologic processes requiring ECMO intervention.

38  Extracorporeal Membrane Oxygenation

Recent medical advances, such as permissive hypercapnea, inhaled nitric oxide and the use of oscillatory ventilation, have spared numerous babies from ECMO, yet many children still benefit from this modality. In summary, any patient with reversible cardiopulmonary disease, who meets criteria, should be considered an ECMO candidate. ECMO provides an excellent opportunity to provide “rest” to the cardiopulmonary systems thus avoiding the additional lung or cardiac injury which otherwise would be associated with maintaining life support.

Further Reading Bailly DK, Reeder RW, Zabrocki LA, Hubbard AM, Wilkes J, Bratton SL et  al (2017) Development and validation of a score to predict mortality in children undergoing extracorporeal membrane oxygenation for respiratory failure. Crit Care Med 45(1):e58–e66

483 Berdajs D (2015) Bicaval dual-lumen cannula for venovenous extracorporeal membrane oxygenation: Avalon(c) cannula in childhood disease. Perfusion 30:182–186 Campbell BT, Braun TM, Schumacher RE et  al (2003) Impact of ECMO on neonatal mortality in Michigan (1980–1999). J Pediatr Surg 38:290–295 Extracorporeal Life Support Organization (2020) International registry report of the extracorporeal life support organization. University of Michigan Medical Center, Ann Arbor Hirschl RB, Bartlett RH (2012) Extracorporeal life support in cardiopulmonary failure. In: Coran AG, Adzick NS, Krummerl T, Laberge JM, Shamberger R, Caldamone A (eds) Pediatric surgery, 5th edn. Mosby, New York, pp 89–102 Jarboe MD, Gadepalli SK, Church JT et al (2017) Avalon catheters in pediatric patients requiring ECMO: placement and migration problems. J Pediatr Surg 53:159–162 Kim ES, Stolar CJ (2000) ECMO in the newborn. Am J Perinatol 17:345–356 Kim ES, Stolar CJ (2003) Extracorporeal membrane oxygenation for neonatal respiratory failure. In: Puri P (ed) Newborn surgery. Arnold, London, pp 317–327

Part V Spina Bifida and Hydrocephalus

Spina Bifida and Encephalocoele

39

Martin T. Corbally

39.1 Introduction Neural tube defects (NTD: spina bifida (SB), encephalocoele) are potentially serious congenital deformities of the spine and spinal cord that can have a major impact on the quality of life, not only for the child but on the entire family. The precise aetiology is uncertain. Although the incidence appears to be decreasing, there remain a significant number of newborns with this condition each year. In the past 10 years, an awareness of the benefits of peri-conceptual folic acid and improved nutrition has significantly decreased the incidence of NTD.  The impact of antenatal screening and therapeutic abortion in some jurisdictions has clearly further reduced the incidence. Surviving children face a varied future, directly related to the severity of their NTD and to the quality of early interventional services and long-­ term support structures. The patient with NTD is likely to require the expertise of many services and specialists over their lifetime including the paediatric surgeon/neurosurgeon/urologist/orthopaedic/ophthalmic surgeon/paediatric radiologist/social workers/continence nurses and many other varied disciplines. Children with SB face the prospect of multiple surgical, urological and orthopaedic interventions for the duration of their

M. T. Corbally (*) Royal College of Surgeons in Ireland, Dublin, Ireland e-mail: [email protected]

lives and must cope with the effects of poor or zero ambulation, bladder and renal failure, hydrocephalus and the complexity of multiple shunt/ bladder or bowel and shunt procedures. The management objective for these children aims to provide as normal a life as possible, to minimise the effect of their disability in areas such as mobility, continence and education. In addition, urinary system monitoring is essential to safeguard against the complications of a neuropathic bladder and renal failure.

39.2 Embryology A NTD is a congenital defect of the spine and neural tube with failure of fusion of the vertebral arches and, to a varying degree, the development of the covering muscles and skin. In some cases, the neural tube will protrude externally as a neural plaque without any covering of skin or muscle, as in a myelomeningocoele, but in others, the neural tube is closed, but there is a defect of the vertebral arch and muscles through which the dura and arachnoid protrude (meningocele) or it is entirely covered by skin (spina bifida occulta). Essentially, the defect arises as an abnormality of fusion of the neural tube. At the start of the 4th week of foetal life, the neural plate (precursor of the neural tube) is a broad flat plate in its cranial portion that will become the brain and a narrow caudal portion that will become the spinal cord.

© Springer Nature Switzerland AG 2023 P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_39

487

M. T. Corbally

488

At about 22 days, the embryo undergoes ventral flexion, and that portion of the neural tube cranial to the point of flexure (the mesencephalic flexure) is recognisable as the future forebrain, the point at the flexure is the midbrain, and the point caudal to this point is the hindbrain. Rapid elongation over the next 7 days occurs with the narrow caudal portion (future spinal cord) occupying up to 60% of the neural tube. One of the most important events of the 4th week is conversion of the neural plate into a neural tube by a series of infolding of the plate called neurulation of the tube. This process begins along the future occipito-cervical region of the plate and progresses caudally. During this process, the lateral edges of the plate meet and fuse in the midline while detaching from the surface ectoderm, which then fuse and so cover the neural tube completely. The tube remains open at both ends during this process through small openings called the cranial and caudal neuropores. The cranial neuropore closes completely at day 24 and the caudal neuropore at day 26. The neural folds have essentially closed by 4 weeks. Subsequently, the mesodermal somites form around the closed cord, and the meninges, vertebral column and muscles result. Failure of part of the neural tube to close disrupts both the process of differentiation of the central nervous system and the induction of the Table 39.1  Types of neural tube defects

Anencephaly Myelomeningocoele

Meningocoele

vertebral arches and can result in a variety of anomalies. This most commonly affects the caudal end of the spinal cord, which affects the lumbar and sacral regions of the central nervous system. Involvement of the cranial end of the tube can result at its most extreme end in anencephaly and form an encephalocoele in less severe cases. Less severe anomalies of fusion are failure of the arch to fuse, with or without meningocele protrusion (spina bifida occulta v meningocele, respectively). More severe defects result in failure of the neuroectoderm with protrusion of the neural tube itself (myelomeningocoele). A similar process in the brain results in an encephalocoele. The process of lack of fusion can occur anywhere along the length of the spinal canal, with varying levels of severity.

39.3 Classification 39.3.1 Anencephaly Failure of the cranial end of the neural tube to close can result in disruption of the differentiation of the CNS and is represented by an exposed mass of undifferentiated neural tissue (Table  39.1). These embryos often survive to late pregnancy but usually do not survive much after birth. Brain and skull poorly developed Failure of closure of neural tube Failure of muscle and skin formation Exposed neural tissue Distal limb innervation affected Neuropathic bladder Failure of spinal fusion Dural sac protrudes

Encephalocoele

Spina bifida

Skin covered defect Usually occipital Defect in cranial bone Herniation of meninges and Brain to varying degree Occulta hamartoma at site Sinus occasionally Skin intact Bony vertebral arch deficient

Death inevitable Significant lesion Variable outcome Urgent closure 90% need VP shunt

Usually no neural consequences Rarely bladder function affected Variable outcome Sometimes shunt needed Excellent outcome

39  Spina Bifida and Encephalocoele

489

Fig. 39.2  Lumbosacral myelomeningocoele

Fig. 39.1  MRI showing occipital encephalocoele

cystic swelling at the site of the lesion are called a meningocoele. This cystic swelling is lined by dura and arachnoid.

39.3.2 Encephalocoele Although it is controversial whether this is truly a NTD defect, it is probably best considered as such for the purposes of this discussion (Fig.  39.1). It results from bony defects in the cranial vault, which lead to a herniation of the meninges, with or without differentiated brain tissue. The condition is more rightly called a meningoencephalocoele if dura and brain tissue herniate or a meningohydroencephalocoele if a part of the ventricular system also herniates.

39.3.5 Myelomeningocoele

39.3.3 Spina Bifida Occulta

39.4 Aetiology

At its mildest extreme, the vertebral arches of a single vertebra fail to fuse, but there is no underlying abnormality of the neural tube. The defect may occur anywhere along the spine but is most commonly found in the lumbosacral region, and its presence may only be signalled by the presence of a small tuft of hair, small dimple, pigmented skin or vascular lesion overlying the lesion.

The aetiology of NTD is clearly multifactorial, and no single agent, either genetic or teratogenic, has been identified. However, there is evidence of a genetic influence in some cases, e.g. SB is more common in some parts of India and Ireland (1.1%) and is relatively rare in African Americans (0.035%). In addition, the presence of NTD in one sibling increases the risk among subsequent siblings to 1 in 20. In families with two NTD siblings, the risk increases to 1 in 8. In certain syndromes, e.g. Meckel syndrome, an autosomal recessive disorder, craniorachischisis may be seen. NTD may also be seen in the Waardenburg syndrome, which may result from Pax-3 gene abnormalities.

39.3.4 Meningocoele Abnormalities of the neural arch without underlying neural tube defects with the formation of a

In this, the severest form of spina bifida, the tissues overlying one or more vertebra are deficient so that the neural tube tissue itself protrudes as a neural plaque to the surface (Fig.  39.2). This plaque may be completely or partially covered by arachnoid or exposed. It most typically involves the lumbar region, but any portion of the spine can be affected.

M. T. Corbally

490

It also seems likely that environmental factors are important in the development of NTD. There has been a significant decline in the incidence of NTDS over the past four decades. Factors noted have been the association of maternal diabetes, the anti-epileptic drug sodium valproate and hyperthermia. Valproate may interfere with folate metabolism, and there is also evidence that a significant number of infants with NTD may have gene mutations that are involved in folate and vitamin B metabolism, especially mutations of 5,10-methylenetetrahydrofolate reductase and methionine synthase reductase. The administration of peri-conceptual folic acid has probably been the single most important factor in the acknowledged decline in the incidence of NTD. Despite increased awareness of the benefit of peri-conceptual folic acid and its widespread use, there remains a significant incidence of the problem. However, in women with a history of folic acid intake and a SB child, it appears that the severity of the lesion is much reduced.

39.4.1 Incidence The worldwide incidence of NTD is reported to be as high as 400,000 per annum. However, the use of peri-conceptual folic acid has reduced the incidence by 70% in the past 20 years and has also reduced the severity of the lesion. In Ireland, the rate fell from 32 in 1979 to 22 per 10,000 in 1982 and continues to fall.

39.5 Diagnosis 39.5.1 Antenatal It is more preferable to make the diagnosis antenatally, which allows for counselling and facilitates transfer to a paediatric surgical centre for appropriate management. In addition, if foetal intervention is available and appropriate, then this can be considered (vide infra). Antenatal diagnosis may be made by careful ultrasound examination, chorionic villous sampling, maternal alpha-fetoprotein (AFP) or amniocentesis.

Varying sensitivities have been reported, but in experienced hands, ultrasound is a sensitive technique to detect a NTD. If antenatal ultrasound is suspicious of a NTD, then maternal AFP combined with amniocentesis for AFP and acetylcholinesterase assay is confirmatory. With improved prenatal care, NTDs are commonly detected before birth, and arrangements can then be made for counselling and for the delivery and care of the infant. Delivery should be scheduled close to a surgical centre and consideration given to delivery by Caesarean section, which may confer a significant functional benefit to the child. Rarely, hydrocephalus may need antenatal drainage to facilitate delivery.

39.6 Clinical Features The diagnosis of a NTD is usually straightforward at birth if the lesion is a myelomeningocoele, meningocoele or encephalocoele. Lesser lesions may not be clinically obvious and require more detailed investigations.

39.6.1 Myelomeningocoele A myelomeningocoele presents as a large open lesion anywhere along the spinal column, although the lumbar and sacral areas are the more frequently involved. It is unfortunately the commonest form of NTD.  Typically, there is a thin membrane covering the exposed neural plaque, which may be intact and appear cystic. Usually, however, the neural plaque is open to the environment, and CSF leaks constantly. If there is little or no CSF leak, the lesion will be raised. There may be occasional hamartomatous lesions associated with it is, such as a haemangioma, a lipoma or a naevus (Fig. 39.3). There may be associated deformities of the lower limbs with hip dislocation or subluxation, hypoplastic lower limbs, genu recurvatum and talipes (Fig. 39.4). In addition, there may be obvious hydrocephalus although this is uncommon at birth. The vertebral anomalies can be significant with severe kyphosis evident at birth. Neurological deficits include

39  Spina Bifida and Encephalocoele

Fig. 39.3  MRI showing lipomyelomeningocoele

491

tion to the external sphincter (puborectalis and pelvic floor muscles) is likely to be lost, which results in a patulous anus. This may result in rectal prolapse and will not have any immediate impact on bowel emptying, but clearly may affect faecal continence later. At least 90% of patients with myelomeningocoele have a neuropathic bladder with disturbances of detrusor and sphincter muscle activity. This is manifest at birth by constant dribbling of urine, but some do have an intermittent urinary stream. The management of the child with a NTD and a neuropathic bladder is quite complex and should involve the surgeon and paediatric continence nurse specialist. Frequent monitoring by renal ultrasound should be performed to allow early detection and intervention in the presence of upper tract dilatation, which occurs as a result of poor bladder compliance. Early introduction of clean intermittent catheterisation (CIC) may be required in some infants.

39.6.2 Meningocoele

Fig. 39.4  Bilateral talipes deformities of the feet

motor and sensory loss to the lower limbs. The effects of neural involvement include paralysis of lower limb muscle groups often with p­ reservation of nerve supply to antagonistic groups, which results in more severe deformity. Occasionally, there will be complete loss of innervation as in a flaccid paralysis, while often, there will be an upper motor neuron lesion and a resultant spastic paresis. While the internal anal sphincter is preserved due to its autonomic nerve supply, the innerva-

A meningocoele is uncommon and presents as a sac and skin covered defect in the lower spinal column with no abnormality of the underlying neural tube. There is usually no neurological defect, and the cord is normal. In addition, the lower limbs are normal. Rarely, some neural fibres may be adherent or contained within the sac, and these require careful dissection from the sac during closure. This may be apparent on ultrasonic exam or MRI pre-operatively.

39.6.3 Spina Bifida Occulta Spina bifida occulta may be difficult to detect on clinical grounds and may only become apparent at a later stage during incidental spinal imaging. However, the presence of a tuft of hair, pigmented naevus or vascular malformation in the midline along the spinal column may indicate the underlying vertebral anomaly of occulta. While the spinal cord is normal and there may be no overt evidence of neurological impairment, it is impor-

492

tant to be aware that functional disorders of the urinary tract may be related to an underlying occulta and would warrant a search for an ­underlying lesion. Occasionally, there may be a sinus-­like tract connecting to the meninges and lying over the spine itself. This is an occasional cause of spinal sepsis and meningitis. It should not be confused with the quite common sacral or posterior anal dimple, which is a skin dimple attached to a normal coccyx and not related to NTD.  It should be remembered that the defect usually includes a vertebral arch abnormality and normal cord and meninges but rarely, the cord may be tethered and may be the cause of gait abnormalities or subsequent bladder abnormalities. Neurosurgical assessment is needed.

39.6.4 Encephalocoele Encephalocoele is a midline defect in the bones of the skull, which allows protrusion of meninges only or gross herniation of brain tissue. In the latter condition, there may also be an associated microcephaly or other macrostructural cerebral anomalies. Often, these include Dandy-Walker cyst formation, hydrocephalus and dysplasia of the cerebellum and optic pathways. The usual bony site is the occiput, but frontal encephalocoeles are more commonly seen in Asia. There may be other congenital lesions, such as NTD at other sites, cleft palate and cardiac, lung and renal anomalies.

39.7 Management 39.7.1 Myelomeningocoele Surgical closure of myelomeningocoele was not regularly attempted until the early twentieth century, when survival of 23% was reported. The advent of asepsis and antibiotics improved survival, and surgical closure became more widely practised. Patients with extensive paralysis and hydrocephalus were not offered closure, but the back lesion was allowed to slowly granulate and epithelialise. However, death from infection and

M. T. Corbally

uncontrolled hydrocephalus was common. Improved surgical and anaesthetic techniques and antibiotics, and the development of reliable valve-regulated shunts, led to more aggressive management of patients even with severe lesions. However, many survivors were noted to have a poor quality of life with mental impairment, shunt and renal problems that made their management difficult and tended to overwhelm existing medical resources. A review of the selection process generated a return to a conservative approach in patients with extensive paralysis, severe hydrocephalus, kyphosis and major associated anomalies, in the firm belief that the severity of the lesion was not compatible with an acceptable quality of life. Reports of unselected treatment for all patients with myelomeningocoele suggested that early mortality, the frequency of mental impairment, poor mobility, pressure sores, incontinence and other issues dictated a selected approach for all patients. However, the continued and increased survival of patients initially regarded as being of poor potential outcome indicated that survival could not always be based on the clinical appearance of the lesion or extent of associated problems alone. Moreover, children surviving this initial conservative approach often suffered greater disabilities as a result of a non-operative attitude. A decision to withhold treatment cannot therefore be supported on clinical or ethical grounds alone. It must be noted, however, that this approach is not universally accepted and that parental wishes must also be considered. Nevertheless, the current standard of practice for children with myelomeningocoele is that the defect should be closed within the first 24–48  h of life, to place a ventriculo-­peritoneal (VP) shunt if hydrocephalus is present and to monitor and treat aggressively their problems for the rest of their life. Patients with myelomeningocoele should be transferred to a paediatric surgical centre and be prepared for early closure of the defect. It is important to protect the defect from contamination with faecal matter so chlorhexidine-soaked gauze is applied to the lesion, and this should be changed frequently. Broad-spectrum antibiotics are usually given, and the baby is allowed to feed

39  Spina Bifida and Encephalocoele

on demand. Upon arrival at the surgical centre, the baby undergoes a variety of investigations, such as a cranial ultrasound, spinal X-ray (include back, pelvis and skull), muscle charting, thorough examination to rule out other problems, orthopaedic assessment and a social work consult. An MRI scan may also be performed at this stage to document the presence or absence of other spinal lesions, although this can be performed at a later date and a spinal ultrasound is probably just as sensitive in the first 6 weeks of life. The surgeon should meet with both parents and discuss the management plan in detail and, in particular, the likely problems that may occur in the future. These include the possible need to treat hydrocephalus with a VP shunt; the possibility of shunt malfunction and its consequences; the likelihood of a neuropathic bladder and its significance; the possibility of orthopaedic treatment for talipes, dislocated hip, etc.; and the issues of continence and intellect. The procedure is then scheduled for the next available time but should generally be within 24–48 h. Although most patients with myelomeningocoele undergo surgical closure after birth, considerable effort has been focused on in utero repair in selected patients. This presents an alternative in their management and may carry significant advantages to the infant in terms of neurological outcome. The proposed mechanism of improved outcome with foetal intervention is to lessen the hindbrain herniation associated with Arnold-­ Chiari malformation and so reduce the frequency of significant hydrocephalus and shunt procedures. The results of randomised trials, such as the MOMS (Management of Myelomeningocoele Study), which offered foetal intervention between 19 and 25 weeks of gestation, have shown that foetal intervention is best performed at 25 weeks of gestation to minimise preterm delivery, chorioamniotic membrane separation and premature labour. While not freely available at every paediatric surgical centre, foetal surgical intervention has some benefits over postnatal surgery as in: 1. Less significant hydrocephalus, as measured by a significant incidence of shunt placement at 12 months of age

493

2. Great chance of independent walking at 30 months 3. Less risk of any degree of hindbrain herniation at 12 months However, it also carries a greater risk to the mother, as in an increased risk of uterine dehiscence and placental abruption. Significant prematurity carries risks to the foetus/baby, as in respiratory distress syndrome. A mini-­hysterotomy may offset some of these adverse outcomes.

39.7.2 Operative Approach The procedure is carried out under general anaesthesia, with the patient prone and in a warm ambient temperature (Fig.  39.5). A small roll may be placed beneath the hips and the lesion and surrounding skin prepped with an aqueous solution. It may be useful to cover the natal cleft with a non-porous tape to exclude the area from the sterile field. Operating loupes are useful during all parts of the procedure. It is wise to plan the orientation of the incision before commencing the procedure as this may impact on the ease of the closure especially with large lesions. The skin edge is incised just at its junction with the lesion and the membranes close to the plaque carefully dissected from the plaque. The plaque should be separated from all epithelial elements so as to prevent a theoretical epithelioid inclusion at a later date. When the plaque has been freed from all local attachments, the dura is incised on its lateral surface in an elliptical manner around the neural plaque. This involves incising down to the underlying fascia and subsequently mobilising the dura so as to allow closure of the dura over the plaque. Closure is effected by a running 6/0 or 7/0 suture throughout the length of the dural sac. If possible, the lumbar fascia can be mobilised to cover the dural repair although this is not essential. A small Redivac drain is left in situ and the skin closed over the repaired defect. The skin is closed using a series of interrupted nylon sutures with alternating Steri-Strips. If the skin closure seems a prob-

M. T. Corbally

494

a

c

b

d

e

Fig. 39.5 (a) Position of the patient on the operating table and an elliptical incision at the junction of the membrane and the skin, (b) membrane being eplised to free the

neural plaque, (c) plaque lying in the dural layer, (d) dura is closed with a continuous suture, (e) skin is closed with interrupted sutures

lem, the skin may be undermined, or very rarely, Z flaps are used. It is normal practice to cover the repair with a semi-permeable dressing. Post-­ operatively, the patient is nursed prone or in a lateral position, the drainage is monitored, and antibiotics are continued until drain removal. If there is excessive CSF leakage, it is generally not due to a problem with dural closure but to increasing hydrocephalus, and a ventriculo-peritoneal shunt is indicated.

appropriate investigations. There is little risk of hydrocephalus although there may be neural elements adherent within the sac and an ultrasound and/or MRI scan should be obtained pre-­ operatively. The skin edge is incised in an elliptical fashion around the defect and the protruding dura exposed. The dural sac is opened vertically on its lateral aspect taking care to avoid any adherent neural tissue (rare) and the sac then repaired, removing the herniating portion. The skin is closed over a drain.

39.7.3 Meningocoele Unlike myelomeningocoele, there is usually no urgency to close a meningocoele. The surgical procedure can be scheduled electively after

39.7.4 Encephalocoele If the defect is small and contains little or no brain tissue, then closure is within the experience

39  Spina Bifida and Encephalocoele

of a paediatric surgeon; however, if there is significant brain tissue and if there is associated microcephaly, then a non-operative approach may be indicated. Although there is generally little urgency about closure of an encephalocoele, there is the potential risk of further herniation, which may compromise the child and make closure more difficult. In general, these should be closed as soon as possible after appropriate imaging (MRI) is performed. Surgery is performed with the child prone and intubated. In the case of occipital lesions, a transverse incision is made over the apex of the lesion. The dural sac is exposed and opened away from the bone edges. Brain tissue should be preserved unless necrotic or grossly dysmorphic or likely to interfere with dural closure. The dura is closed with a continuous suture and a small drain left in situ. An acute rise in intracranial pressure may require an urgent VP shunt following repair. Anterior encephalocoeles and meningocoeles are complex and may require the input of other specialist services, such as neurosurgery and/or otolaryngology.

495

the cerebellum and medulla through the foramen magnum, elongation of the aqueduct of Sylvius making it liable to blockage and various bony defects of the upper cervical vertebra and occiput.

39.7.6 Clinical Features

The most obvious is that of a symmetrically enlarged head, either at birth or developing over the next few weeks. An ultrasound examination will easily show dilated ventricles, and serial measurement of the head circumference will show increasing deviation over the standard measurements. The anterior fontanelle is wide and bulging, and the sutures will appear separated. Consideration should be given to ventriculo-­ peritoneal shunt insertion when there is a rapid increase in head circumference or when there is clear evidence of significant hydrocephalus on ultrasound, CT or MRI. Newborn infants with a NTD and hydrocephalus tend not to have many symptoms of increased intracranial pressure as the open fontanelles and sutures can accommodate to some extent. However, internal strabismus and setting-sun sign due to pressure on soft 39.7.5 Hydrocephalus orbital plates are seen with significant and untreated hydrocephalus. In addition, optic nerve Hydrocephalus results from an imbalance in the damage or occipital lobe damage may result in production and absorption of cerebrospinal fluid. visual deficits if the hydrocephalus is not treated. Obstruction to the flow of CSF out of the venA description of the technique of VP shunt tricular system by the Arnold-Chiari malforma- insertion is beyond the scope of this chapter. tion, tumour, aqueductal stenosis, haemorrhage However, shunt valves are selected on their openor obstruction of the fourth ventricle (Dandy-­ ing pressure, which is the pressure that the valve Walker cyst) causes a non-communicating hydro- will open to allow CSF leave the ventricular syscephalus and is the most common type seen. Free tem and enter the peritoneal cavity. Since it is flow of CSF due to lesions of the choroid plexus possible to overdrain and cause a slit ventricle or following inflammatory conditions causes syndrome, it is probably best not to use low prescommunicating hydrocephalus. In the newborn, sure systems except in the very small, preterm increasing pressure within the ventricles and cra- infant. The author’s preference is to use a unitised nial vault is somewhat compensated by the open single medium pressure system in the majority of fontanelles. Most patients with myelomeningo- cases. coele have hydrocephalus, and approximately Shunts are mechanical devices and are subject 90% of these will ultimately require insertion of to problems such as blockage, breakage, a VP shunt to control it. In this group, hydroceph- ­malfunction and infection. Many or all of these alus is associated with the Arnold-Chiari malfor- problems can be found in the life of a single mation, which includes caudal displacement of shunt.

M. T. Corbally

496

39.8 Long-Term Management Most patients survive the trauma of back closure and surgical treatment of hydrocephalus. However, as many as 23% may die within 1 year of birth. Surviving patients face ongoing problems in relation to mobility, shunt issues, problems with faecal and urinary continence, problems specific to their neuropathic bladder and educational, intellectual and social issues. As many as 75% have normal intelligence, but a significant number require special educational support. Many of these problems can and perhaps should be addressed at a special clinic to cope with the needs of this particular group. Individual care plans can readily be worked out for each patient, and this can be modified on a fluid basis at the SB clinic. It can often be difficult to have all involved specialists attend such a clinic, but the benefits to the parents and child are significant. Increasingly, local agencies provide comprehensive care to NTD patients and their families, such as Enable Ireland, and the impact of a variety of voluntary agencies, such as “The Association for Spina Bifida and Hydrocephalus”, cannot be overstated. Of particular importance is the ongoing surveillance of the urinary tract, especially the results of 6 monthly renal ultrasound exams. This is chiefly to detect the early development of upper urinary tract dilatation due to a non-compliant neuropathic bladder. When this occurs, the parents are instructed in the technique of clean intermittent catheterisation (CIC), which is performed every 4 h on average and serves to empty the bladder of urine and prevent reflux of static urine from high vesical pressures. Approximately 10–15% of patients will not have their high pressure bladder controlled by CIC, and a vesicostomy may be necessary. Older children and children with small volume high pressure bladders will need the expertise of a paediatric urologist to assess the need for bladder augmentation. The input of paediatric continence nurses is invaluable, especially in the performance of urodynamic studies and also in instructing older children in the often difficult task of overcoming manipulative skills to facilitate self CIC. Bowel problems are generally treated using medications such as stool softeners and/or regu-

lar stimulant enemas; however, social “continence” can be achieved in up to 85–90% of children using a regular washout enema containing 200–300 ml of water and a stimulant such as toilax. This generally gives clean results lasting up to 24 h. When this fails, consideration should be given to performing an antegrade colonic enema (ACE) procedure using the appendix as a catheterisable conduit. Ongoing issues of mobility and joint deformities need the continued input of orthopaedic surgeons and occupational physiotherapists. In addition, social workers play a significant role in helping the family adapt the home environment to cope with mobility and toileting issues and to secure proper state funding for their needs. The incidence of NTD continues to decline, but for those born with this condition, it can impose severe restrictions on the quality of their lives. The current standard of practice is that all patients with myelomeningocoele should be offered surgical repair within the first 24–48 h of life. Improvements in valve regulated ventriculo-­ peritoneal shunts have contributed greatly to quality of life. Long-term review in special multidisciplinary clinics facilitates review of renal function, status of urinary tracts, status of their valve-shunt and management of continence and education and social issues.

39.9 Conclusion The incidence of NTDs continues to decline. This may be due to the combined impact of peri-­ conceptual folic acid, antenatal diagnosis and termination of pregnancy. Early surgical closure is important to reduce infective complications. Foetal surgery between 19 and 25 weeks has a role and can improve overall outcomes and performance; however, it is limited to select centres and includes the risk of premature delivery with its attendant problems. Following the initial measures to close the back and secure adequate CSF drainage, the myriad of issues that can envelop a child with a NTD and the family are best dealt with at a multidisciplinary clinic. This is especially important to provide a one-stop assessment

39  Spina Bifida and Encephalocoele

of shunt performance, mobility and continence issues. Early back closure is important to prevent complications such as infection. Monitoring for hydrocephalus is important in order to insert a shunt at an appropriate time and thus preserve function.

Selected References Corbally MT (2006) Spina bifida. In: Puri P, Höllwarth ME (eds) Pediatric surgery. Springer surgery atlas series. Springer, Berlin, pp 419–426 Martin T. Corbally (2009) in Pediatric Surgery, Eds Puri and Hollwarth. Chapter 79. 765–774, 2009. Pubs Springer Dias MS (2005) Neurosurgical management of myelomeningocoele. Pediatric Rev 26(2):50–60

497 Finnell RH, Gould A, Spiegelstein O (2003) Pathobiology and genetics of neural tube defect. Epilepsia 44(Suppl 3):14–23 Hunt GM (1990) Open spina bifida: Outcome for a complete cohort treated unselectively and followed into adulthood. Dev Med Child Neurol 32:108–118 Larsen WJ (2001) In: Sherman LS, Potter SS, Scott WJ (eds) Human embryology, 3rd edn. Churchill Livingstone, New York Lorber J (1971) Results of treatment of myelomeningocoele: An analysis of 524 unselected cases, with special reference to possible selection for treatment. Dev Med Child Neurol 13:279–303 Mitchell LE, Azdick NS (2004) Spina bifida. Lancet 364:1885–1895 Sacco A, Ushakov F, Thompson D et  al (2019) Fetal surgery for open spina bifida. Obstet Gynaecol 21:271–282 Walsh DS, Adzick NS (2003) Foetal surgery for spina bifida. Semin Neonatol 8:197–205

40

Hydrocephalus Geraint Sunderland, Jonathan Ellenbogen, and Conor Mallucci

40.1 Introduction The central dogma of hydrocephalus is a simple one: cerebrospinal fluid (CSF) is constantly produced and reabsorbed. Disturbances in CSF flow and absorption lead to an accumulation within the ventricular system. Untreated, this leads to rising intracranial pressure (ICP) and active distension of ventricles in the majority of cases (Rekate 2011). The management of hydrocephalus is far from simple, however, as causes are numerous and heterogeneous. There are many nuances, and expert management is essential. Patients will often have multi-system disorders and complex needs, and thus there are some fundamental principles all clinicians involved in their care should know. Hydrocephalus is really a catch-all term encompassing many disparate aetiologies, each with their own vagaries of treatment. Defining hydrocephalus is thus problematic; however, the vital concern is whether there is associated raised ICP.  Ventricular volume, while a useful adjunct especially in the radiologically surveillance of patients or in the acute setting, is secondary. Small ventricles ↑ normal intracranial pressure

G. Sunderland (*) · J. Ellenbogen · C. Mallucci Department of Neurosurgery, Alder Hey Children’s NHS Foundation Trust, Liverpool, UK e-mail: [email protected]; [email protected]

Radiological definitions, therefore, such as Evans ratio (ventriculomegaly=(maximal frontal horn diameter)/(transverse inner skull diameter) ≥ 0.3), the diameter (>2 mm) of the temporal horns, rounded third ventricle walls, etc. are crude guides and not universally applicable. The aim of this chapter is not to provide an exhaustive guide to hydrocephalus in all its guises, but rather to provide the non-specialist with a practical approach to the management of patients with hydrocephalus, both treated and untreated.

40.2 Historical Overview Hydrocephalus has been recognised as a pathological entity and studied since the time of Hippocrates in the fifth century BC. All the key figures of the history of medicine, including, but not limited to, Galen, Vesalius, Sylvius and Abulcasis, have at one point turned their enquiring minds to its study. Early writers sought to understand its origin, believing the soul to reside within the ventricles (Aschoff et al. 1999). Progress in understanding the pathology did not occur until the physiology of CSF production was explored, firstly by Thomas Willis in the seventeenth century who identified the choroid plexus as the organ of origin. The CSF circulation was defined between the nineteenth and into the twentieth century with two of the contributors

© Springer Nature Switzerland AG 2023 P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_40

499

500

immortalised in anatomical nomenclature: Magendie and Luschka. In the early twentieth century, the practice of neurosurgery as we know it today was born due to pioneers such as Harvey Cushing, Sir Victor Horsley and Walter Dandy who all had a hand in developing hydrocephalus surgery (Lifshutz and Johnson 2001). The first documented shunt surgery was performed in 1905, but the technique subsequently fell into disrepute due to high mortality. There was a resurgence in the 1950s, thanks to Frank Nulsen and Eugene Spitz ,who pioneered ventriculoatrial shunting and subsequently ventriculoperitoneal shunting (Rachel 1999). Spitz also developed the first working valve alongside hydraulics engineer, John Holter. Numerous subsequent valve designs have been developed including the Wade-Dahl-Till valve named after its designer’s hydraulics engineer, Stanley Wade, world-famous children’s author Roald Dahl and neurosurgeon Kenneth Till. The development of CSF shunting revolutionised the management of hydrocephalus, transforming it from a disabling and often fatal condition to an eminently treatable one.

40.3 Incidence The prevalence of hydrocephalus in paediatric populations is 88/100,000 globally (Dewan et al. 2018a). Hydrocephalus is less prevalent in younger adults (11/100,000), but peaks in the elderly (175/100,000) in the over 65s and is >400/100,000  in those over 80, due to the occurrence of normal pressure hydrocephalus (NPH). It is also associated with congenital malformations of the brain and spinal cord in a significant proportion of cases in infants. Timely and effective surgical treatment is essential, as complications related to delay in treatment can be life-threatening. Once treated, the majority of patients will require lifelong support and follow-­up, with up to 85% requiring further surgical intervention in their lifetime (Stone et al. 2013).

G. Sunderland et al.

40.4 Aetiopathogenesis Doctors love good taxonomy, and those interested in hydrocephalus are no different; thus, numerous classifications have been devised to varying success. Classically, hydrocephalus was either ‘obstructive’ or ‘communicating’, terms coined by one of the fathers of modern neurosurgery Walter Dandy in 1914. Subsequent grouping by aetiology into ‘congenital’ or ‘acquired’ causes has been used, but has little clinical relevance (Dandy and Blackfan 1914). The ‘obstruction’ in obstructive hydrocephalus relates to a mass lesion (tumour, swelling) or some physical impediment (pus, blood) to the CSF circulation pathway prior to the point of reabsorption. Communicating hydrocephalus occurs due to obstruction also, though the obstruction is more functional, existing at the point of reabsorption, the prime example of communicating hydrocephalus being blood and blood breakdown products occluding the fine filtration substrate of the arachnoid granulations in cases of intraventricular haemorrhage. The terms obstructive and communicating are increasingly outdated and, in fact, give a gross over-­ simplification of a hugely complex and still poorly understood physiology (Tomycz et  al. 2017). They do, however, provide a workable scheme around which we can make clinical decisions. The exception to this two-classification rule is where CSF production outstrips reabsorption. This is very rare, however, limited to choroid plexus tumours and in some circumstances of infection, such as ventriculitis, due to inflammation and increased ependymal blood flow. In practical terms, whether ‘obstructive’ or ‘communicating’, CSF accumulates within the cranial CSF compartment, and pressure rises. The normal CSF pressure is 10–15  cm/H2O, as measured in the lumbar theca in the lateral position (7–12 cm/H2O in infants). Unchecked elevations in ICP can lead to coma and eventually death. Hydrocephalic aetiology and age are inextricably linked resulting in the three peaks in inci-

40 Hydrocephalus

dence. The vast majority of neonates and infants will develop hydrocephalus either secondary to intraventricular haemorrhage (IVH) of prematurity or in association with congenital malformations, chief amongst which is spina bifida. The rest of the cases (at least in the developed world) are due to tumours, with a small number occurring following episodes of meningoencephalitis (the second leading cause globally) (Dewan et al. 2018b). During adulthood and middle age, the incidence is low and typically limited to cases of subarachnoid haemorrhage and as a consequence of obstructive tumours. With advancing years, the neurodegenerative condition, idiopathic normal pressure hydrocephalus (NPH), results in a third and final peak of hydrocephalus cases, summarised in Table 40.1. This division into different age groups and how this impacts on management neatly demonstrates the importance of understanding the pathoanatomical basis of a patient’s hydrocephalus. Careful consideration and understanding of this is crucial to making rational management decisions. Table 40.1  Common aetiologies causing hydrocephalus by age group Age group Neonates/ infants

Children

Adults (65)

Common aetiologies • Post-haemorrhagic (IVH of prematurity) • Post-infective (meningitis/ventriculitis) • Spina bifida and related neural tube defects • Other congenital malformations (craniofacial syndromes, arachnoid cysts, Dandy-Walker malformation, X-linked aqueductal stenosis, etc.) • Tumours (benign/malignant) • Post-haemorrhagic (vascular malformations) • Aqueduct stenosis • Post-infective (meningitis/ventriculitis) • Idiopathic intracranial hypertension • Post-haemorrhagic (aneurysmal subarachnoid haemorrhage) • Tumours (benign/malignant) • Idiopathic intracranial hypertension • Trauma • Idiopathic normal pressure hydrocephalus • Post-haemorrhagic (aneurysmal subarachnoid haemorrhage)

501

40.5 Pathophysiology Understanding the pathophysiology of hydrocephalus requires some understanding of the physiology of CSF production and circulation. Approximately 80% of CSF is produced by the choroid plexus; tufts of capillaries enclosed in modified ependyma (membranous lining of the ventricles). This active process is supplemented (20%) by passive egress of fluid from the brain substance into the interstitial space, perivascular ‘Virchow-Robin’ spaces and onwards then to the body of CSF within the ventricular system. CSF is produced at around 0.33 ml/min, resulting in a total of 20 ml/h and about 500 ml/day in an adult (Kimelberg 2004). At any point, there is about 150 ml of CSF within the neuroaxis, half of which is in the intracranial compartment. Volumes are obviously less in infants, proportional to body weight, but approximate adult values are reached by 5 years of age. Choroid plexus is found predominantly within the lateral ventricles but also within the third and fourth ventricles. From the point of secretion, CSF normally circulates from the lateral ventricles through the foramina of Monro into the third ventricle and then onwards through the aqueduct of Sylvius into the fourth ventricle in the posterior fossa. From the fourth ventricle, the CSF exits into the spinal subarachnoid space either via the midline foramen of Magendie or via the lateral foramina of Luschka into the basal cisterns and around the cerebral cortex in the subarachnoid space (Sakka et al. 2011). CSF is then reabsorbed into the cerebral venous system, via outpouchings of arachnoid mater known as arachnoid granulations. The arachnoid granulations bulge into the dural venous sinuses, where CSF is resorbed depending on a hydrostatic pressure differential (Khasawneh et al. 2018).

40.6 Pathology One of the key factors to consider, certainly when timing surgical intervention in young children, is whether the skull plates are fused at the cranial

502

sutures. A child with a closed fontanelle and fused sutures will not be able to accommodate for a relative increase in CSF volume without incurring dangerous elevations in ICP, for example, as their ICP dynamics conform to the Monro-Kellie doctrine (Fig. 40.1). Conversely, a premature neonate may accommodate, usually with a rapidly increasing head circumference. This can be temporised relatively safely with episodic ventricular taps.

Fig. 40.1  Monro-Kellie doctrine

G. Sunderland et al.

The cranial vault is a fixed volume (approx. 1700 ml). The contents are the brain (1400 ml), CSF (150 ml) and blood (150 ml). An increase in one of these components requires displacement, and a reduction in one or both of the other two components or the intracranial pressure will rise.

40 Hydrocephalus

40.7 Diagnosis 40.7.1 Clinical Presentation Clinically speaking, the vast majority of hydrocephalus will present one of five ways: 1. On serial imaging of the premature neonate at risk for germinal matrix haemorrhage and intraventricular haemorrhage 2. Following investigations for expanding head circumference—usually instigated in the community by a GP or health visitor 3. Following imaging and investigations performed for symptoms of raised ICP (Table 40.2) Table 40.2  Symptoms and signs of raised ICP by age group Age group Neonates and infants

Young children

Symptoms and signs • Irritability • Floppy, peripherally cool • Reduced responsiveness • Vomiting • Posturing (arching back) • Sunsetting eyes (combination of upper eyelid retraction and failure of upgaze) • Distended scalp veins • Bulging/tense anterior fontanelle • Splaying of suture lines • Crossing centiles on head circumference chart • Desaturation/bradycardic episodes • Headache • Clutching head • Vomiting • Non-specific behavioural change • Change in appetite/anorexia • Collapse (hydrocephalic attack— associated with transient increases in ICP: coughing, sneezing, etc.) • Sunsetting eyes • Crossing centile lines on head circumference chart • Head circumference out of proportion with body weight/ height • Visual disturbance (blurring/visual field defect) • Papilloedema

503 Table 40.2 (continued) Age group Symptoms and signs Older children • Headache and adolescents • Vomiting • Collapse (hydrocephalic attack— associated with transient increases in ICP: coughing, sneezing, etc.) • Visual disturbance (blurring/visual field defect) • Papilloedema

4. Subsequent to routine fundoscopic examination by the child’s optometrist and identification of papilloedema 5. Incidentally discovered when investigated for an unrelated cause, e.g. trauma Radiological imaging forms the foundation of neurosurgical diagnosis. Cranial imaging is essential to the diagnosis of hydrocephalus, typically based on the presence of ventriculomegaly. It bears repeating, however, that normal-sized ventricles do not necessarily mean there is no problem with CSF flow and/or raised intracranial pressure (Dinçer and Özek 2011).

40.7.2 Plain Radiography Plain radiographs of the skull do not have a role in the routine workup of hydrocephalus, although they can give clues to the underlying pathology. Signs of raised pressure include a ‘copper beaten’ or ‘thumbprinted’ appearance of the calvarium and splaying of the cranial sutures (Fig. 40.2). In addition, the aetiology of the raised pressure may be noted, e.g. traumatic fractures or abnormal calcification associated with a pineal region mass obstructing the aqueduct. Plain radiographs are, however, vital in the longer-term management of the hydrocephalus patient. Shunt series XR (skull, neck soft tissues, chest and/or abdomen) allow clinicians to follow the tract of the shunt (radio-opaque) to ensure all components are in continuity and exclude a shunt ‘fracture’ or disconnection, which could compromise shunt func-

504

G. Sunderland et al.

maturity, will be echo-bright and should be obvious to all (Fig.  40.4). Skilled and experienced operators can identify other underlying aetiologies, such as tumours or anatomical abnormalities; however, ultrasound is not ideally suited to this, and confirmatory cross-sectional imaging would be required (Dorner et al. 2018).

40.7.4 Computed Tomography (CT)

Fig. 40.2  Plain radiograph of infant skull demonstrating the ‘copper beaten’ or ‘thumbprinted’ appearance of the calvarium due to moulding of the inner table over the cerebral cortex reflecting high ICP.  Note the prominent coronal suture. Also note the discontinuity in the shunt (arrow) where the distal catheter has disconnected from the valve

tion (Fig. 40.3). Furthermore, a skull XR permits accurate evaluation of the setting in programmable shunt valves (see Sect. 40.9.1).

40.7.3 Ultrasonography Transcranial ultrasound is a very useful screening tool in neonates. It requires an open anterior fontanelle, which provides a window to assess intracranial structures. With minimal training and experience, operators (typically neonatologists) can achieve a diagnosis of intracerebral/ intraventricular haemorrhage, with or without ventriculomegaly. It is vital to record the ventricular index (VI) to communicate effectively with colleagues. VI is defined as the width (in mm) from the falx to the lateral wall of the lateral ventricle at the level of the foramen of Monro, when viewed in the coronal plane (Fig. 40.4). As there are both left and right ventricles, a value is given for each as these are frequently subtly different. Blood within the ventricles, for example, in cases of IVH of pre-

Computed tomography (CT) is a ubiquitous, rapid and easy-to-interpret tool in the diagnosis of hydrocephalus. A plain unenhanced CT will successfully diagnose hydrocephalus and the underlying aetiology in the vast majority of cases. CT diagnosis of hydrocephalus is based primarily on the presence or absence of ventriculomegaly (Fig. 40.5). Features suggestive of ventriculomegaly include ballooning of the frontal horns of the lateral ventricles, the presence of dilated temporal horns of the lateral ventricles (not typically visible in health) and periventricular low density suggesting transependymal passage of CSF under pressure. Other features include enlarged and rounded third ventricle, effacement of the cortical sulcal pattern and surface subarachnoid spaces especially at the vertex of the skull and Sylvian fissure and upward bowing of the corpus callosum. Radiological criteria quoted in the literature, such as Evan’s ratio, are not routinely used in current clinical practice, and imaging must be interpreted in the clinical context (history and examination) and with reference to previous imaging (Dinçer and Özek 2011). CT is also frequently used to assess patients after insertion or revision of ventricular catheters for CSF diversion. The scan is useful to confirm satisfactory positioning as well as the state and size of the ventricles and can also reassure surgeons of the absence of complications, e.g. post-­ operative haematoma or IVH, in the event of difficult shunt revision surgery. Many neurosurgeons will perform a baseline or ‘well’ CT several weeks after recovery from shunt insertion or revision to capture the ventricular configuration at baseline. This can be used as a comparator

40 Hydrocephalus

505

a

b

c

d

Fig. 40.3 (a–d) Plain radiograph ‘shunt series XR’ demonstrating the course of a right parietal VP shunt transiting from the cranium to abdomen. Note the continuity of the

tubing through its course, the programmable valve evident on the cranial XR and the distal tubing coiled in the abdomen

G. Sunderland et al.

506

should the patient present with suspicion of shunt dysfunction in the future (Pople 2002). Despite the many advantages, CT is unfortunately less good at defining underlying causes

than is magnetic resonance imaging (MRI). In addition, it exposes the child to radiation, the cumulative doses of which can be significant across a lifetime of treatment.

40.7.5 Magnetic Resonance Imaging (MRI) VI

Fig. 40.4  Cranial ultrasound taken in the coronal plane via the anterior fontanelle in a premature neonate. Note intraventricular haemorrhage (yellow arrow), dilated ventricles (red arrows) and ventricular index (VI) demonstrated

MRI is the investigation of choice for delineating the ventricular size and the pathoanatomical substrate of hydrocephalus. It has many advantages over CT, not least of which is the avoidance ionising radiation exposure. As with CT scanning, multiplanar imaging allows the assessment of anatomical structures but with greater tissue determination. Fine membranous structures, such as the floor of the third ventricle, can be identified; useful as this is typically bowed down and backwards into the pre-pontine space, particularly in cases of obstructive hydrocephalus where the obstruction is at the level of the aqueduct of Sylvius or fourth ventricle. Modern scanning technology permits high-resolution and accurate

Fig. 40.5 Axial unenhanced CT demonstrating gross acute hydrocephalus. Note ballooning of the frontal horns of the lateral ventricles, dilated rounded third ventricle

and prominent occipital horns. Note the periventricular lucency around the frontal horns (white arrows), which represents transependymal flow of CSF under pressure

40 Hydrocephalus

507

a

c

b

d

Fig. 40.6 (a) Axial T2-weighted MRI demonstrating multi-loculated hydrocephalus due to multiple septations (red arrows). (b) Axial T2 FLAIR demonstrating hydrocephalus with periventricular oedema due to transependymal flow. (c) Coronal T2-weighted MRI showing marked

chronic hydrocephalus with expanded temporal horns (yellow arrows). (d) Sagittal T2-weighted MRI.  Note expanded third ventricle with thinned corpus callosum and downward bowing of the third ventricle floor (blue arrows)

image representation of CSF spaces and the ability to perform assessments of CSF flow. This can allow the identification of septations or cysts within the ventricles, common following intraventricular haemorrhage and infection. These would need to be traversed and fenestrated in order to successfully treat that patient’s hydro-

cephalus (Fig.  40.6). Pathological lesions, such as tumours, vascular malformations (e.g. vein of Galen malformation), cysts and even anatomical variants, are identified with an extremely high degree of accuracy by MRI; thus, for the assessment of de novo hydrocephalus, MRI is mandated.

508

Like CT, MRI has a role in the post-operative imaging of patients with hydrocephalus; it is particularly useful in the assessment of internal CSF diversion procedures, such as endoscopic third ventriculostomy, where particular imaging ­paradigms (e.g. time-resolved 2D phase contrast) can be applied directly to confirm and even quantify CSF flow, for example, across a newly formed stoma or through a pre-existing CSF pathway, e.g. across the craniocervical junction, in cases of Chiari malformation (Fig.  40.13) (Dinçer et al. 2011).

40.8 Differential Diagnosis The differential diagnosis for a child presenting with symptoms and signs of raised intracranial pressure is wide. Hydrocephalus is rarely the primary disease entity but rather a consequence of other underlying pathology. Accurate diagnosis obtained from imaging, blood tests, clinical history and examination will identify an underlying cause in the vast majority of cases.

40.9 Management Hydrocephalus is an eminently manageable condition surgically speaking, albeit this may be complicated in a small proportion of cases. The treatment involves CSF diversion or ‘shunting’ from the intraventricular compartment of the brain to another compartment either intra- or extracranial. These shunts may be physically implanted or internally formed by creating an ostomy between cavities. The absolute indication for the treatment of hydrocephalus is the presence of signs and symptoms of raised intracranial pressure (Table 40.2). Caution should be applied and consideration of alternate management strategies made where available, especially in the presence of active/proven CNS/CSF infection, low body weight or other remediable causes. The mode of treatment is dependent on a multitude of factors outlined below. Lesional hydrocephalus, i.e. secondary to a focal mass lesion causing CSF pathway obstruc-

G. Sunderland et al.

tion, is usually best treated by the removal of the causative lesion. Examples of these might include an enlarging colloid cyst obstructing both foramina of Monro or a tumour within the posterior fossa compromising CSF egress via the fourth ventricle. In the case of acute symptomatic hydrocephalus, secondary to an operable or potentially operable lesion, a short-term CSF diversion may be employed to gain control of the immediate situation allowing time for further assessment. Insertion of an external ventricular drain (EVD) into the frontal horn of the lateral ventricle (by convention the right side as this is non-dominant in the majority of individuals) allows control of raised ICP and more detailed investigations and definitive treatment. Occasionally, despite resection of the offending lesion, the hydrocephalus persists, and in these cases, diversion of CSF to an alternative site of reabsorption is required.

40.9.1 Implantable CSF Shunts Implantation of a CSF shunt is one of the most commonly performed procedures in neurosurgery worldwide. While one of the core procedures a neurosurgeon masters at the very beginning of their training and, thus, often disdained as ‘just a shunt’, one needs to be wary of creeping complacency. The consequences and potential long-term sequelae of a poorly thought-­ out or carelessly performed operation are serious and potentially dangerous. Complications of shunt surgery are detailed later in this chapter. CSF flow is most frequently diverted into the peritoneal cavity via a ventriculoperitoneal or ‘VP’ shunt (VPS). Alternatives include direct drainage into the central venous circulation, a ventriculoatrial or ‘VA’ shunt (VAS) and much less frequently the pleural cavity (ventriculopleural shunt). Shunts consist of a proximal (ventricular) catheter, a distal catheter and an interposed one-­ way valve. A valveless system may be implanted, though outside of the developing world these are limited to complex cases, usually following multiple failed revision surgeries. Shunt catheters are

40 Hydrocephalus

509

manufactured from hydrophobic silicone which is biologically inert, resists bacterial colonisation and retains its flexibility over decades, all crucial to longevity. Over the last two decades, manufacturers have devised numerous ways to improve the durability of shunts and avoid the two main complications, blockage and infection. There is level I (double-blind RCT) evidence to support the preference of antibiotic (rifampicin and clindamycin)-impregnated tubing to reduce the rate of infection-related failure and revision surgery (Mallucci et al. 2019). The major differences between shunt systems are in the valve, and there is a wealth of choice available on the market. It should be stated at the outset that all evidence to date, including one RCT published in 1998, has failed to categorically establish superiority of one valve design over another (Drake et  al. 1998; Kestle et  al. 2000). More research in this area is needed. The most frequent valves encountered are differential pressure valves (DPVs). These rely on a hydrostatic pressure gradient across the valve from inlet to outlet to drive valve opening and CSF flow. The alternative, flow regulating valves, are more rarely used. DPVs may have a fixed opening pressure (most common); manufacturers a

b

Fig. 40.7  Pictorial illustration of the process of shunt valve reprogramming using the MIETHKE proGAV 2.0 system, one of the numerous programmable shunts on the

supply valves with a variety of pressure settings or that alternatively have a programmable function whereby the opening pressure can be adjusted manually by the use of a transcutaneous handheld device (Fig.  40.7). It is important to realise that although valves may be of equivalent opening pressures, they may have different internal resistances and behave quite differently in vivo. Numerous other adjustments and technical features, including gravity-assisted valves and anti-syphon devices, are available, but their discussion alone could fill a book and is thus outside of the scope of this chapter. Selection of valves and shunt types is ultimately a matter of preference and tailoring to patient and clinician requirements. To the untrained observer, these choices may seem random but are often based on individual or departmental collective experience (sometimes painful). Programmable valves, for example, may be inserted to allow adjustment in demand as young children (especially neonates) grow or in the elderly with NPH to prevent CSF overdrainage and complications thereafter (Serarslan et  al. 2017). Gravity-assisted valves might be preferred again in younger patients to accommodate for rapid growth and change from predominantly c

market. (a) The setting is first confirmed. (b) A strong magnetic ring is used to reprogram the valve to the desired setting. (c) Post-reprogramming setting confirmation

510

recumbent to upright posture and the effect these have on the ‘syphoning’ effect (Schatlo et  al. 2013). Finally, pragmatic concerns are also valid; considerations such as the size of the valve and the risk of tissue breakdown over it (particularly in neonates), or favouring one programmable interface over another, are taken into account and weighted appropriately. For general surgeons who may come across intraperitoneal catheters, it is important to be aware that a significant proportion of resistance to CSF flow comes the distal catheter and is very much related to its length. Shortening a catheter in the peritoneum can have a marked impact on the function of shunt system so care should be taken to preserve shunt catheters when encountered.

40.9.2 Insertion of VP Shunt: The Technique After induction of general anaesthesia and endotracheal intubation, the patient receives prophylactic intravenous antibiotics and is positioned on the operating table. The positioning should allow access to the insertion site on the cranium (usually parieto-occipital or frontal) and the site of distal implantation (neck, chest or abdomen). The ideal insertion point permits a catheter trajectory which accesses a body of CSF (typically the lateral ventricle) but which traverses the least amount of brain, avoiding eloquent cortex and crucial deep brain structures (basal ganglia). The tip of the catheter bears multiple perforations, and these should be lying free in CSF away from the ventricle wall and choroid plexus to prevent ingrowth of tissue and thereby catheter blockage. In the circumstance of small or ‘slit’ ventricles, this may not be possible. A sub-optimally placed shunt may still work but will be more prone to failure. A number of ‘classical’ cranial insertion points have been described based on surface anatomy; however, these pre-date the CT scan and image guidance era and so are now used to approximate insertion only. The commonly used reference points include: In the parietal position: Keen’s point, three finger’s breadths above and behind the pinna and

G. Sunderland et al.

Frazier’s point, 6 cm cephalad from the inion and 3 cm from the midline. Frontally: Kocher’s point dictates that burr holes are placed in the midpupillary line just anterior to the coronal suture. As noted above, the standard of care now is to insert ventricular catheters using image guidance. There are a number of systems available, but broadly speaking, they utilise either real-time ultrasound guidance or cross-sectional (CT or MRI) data. Ultrasound guidance permits accurate localisation of the fluid-filled ventricles, and operators can first adjust to a suitable trajectory and then guide the echo-bright catheter to the desired target. Neuronavigation systems require volume cross-sectional imaging to be performed pre-operatively. This is then uploaded to the image guidance console where entry point, target and resulting trajectory can be planned. Avoidance of blood vessels and vital structures is thus ensured. These systems may rely on optical or electromagnetic (EM) technology. Optical systems use an infrared camera which ‘sees’ reflective markers attached to working instruments and a reference array which remains fixed relative to the patient’s head. EM technology uses an EM emitter, which creates a magnetic field around the target. The relative positions of a reference probe (attached to the patient) and working instruments are then computed. The results are equivocal; a virtual, real-time representation of your instrument’s position in 3D space relayed on a multiplanar (axial/coronal/sagittal or even 3D) reconstruction of the patient’s scan with sub-­ millimetric accuracy. Shunt insertion is best done by two surgeons, with the opening of the cranium and the peritoneal cavity done simultaneously to shorten operative time. The cranial opening should be done with a small burr hole and a minimal dural opening, just wide enough to admit the catheter to minimise CSF bypass. In neonates, a drill is often not required to open the soft calvarium. The abdominal exposure is done carefully in layers to avoid visceral injury. Leaflets of the posterior rectus sheath and peritoneum are reflected and held in haemostats to identify and preserve the

40 Hydrocephalus

opening. The catheter is then tunnelled in the subcutaneous layer with care paid to avoiding vascular structures in the neck and inadvertent entry into the thoracic compartment. The distal catheter is placed into the peritoneal compartment under direct vision ensuring free passage and the wounds are closed in layers with absorbable sutures. Post-operative radiological evaluation of a shunt’s placement may be performed with CT or MRI. There is unlikely to be significant improvement in the degree of ventriculomegaly at this stage. In circumstances of VP shunt revision, post-operative cross-sectional imaging can reassure there is no haemorrhage associated with the removal of the existing shunt—a common cause of early failure of revised shunts (Fig. 40.8).

Fig. 40.8  Post-operative axial CT scan in a patient with symptoms of early shunt failure following a ventricular catheter revision surgery. CT demonstrates intraventricular haemorrhage along the shunt trajectory most commonly due to bleeding from avulsed choroid plexus that was adherent to the revised catheter

511

40.9.3 Complications of CSF Shunts Shunt surgery is the mainstay of hydrocephalus management. It has been hugely successful since its adoption some seven decades ago. It has, however, been plagued by a number of complications some of which persist despite numerous iterations of technique and improvements in implanted technology. Between 11% and 25% of all patients will require shunt revision surgery within the first year of implantation and 85% during their lifetime, the most common and important reasons being mechanical failure (obstruction or overdrainage) and infection (Khan et  al. 2013; Wu et al. 2007; Reddy et al. 2014). Mechanical Dysfunction and Blockage Any shunt may become obstructed at any point along its length, and this is the most common type of shunt malfunction accounting for approximately three quarters of all failures (Kestle et al. 2000). The most common site is the proximal catheter, where it is often noted at the time of revision that fronds of choroid plexus have migrated into the perforations at the tip of the catheter (Paff et al. 2018). There are various theories as to why this may happen but none have been definitively proven (Harris and McAllister 2nd. 2012). Accurate placement with the distal tip floating free within the body of CSF and not in contact with ventricle walls is believed to reduce the risk of proximal obstruction (Hayhurst et al. 2010). For years, it was believed that frontally placed catheters were less prone to this complication; however, this has not been born out in the literature (Dickerman et al. 2005). The valve is the next most common site of shunt failure. Valve failure can result in either over- or more typically underdrainage. Implantation of a valve with too low a fixed pressure will result in low-pressure symptoms and require revision. This still qualifies as a shunt failure even if the system is working as designed. One study has demonstrated superiority in terms of survival for

512

G. Sunderland et al.

programmable shunts over non-­ programmable shunts when corrected for age and hydrocephalus aetiology; this has not been replicated in other studies, however (McGirt et al. 2007). A number of theories regarding valve occlusion and the reasons underlying it exist. In reality, a number of factors will contribute. In the acute phase, active bleeding within the ventricular system puts the valve at high risk of failure. The blood can coagulate under low flow conditions within the fine calibre valve mechanism. This is not uncommon following proximal revisions as friable adherent choroid plexus has a propensity to avulse and bleed when existing ventricular catheters are removed. CSF infections predispose shunts to fail, and commonly, this is at the level of the valve. The pathological process is not fully understood, but it is likely that the innate immune response and resulting inflammatory reaction to infection cause the build-up of cellular debris within the shunt. Distal occlusions (beyond the valve) are uncommon but do occur, and distal patency must therefore always be confirmed at the time of shunt exploration surgery. Common causes for distal failure are coiling within the layers of the abdominal wall, due to either tube expulsion by bowel peristalsis or, more likely, poor implantation technique. Entrapment within an abdominal pseudocyst or coiling within an intra-abdominal compartment due to adhesion formation is also commonly noted at distal revision. Finally, raised intra-abdominal pressure, for example, due to chronic constipation or mass lesion, has also been reported as a cause of shunt failure.

in the upright position. Occasionally, mechanical failure of shunt mechanisms can occur wherein the resistance to drainage fails and patients experience low-pressure symptoms. Chronic overdrainage of the ventricular system, especially in the context of previous infection, can result in small, non-compliant ventricles that do not expand when the shunt is blocked even when the pressure is very high. This is a particularly difficult problem termed the ‘slit ventricle syndrome’ and is one reason that ventricular volume is not a sure determinant of shunt patency. Chronic low pressure may be manifest in a thickened skull vault as well as thickened, enhancing meninges visible on post-contrast MRI. Subdural fluid collection (hygromas) may also be visible. Mechanical failure in children is notoriously difficult to accurately diagnose except in cases of acutely raised ICP where children may present vomiting, lethargic or even in coma. More subtle presentations with headache, behavioural change, abnormal posturing and change in appetite/feeding may be all that is evident. Remember that a large proportion of shunted patients have complex needs and often global developmental delay and so are unable to volunteer much information. Parents of shunted children are coached to have a low threshold for presentation to hospital if they are concerned. Parents are also very sensitive to changes in behaviour and are often very experienced in identifying subtle changes that may be due to shunt malfunction. The emergency clinician or junior neurosurgeon dismisses parental concerns at their peril!

Overdrainage Overdrainage of the ventricular system can occur when an inappropriately low-pressure valve is inserted into a shunt system, such that the cerebral mantle is allowed to collapse away from the overlying dura. This can result in tearing of delicate traversing draining veins and the formation of subdural haematomas. Similarly, overdrainage can be a positional phenomenon due to the siphoning of fluid into the peritoneal cavity when upright. This can be limited by utilising shunt designs with incorporated anti-siphon systems or gravitational valves which mitigate overdrainage

Disconnection Shunt components have a frustrating habit of disconnecting, even in cases where they were properly and diligently secured (with a non-absorbable braded tie). Shunt material can become tethered along its length by adjacent tissue reaction. Older materials had a susceptibility to become calcified and stiff also. Rapid growth with tethering of distal/proximal components or rapid/extreme neck movements, such as those witnessed in generalised tonic-clonic seizures, can test shunt connections or fracture old and brittle tubing. In these cases, the presentation is often more sub-

40 Hydrocephalus

513

acute with subtle symptoms or signs of over- or underdrainage. Patients may even complain of pain, redness or swelling at the site of disconnection or fracture. Infection Infection remains an important complication, though rates have significantly improved in the last four decades, from between 14% and 24% reported in the mid-1970s to between 2% and 10% quoted in contemporary literature. Inoculation of the shunt almost universally occurs at the time of surgery, and infections present within 30 days of the index operation in the vast majority of cases (Kulkarni et al. 2001). It is usually due to skin commensal organisms, most often Staphylococcus species (Bayston 2018). Risk factors for shunt infection include young age, in particular premature neonates, post-operative CSF leak, improper technique and excess handling of the shunt prior to insertion and previous shunt infection (Simon et al. 2009; McGirt et al. 2003). Patients with infection may present with obvious symptoms and signs of CNS infection: pyrexia, nausea and vomiting, headache, ­meningism and confusion. In addition, anorexia and abdominal pain may reflect a peritoneal reaction to infected CSF. There may also be tenderness and erythema tracking along the shunt tract. More often, however, patients present with a less well-defined illness, and often the only symptoms of infection are due to partial or complete shunt obstruction and resulting raised ICP. Diagnosis can therefore be challenging, and a ‘shunt tap’ is often advocated (Fig. 40.9). CSF aspirated aseptically via the shunt reservoir is sent for microscopy, gram stain, cell count and culture. Unfortunately, due to low numbers of viable bacteria present in the CSF of infected systems, microbiological culture can be negative in up to 25% of ‘confirmed’ infected cases. The CSF cell count is most instructive in this scenario, and a pleocytosis with predominant neutrophilia is highly suggestive of shunt infection particular with concordant clinical details (recent surgery, unwell patient). Staphylococci that colonise a shunt survive in a biofilm and in deeper layers downregulate their

Fig. 40.9  Photograph demonstrating the technique of ‘shunt tap’ via a butterfly needle inserted into the subcutaneous shunt reservoir. Strict adherence to aseptic precautions is essential to prevent inadvertent bacterial inoculation of the shunt system

metabolic activities and cell division markedly (Bayston 2018). The result is a relative resistance to antibiotics, and surgical explantation is necessary to manage the infection. The shunt is replaced with an EVD to maintain CSF diversion, and systemic intravenous antibiotics, with or without intrathecal antibiotics, are administered for 10–14 days before reimplantation (Fig. 40.10 (EVD)). Improvements in operative technique, standardisation of practices, antimicrobial stewardship and most recently the adoption of antibiotic impregnated catheters have all contributed to improvements in infection rates. The considerable morbidity associated with infection mandates continued vigilance however, and timely and effective treatment of shunt infections is essential. Ventriculoatrial Shunts VA shunts are indicated in patients with concomitant intra-abdominal pathologies precluding the use of the peritoneum as a drainage site. Common coexistent pathologies include necrotising enterocolitis, peritonitis and extensive abdominal surgery. Complications specific to VA shunts include the need for repeated lengthening of the short distal catheter, higher risk of bacteraemia

514

G. Sunderland et al.

Fig. 40.10  External ventricular drainage

and sepsis as well as the risks of specific vascular complications, such as thrombosis, microemboli with resultant pulmonary hypertension, macroemboli with pulmonary embolism and vascular perforation.

40.9.4 Ventriculoperitoneal Shunts and Abdominal Surgery Paediatric general surgeons will at times have to assess patients with VP shunts and potential abdominal pathology in both the acute and elective settings. If, during an elective procedure without infection or contamination of the peritoneum, a peritoneal catheter is encountered, then it should merely be pushed aside gently and excluded from the field. Significant contamination of the field mandates externalisation of the shunt. The distal shunt tubing is assumed to be contaminated and

is delivered externally through the skin to continue draining into a collection bag. Once the abdominal infection is treated, the external length of tubing is cut, and a new clean length of distal shunt tubing can be safely reimplanted. Laparoscopic techniques are now commonplace, and experience has shown these to be entirely safe in the presence of a VP shunt (Rosenfeld et al. 2019; Walker and Langer 2000; Jackman et  al. 2000). Early animal studies reported significant rises in ICP on induction of pneumoperitoneum as one would see in an ICP trace anytime a patient coughed or strained (Rosenthal et  al. 1997). There was concern that prolonged raised intra-abdominal pressure could significantly impair valve or shunt function or potentially even lead to retrograde flow and significant pneumocephalus. Studies of shunted patients during laparoscopic surgery have not identified such risk or adverse events, and there are only scattered case reports of shunt failure

40 Hydrocephalus

associated with laparoscopy amongst the many thousands of cases performed in the last 20+ years. In vitro studies of shunt valves show them to be resistant to retrograde flow at pressures far in excess of that used for pneumoperitoneum (Neale and Falk 1999). While there is no good evidence to support safety concerns in this patient cohort, caution is still advised over high insufflation pressures (>16 mmHg) and prolonged surgery (>3 h) (Sankpal et al. 2011). Routine discussion with a neurosurgeon regarding the safety of laparoscopy in shunted patients is probably not necessary unless significant abdominal contamination is anticipated, in which circumstance externalising the shunt could be considered in advance. Assessment of the acute abdomen with a shunt in situ is a difficult task and requires a holistic approach. Children may present with abdominal symptoms secondary to shunt malfunction, and it is important to define early if there are clinical or radiological signs of shunt dysfunction and not concentrate solely on the abdomen; similarly, it is vital that neurosurgeons seek an experienced opinion early in cases where the shunt seems to be functioning but the patient is symptomatic. If there are abdominal symptoms and signs of peritonitis mandating laparotomy, then the shunt should be externalised, the CSF cultured, and antibiotics started. If the shunt is obviously infected in addition, i.e. signs of raised ICP, meningism and erythema tracking along the shunt, then the shunt is removed in its entirety and only reimplanted once effective antibiotic treatment of both the abdominal and CNS infection is completed. Appendicitis is a common acute abdomen presentation, and children with VP shunts are not excepted from this. There is no evidence to suggest these patients have a worse outcome than non-shunted patients, and with caution, there should be no shunt-related complications (Ein et  al. 2006; Barina et  al. 2007). In an emergent appendicectomy, if the appendix is inflamed but not perforated, it is reasonable to leave the shunt in situ; if, however, there is any peritoneal soiling, then the shunt must be externalised. Other common, but usually non-emergent, complications of peritoneal catheters include

515

peritoneal pseudocysts and ascites. Pseudocysts are wall-less fluid collections accumulating between matted bowel loops; they may be complicated by infection. If there are no features of infection, then simple repositioning of the catheter in another portion of the peritoneum is all that is required.

40.9.5 Endoscopic Third Ventriculostomy (ETV) Endoscopic exploration of the ventricular system with therapeutic intent had been explored in the early twentieth century (Demerdash et al. 2017). Early forays into neuroendoscopy were aimed at the management of hydrocephalus but had limited success. There were a handful of stumbling attempts to revive the technique in the subsequent decades, but these never translated to routine practice. These early failures were due largely to technological limitations, and so unsurprisingly, it wasn’t until the 1970s with the inception of the Hopkin’s rod endoscope that the technique really developed. Endoscopic applications have since exploded throughout the whole breadth of surgical practice, and neurosurgery is no exception. The re-introduction of neuroendoscopy at this time was followed by a rapid broad adoption, and since then, a multitude of applications have been developed focused on, but not limited to, the management of hydrocephalus. The most ubiquitous of these is the endoscopic third ventriculostomy (ETV). ETV is primarily indicated in cases of obstructive hydrocephalus, the aim being to provide a route whereby CSF can bypass the point of obstruction and enter the subarachnoid space, creating a functional shunt from the third ventricle into the basal cisterns, from where it may circulate and be reabsorbed. The major advantages of ETV over an implanted shunt are that it obviates the need for an implanted foreign body, thus avoiding risks of infection, material degradation, disconnection, misplacement, etc., and it maintains a more ‘physiological’ CSF dynamic, something that even modern shunt valves struggle to replicate.

G. Sunderland et al.

516

40.9.6 ETV: The Technique The technique of ETV is demonstrated in Fig.  40.11. A rigid endoscope is navigated into the frontal horn of the lateral ventricle through a frontal pre-coronal burr hole placed in the midpupillary line. After ventricular cannulation, the endoscope is introduced, and the operator oriena

tates themselves noting the thalamic veins running toward the foramen of Monro. The foramen of Monro transmits CSF from the lateral to third ventricle and is bounded posteriorly by the anterior pole of the ipsilateral thalamus and anteriorly by the fornix—an important part of the limbic system that links the hippocampus to mammillary bodies. The fornices are integral to working b

b

b d

c

a

d

c

a

a

c

d

a

a

Fig. 40.11  Endoscopic images representing key stages and anatomy encountered at ETV. (a) View from the endoscope within the right lateral ventricle. Note the choroid plexus and posterior caudate vein (a), anterior caudate vein (b) and thalamostriate vein (c) converging towards the foramen of Monro (d). (b) Endoscopic image from within the third ventricle. Visible are the mammillary bodies (a) and pituitary infundibulum (b) bordering the thin

membranous tuber cinereum (c). Just visible is the basilar artery (d) through the tuber cinereum. (c) A Fogarty catheter is used to perforate the floor of the third ventricle and the balloon (a) inflated to expand the ‘stoma’. (d) View with Fogarty catheter withdrawn demonstrating a perforation (a) through to underlying pre-pontine cistern. Note the fronds of arachnoid visible

40 Hydrocephalus

memory function and thus must not be injured or put under excessive stretch. The endoscope is then navigated through the foramen of Monro into the third ventricle. On the third ventricle floor, the landmarks are reliably seen, the mammillary bodies posteriorly and the pituitary infundibulum (a red blush at the anterior apex of the third ventricle). Within a triangle drawn between these points is the tuber cinereum where the membranous floor of the third ventricle is thin and through which it is safe to pass. The tuber cinereum is punctured, and this ‘stoma’ then expanded sequentially with balloon or endoscopic forceps. Further membranes, in particular the membrane of Liliequist, should be identified and fenestrated, and a clear view to the pre-­ pontine cistern and the basilar artery is seen. The floor of the third ventricle is seen to billow as if, in the wind when the endoscope is withdrawn, a reassuring sign that the intervention has been successful (Fig. 40.11). Post-operatively, ventriculostomy patency can be investigated using phase contrast MRI or ‘CSF flow study’. On a midline sagittal cut, the flow of CSF across the stoma and turbulence within the pre-pontine cistern can be identified via a dark ‘flow void’ in this path (Fig. 40.12) (Dinçer et al. 2011).

517

structure, and interfering with it risks causing IVH.  Furthermore, the choroid plexus is not merely a CSF ‘pump’ but generates intracranial pressure (ICP), maintains CSF homeostasis and provides micronutrients, proteins and hormones for neuronal and glial development, maintenance and function. Concerns regarding long-term neurocognitive outcome and the potential neurodevelopmental impairment exist (Spector et  al. 2015).

40.9.8 Indications for ETV

ETV is considered in cases of obstructive hydrocephalus including obstructions in the caudal portion of the third ventricle (pineal region), aqueduct, fourth ventricle and foramen magnum. It has an established role in the primary management of aqueduct stenosis (Kulkarni et al. 2018). ETV enjoys an overall success rate of 60–90%, the significant variety due to age and aetiological factors. Rates improve with advancing age (children > 1 year) and with specific aetiologies, i.e. those with an anatomical obstruction such as aqueduct stenosis, posterior fossa tumours and Dandy-Walker malformation. Aetiologies with inferior rates of success are those classically termed ‘communicating’, i.e. post-haemorrhagic 40.9.7 ETV with Choroid Plexus and post-infective. Coagulation The commonest causes seen in neonates are germinal matrix IVH and meningitis; thus, ETV In a technique akin to that pioneered in the 1920s, is not a viable option in this cohort. The choroid plexus coagulation sees the surgeon Hydrocephalus Research Network in the USA endoscopically cauterising the choroid plexus developed and in 2010 published the ETV with the aim of reducing CSF production. This Success Score (ETVSS), a reliable and practical may be done in conjunction with an ETV (ETV-­ predictive score based on age, aetiology and prior CPC), a sort of ‘belt and braces’ approach. Trials CSF shunt (Table 40.3) (Kulkarni et al. 2010). run in Uganda and subsequently in the USA in the 2000s showed advantage over ETV alone especially in younger children ( 50, central obesity, smoking history, Caucasian, and first-degree relative with BE or esophageal adenocarcinoma. The guidelines suggested by the American College of Gastroenterology for GERD at risk for BE (intestinal metaplasia) recommend the following monitoring strategy: (a) In the case of dysplasia not demonstrated on two occasions: control endoscopies every 2–3 years. (b) In the case of low-grade dysplasia demonstrated on two occasions: control endoscopies every year. (c) In the case of high-grade dysplasia: selective resection or control endoscopies at 3-month intervals. Today, intensive clinical research is focused on risk stratification to perform screening endoscopy in the appropriate patient’s group (Steele et al. 2019). Analysis of the results of esophagogastroduodenoscopies for uncomplicated GERD showed that only 5.6% had suspected BE of any

58  Gastroesophageal Reflux Disease

length and 1.4% has suspected BE >3 cm length (Lin et al. 2019). Raicevic and Saxena performed a systematic review in children and reported that out of 130 cases of BE, 61.5% were associated with GERD, 10% with esophageal atresia and reflux, and the other patients with a variety of different diseases (Raicevic and Saxena 2018). Adenocarcinoma after BE is rare in children; however, individuals born preterm or small for gestational age have a threefold increased risk for developing BE as adults (Forsell et al. 2013). The presence of high-­ grade dysplasia or adenocarcinoma is rare in children and has been reported in only a few isolated cases. Radiofrequency ablation is highly effective and safe for treatment of BE with dysplasia or early stage cancer.

58.7.3.5 Hiatal Hernia Any upward shift of portions of the stomach into or beyond the esophageal hiatus is referred to as a hiatus hernia (HH). A distinction is made between axial and paraesophageal hiatal hernias. The axial HH is either a sliding phenomenon or is fixed in the chest region (Fig. 58.12). A sliding

Fig. 58.12  Typical hiatal hernia seen during endoscopy. The lower “ring” corresponds with the diaphragm; high up one can see the LES. (With permission from Höllwarth ME: Gastroesophageal reflux disease. In Coran A (editor in chief) Pediatric Surgery, seventh edition 2012 by Saunders, an imprint of Elsevier)

767

HH is much rarer in children than in older adults (it occurs in approximately 60% of adults). A fixed HH is occasionally seen in children after correction of an esophageal atresia and/or diaphragmatic hernia, in children with mental disability, congenital short esophagus, and esophagitis-related stenosis with a shortening of the esophagus due to scarring. In infants or young children with reflux disease and HH, the condition cannot be expected to resolve spontaneously. Therefore, conservative therapy has no chance of success in this setting and surgery is indicated in all cases. Paraesophageal hernias are a typical postoperative complication after fundoplication (4.5%) and occur less often after suture closure of the hiatus (3%) than without (10%) (Fig.  58.13) (Avansino et al. 1999). A part of the stomach slips through the hiatus into the chest sideways, from the esophagogastric junction located in normal position. When a paraesophageal hernia is not combined with a pathological recurrent reflux and no further symptoms are present (such as a sensation of postprandial pressure or pain, dysphagia, vomiting, or gastric obstruction), surgical correction is not required in every case. The congenital upside-down stomach is an extreme form of a paraesophageal hernia and must be corrected surgically.

Fig. 58.13  Typical endoscopic view of a paraesophageal hernia. (With permission from Höllwarth ME: Gastroesophageal reflux disease. In Coran A (editor in chief) Pediatric Surgery, seventh edition 2012 by Saunders, an imprint of Elsevier)

768

58.8 Management 58.8.1 Conservative Treatment The North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN) along with the European Society for Gastroenterology, Hepatology, and Nutrition (ESPGHAN) published 2018 clinical practice guidelines for children with GER, in regard to diagnostic and therapeutic strategies. The following management strategies are based on these recommendations. Additionally, patient and parental education, guidance, and support are always required as a part of the treatment of GERD (Rosen et al. 2018; Poddar 2019). In this age group, it is important to avoid the use of disease labels because they may promote overtreatment (Vandenplas and Hauser 2015). Parents of healthy children should be assured that symptoms of GER are common in babies while on liquid nutrition and may resolve spontaneously after a year. However, we always tell the parents that a pathological reflux can still persist, but the refluxed material doesn’t reach the mouth. Thus, a pH/MII control is advisable at the age of 1 or 2  years in order to prove that no pathological reflux pattern is going on.

58.8.1.1 Babies and Small Infants Conservative therapy is the primary strategy focused on feeding habits in this age group. A reduction of the ingested volume per feeding can already avoid an overloading of the stomach, but at the same time, the frequency of feedings is increased to allow the adequate and age-appropriate nutrition. Additional thickening of feedings with cereals improves visible regurgitation. Despite the safety concern of the FDA due to elevated levels of inorganic arsenic in all forms of rice, rice gruel still has the advantage of a good dissolving thickener and it is affordable. Of course, breastfeeding should be supported at least for the first 6 months. In breastfed babies with significant GER, the breast milk can be thickened with carob bean too, which is approved for infants after 42 weeks of gestation. Feeding modifications are recommended as the

M. E. Höllwarth and V. Solari

first therapeutic strategy before other costly or risky interventions. Investigations have shown that lying in prone position, possibly with a raised torso, prevents reflux most effectively. However, it also involves a much greater risk of sudden infant death, due to vomiting in the sleep and apnea events. Therefore, the supine sleeping position is recommended. Infants with an allergy to cow’s milk protein (CMPA) often show regurgitation and vomiting typical of GERD.  Vomiting frequency significantly decreases within 2  weeks of cow’s milk protein elimination from the diet. Thus, infants with suspected CMPA should receive formula feeding with hydrolyzed protein.

58.8.1.2 Conservative Therapy in Older Children and Adolescents As the child’s food starts to resemble that of adults, there is an increasing quantity of acid secretion from the stomach after meals. Therefore, the majority of postprandial refluxes are now accompanied by a pH reduction to below 4, and, in cases of GERD, the possibility of esophagitis is increased. Recommendations for lifestyle and behavioral changes include avoiding food and beverages (caffeine, chocolate, carbonated beverages, spicy or acid food) that trigger reflux symptoms. A metaanalysis showed that weight loss and head of the bed elevation and left lateral decubitus are the only effective lifestyle interventions (Kaltenbach et al. 2006). However, obesity is still a matter of discussion. A retrospective study did not show that obesity is a risk factor in 738 children with reflux esophagitis (Elitsur et al. 2009), while other studies found an association between childhood obesity and the risk for developing a GERD (Koebnick et al. 2011; Quitadamo et al. 2012). 58.8.1.3 Pharmacologic Treatment The aim of drug therapy is to reduce acid exposure to the esophagus and thus avoid or treat esophagitis. A distinction should be made here between drugs that protect the mucosal surface and those that reduce or hinder the production of gastric juices.

58  Gastroesophageal Reflux Disease

Alginates Alginates are aluminum, magnesium, or calcium salts of alginic acid which are used to treat acid-­ related disorders such as heartburn or dyspepsia. Alginates are often combined with sodium/potassium bicarbonate. They buffer gastric acid and improve the symptoms of esophagitis and reduces signs of inflammation. Short-term treatment with alginates seems to have no side effects, but long-­ term use of aluminum-containing alginates (sucralfate) may lead to increased aluminum plasma concentration in infants. NICE (National Institute for Health and Care Excellence) guidelines recommended alginates as an alternative to feed thickeners in breastfed infants or as a trial in infants in whom symptoms persist despite conservative measures (Davies et al. 2015).

769

acid or volume clearance takes a very long time. This happens in children who have undergone surgery for esophageal atresia, especially when the lower segment has to be mobilized extensively, in children who have experienced diaphragmatic hernia, in those with an incompetent esophagogastric junction, as well as children with severe mental disorders. Children frequently require a much higher PPI dose compared to adults (0.7–3.5  mg/kg/day BW). The efficacy of gastric acid blockade must be checked by performing pH monitoring about 2 weeks after the start of therapy, as the dose may need to be modified significantly in individual cases. Investigations after long-term treatment with PPIs in adults revealed a 30% reduction in vitamin B12 levels, atrophic gastritis, and several weeks of acid hypersecretion after discontinuation of the medication. As mentioned earlier, Proton Pump Inhibitors (PPI) PPIs are the most potent pharmacologic agents long-term therapy with acid inhibitors in children suppressing acid production by inhibiting H +/K is indicated only in exceptional cases, although + -ATpase in parietal cells within the stomach. In positive results have been reported in the pubfact, PPIs reduce the acidity of gastric contents lished literature in this regard (Hassall et al. 2007; and refluxes, thereby avoiding erosions of the Tolia and Boyer 2008). Side effects are rare and esophageal mucosa but does not necessarily mostly mild (headache, constipation, diarrhea, reduce the total number of refluxes. As long as nausea, hypomagnesemia) (Carroll and Jacobson this is the case, the vicious cycle between severe 2012). However, long-term treatment with PPIs esophageal inflammation leading to a high num- is associated with a higher risk due to the prober of TLESR exists. Thus, respiratory tract-­ longed acid suppression and hypochlorhydria, related symptoms due to recurrent aspiration which impairs vitamin B12, calcium, and iron absorption. Furthermore, long-term acid suppresduring sleep may still be present. Infants: Most milk-fed infants usually do not sion may lead to a significant change of the small require a drug-based treatment of reflux. The bowel microbiota and overgrowth with pathoabove-mentioned guidelines do not recommend genic bacteria causing gastroenteritis, increased PPI for crying distress and also not for the treat- risk of community-acquired pneumonia, and ment of visible regurgitation in otherwise healthy Clostridium difficile infections (Moussa and infants. However, if recurrent unquietness, dis- Hassan 2017; Nylund et  al. 2014; Berni and turbed sleep, and pain are the symptoms of a Terrin 2010; Lo and Chan 2013; Poddar 2019). GERD, PPI therapy is indicated. Studies in infants 80  cm  +  ileocaecal valve), 5–15  cm bulbous hypertrophied proximal bowel is resected back to the near normal calibre of the bowel. In performing the resection, one should preserve as much of the mesentery as possible for later use to fill in any gaps in the small bowel mesentery left after end-to-end anastomosis. Having decided on the site of transection, the bowel is then divided at right angles leaving an opening of approximately 0.5–1.5 cm in diameter. The blood supply should be adequate to ensure a safe anastomosis. If, however, an extensive cutback resection is contraindicated because of insufficient residual bowel length, the bulbous portion alone or any compromised bowel should be resected. The proximal bowel should then be tapered obliquely hand-sewn or using a GIA stapler leaving the proximal bowel opening a similar size to the distal bowel lumen to facilitate an end-to- end primary anastomosis. Proximal bowel resection is followed by very limited distal small bowel resection over a length of 2–3 cm. The resection line should be slightly

a

oblique towards the anti-mesenteric border (fish-­ mouth) to ensure that the openings of the proximal and distal bowels are of approximately equal size to facilitate easy end-to-end or rarely an end-­ to-­back (Denis-Browne) single-layer anastomosis; 5/0 or 6/0 absorbable suture material is used (Fig. 64.8). Alternatively, an extra-mucosal end-­ to-­end anastomosis can be performed, placing sutures on the proximal bowel further apart so as to accommodate the discrepancy in the bowel lumen diameters. The mesentery is approximated with interrupted sutures, which may be difficult with large mesenteric defects. A side-to-side anastomosis should not be performed as it can lead to a blind loop syndrome. There is little place for routine gastrostomy or transanastomotic feeding tubes where facilities for parenteral nutrition are available.

64.7.5.2 Special Considerations Atresia type I and stenosis are best dealt with by primary resection and end-to-end anastomosis. Procedures such as simple transverse enteroplasties, excision of membranes, bypass techniques or side-to-side anastomosis are no b

Fig. 64.8  End-to-end (a) or end-to-back (b) single-layer bowel anastomosis

64  Jejuno-Ileal Atresia

longer utilized. They fail to remove the abnormal dysfunctional segments of intestine, thus increasing the risk of the blind loop syndrome and dysmotility. Atresia types II and IIIa are managed in the same manner as type I with back resection and primary end-to-end anastomosis. The conservation of bowel length is mandatory. Multiple membranous diaphragms (type I atresias) can be successfully perforated by transluminal bouginage done along the entire length of the affected small bowel (Romao et al. 2011). High jejunal atresia: With type IIIb or high jejunal atresia, the proximal bowel should be derotated, the ligament of Treitz, if present, should be taken down, and resection of the bulbous portion may be extended into the second part of the duodenum taking care to stay well clear of the ampulla of Vater. This is followed by an antimesenteric tapering duodenojejunoplasty (Kling et  al. 2000). Bowel tapering can safely be done over a length of between 20 and 35  cm. This is done to conserve bowel length, to reduce disparity in anastomotic size and to improve duodenal and proximal jejunal prograde peristaltic function. At completion, the bowel is left in a position of derotation with the duodenum-jejunum positioned on the infant’s right side, the mesentery broad-based and the caecum lying anterior, to the left of the midline in the upper abdomen. These additional maneuvers induce rapid return of prograde intestinal function, and the neonates are usually able to tolerate graded to full oral intake within 14 days. With type IIIb atresias, the distal ‘apple peel’ component of the bowel is gently laid out so that the inner margin of mesentery is clearly visible from its right colon origin to the atretic blind end. The division of avascular restricting mesenteric bands along the free edge of the distally coiled narrow mesentery may be required, thereby releasing kinking and interference with the bowel blood supply. The large mesenteric defect may be left open, but where possible the preserved mesentery can

877

be used to fill in the defect. Furthermore, to prevent kinking of the marginal artery after the completion of the anastomosis, the bowel is replaced carefully into the peritoneal cavity in the position of non-rotation. There is no need to remove the appendix. Multiple type IV atresias, present in 20% of cases, are often localized to a segment facilitating an en-bloc resection with a single anastomosis in preference to multiple anastomosis. If multiple anastomoses are deemed necessary because of insufficient bowel length, it is useful to railroad each bowel segment to be anastomosed onto a silastic feeding tube before completing the anastomoses to avoid torsion and ensure correct orientation and continuity of the bowel. This tube may be left in situ for a time to serve as a stent until bowel function has returned. It is always important to ensure the preservation of the maximum bowel length to avoid the short bowel syndrome; for this reason the following techniques may be necessary. Tapering enteroplasty is indicated when the ischemic insult has resulted in an atresia with markedly reduced intestinal length (3.5 mmol/L) and to inhibit inappropriate insulin secretion. Therapeutic strategies can be medical, surgical, or combined.

84.5.4 Medical Treatment In the emergency setting, such as in the case of seizures, the infant should receive an intravenous bolus of 2ml/kg of 10% glucose, followed by a glucose infusion at >6–8 mg/kg/minute. In the absence of intravenous access, glucagon can be administered via intramuscular injection to stimulate glycogenolysis and glucose release from the liver. Patients will also be started on a continuous or frequent feeding regime. Drug treatment includes diazoxide with chlorothiazide, nifedepine, octreotide, and glucagon. Diazoxide (K-ATP agonist) is an inhibitor of insulin secretion and it is used in all types of CHI as first-line therapy. Diazoxide binds to the SUR1

1160

E. Zani-Ruttenstock and A. Zani

subunit of the K-ATP channel, which activates 84.5.5 Surgical Treatment intact K-ATP channels and reduces insulin secretion. To be effective, both subunits of the channel In the case of focal CHI, a partial pancreatectomy must be structurally and functionally normal. with clear margins will be curative. Intraoperative However, patients with diffuse high-resolution ultrasound can help in localizing CHI secondary to ABCC8 or KCNJ11 muta- focal lesions and can reduce the number of frozen tion and those with focal lesions are unresponsive section biopsies. Furthermore, it can be useful to to diazoxide, as their K-ATP channels are not identify the pancreatic duct. structurally normal. Since the most common Small and superficial lesions in the body or tail cause of CHI is a mutation in the SUR1/KIR6.2 of the pancreas are treated by simple resection, gene complex, about two-thirds of CHI patients whereas deep, periductal lesions are treated by disdo not respond to diazoxide. To identify nonre- tal pancreatectomy. Small and superficial lesions sponders, patients undergo a diazoxide response in the head of the pancreas can also be treated by test, which consists of a five-day administration simple resection, whereas deep lesions are difficult of the drug followed by a fasting test in the to resect without causing damage to the CBD or absence of glucose infusion or other medications. pancreatic duct. To ensure a complete resection of Patients who maintain a plasma glucose level these more challenging lesions, the majority of the >70 mg/dL for ≥12 hours are considered “diazox- pancreatic head will be removed, and a Roux-en-Y ide responsive.” These patients are subsequently pancreatico-jejunostomy will be fashioned to managed by a regimen of frequent feedings and allow drainage of the pancreatic body and tail long-term diazoxide. Conversely, diazoxide-­ (Laje et al. 2012). In the case of extension of the resistant patients generally require surgery; prior focal pancreatic head lesion into the duodenal to surgery, to maintain euglycemia, they can use wall, a pancreaticoduodenectomy (i.e., Whipple octreotide, glucagon, and nifedipine as an procedure) is the treatment of choice. alternative. Laparoscopic surgery is especially suitable Side effects: diazoxide can causes severe for focal lesions in the pancreatic body or tail. sodium and water retention that may result in In cases of diffuse CHI that fail to respond to congestive heart failure, especially in neonates. medical management, an open or laparoscopic For this reason and for its synergistic effect on near-total pancreatectomy (95%) is considered as suppressing insulin secretion, chlorothiazide, a the gold standard (Fig. 84.2). A near-total pancrethiazide diuretic is used in combination with diazoxide. Octreotide is a long-acting analogue of somatostatin, which naturally inhibits the insulin release from the pancreatic β-cells. Octreotide activates the somatostatin receptor 5 (SSTR5), thus inhibiting calcium mobilization and acetylcholine activity and decreasing insulin gene promoter activity, which reduces insulin biosynthesis and insulin secretion. Glucagon is a natural insulin antagonist that is mainly used to promptly reverse severe hypoglycemic episodes. Nifedipine is a calcium-channel blocker that inhibits insulin secretion by inactivation of 84.2  Near-total pancreatectomy: the head, body, voltage-­gated calcium-channels. However, it is Fig. and tail of the pancreas including the uncinate process are only effective in a minority of patients, and dos- resected, and only a small portion (5%) of pancreatic tissue is left in situ age is often limited by hypotension.

84  Pancreatic Disorders

atectomy consists of resection of the tail, body, uncinate process, and most of the pancreatic head, leaving a rim of pancreatic tissue surrounding the CBD along the duodenum. A near-total pancreatectomy can be performed with an open or laparoscopic approach.

84.5.6 Postoperative Management Postoperatively, an intravenous glucose infusion is restarted at a low rate as surgical stress induces hepatic glycogenolysis (Laje and Adzick 2020). In the case of persistently high plasma glucose levels, an intravenous insulin infusion is also started. Upon return of bowel function, enteral feeds are resumed and advanced gradually, with subsequent weaning of the intravenous glucose infusion. Postoperative pain after neonatal pancreatectomy is controlled with an epidural catheter or intravenous narcotics. Postoperative complications specific to pancreas surgery, such as pancreatic leak, lymphatic leak, and CBD injury are reported with an incidence of 95% of cases cured after surgical excision. Conversely, following the near-total pancreatectomy, 50% of patients with diffuse CHI have persistent hypoglycemia, 25% are cured and have normoglycemia, and 25% develop diabetes that require insulin (Laje and Adzick 2020). Moreover, children who undergo near-­ total pancreatectomy for diffuse CHI commonly have neurobehavioral deficits (Lord et al. 2015).

84.6 Pancreatitis Pancreatitis is defined as an inflammatory process secondary to the intraductal activation and secretion of pancreatic enzymes, which results in the digestion of the parenchyma. With an estimated incidence of 3–13 cases per 100,000 children

1161

(Keim et al. 2003; Rebours et al. 2009), this condition is far less common in children than in adults. Recognizing the limited literature, especially for the more complex forms, the “International Study Group of Pediatric Pancreatitis: In Search for a Cure” (INSPPIRE) consortium formed as a multicenter approach to systematically characterize pancreatitis in children. In 2012, the consortium reported a consensus statement on the definitions of childhood-onset acute pancreatitis, acute recurrent pancreatitis, and chronic pancreatitis (Morinville et al. 2012) (Table 84.2).

84.6.1 Acute Pancreatitis The most common causes of acute pancreatitis in children are summarized in Table  84.3 (Diesen 2020a, b). Although acute pancreatitis is rare in childhood, it appears that the incidence of pediatric pancreatitis has increased over the last two decades (Diesen 2020a, b). This is likely due to a combination of factors, including the increased incidence of obesity in the pediatric population, drug-associated pancreatitis, and improvements in its diagnosis (Poddar et al. 2017).

Table 84.2  Definition of pancreatitis in children (modified from Morinville et al. 2012) Acute pancreatitis (AP) One of the three criteria: • Abdominal pain suggestive of AP (acute onset in the epigastrium) • Serum lipase or amylase at least three times the upper limit of normal • Imaging findings characteristic of AP (any modality)

Acute recurrent pancreatitis (ARP) • AT least two distinct attacks of AP N.B.: Attacks must be separated by either ≥1-month pain-free interval or pain resolution and normalization of serum amylase and lipase before the subsequent attack

Chronic pancreatitis (CP) • Imaging findings of CP AND one of the following three: • Abdominal pain suggestive of pancreatitis • Exocrine pancreatic insufficiency • Endocrine pancreatic insufficiency

E. Zani-Ruttenstock and A. Zani

1162 Table 84.3  Causes of acute pancreatitis in the pediatric population (modified from Diesen 2020a, b) Causes of pancreatitis Trauma

Choledocholithiasis

Congenital anomalies

Systemic and genetic diseases

Infections

Drugs

Blunt/penetrating Non-accidental trauma Motor vehicle collision Iatrogenic Abdominal surgery Endoscopy/ERCP Cholesterol stones—obesity Hemolytic stones—spherocytosis  • alpha-thalassemia  • sickle cell disease Parenteral nutrition Annular pancreas Pancreas divisum Pancreaticobiliary malunion Choledochal cyst Hypertriglyceridemia Hypercalcemia Cystic fibrosis Crohn’s disease Ulcerative colitis Systemic lupus erythematosus Familial pancreatitis Shwachman-Diamond syndrome Mumps Rubella Coxsackie B HIV Parasites Valproic acid l-asparaginase Steroids Metronidazole Macrodantin Azathioprine Tetracycline

Toxin exposure Liver transplant Autoimmune pancreatitis Idiopathic

84.6.1.1 Diagnosis and Differential Diagnosis The diagnosis of acute pancreatitis is based on history and physical examination and confirmed by the results of both laboratory and radiological investigations. Acute pancreatitis classically presents with epigastric pain that may radiate to the upper quadrants and/or the back. Sometimes, the pain

may be less well localized, and particularly in younger children, nausea and vomiting are often the predominant features. To determine the cause of pancreatitis, it is important to elucidate a history of recent abdominal trauma or surgery, recent infections or sick contacts, use of new medications, toxin exposure, history of blood disorders, gallstones, and family history of medical problems involving abdominal pain, surgery, or pancreatitis (Table  84.3). Physical examination may elicit a range of signs and symptoms, including low-grade pyrexia, jaundice, diarrhea, irritability, lethargy, epigastric tenderness, and generalized peritonitis. In severe forms of pancreatitis, the child may present with hypovolemic shock or multi-organ failure with dyspnea, oliguria, and mental status change. In cases of necrotizing or hemorrhagic pancreatitis, patients may have abdominal discoloration at the umbilicus (Cullen sign) or flank (Grey Turner sign) due to internal bleeding (Diesen 2020a, b). The key laboratory tests for the diagnosis of pancreatitis are serum amylase and lipase. Amylase rises quicker than lipase, whose levels remain elevated a few days longer. Complete blood count and electrolyte levels are indicative of general clinical status and useful to determine the need for resuscitation. Serum triglyceride levels may indicate that the pancreatitis is the result of hyperlipidemia, whereas liver function tests are important to rule out biliary obstruction in patients with gallstones. A plain abdominal X-ray (AXR) is often the first imaging study obtained, especially in patients with abdominal pain of unknown etiology. However, this study will not show any specific signs of acute pancreatitis. Similarly, a chest X-ray may demonstrate pulmonary edema or pleural effusion, which may be associated with pancreatitis. Abdominal ultrasonography can detect some signs of acute pancreatitis, such as diffuse or focal enlargement and altered echogenicity, and can be useful to detect dilation of the pancreatic duct and the presence of stones. Nonetheless, it is not uncommon that the pancreas appears normal despite the presence of inflammatory changes (Restrepo et al. 2016). An abdominal CT scan with contrast provides a more

84  Pancreatic Disorders

Fig. 84.3  Abdominal CT scan with double contrast of a 12-year-old girl presenting with acute necrotizing pancreatitis secondary to gallstones. The scan demonstrates a diffusely swollen pancreas with 75% necrosis

accurate picture of the severity of pancreatic damage (enlargement, peripancreatic stranding, pancreatic necrosis, and accumulation of peripancreatic fluid) and a better visualization of the ductal anatomy and surrounding vasculature (Restrepo et al. 2016) (Fig. 84.3). The MRCP is useful to study anomalies of the pancreatic ducts (e.g., strictures, stones, tumors) and CBD, and it is particularly indicated in patients with recurrent pancreatitis of unknown etiology. In cases of acute pancreatitis, an ERCP is typically not indicated, also because it has an associated risk (5–10%) of post-procedural pancreatitis. However, an ERCP can be useful in patients with pancreatic duct disruption secondary to trauma for diagnosis and management (stent placement), in cases of idiopathic recurrent pancreatitis, and in patients with pancreatitis associated with cholangitis/cholelithiasis, if the anatomy is not well characterized on MRCP.

84.6.1.2 Management Children with acute pancreatitis require very close monitoring throughout their admission and sometimes may even need transfer to a higher level of care (e.g., intensive care unit). As children with acute pancreatitis may deteriorate rapidly, early signs of multi-organ failure should be treated aggressively with ventilation support and active fluid resuscitation as needed. Pain control

1163

is usually achieved with nonsteroidal anti-­ inflammatory agents and narcotics. Antibiotics are typically not needed, unless the patient has evidence of infected pancreatic necrosis (Villatoro et  al. 2010). In contrast to previous teachings, enteral nutrition should be provided from early on, regardless of the feeding route, i.e., oral, gastric, or post-pyloric (Márta et  al. 2016; Poropat et al. 2015). Conversely, parenteral nutrition should be considered only if the patient is unable to tolerate feeds. Nasogastric decompression is needed only if the child is vomiting. If the acute pancreatitis is secondary to a stone, an MRCP or an ERCP may be needed to identify or remove the obstruction. If there is strong suspicion of ductal obstruction on ultrasound, the ERCP is the next logical step to remove the stone, dilate, and/or stent a duct stricture, biopsy a pancreatic mass if identified, perform a sphincterotomy. If the patient has gallstones, once the abdominal pain has resolved, a cholecystectomy prior to discharge from the hospital is warranted to prevent early recurrence of pancreatitis.

84.6.2 Complications of Acute Pancreatitis 84.6.2.1 Pancreatic Pseudocyst A pseudocyst results from the accumulation of leaked pancreatic enzymes enclosed within an inflammatory non-epithelial lining (Fig.  84.4). Pseudocysts initially appear as peripancreatic fluid collection 10–14 days after the onset of acute pancreatitis and should be suspected if elevated amylase levels reappear after having settled. Most pseudocysts develop soon after the insult and resolve spontaneously in 75,000) 3 mg/kg/day OD orally/NGT (maximum dose 75 mg)  Dipyridamole (if platelets > 50,000)  If patient weighs < 10 kg 25 mg tds orally, if weighs >  10 kg 50 mg tds orally (vii) Heparin infusion  For vascular anastomosis at risk (complex vascular reconstruction or small diameter arteries or portal vein) (viii) Analgesia and sedation  Analgesia is achieved with morphine in the routine transplant patient with reasonable graft function and is titrated against pain level be commenced within 72 h of surgery and may be supplemented by nasogastric feeding or parenteral nutrition in the early phase if there is a delay in restoration of bowel function. Phosphate and magnesium deficiency is common and requires replacement therapy in nearly all patients.

Liver ultrasound with colour flow Doppler is performed for the first 5 days and later, as clinically indicated, to confirm vascular patency and the absence of biliary dilatation. Hypertension is almost universal in pediatric transplantation and can initially be managed with nifedipine sublingually in conjunction with diuretic agents. Subsequently, calcium channel blockers may be given in appropriate dosage. Aspirin 3 mg/kg given on alternate days is used as prophylaxis against arterial thrombosis and a proton pump inhibitor is given for gastric mucosal protection.

87.10 Immunosuppression There is considerable variation in the selection of immunosuppressive agents. Most protocols currently employ triple therapy with tacrolimus, methylprednisolone and a monoclonal interleukin-­2 inhibitor (CD 25) antibody. Some centres use steroid-free immunosuppression. In addition, there are a number of other strategies in place to reduce the amount of nephrotoxicity, which is a toxic side effect to both calcineurin inhibitors. Thus, mycophenolate mofetil, an inosine monophosphate dehydrogenase inhibitor, may be used

1206

instead from early on in what is called ‘nephron sparing immunosuppression’. Rapamycin, a drug structurally similar to tacrolimus, which prevents proliferation of T cells but acts at a different stage of T cell activation from either cyclosporin or tacrolimus, has the advantage that it is not nephrotoxic and does not interfere with transcription and production of interleukin 2, rather it antagonizes the action of interleukin 2 on its receptor. It has no adverse effects on liver function and may be synergistic with cyclosporin. The recent consensus conference of the International Liver Transplantation Society developed guidelines on immunosuppression in liver transplants recipients (Charlton et al. 2018). Recently, significant research is focused on immune regulation with regulatory T-cells (Safinia et al. 2018). The methylprednisolone dosage is reduced over the first week to about 1 mg/kg/day for the first month and then reduced to a level of 0.3 mg/ kg/day to 0.2 mg/kg as maintenance. This can be later reduced in some patients to alternate day therapy or even withdrawn completely. Both mycophenolate mofetil and rapamycin/sirolimus can be used as renal sparing should nephrotoxicity become evident. Use of humanised anti-CD25 monoclonal antibodies given before and during the first week of the transplant has reduced the incidence of acute rejection in the first 3 months by around 30%, but long-term graft survival is essentially the same as when these agents have not been used. The other polyclonal anti-­ lymphocyte immunoglobulins are rarely used.

87.11 Anti-Infection Agents

K. Sharif and A. J. W. Millar

day in two divided doses 3 days a week for the prevention of pneumocystis carinii infection for at least the 1st year. Intravenous ganciclovir 5 mg/kg/dose 12 hourly is used as prophylaxis against cytomegalovirus (CMV) and Epstein Barr virus (EBV) infection, initially for 2 weeks, and this may be extended for up to 3 months in high-risk patients, who have not previously been exposed to CMV or EBV but have received a donor graft with previous exposure. This considerably reduces the incidence of both cytomegalovirus disease and post-transplantation lymphoproliferative disorder. Either hyperimmune cytomegalovirus globulin or immunoglobulin is also given to assist viral prophylaxis. Leucocyte filtered blood products are used throughout to reduce CMV load. Prophylactic antibiotics are given with induction of anaesthesia and continued for 3–5 days. These are changed according to cultures taken of blood, secretions, sputum and urine. Anti-tuberculosis prophylaxis is given only if evidence of tuberculosis is found before surgery and if a close family contact has tuberculosis. Ofloxacin, rifampicin and ethambutol or ethionamide may be used in addition to isoniazid but very careful monitoring of liver function tests is required because all of these drugs may be hepatotoxic and particularly rifampicin may result in a decrease in cyclosporin or tacrolimus levels due to enzyme P450 induction with increased drug metabolism.

87.12 Surgical Complications

Surgical complications may be reduced to an absolute minimum with meticulous technique. Immunosuppression naturally leads to suscepti- These may present early or late, as summarized bility to bacterial, fungal and viral infections. in Table 87.4. Fungal infection is a major and potentially fatal Most common surgical complications are as complication in liver transplantation. Fungal pro- follows: phylaxis is given as mycostatin orally before the Biliary complications continue to be a signifitransplant, to reduce Candida colonization of the cant problem, with an overall incidence of gut and amphotericin after the transplant and between 10–20%, particularly in living related continued for a period of several months. From left lateral segment grafts and split liver transthe first week after the transplant, trimethoprim-­ plants. These complications include bile leak, sulphamethoxazole is given at a dose of 6 mg/kg/ anastomotic strictures, and non-anastomostic

87  Pediatric Liver Transplantation

1207

cations are best treated by immediate surgery and re-anastomosis, if a major leak is suspected. Minor leaks from the cut surface of the graft are generally managed with drainage. Late stricture formation may be satisfactorily dealt with by endoscopic or percutaneous balloon dilatation or stenting. Graft ischemia either from hepatic artery thrombosis or portal vein thrombosis can be a devastating complication. Hepatic artery thrombosis represents a significant cause of graft loss and mortality after pediatric liver transplantation. The reported incidence of this complication is 5–7%. The incidence is much less frequent with the use of reduced-size liver transplants with larger size vessels and microsurgical vessel anastomosis techniques used for living donor transplants. Most centres recommend routine Doppler ultrasound in the early post-operative period ranging from 3–7 days to confirm the patency of these vessels. Consequences of vascular thrombosis are graft necrosis, intra-hepatic abscess, biliary necrosis and bile leakage. A massive rise in enzyme activity, particularly in the first few days after transplant, may be the first signs. Immediate intervention with thrombectomy and re-anastomosis may be successful if the diagnosis and treatment are carried out as soon as the complication is diagnosed in the first 3–5 days. If thrombectomy fails, urgent retransplant is required. Late thrombosis may be asymptomatic and if so can be ignored. Although technical factors usually account for most cases, it is advisable to maintain the hematocrit at around 30 to strictures of the donor bile duct with sludge for- improve microvascular flow, and most centres mation. Most biliary complications (72%) occur use aspirin and dipyrimidole as long-term prophylaxis, and in addition, some centres use intra(Kochhar et al. 2013). Ultrasound and cholangiography are the prin- venous or subcutaneous heparin in the first week cipal imaging modalities used for detection of post-transplant Portal vein thrombosis can occur early or late. these complications. It is imperative with all suspected biliary complications to ensure that the Early thrombosis usually presents with a degree hepatic artery is patent using Doppler ultrasound of liver dysfunction with prolonged clotting or angiography as hepatic artery thrombosis will while late thrombosis presents with portal hypercause ischemia and necrosis of the biliary tree. tension, which may be heralded by an esophageal Simple bile leaks are diagnosed in the early post-­ variceal bleed. Immediate thrombectomy may be operative period by the presence of bile in drain- successful. Where graft portal vein thrombosis is age fluid or in percutaneous aspirate of fluid established, a meso-portal (Rex) shunt, with a collections around the liver. Early biliary compli- vein graft taken from the internal jugular vein of Table 87.4  Summary of common post-operative problems 1. Biliary tract  Stenosis or stricture  Anastomotic leak—often associated with hepatic artery thrombosis  Infection 2. Rejection  Acute  Chronic (vanishing bile duct syndrome) 3. Infection—bacterial, viral, (CMV, EBV, Herpes Zoster, hepatitis B). fungal (Candida, Aspergillus), parasitic (pneumocystis)  Abdominal (peri- or intra-hepatic abscess)  Biliary tree  Pulmonary  Re-activated virus  Gastrointestinal tract  Catheter associated (intravenous, urinary tract) 4. Graft vascular injury (thrombosis, stenosis)  Hepatic artery  Portal vein  Inferior vena cava (supra and infra-hepatic)  Hepatic vein (left lateral segment grafts), Budd-­ Chiari recurrence, hepatic venous outflow obstruction (HVOO) 5. Renal dysfunction  Tacrolimus/cyclosporin or other drug-induced injury  Tubular necrosis due to hypo-perfusion  Pre-existing disease (hepato-renal syndrome)  Hypertension 6. Miscellaneous  Encephalopathy (cyclosporin, tacrolimus, hypertensive, metabolic)  Bowel perforation (steroid, diathermy)  Diaphragm paresis/paralysis  Gastrointestinal haemorrhage (peptic ulceration, varices)  Obesity (steroids)  Other drug side effects in the first 2 weeks following transplantation

1208

the patient or donor veins from a vascular bank (if available) and interposed between the superior mesenteric vein and the left branch of the portal vein, may be curative. Significant risk factors for portal vein thrombosis are young age and low weight at the time of liver transplantation, small portal vein and in urgent transplants, when the patient is in a generally poor condition. Overall risk of portal vein thrombosis (PVT) is 2–5%. Bowel perforation is a well-recognized complication following liver transplantation (+/-7%). Contributory factors include previous operation, steroid therapy, hypoxemia in porto-pulmonary syndrome and viral infection. The incidence is higher in children who underwent transplantation for biliary atresia after a previous Kasai portoenterostomy. Diagnosis may be difficult and a high index of suspicion is needed. Post-operative fluid collections arising from the cut surface of the liver has the reported incidence of 30% and 40% in which nearly half required intervention. These collections can be due to biliary anastomosis leaks, bile leaks from the cut surface of a partial liver graft or bowel perforation. Late presentations may be less acute and typically present with gram-negative sepsis, liver abscess or biliary complications. Inferior vena cava thrombosis may develop either in the immediate post-operative period presenting with ascites and lower body edema or later on due to regeneration of the graft and twisting of the caval anastomosis. Thrombolytic therapy may be successful in late thrombosis but should be avoided in early thrombosis because uncontrollable bleeding may occur from raw surfaces, particularly if a reduced/split liver was transplanted. Hepatic venous outflow obstruction (HVOO) is not an infrequent complication. This can be due to redundancy of hepatic vein (when the graft hepatic vein is kept long) or torsion in positioning of a partial graft. The correction of the redundancy or torsion is made by pulling the graft caudally and to the left or right side of the abdominal cavity as determined by Doppler ultrasonography and stabilizing the graft by suture of the falciform ligament to the diaphragm. HVOO can also be suspected if there is persistence of ascites

K. Sharif and A. J. W. Millar

in the early post-transplant period. This is usually confirmed by either an angiography or by liver biopsy findings of congestion and red cell extravasation around central veins. Diaphragmatic paresis and hernia are rare complications of liver transplantation. The possible role of several contributing factors includes cross-clamping of the IVC at the level of the diaphragmatic hiatus, trauma at operation (dissection and diathermy) and diaphragm thinness related to low weight and malnutrition.

87.13 Common Medical Complications Most patients can be discharged from the intensive care unit within the first week after transplantation. Common medical complications of transplantation include bacterial, viral, fungal and opportunistic infections, renal function impairment, hypertension, rejection and of particular concern is the post-transplant lymphoproliferative syndrome. Infections: The reported incidence of infection in the liver transplant population is 1.36 infection/ patient. The most common sites of infection are bloodstream (36%) and abdomen (30%). Gram-positive bacteria (78%) predominated over gram-negative bacteria (22%). Detailed analysis of risk factors shows that age < 1 year, body weight < 10 kg, extra-hepatic biliary atresia, intraoperative transfusion > 160 ml/ kg, mechanical ventilation > 8 days and PICU stay > 19 days are associated with a higher risk of infection. Acute rejection: Despite the availability of potent immunosuppressive drugs, rejection after organ transplantation in children remains a serious concern and may lead to significant morbidity, graft loss, and death of the patient. Diagnosis of rejection can be made on the basis of clinical, biochemical and histologic changes and usually presents in the first few weeks after transplant with fever, malaise, a tender graft and loose stools. Diagnosis is confirmed by liver biopsies performed using the Menghini technique Hypafix needle (Braun) or spring loaded Tru-Cut needles, unless biliary dilatation is observed on ultraso-

87  Pediatric Liver Transplantation

nography. Biopsy tissue is routinely assayed for viral and bacterial activity. The grade of rejection is assessed, according to established Banff histological criteria, on a scale of 0–4. Some centres are trying to evaluate non-invasive tools to diagnose acute rejection, such as radiologic findings on post-transplant Doppler ultrasound. Others are using Interleukin 5 (IL-5), it is produced in the liver and is a T cell-derived cytokine that acts as a potent and specific eosinophil differentiation factor in humans. During liver allograft rejection, intragraft IL-5 mRNA and eosinophilia have been observed. It may be useful as a specific marker of allograft rejection. Once diagnosed, acute rejection is treated with three doses of methyl prednisolone 10 mg/kg given intravenously on successive days with adjusted baseline immunosuppression. Some patients experience corticosteroid-resistant acute rejection, the management of which is not standardized. Various agents used include the addition of myco-­ phenolate mofetil or sirolimus. Other options are the use of anti-thymocyte globulins (ATG) or monoclonal anti-CD3 antibodies, muromonab CD3 (OKT3). In patients with refractory rejection despite therapeutic escalation, the risks of over-immunosuppression, including opportunistic infections and malignancies (especially the Epstein-Barr virus related post-transplant lymphoproliferative disorder) have to be balanced with the consequences of graft loss due to rejection. Late acute cellular rejection: Although acute rejection is mostly encountered during the first 3 months after liver transplant, it may occur later on. Late cellular rejection in children is usually due to low or decreased immunosuppression and is associated with long-term complications. Prompt intervention to correct inadequate immunosuppression and careful follow-up to identify other treatable conditions is essential. Antibody mediated rejection (AMR) is a well-­ known entity in kidney, heart, and pancreas transplants. AMR in the liver transplant population is a rare and possibly underdiagnosed condition. In 2016, the Banff working group on Liver Allograft Pathology published consensus guidelines to diagnosed AMR in liver transplants. AMR should

1209

be suspected in children with steroid-refractory rejection, especially in the presence of class II donor-specific antibodies (DSA) and presence of C4d staining, along with other features. Chronic rejection is an irreversible phenomenon, which is chiefly intrahepatic and ductular rather than a vascular phenomenon in contrast to other organ transplants. This is usually manifested by the disruption of bile duct radicals with development of the vanishing bile duct syndrome. The incidence seems less frequent with tacrolimus based immunosuppressive regimens as opposed to cyclosporin where an incidence of up to 10% has been recorded. Late chronic rejection may also be associated with a vasculopathy affecting larger arteries. Chronic graft hepatitis occurs in 20–30% of children after liver transplantation, but the prevalence and causes are not known. Serum liver associated autoantibodies are often positive. It is most frequently seen in children transplanted for cryptogenic cirrhosis (71%). However, neither hepatitis C nor hepatitis G infection was associated. Management is with re-introduction or increase in steroid dose. Cytomegalovirus (CMV) infection: Cytomegalovirus (CMV) infection (seroconversion or virus isolation) and CMV disease (infection plus clinical signs and symptoms) have a reported incidence of 37% and 12% respectively with significant morbidity and mortality. The high prevalence of CMV infections supports the view that clinical signs alone are inadequate to direct investigations for CMV. Cytomegalovirus (CMV) infection is best monitored with PP65 antigen and polymerase chain reaction (PCR) measurement of the virus. Ganciclovir/valgancyclovir remains an important therapeutic option for the prevention and treatment of CMV disease in transplant recipients. Prophylactic treatment with ganciclovir appears the best strategy to implement in high-risk patients. A rare association with cytomegalovirus (CMV) reactivation is haemophagocytic syndrome (HPS). It is a rare event, which is often fatal. These patients are treated with a combination of antiviral agents, immunomodulatory and supportive therapy.

1210

Epstein-Barr virus (EBV) and post-transplant lymphoproliferative disorder (PTLD). EBV infection is the main cause of PTLD. Since many infants are EBV seronegative at the time of transplantation, PTLD is a major concern for these patients. Post-transplantation lymphoproliferative disorder (PTLD) presents from the first few weeks after transplant to several years later with a mean time of onset around 9 months. First manifestations of PTLD are adenoidal and/or tonsillar involvement. A typical presentation is usually with acute membranous tonsillitis and associated cervical lymphadenopathy, which is resistant to antibiotic therapy. It is important to remember that tonsillar enlargement in pediatric liver transplant patients does not necessarily imply a diagnosis of PTLD.  Furthermore, the presence of increased numbers of EBV infected cells in tonsils from liver transplant recipients by itself does not indicate an increased risk of developing PTLD. However, the disease may be widespread and gastrointestinal and central nervous system involvement is common. Currently, there are no tests to accurately identify pediatric liver transplant patients at risk for post-transplant lymphoproliferative disorder (PTLD). Attempts have been made to use cytokine polymorphisms and real-time quantitative polymerase chain reaction (qPCR) Epstein-Barr virus (EBV) viral load to identify patients at risk for PTLD development. Use of cytokine genotyping, in conjunction with qPCR for EBV viral load, can significantly improve the predictive value of diagnostic tests for identification of patients at high risk for PTLD. Management strategies include reduction of immunosuppression, which may require complete withdrawal along with standard anti-­ lymphoma chemotherapy, particularly with the monoclonal type. Mortality varies from 20% to 70% or more. Prophylactic intravenous ganciclovir given for a prolonged period (two weeks minimum) may be effective in preventing EBV activation which is the promoter of PTLD in most cases. Rituximab, an anti-CD 20 monoclonal antibody has been used with good effect. As B

K. Sharif and A. J. W. Millar

cells are largely ablated, replacement immunoglobulin therapy is required until B cell recovery has occurred. Renal impairment: A degree of renal impairment is almost inevitable in those patients suffering from chronic liver disease and with the additional burden of the use of nephrotoxic immunosuppressive drugs, such as cyclosporin and tacrolimus with other nephrotoxic antibiotics and antifungal agents, may result in significant renal impairment of function in the long term. The importance of renal sparing strategies in immunosuppression is becoming increasingly evident as long-term survivors present with drug-­ induced renal failure. Retransplantation: Ten to 15% of patients may suffer graft failure at some time and need retransplantation. Early indications may be primary non-function, early hepatic arterial thrombosis, severe drug-resistant acute rejection and established chronic rejection. Early retransplantation is technically a much less traumatic procedure than the original transplant, although the patient may be in a poorer clinical condition. Outcome largely depends on the indication for retransplantation and is quite good for technical causes but less satisfactory for rejection and infection. An increasingly poorer outcome can be expected after third and fourth retransplants and the efficacy and ethics of these interventions are in question.

87.14 Long-Term Survival and Quality of Life One-year survival of > 95% is being achieved in the best centres, with predicted 10-year survivals of around 85–90% (Ramos-Gonzales et al. 2019; Kohli et al. 2018). Patients grafted for acute liver failure have done less well, with a higher early death rate usually associated with cerebral complications and multi-organ failure. Excellent quality of life can be achieved and most children are fully rehabilitated. It is, however, increasingly evident that prolonged cholestatic jaundice

87  Pediatric Liver Transplantation

and malnutrition in infancy may have late effects and despite good physical rehabilitation evidence of significant cognitive deficits, which present during early schooling as learning difficulties and attention deficit disorder, are common. Quality of life may not reach perfection, and also depends on the way society accepts these imperfections. As with any immunosuppressed patient, the incidence of neoplasia in a lifetime is greatly increased.

87.15 Conclusion Careful planning, extensive preparation of personnel and a broad base of skills, along with good teamwork between health professionals, are required for the development of a successful pediatric transplant program. These services are best provided within a pediatric hepatology and transplant centre. Surgical technique, anaesthetic skills, and medical care of the highest order are essential. A patient with a liver transplant is a patient for life and requires complete commitment from the transplant medical and surgical team, which cannot be abrogated after discharge from hospital. Endemic viral and bacterial infections particularly HBV, CMV, EBV and PTLD impact negatively on any program. Extended hospital stay may be required, and this, along with long-term therapy, may be extremely expensive. The need for pediatric liver transplants has been assessed at approximately 1–5 children per million per year. Thus, transplant activity should be concentrated in specific centres preferably doing more than 12 transplants a year. The shortage of donor organs will continue and future efforts must be focused on maximum use of cadaver donors and increasing living related donation. Transplant activity is rapidly increasing throughout the developing world. This endeavour should be strongly supported as poor socio-economic status and not a contraindication to transplantation. Parents with relatively few material resources have been shown to be able to diligently care for their children. No child

1211

with end-stage liver disease should be denied the opportunity of receiving appropriate treatment. As with any new development, knowledge and experience improve, costs decline and excellent outcomes are achieved. These challenges must be met to offer any infant or child requiring liver replacement, a chance of a life. The ultimate aim is to restore the child to normal health such that he/she can grow up into a productive healthy adult, who can make his/her contribution to society and develop all of his/ her human potential. A normal pregnancy can be expected in women who received transplants as children.

References Brenner DA, Thomas E (2017) Starzl: transplantation pioneer. Proc Natl Acad Sci U S A 114:10808–10809 Charlton M et al (2018) International Liver Transplantation Society consensus statement on immunosuppression in liver transplant recipients. Transplantation 102:727–743 Cuenca AG et  al (2017) Pediatric liver transplantation. Semin Pediatr Surg 26:217–223 de Ville de Goyet J et  al (1995) Standardized quick en-­ bloc technique for procurement of cadaveric liver grafts for pediatric liver transplantation. Transpl Int 8:280–285 Ferreira-Meirelles R et  al (2015) Liver transplantation: history, outcomes and perspectives. Einstein (Sao Paulo) 13:149–152 Fischler B et  al (2019) Similarities and differences in allocation policies for pediatric liver transplantation across the world. J Pediatr Gastroenterol Nutr 68:700–705 Kochhar G et  al (2013) Biliary complications following liver transplantation. World J Gastroenterol 19:2841–2846 Kohli R et  al (2018) Liver transplantation in children: state of the art and future perspectives. Arch Dis Child 103:192–198 Mazariegos G et al (2014) Liver transplantation for pediatric metabolic disease. Mol Genet Metab 111:418–427 Miller C et al (1988) Rapid flush technique for donor hepatectomy: safety and efficacy of an improved method of liver recovery for transplantation. Transplant Proc 20:948–950 Montenovo MI et al (2019) Living liver donation improves patient and graft survival in the pediatric population. Pediatr Transplant 23:e13318. https://doi.org/10.1111/ petr.13318

1212 Otte JB (2002) History of pediatric liver transplantation. Where are we coming from? Where do we stand? PediatrTranspl 6:378–387 Ramos-Gonzales G et  al (2019) Predictors of need for liver transplantation in children undergoing hepatoportoenterostomy for biliary atresia. J Pediatr Surg 54:1127–1131 Safinia N et  al (2018) Cell therapy in organ transplantation: our experience on the clinical translation of

K. Sharif and A. J. W. Millar regulatory T-cells. Front Immunol 9:e354. https://doi. org/10.3389/fimmu.2018.00354 Shneider BL (2002) Pediatric liver transplantation in metabolic disease: clinical decision making. Pediatr Transplant 6(1):25–29 Yersitz H et  al (2003) One hundred in situ split-liver transplantations. A single center experience. Ann Surg 238:496–507

Part X Genitourinary Disorders

Urinary Tract Infection

88

Thomas de los Reyes and Martin A. Koyle

88.1 Introduction UTIs may result from a multitude of bacterial pathogens, usually gram-negative bacteria, and although they can occur anywhere within the urinary tract, the kidneys (pyelonephritis) and bladder (cystitis) represent the most common sites manifested clinically. Anatomically, pyelonephritis is considered to be an upper UTI, while cystitis is considered to be a lower UTI. Pyelonephritis is most commonly associated with fever. It has been suggested that the younger the patient, infants and neonates, that are untreated, especially repeated episodes of pyelonephritis, can lead to renal scarring and permanent renal damage (Jacobson et al. 1989). Up to 40% of infants with a bout of pyelonephritis will show evidence of renal scarring upon subsequent radioisotope Tc99 DMSA nuclear scintigraphy of the kidneys (Hewitt et al. 2008). For this reason, the differentiation between upper and lower tract infections is critical. However, it is important to appreciate the limitations behind the evidence that shaped our initial understanding of the relationship between pyelonephritis, scarring, and long-term effects. More contemporary studies have shown that some scars

T. de los Reyes · M. A. Koyle (*) The Hospital for Sick Children and University of Toronto, Toronto, ON, Canada e-mail: [email protected]; [email protected]

noted on DMSA nuclear scintigraphy may be congenital, rather than as a result of infection (Pohl and Belman 2009). The degree of scarring may also be more relevant, rather than the presence of a scar in itself. In a long-term follow-up of patients with renal scarring, bilateral scarring was associated with a decrease in renal function, while unilateral scars were no different from those with no scarring (Wennerström et  al. 2000). It also appeared that the presence of renal scarring did not affect mean 24-h ambulatory blood pressure measurements (Wennerström et al. 2000). In addition to relating UTIs to the anatomic location, UTIs can also be classified in several other ways. A UTI can be categorized as complicated or uncomplicated or whether it is the first infection or a recurrent episode (Table 88.1).

88.1.1 Complicated vs. Uncomplicated UTI A UTI is considered complicated if the patient has any of the following factors: anatomic or functional abnormalities of the genitourinary system, recent instrumentation, retained foreign body such as urethral catheters, a documented fever, or occurring in an infant or neonate. Conversely, uncomplicated UTIs are infections that arise in the absence of any of these factors. Another important consideration that affects the management and clinical evaluation of a

© Springer Nature Switzerland AG 2023 P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_88

1215

1216 Table 88.1  Classification of UTI Anatomic location  Upper—kidney (pyelonephritis)  Lower—bladder (cystitis), urethra (urethritis) Complicating factors  Complicated infection—infant or neonate, fever, foreign body or stone, anatomic or functional abnormalities  Uncomplicated infection—simple lower tract infection without fever Initial or recurrent infection  First infection—evaluate if complicated or uncomplicated recurrent infection  Unresolved—urine cultures are always + with the same pathogen  Bacterial persistence—urine cultures become sterile after initial infection; recurrent infection occurs with the same bacteria  Re-infection—the original infection is eradicated, and then, a different bacterium is isolated with recurrent infection

child is whether the UTI is the first or a recurrent infection. Recurrent infections fall into three categories, namely, unresolved infections, infections with bacterial persistence, and true re-infections.

88.1.2 Unresolved Infection In patients with unresolved bacteriuria, initial urinary cultures, as well as all subsequent cultures, will show persistence of the same bacteria. In these cases, the original bacterial infection has never been cleared from the urinary tract. This is often the result of inadequate treatment, due to either poor compliance or antibiotic resistance. In recent years, uropathogens have developed increased resistance to commonly prescribed antibiotics for UTI.  Notable examples of this increasing resistance are Escherichia coli’s resistance to ampicillin and vancomycin-resistant enterococcal infections. The urine culture, at the time of initial presentation, will reveal not only the type of bacteria but also the sensitivity and resistance patterns to various antibiotics. A follow-­up of the results of a urine culture by the clinician can help prevent inadequate and inappropriate antibiotic treatments of the infection. Proof of cure urinalysis and culture, how-

T. de los Reyes and M. A. Koyle

ever, are typically not indicated in the vast majority of patients who have otherwise responded to initial therapy. Asymptomatic bacteriuria, in which a positive urine culture is obtained in patients who lack symptoms or signs of an infection, can result in unnecessary treatment which in turn can lead to antibiotic resistance, adverse reactions, and increasing healthcare costs. Counseling of the patient and caregiver is also important to ensure that prescribed antimicrobials be taken for the entire prescribed course even if symptoms resolve prior to course completion.

88.1.3 Bacterial Persistence In patients with bacterial persistence, the same bacteria are cultured in the urine, despite the initiation of and adherence to sensitivity-adjusted therapy. Bacterial persistence can indicate an occult nidus or reservoir for the infective process. The nidus for persisting infection can be retained foreign body, such as a piece of catheter or ureteral stent, or a chronically infected stone. In addition to an actual physical nidus for recurrent infection, an anatomic or functional obstruction can also act as a reservoir for these infections. Obstruction of urinary flow eliminates the normal antegrade flow of urine and promotes urinary stasis and chronic infection. Anatomic obstruction of the urinary system may occur at different levels of the urinary tract. Supra-vesical obstruction includes congenital anomalies, such as ureteropelvic junction obstruction or obstructed megaureter, while bladder outlet obstruction can occur from posterior urethral valves, to name a few. Functional obstruction can include conditions such as neurogenic bladder, whereby both bladder storage and emptying can be affected. This would be in the form of elevated filling pressures resulting in hydroureteronephrosis or impaired bladder contractility resulting in urinary stasis. Such occurrences, in turn, can serve as a reservoir for infection in either the upper or lower urinary tract. Both functional and anatomic obstructions promote recurrent infections by eliminating the nor-

88  Urinary Tract Infection

mal antegrade flow of urine which is an important physiologic host defense factor.

88.1.4 Re-infection In bacterial re-infection, the initial infection is resolved, and the patient has another documented UTI. Re-infection can occur with the same pathogen but of differing serologic strain or with a different pathogen as a whole. Re-infection may be a recurrent problem in patients who are genetically susceptible to UTIs. Treatment strategies for patients with recurrent infections include treatment with appropriate antibiotics, surgery to address anatomic anomalies if present, and prophylaxis with daily antibiotics if the re-infections are common or accompanied by complicating factors.

88.2 Historical Overview UTIs and subsequent sepsis carried a high risk of death in children prior to the advent of effective antibiotic therapy, with mortality rates as high as 20% previously described (Zorc et al. 2005). The development of antimicrobials, improved diagnostic capabilities including readily available point-of-care screening, and an overall understanding of sepsis management mean that UTIs are no longer the fatal illness it once was. However, UTIs continue to cause significant morbidity warranting adequate understanding of clinicians in its diagnosis and management.

88.3 Incidence During the first few months of life, several studies have shown that boys have a higher incidence of UTI at up to 14% compared to girls at 7% (Shaikh et al. 2008). It has been substantiated that uncircumcised boys, in particular, are the most prone to UTI within the first year of life, with a tenfold risk compared to circumcised males (Shaikh et  al. 2008). Despite this, circumcision remains a controversial subject as a protective element for this

1217 Table 88.2  Yearly incidence of UTI in pediatric patients Age 0–1 year 1–5 years 6–16 years

Boys% 2.7 0.1–0.2 0.04–0.2

Girls % 0.7 0.9–1.4 0.7–2.3

Source: Foxman B. Epidemiology of urinary tract infections: incidence, morbidity, and economic costs. Am J Med. 2002; 113(1A):5S–13S

indication only. In high-risk situations, however, the surgeon must weigh the risks and benefits. After the first few months of life and continuing into adulthood, girls have a higher overall incidence of UTI. In children aged 1–5 years, the incidence of UTI per year in boys is 0.1–0.2%, and in girls, it is 0.9–1.4% (Foxman 2002). From the age of 6 to 16, the reported incidence of UTI is stable. In boys, it is up to 0.4% per year, and in girls, it approaches 2.3% per year (Table  88.2; Foxman 2002). With the onset of sexual activity, this incidence climbs to 11% in women, but is not very prevalent in men. Overall, about two-thirds of women will have a urinary tract infection throughout their life (Foxman 2002).

88.4 Etiopathogenesis Some hypothesize that in the first several months of life, UTIs may be caused by hematogenous spread of pathogens from transient bacteremia. After this time period, however, most accept that UTIs more frequently arise from fecal pathogens. In these infections, ascending bacteriuria from contamination and colonization of the perineum and urethral meatus is thought to be the main factor in the development of UTI. This may offer an anatomic explanation regarding the increased incidence of UTIs in females. The peri-urethral folds and the moist environment of the vagina promote bacterial colonization around the urethral meatus. In addition, the shorter length of the urethra allows ascending infection to spread more easily to the bladder than in the male. Bacterial virulence factors enable some to adhere to the urothelial lining of the genitourinary tract. The process of colonization and subsequent proliferation is a complex balance between

T. de los Reyes and M. A. Koyle

1218 Table 88.3  Common uropathogens Bacteria Escherichia coli Enterococcus spp. Klebsiella spp. Serratia spp. Staphylococcus aureus Pseudomonas aeruginosa Enterobacter cloacae Streptococcus spp. Proteus spp.

Incidence % 50–93 0–17 0–10 ~1 ~1 ~1 ~1 ~1 ~1

Source: Edlin RS, Shapiro DJ, Hersh AL, Copp HL. Antibiotic resistance patterns of outpatient pediatric urinary tract infections. J Urol. 2013;190(1):222–227

the host’s immune factors and the virulence factors expressed by the bacteria. Proximal involvement of the kidneys is believed to result from an ascending infection originating from the bladder, either by vesicoureteral reflux (VUR) or in a third or more patients without VUR, due to the complex relationship between host immunity and susceptibility and innate pathogenicity and properties of that bacteria, as discussed below (Montini et al. 2011). Numerous organisms including viruses, yeast, and bacteria can cause UTI. Bacterial infection, however, is the most common cause of UTI in an otherwise healthy, immunocompetent child. Of the uropathogenic bacteria, E. coli is the most common cause of UTI in both adult and pediatric populations. E. coli account for up to 83% of bacterial UTIs (Edlin et al. 2013; Shaw et al. 1998). Other common bacteria that cause urinary tract infections include Enterococcus spp., Klebsiella spp., Enterobacter cloacae, Serratia spp., Staphylococcus aureus, Pseudomonas aeruginosa, Streptococcus spp., and Proteus spp. (Table 88.3).

88.5 Risk Factors 88.5.1 Gender Gender plays a large role in susceptibility to UTI. During the first year of life, UTIs are more common in boys than in girls. This is more pro-

nounced in febrile infants within the first 6 months after birth, where the incidence is comparatively greater for uncircumcised boys as noted earlier. Thus, the foreskin, especially the unretractile prepuce that is normal in neonates and infancy, appears to be a harbinger for early male UTI.  After the first year of life, the incidence of UTI changes with girls having a higher incidence than boys well into adulthood.

88.5.2 Circumcision Status Circumcision remains the most common surgical procedure performed in the United States (Morris et al. 2014). There is evidence that show the benefits of circumcision in decreasing the risk of sexually transmitted diseases, HIV infection, development of penile carcinoma, and UTIs (Morris et  al. 2014). In uncircumcised boys, the risk of UTI is approximately ten times more in the first year of life compared to their circumcised counterparts. The physiologic reason for this increased incidence of infection is well elucidated. During the first year of life, there is increased colonization of the peri-urethral tissue of the inner prepuce, allowing the retrograde ascent of infection more readily into the bladder. The advantage conferred by circumcision is diminished in older boys as the overall incidence of UTI decreases with increasing age. In a meta-­analysis looking into circumcision for the prevention of urinary tract infections published in 2005, 111 circumcisions would need to be performed to prevent 1 UTI (Singh-Grewal et  al. 2005). Routine circumcision, however, remains controversial for a variety of cultural, psychosocial, and ethical reasons. Combined with a relatively low reported UTI incidence of 1–3% in boys less than 1 year of age, many practitioners view the psychosocial aspects of routine circumcision may outweigh any potential medical benefits from the procedure. Although not the purpose of this chapter, the absence of high-­quality data, as reflected by the unsuccessful attempt by the Cochrane collaboration in 2012, means that cir-

88  Urinary Tract Infection

cumcision continues to be a contentious issue with conflicting guidelines (American Academy of Pediatrics Task Force on Circumcision 2012; Jagannath et al. 2012).

88.5.3 Previous Infection One of the most significant risk factors for UTI is a previous episode of infection. Of children who have a UTI in the first year of life, 23% of boys and 12% of girls are fated to have a further UTI in the several months following their initial ­infection (Shim et al. 2009). The number of previous infections also increases the likelihood of recurrent infection (Keren et al. 2015). This risk may be reflective of changes that occur physiologically in the urinary tract after a UTI which predispose patients to recurrent infections or may also be secondary to selection of patients who are at high risk for UTI due to genetic or anatomic factors.

88.5.4 Bladder and Bowel Dysfunction Bladder and bowel dysfunction (BBD), previously termed dysfunctional voiding and dysfunctional elimination syndromes, describes any abnormalities in the storage or emptying of urine, but recognizes the importance of fecal elimination, specifically constipation (Austin et  al. 2016). Increased fecal load is thought to affect bladder storage and emptying, possibly through direct mechanical compression or through inducing changes in physiological neural stimuli leading to decreased urge to evacuate (Malykhina et al. 2017). The resultant urinary stasis may then increase the risk of UTIs. The Careful Urinary Tract Infection Evaluation (CUTIE) study enrolled 195 children without VUR who had 1 or 2 febrile or symptomatic UTI. This study found

1219

the risk of febrile or symptomatic UTI recurrence was 35% in children with BBD (Keren et  al. 2015; Shaikh et al. 2016). Multiple studies of differing quality have shown the same conclusions regarding the importance of BBD and UTI recurrence (Meena et al. 2020).

88.6 Pathophysiology Most pathogenic bacteria that cause UTI arise from a reservoir in the intestinal tract. As mentioned, E. coli is by far the predominant bacterium that causes UTI. One way of differentiating E. coli strains is based upon differences in the antigens expressed on the polysaccharide capsule that surrounds the bacteria (Tullus et  al. 1991). These antigens are known as K antigens, and it has been demonstrated that certain K antigenic E. coli have a much higher propensity for causing UTI than other strains. Another predictor of a bacteria’s uropathic potential is its ability to adhere to the urothelium where they cause infection. Pili or fimbriae are long filamentous appendages composed of protein that project from the bacterial surface and allow this adhesion to take place. In E. coli, type 1 pili are highly associated with bacteria that cause UTI.  Type 1 pili bind uroplakin, a protein cap that is expressed by the urothelial cells. Another form of pili, P pili, is highly associated with strains of E. coli that cause pyelonephritis. Bacterial adherence, colonization, and subsequent infection are a complex process that involves a balance between bacterial virulence factors and the host’s immune response to invasive bacterial infection and colonization.

88.7 Pathology The urothelium of the collecting system is comprised of a layer of superficial umbrella cells, an intermediate layer, and basal layers. The latter is

1220

thought to be the location of stem cells that enable umbrella cell regeneration following regular turnover similar to epithelial tissue in other parts of the body. Urinary pathogens, such as E. coli, are found exclusively in the umbrella cell layer in part due to adherence factors described above (Hickling et  al. 2015). A gram stain on cross sections of pathologic specimens enables the detection and differentiation of bacteria based on its staining properties and morphologic appearance.

88.8 Diagnosis Diagnosis of UTI in an infant and neonate may be quite difficult as the signs and symptoms of a UTI are nonspecific. Signs and symptoms of UTI in a febrile infant can include abdominal, suprapubic, or flank pain, poor appearance, jaundice, foul smelling urine, and lack of other identifiable etiology for an unexplained fever (Hoberman et al. 1993). Among febrile infants in the first 2 years of life without an identifiable etiology for fever and a temperature greater than 38.0 °C, it was found that the incidence of UTI was 3.3% (Craig et al. 2010). UTI is one of the more common serious causes of infant and neonatal sepsis, and in the first 8 weeks of life, the incidence of UTI in a child with an unexplained febrile illness is as high as 13% (Crain and Gershel 1990). In any unexplained febrile illness in a child who is not capable of verbalizing the symptoms, a high suspicion of UTI must be maintained.

88.8.1 Urinalysis, Microscopy, and Culture A urine analysis (UA) will provide information about the urine which can suggest the present of UTI. Nitrite, when present, suggests bacteria in the urine as it is a byproduct of bacterial metabolism of nitrates. Although many gram-positive and gram-negative bacteria are capable of this conversion, Enterococcus spp., the second or third leading cause of UTI in pediatric patients, do not possess this ability. Thus, the test is spe-

T. de los Reyes and M. A. Koyle

cific but not sensitive. A UA in a patient with a UTI will often also show the presence of leukocyte esterase. Leukocyte esterase is an enzyme released when white blood cells are lysed in urine and indicates a significant degree of leukocyturia. An additional important part of the UA is the microscopic examination of the urine. The collected urine sample is typically centrifuged at 2000 revolutions per minute for 10 min (Lin et al. 2000a). The resultant pellet is then examined under microscope. When there are greater than five white blood cells (WBC) per high-power field (hpf), significant leukocyturia exists. As the number of WBC/hpf increases, the positive predictive value of the microscopic exam for UTI is greater (Lin et al. 2000b). Additional sensitivity can be achieved with the use of a hemocytometer, which measures the number of WBCs in an uncentrifuged specimen of urine. The sensitivity of predicting UTI accurately by finding greater than 10  WBC/hpf using the hemocytometer is 23% higher as compared to finding greater than 5  WBC/hpf on microscopic exam (Lin et  al. 2000a, b). Overall, the positive predictive value for a UTI when there is nitrite and leukocyte esterase on urinalysis coupled with bacteria and leukocytes in microscopy approaches 100% (Lohr et al. 1993). Conversely, the negative predictive value for UTI when these factors are not present also approaches 100% (Lohr et al. 1993). When a UTI is suspected, based upon clinical suspicion, and a UA with microscopy suggests the diagnosis, bacterial culture of the urine remains the gold standard for establishing the diagnosis of UTI. A culture should be sent before the initiation of antibiotic therapy. However, the initiation of antimicrobial therapy should not be delayed as urine culture results typically return 24–48 h after collection. Traditionally, growth of greater than 105 colony-forming units (CFU) of bacteria within a urine culture indicates significant infection. This limit has been challenged in adults and many symptomatic women who are found to have fewer bacteria than 105 CFU and often develop significant UTI if they are not treated. Similarly, significant UTI can be present in febrile children, despite having fewer than 105 CFU.  In one study, up to

88  Urinary Tract Infection

27% of febrile infants presenting with UTI only grow 104–105 CFUs upon urinary culture (Lin et  al. 2000a, b). For this reason, the American Academy of Pediatrics guidelines on urinary tract infections, published in 2016, have changed the diagnosis definition to the presence of pyuria and at least 50,000 CFUs/mL of a single uropathogen (Subcommittee on Urinary Tract Infection AAP 2016). The urine culture will also reveal the bacteria responsible for the UTI as well as the antibiotic sensitivity and resistance pattern of the bacteria. This then allows for focused antibiotic treatment strategies.

88.8.2 Urine Collection The simplest collection method for a sample of urine for analysis and bacterial culture is a bag taped to the perineum. This method is helpful if the urine analysis is negative, making infection very unlikely. However, this method is not ­helpful if the urine suggests infection since genital, perineal, and fecal contamination can easily give a false positive UA.  Two other methods have been commonly employed to obtain a sterile sample of urine when the proper diagnosis is critical, such as in an infant or neonate with unexplained febrile illness. These methods are suprapubic aspiration of the bladder and sterile urethral catheterization. To obtain a urine sample by suprapubic aspiration, a fine needle (23 or 25 gauge) is passed into the bladder by puncture directly above the pubic symphysis. The needle is directed perpendicularly or with a slight caudal angle and aspirated as it is advanced until urine is obtained. Point-of-care ultrasonography may also help with identification of the bladder and needle guidance. Urethral catheterization is best accomplished with a 5 French feeding tube or specifically designed urethral catheters which are usually designed with a more compliant material. The technique should be performed with as strict adherence to sterile conditions as possible to prevent contamination of the sample. Both suprapubic aspiration and urethral catheterization yield acceptable sensitivity for UTI. In

1221

a recent analysis of pain perceived by caregivers and parents, urethral catheterization appeared to be better tolerated by infants and neonates than suprapubic aspiration (Kozer et  al. 2006). Urethral catheterization is a simple method which may be better tolerated by patients and is possibly also perceived less invasive by parents. Urethral catheterization may not be feasible when it is complicated by anatomic problems such as a dilated urethra or labial adhesions. In some instances, urethral catheterization may be aided by perineal pressure or gentle flushing of the catheter with sterile saline as it is advanced. When urethral catheterization is not achievable, a urine sample can usually be easily obtained by suprapubic aspiration.

88.8.3 Renal-Bladder Ultrasound A renal-bladder ultrasound (RBUS) should be performed in patients with a febrile urinary tract infection (Subcommittee on Urinary Tract Infection AAP 2016). The purpose of the ultrasound is to assess anatomic features, such as renal size, architecture, and possible abnormalities such as hydronephrosis or scarring that may require further investigations. The timing of when a RBUS should be performed remains controversial. Based on the recommendations by the American Academy of Pediatrics (AAP) and the UK National Institute for Health and Clinical Excellence (NICE), a RBUS should be performed in patients who are seriously ill or are not exhibiting the expected response to treatment within the first 48 h of initiation. If a child responds well to antibiotics, imaging can be postponed to within 6 weeks after the infection.

88.8.4 Further Work-Up The full spectrum of imaging to identify urologic pathology, such as a voiding cystourethrogram (VCUG) and renal scintigraphy, is beyond the scope of this chapter and will be addressed later in the text. It is pertinent to note, however, in the context of febrile urinary tract infections that

1222

recent guidelines including from the AAP recommend performing a VCUG in patients with abnormalities on ultrasound, such as hydronephrosis, or in those with recurrent febrile UTI (Subcommittee on Urinary Tract Infection AAP 2016). This is due to the association between vesicoureteral reflux and pyelonephritis and potential for subsequent renal damage.

T. de los Reyes and M. A. Koyle

There are several controversies surrounding the treatment of pediatric UTI.  These controversies include the timing of treatment initiation, the time course of adequate treatment, treatment with oral versus parenteral antibiotics, and the need and timing of radiologic imaging. An approach to treatment of pediatric UTI can be based on whether it is complicated or uncomplicated (Koyle & Shifrin 2012).

In a Cochrane collaboration review between short-term (2–4 days) and long-term therapy (7–10 days), there was no evidence that longterm therapy was advantageous (Michael et  al. 2003). The recurrence rate and re-infection rate remained the same up to 15 months after treatment with either short- or long-term therapy (Michael et  al. 2003). A more recent Cochrane collaboration review published in 2012 compared conventional 10-day antibiotic therapy to singledose therapy (Fitzgerald et  al. 2012). This had similar outcomes to the aforementioned study, with regard to having no difference in recurrence or re-­infection rate. However, persistent bacteriuria at the end of treatment was reported in 24% of children receiving single-dose therapy compared to 10% of children who received conventional 10-day therapy (Fitzgerald et  al. 2012). Although there is not enough evidence that definitively shows superiority of one duration versus another, the totality of available evidence points toward a 3–4-day course as reflected by the European Society of Pediatric Urology/European Association of Urology guidelines which is the most recently available across all relevant bodies (Stein et al. 2015). It is important to emphasize that children with uncomplicated UTI are more likely under the care of a primary care provider. The surgical consultant is usually involved with a child with a febrile UTI and a known or suspected genitourinary surgical issue.

88.10.1 Uncomplicated UTI

88.10.2 Complicated UTI

When an uncomplicated UTI has been identified based upon symptoms, a UA suggesting infection, or a positive urine culture, oral antibiotic therapy should be initiated. The most common pathogen is E. coli, and empiric antibiotic therapy should be guided by local resistance patterns. Options for empiric therapy include narrow-­ spectrum antibiotics such as nitrofurantoin, sulfonamide-­ based antibiotics such as trimethoprim-­ sulfamethoxazole (TMPSMX), or cephalosporins such as cefuroxime. The duration of oral antibiotic therapy has been an area of controversy in pediatric UTI treatment.

A common scenario presented to the consulting surgeon will be an unexplained high fever in a neonate or infant (Fig. 88.1). In this scenario, the clinical should make a clinical judgment about the need for admission to the hospital. If the child shows evidence of dehydration and toxemia, and if admission to the hospital is warranted, parenteral antibiotics are typically administered. Broad-spectrum antibiotics are initiated after proper urine collection for urinalysis and culture and sensitivity has been performed. If the child improves in 24–48  h, these can be switched to oral antibiotics to complete a full course of treat-

88.9 Differential Diagnosis Pediatric fever of unknown origin has a broad differential including infectious, autoimmune, or oncologic. Infectious causes can be further divided either into viral, bacterial, fungal, and other sources or by organ system, such as pneumonia, meningitis, or urinary tract infections.

88.10 Management

88  Urinary Tract Infection

1223

2 month – 2 year old with unexplained fever

Consider UTI

Does the infant’s “toxicity” warrant immediate antimicrobial therapy?

YES

Obtain urine specimen by SP aspiration or urethral catheterization; culture urine

NO Option: Perform urinalysis on specimen collected by most convenient method

Option: Obtain urine for culture by SP aspiration or urethral catheterization

Initiate antimicrobial therapy parenterally; consider hospitalization

YES Urinalysis positive for leukocytes, nitrite or WBCs? NO UTI unlikely in the absence of specific symptoms. Follow clinical course. Reconsider UTI if fever persists.

Urine Culture positive?

YES

7-14 day of antimicrobial therapy guided by culture results

NO Clinical response within 48 hours?

NO UTI

YES

Renal and Bladder Ultrasound as soon as convenient

NO Renal and Bladder Ultrasound now; consider repeating urine and blood culture

VCUG or Renal Nuclear scan if ultrasound shows abnormalities or recurrent/ atypical infection

Fig. 88.1  Algorithm for the treatment of infants with unexplained fever

ment spanning 7–14 days. If the urine culture proves to be positive, antibiotic therapy can be tailored to the sensitivities of the bacteria responsible for the episode. Many antibiotic regimens have been utilized for the treatment of pyelonephritis. The Cochrane collaboration published in 2014 reviewed trials that compared regimens and outcomes of pyelonephritis (Strohmeier et  al. 2014). There was no difference in the various treatment regimens published for the treatment of pyelonephritis (Strohmeier et al. 2014). Most clinicians utilize a third- or fourth-generation cephalosporin such as ceftriaxone possibly in conjunction with an aminoglycoside during the acute phase of treatment. The patient is then tran-

sitioned to an equivalent oral medication once fevers have abated, usually after approximately 48  h. By this time, urine culture results should also be available to tailor antibiotic therapy based on sensitivities. In the same Cochrane review above, the evidence also showed that oral ­antibiotics alone are as effective as a short course (3–4 days) of IV antibiotics followed by oral therapy for a total duration of 10–14 days in appropriately selected patients (Strohmeier et al. 2014). Typically, these are children who have a fever but are otherwise able to tolerate oral hydration and non-toxic appearing. Oral antibiotic agents that concentrate in the urine but obtain poor tissue and blood levels should be avoided in

1224

patients who have had a febrile component to their UTI. Nitrofurantoin is one of these antibiotics, which remains a good choice for uncomplicated UTI and UTI prophylaxis.

88.10.3 Catheter-­Associated UTI Catheter-associated UTI (CAUTI) is among the most frequent cause of healthcare-associated infections in the pediatric setting. CAUTI is defined by the United States Centers for Disease Control and Prevention as “a UTI where an indwelling urinary catheter was in place for more than 2 consecutive days in an inpatient location on the date of the event, with the day of device placement being day 1, and an indwelling catheter was in place on the date of event or the day before.” Complications of CAUTI include ­pyelonephritis, sepsis, and death. Risk factors for CAUTI include duration of catheterization, with those having a catheter in place for a longer period at an increased risk of 8% for each additional day that the catheter remained in place (Gould et al. 2010). A history of prior catheterization has also been shown to increase the risk of CAUTI, possibly due to the colonization of the urethra persisting after catheter removal (Gould et al. 2010). This then serves as a nidus for infection upon re-catheterization due to disruption of the urothelium. It is not clear, however, as to what time interval constitutes a positive history for prior catheterization. Treatment strategies to tackle CAUTI include using urinary catheters only when indicated, aseptic techniques in catheter insertion, and daily evaluation regarding the need for a catheter and potential catheter removal (Gould et al. 2010). Unfortunately, most prevention strategies are evaluated as a CAUTI prevention bundle rather than the different components

T. de los Reyes and M. A. Koyle

of the above. Therefore, it is difficult to discern which element has the greatest impact in CAUTI prevention or reduction.

88.10.4 Antibiotic Prophylaxis The use of continuous antibiotic prophylaxis in pediatric UTI remains a controversial issue, especially in the era of increased antimicrobial resistance by common uropathogens such as E. coli. In several randomized trials involving patients with known vesicoureteral reflux, the use of prophylactic antibiotic therapy has been shown to decrease the incidence of recurrent febrile UTI (Brandström et  al. 2011; RIVUR Trial Investigators et al. 2014). However, the benefits of antibiotic prophylaxis in preventing renal injury are controversial. Thus, most clinicians would agree that beginning antibiotic prophyTable 88.4  Surgical conditions leading to febrile UTI Anatomic Vesicoureteral reflux Ureterocele Ureteropelvic junction obstruction Posterior urethral valves Ureteral or renal calculi Obstructed megaureter Functional Neurogenic bladder—Myelomeningocele Dysfunctional voiding—Hinman’s syndrome

laxis is warranted in young infants after an epiTable 88.5  Prophylactic antibiotic regimens for prevention of UTI Trimethoprim—1–2 mg/kg daily Trimethoprim-sulfamethoxazole (TMP-SMX)—1–2 mg/ kg TMP, 10–15 mg/kg SMX Nitrofurantoin—1–2 mg/kg daily Cefixime—2 mg/kg daily

88  Urinary Tract Infection

sode of pyelonephritis until imaging studies have been completed to rule out surgical conditions leading to febrile UTI (Table 88.4). Many different antibiotics have been utilized to prevent recurrent UTI.  The ideal antibiotic would have little effect upon the flora of the gut and be easy to administer for the child and the parent since compliance is a major issue with these regimens. Common regimens used include nitrofurantoin, cefixime, or trimethoprim (Table 88.5).

1225

trolled trial in children with dilating vesicoureteral reflux. J Pediatr Urol 7(6):594–600 Craig JC, Williams GJ, Jones M et  al (2010) The accuracy of clinical symptoms and signs for the diagnosis of serious bacterial infection in young febrile children: prospective cohort study of 15 781 febrile illnesses. BMJ 340:c1594 Crain EF, Gershel JC (1990) Urinary tract infections in febrile infants younger than 8 weeks of age. Pediatrics 86(3):363–367 Edlin RS, Shapiro DJ, Hersh AL, Copp HL (2013) Antibiotic resistance patterns of outpatient pediatric urinary tract infections. J Urol 190(1):222–227 Fitzgerald A, Mori R, Lakhanpaul M, Tullus K (2012) Antibiotics for treating lower urinary tract infection in children. Cochrane Database Syst Rev (8):CD006857 88.11 Conclusion Foxman B (2002) Epidemiology of urinary tract infections: incidence, morbidity, and economic costs. Am UTI in pediatric patients can arise from a variety J Med 113(1A):5S–13S of causes, encompassing simple cystitis to life-­ Gould CV, Umscheid CA, Agarwal RK, Kuntz G, Pegues DA, Healthcare Infection Control Practices threatening urosepsis secondary to surgical Advisory Committee (2010) Guideline for prevention anomalies of the urogenital tract. There are a of catheter-­associated urinary tract infections 2009. number of complex host and bacterial virulence Infect Control Hosp Epidemiol 31(4):319–326 factors that also make children susceptible to Hewitt IK, Zucchetta P, Rigon L et al (2008) Early treatment of acute pyelonephritis in children fails to reduce UTI. It is important that the clinician be familiar renal scarring: data from the Italian Renal Infection with the spectrum of UTI and children at risk of Study Trials. Pediatrics 122(3):486–490 UTI, so that infants and children who need to be Hickling DR, Sun TT, Wu XR (2015) Anatomy and physievaluated for anatomic or functional abnormaliology of the urinary tract: relation to host defence and microbial infection. Microbiol Spectr 3(4). https://doi. ties of the urinary tract are correctly identified. org/10.1128/microbiolspec.UTI-­ 0016-­2012 The hallmark sign of one of these children is an Hoberman A, Chao HP, Keller DM, Hickey R, Davis HW, infant in the first few years of life with evidence Ellis D (1993) Prevalence of urinary tract infection in of a UTI and high fever. This child must be confebrile infants. J Pediatr 123(1):17–23 sidered to have pyelonephritis and an anatomic Jacobson SH, Eklöf O, Eriksson CG, Lins LE, Tidgren B, Winberg J (1989) Development of hypertension and abnormality, until proven otherwise. uraemia after pyelonephritis in childhood: 27 year follow up. BMJ 299(6701):703–706 Jagannath VA, Fedorowicz Z, Sud V, Verma AK, Hajebrahimi S (2012) Routine neonatal circumciReferences sion for the prevention of urinary tract infections in infancy. Cochrane Database Syst Rev 11:CD009129 American Academy of Pediatrics Task Force on Keren R, Shaikh N, Pohl H et al (2015) Risk factors for Circumcision (2012) Male circumcision. Pediatrics recurrent urinary tract infection and renal scarring. 130(3):e756–e785 Pediatrics 136(1):e13–e21 Austin PF, Bauer SB, Bower W et al (2016) The standard- Koyle MA, Shifrin D (2012) Issues in febrile urinary ization of terminology of lower urinary tract functract infection management. Pediatr Clin North Am tion in children and adolescents: Update report from 59(4):909–922 the standardization committee of the International Kozer E, Rosenbloom E, Goldman D, Lavy G, Rosenfeld Children’s Continence Society. Neurourol Urodyn N, Goldman M (2006) Pain in infants who are younger 35(4):471–481 than 2 months during suprapubic aspiration and transBrandström P, Jodal U, Sillén U, Hansson S (2011) The urethral bladder catheterization: a randomized, conSwedish reflux trial: review of a randomized, controlled study. Pediatrics 118(1):e51–e56

1226 Lin DS, Huang FY, Chiu NC et  al (2000a) Comparison of hemocytometer leukocyte counts and standard urinalyses for predicting urinary tract infections in febrile infants. Pediatr Infect Dis J 19(3):223–227 Lin DS, Huang SH, Lin CC et  al (2000b) Urinary tract infection in febrile infants younger than eight weeks of Age. Pediatrics 105(2):E20 Lohr JA, Portilla MG, Geuder TG, Dunn ML, Dudley SM (1993) Making a presumptive diagnosis of urinary tract infection by using a urinalysis performed in an on-site laboratory. J Pediatr 122(1):22–25 Malykhina AP, Brodie KE, Wilcox DT (2017) Genitourinary and gastrointestinal co-morbidities in children: The role of neural circuits in regulation of visceral function. J Pediatr Urol 13(2):177–182 Meena J, Mathew G, Hari P, Sinha A, Bagga A (2020) Prevalence of bladder and bowel dysfunction in toilet-­ trained children with urinary tract infection and/or primary vesicoureteral reflux: a systematic review and meta-analysis. Front Pediatr. 8(84) Published 2020 Mar 31 Michael M, Hodson EM, Craig JC, Martin S, Moyer VA (2003) Short versus standard duration oral antibiotic therapy for acute urinary tract infection in children. Cochrane Database Syst Rev (1):CD003966 Montini G, Tullus K, Hewitt I (2011) Febrile urinary tract infections in children. N Engl J Med 365(3):239–250 Morris BJ, Bailis SA, Wiswell TE (2014) Circumcision rates in the United States: rising or falling? What effect might the new affirmative pediatric policy statement have? Mayo Clin Proc 89(5):677–686 Pohl HG, Belman AB (2009) The “top-down” approach to the evaluation of children with febrile urinary tract infection. Adv Urol 2009:783409. https://doi. org/10.1155/2009/783409 RIVUR Trial Investigators, Hoberman A, Greenfield SP et  al (2014) Antimicrobial prophylaxis for children with vesicoureteral reflux. N Engl J Med 370(25):2367–2376

T. de los Reyes and M. A. Koyle Shaikh N, Hoberman A, Keren R et al (2016) Recurrent urinary tract infections in children with bladder and bowel dysfunction. Pediatrics 137(1):e20152982 Shaikh N, Morone NE, Bost JE, Farrell MH (2008) Prevalence of urinary tract infection in childhood: a meta-analysis. Pediatr Infect Dis J 27(4):302–308 Shaw KN, Gorelick M, McGowan KL, Yakscoe NM, Schwartz JS (1998) Prevalence of urinary tract infection in febrile young children in the emergency department. Pediatrics 102(2):e16 Shim YH, Lee JW, Lee SJ (2009) The risk factors of recurrent urinary tract infection in infants with normal urinary systems. Pediatr Nephrol 24(2):309–312 Singh-Grewal D, Macdessi J, Craig J (2005) Circumcision for the prevention of urinary tract infection in boys: a systematic review of randomised trials and observational studies. Arch Dis Child 90(8):853–858 Stein R, Dogan HS, Hoebeke P et al (2015) Urinary tract infections in children: EAU/ESPU guidelines. Eur Urol 67(3):546–558 Strohmeier Y, Hodson EM, Willis NS, Webster AC, Craig JC (2014) Antibiotics for acute pyelonephritis in children. Cochrane Database Syst Rev (7):CD003772. Published 2014 Jul 28 Subcommittee on Urinary Tract Infection (2016) Reaffirmation of AAP clinical practice guideline: the diagnosis and management of the initial urinary tract infection in febrile infants and young children 2-24 months of age. Pediatrics 138(6):e20163026 Tullus K, Jacobson SH, Katouli M, Brauner A (1991) Relative importance of eight virulence characteristics of pyelonephritogenic Escherichia coli strains assessed by multivariate statistical analysis. J Urol 146(4):1153–1155 Wennerström M, Hansson S, Hedner T, Himmelmann A, Jodal U (2000) Ambulatory blood pressure 16-26 years after the first urinary tract infection in childhood. J Hypertens 18(4):485–491 Zorc JJ, Kiddoo DA, Shawn KN (2005) Diagnosis and management of pediatric urinary tract infections. Clin Microbiol Rev 18:417–422

Imaging of the Paediatric Urogenital Tract

89

Michael Riccabona

Abbreviations (a)CDS (Amplitude-coded) colour Doppler sonography (a)PN (Acute) Pyelonephritis CA Contrast agent ce Contrast-enhanced ce-US Contrast-enhanced ultrasound ce-VUS Contrast-enhanced voiding urosonography CM Contrast media CT Computed tomography DMSA Dimercaptosuccinic acid IVU Intravenous urography KUB Kidney–ureter–bladder abdominal film MAG3 Dynamic/diuretic renal scintigraphy MRT Magnetic resonance tomography MRU Magnetic resonance urography MU Megaureter PCN Percutaneous nephrostomy PUJ(O) Pelvi-ureteral junction (obstruction) PUV Posterior urethral valve RNC Radionuclide cystography Tbc Tuberculosis Tu Tumour UCA Ultrasound contrast agents UG(T) Urogenital (tract) US Ultrasound M. Riccabona (*) Department of Pediatric Radiology, University Clinic of Radiology, Medical University Graz, Graz, Austria e-mail: [email protected]

UT UTI UVJ(O) VCUG VUR XPN

Urinary tract Urinary tract infection Ureterovesical junction (obstruction) Voiding cystourethrography Vesico-ureteral reflux Xanthogranulomatous pyelonephritis

89.1 Introduction When discussing imaging of the paediatric urogenital (UG) system, the basic rationale of any imaging should be reconsidered and remembered: imaging should only be used if it has an impact on diagnosis, treatment, further management and/or patient outcome. Particularly in paediatrics, the imaging modality should be selected considering the specific paediatric needs and diseases: imaging should try to avoid ionising radiation whenever possible, the least invasive procedure applicable for the individual query should be selected (e.g. try to avoid catheterism if possible, reduce examinations that need sedation etc.) and any imaging that does not impact management should be avoided. Furthermore, the recommended imaging method should be easily accessible and available. Numerous recommendations exist on how to image when and for what; those from the ESPR abdominal imaging task force are listed in the Further Reading section— as a help to select the appropriate approach and

© Springer Nature Switzerland AG 2023 P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_89

1227

1228

then tailor the necessary imaging individually in terms of a more personalised medicine. And one more basic comment on imaging in children: children are not small adults, therefore state-ofthe-art imaging of infants and children requires dedicated techniques (equipment, transducer, filtering ...), as well as age-adapted investigation protocols (ultrasound, CT, MR etc.). This means that everyone who deals with children needs to have not only the adequate equipment available, but also has undergone dedicated training in paediatric imaging.

89.2 Imaging Methods There are a number of imaging methods that can be used and are applied for imaging the paediatric UG-tract (UGT): ultrasound (US), abdominal plain film, intravenous urography (IVU) (modified and adapted), voiding cystourethrography (VCUG), as well as other forms of urethra- or uretero-pyelography (antegrade or retrograde), including the therapeutic percutaneous nephrostomy (PCN), scintigraphy, computed tomography (CT) or magnetic resonance tomography (MRT). The details of specific imaging, such as scintigraphy, CT or MRT in children are not discussed here, but the chapter briefly outlines the basic requirements for more commonly performed procedures such as US, IVU or VCUG, which, in some centres, are in the hands of paediatric surgeons or paediatric nephro-/urologists. Interventional radiology in the paediatric UGT is discussed in the respective chapters. Note that the role and potential of various imaging modalities are constantly evolving, as well as the patient treatment concepts and management strategies; this naturally has implications on imaging algorithms. Additionally, whoever performs paediatric uroradiologic studies should have undergone dedicated training in paediatric imaging techniques to grant the optimal diagnosis at the lowest achievable invasiveness by high-quality imaging and child-adapted procedures. Ultrasound (US) has emerged from an initially orienting imaging tool to a dedicated high-­ end examination method and a range of

M. Riccabona

age-adapted sector, vector, curved array and linear transducers are necessary to reliably perform the task. Today, high-frequency, high-resolution transducers and some modern US techniques are applied routinely, such as (amplitude-coded) colour Doppler Sonography [(a)CDS], motion mode, harmonic and high-resolution imaging, as well as image compounding. Other modern methods, such as the three-dimensional US (e.g. for conspicuously displaying the rendered dilated collecting system, for calculating renal parenchymal volume, for virtual cystoscopy or assessment of particularly uterine anomalies) or US elastography (e.g. for chronic kidney disease or transplant queries) have not found a place in routine paediatric UG US, whereas contrast-enhanced (ce) US has become an increasingly appreciated tool for many conditions, such as intraluminal queries (VUR = ce-VUS, US-guided nephrostomy and drain placement, US-genitography) or iv. applications (e.g. assessment of complicated cysts or renal trauma) (Fig. 89.1). These applications may help to reliably establish a final diagnosis in many conditions, reducing the need for other (irradiating) imaging in many conditions. However, to properly explore this quickly growing, vast potential of US, some prerequisites need to be considered: • Investigations should be performed in physiologically hydrated patients with a sufficiently filled bladder. • Every US investigation includes the entire urinary tract (UT) and possibly the genital tract (GT), supplemented by an abdominal survey. • For a thorough evaluation, a detailed study of the renal parenchyma using linear transducers from a lateral, ventral or dorsal approach is mandatory, as well as pre- and post-void assessment of the bladder, the retrovesical space and the kidneys/the upper collecting system. • (a)CDS is very helpful in various conditions, such as urinary tract infection (UTI). • Kidney volume should be calculated and assessed in relation to growth charts. • Note that particularly initial studies should always include a post-void assessment.

89  Imaging of the Paediatric Urogenital Tract

1229

b

a

c Bl

Bl U

Bl

d

PCD

f

e PCD U

g

h Bl

Fig. 89.1  Applications of intracavitary contrast-enhanced Ultrasound in paediatric urogenital radiology: some typical examples. (a) Echogenic appearance of the urinary bladder ((Bl), axial section) filled with ultrasound contrast agent. (b) Parasagittal oblique section depicting a contrast filled ureter (U) behind the contrast filled urinary bladder (Bl) on ce-VUS; note the wide-open bladder neck and the pathological course of the transmural ureter. (c) Perineal US during voiding during ce-VUS nicely demonstrating the normal male urethra (arrow). Bl: urinary bladder. (d) Coronal view of the kidney on ce-VUS; dual image display (left side: contrast image, right side: fundamental US image) Contrast depicted in the renal pelvis, without dilatation of the calices, in middegree VUR (III°); PCD: pelvicalyceal system. (e) High-grade reflux into a dilated pelvicalyceal system (PCD) with a dilated and wide ureter

Bl

*

at the ureteropelvic junction. (f) Coronal view of the kidney on ce-VUS: Intrarenal reflux displayed in the right upper corner image (arrows) seen as echogenic bubbles beyond the caliceal border in the renal medulla. (g) Axial pelvic section — dual image display (left side: contrast image, right side: fundamental US image) during ce-VUS showing a few contrast bubbles in a dilated ureter (arrow) behind the bladder (Bl) — as such VUR is diagnosed which was invisible on VCUG (as the dilution effect of the radiopaque contrast did not create enough contrast to be visualized on fluoroscopy). (h) ce-VUS with US genitography, perineal view: during voiding after filling the urinary bladder (Bl) contrast influx from the urethra (arrow) into the vagina (arrowhead) through a urethrovaginal sinus tract (Asterix) is depicted, in a baby girl with adrenogenital syndrome

M. Riccabona

1230

In dilated or distended pelvi-caliceal systems or UTs, formerly called “hydronephrosis”, now replaced by the terms PCD (= pelvi-caliceal distention) or UTD (=UT dilatation), standardised (mostly axial) measurements of caliceal dilata-

tion and the width of the ureter and renal pelvis are mandatory; grading of PCD/UTD should be performed using standardised classifications, such as the ESPR or the new American suggestions (Tables 89.1 and 89.2).

Table 89.1  Grading of pelvi-calyceal respectively urogenital tract dilatation. Pelvi-caliceal Distention (PCD) grading by US (has replaced the old term “Hydronephrosis”) as used in Europe and suggested by the ESPR task force

Typ 0

Typ I

Typ II

Typ III

Typ IV

Typ V

PCD 0: no collecting system visible, usually considered normal PCD I: just the renal pelvis visible with an axial diameter less than 5–7 mm, usually considered normal PCD II: axial renal pelvis diameter less than 5/7–10 mm, some calyces with normal forniceal shape visible PCD III: marked dilatation of the renal calyces and pelvis larger than 10 mm with reduced forniceal and papillar differentiation without parenchymal narrowing, often seen in dilating VUR and obstructive uropathy PCD IV: gross dilatation of the collecting system with narrowing of the parenchyma, in high grade and obstructive uropathy PCD V: in some places used additionally to communicate an extreme PCD/UTD with only a thin, membrane-­like residual renal parenchyma. Differentiation from MCDK is commonly achievable, as in MCDK the residual dysplastic parenchyma is usually positioned centrally, with surrounding and exophytic cysts of varying size Adapted suggestion, based on Hofmann V. and on Fernbach SK et al. Ultrasound grading system of hydronephrosis: introduction to the system used by the Society for Fetal Urology. Pediatr Radiol 1993, 478–480) (Riccabona et  al., Pediatr Radiol 2008, 38: 138–145, updated 2017; see also Riccabona M (ed) Paediatric urogenital Radiology 3rd edn, Springer 2019; and Riccabona M, Ultrasound of the Urogenital tract in neonates, infants and children. In Riccabona M (ed) Pediatric Ultrasound, requisites and applications, 2nd edn, Springer 2020

89  Imaging of the Paediatric Urogenital Tract

1231

Table 89.2  Grading of pelvi-calyceal respectively urogenital tract dilatation. Urinary tract dilatation (UTD) grading issued by the American Societies (has replaced the old term “Hydronephrosis”) more commonly used in the USA— introduced by a multi-society consensus statement (adapted suggestion, based on Nguyen HT et al., Multidisciplinary consensus on the classification of prenatal and postnatal urinary tract dilation—UTD classification system. J Pediatr Urol 2014) composed of pre- and postnatal presentations leading to a risk-based assessment guiding management and respective imaging strategy Prenatal presentation Normal [ap pelvis 4–7 (28 w) mm] if no other pathology (normal parenchyma, slim ureter, bladder normal, slim calices) + no oligohydramnios UTD A1 = low risk

Abnormal [ap pelvis >7 (10 (>28 w) mm] other pathology (cysts or narrow/dysplastic parenchyma, abnormal ureter, bladder pathology, dilated/clubbed calices) + oligohydramnios UTD A2/3 = increased/high risk

Postnatal presentation No or some dilatation (>48 h) [ap pelvis 48 h) [ap pelvis >15 mm]

No or some dilatation (>48 h) [ap pelvis >15 mm]

Peripherily dilated calices Normal parenchyma Abnormal ureter Bladder normal

Peripherily dilated calices Parenchymla pathology Abnormal ureter Bladder abnormal

UTD P1 = low risk

UTD P2 = intermediate risk

UTD P3 = high risk

Risk based management UTD A1

UTD P1

UTD P2

UTD P3

Additional US at-4–6 w

-

-

-

US at 2 d–1 mo depending on findings

Follow-up US 1–6 mo

Follow-up US 1–3 mo

Follow-up US at 1 mo

VCUG at discretion of Clinician

VCUG at discretion of Clinician

VCUG recommended

Antibiotics at discretion of clinician

Antibiotics at discretion of clinician

Antibiotics Recommended

Functional scan not recommended

Functional scan at discretion of clinician

Functional scan at discretion of clinician

UTD A2/3

Prenatally 1 additional US Postnatally US at 2 d–1 mo US at 6 w–6 mo

Others: Risk adaption if indicated

Others: Specialist consultation

Abbreviations: ap anterior-posterior (axial) section, d days, MCDK multicystic dysplastic kidney, mo months, PCD pelvo-caliceal dilatation/distention, US ultrasound, UTD urinary tract dilatation, w weeks

1232

Assessment of vesico-ureteral reflux (VUR) based on changing dilatation of the ureteropelvic system is unreliable, only the use of US contrast agents (UCA) instilled into the bladder (called cevoiding urosonography = ce-VUS) enables a reliable and adequate sonographic assessment of VUR (see Fig.  89.1a–g). Include surrounding body compartments (i.e. perform an orienting overview (“sonoscope”) of the other pelvic structures) as well as of the liver, the pancreas, the spleen, the adrenal glands and adjacent intestinal parts, in order to not miss a potential disease that is clinically mimicking UT pathology or that is part of an underlying systemic or syndromal condition (e.g. US of the neonatal spine in babies with neurogenic bladder). Furthermore, it is recommended to thoroughly evaluate the inner genitalia in female neonates with severe UT malformations or single kidneys, as there is a relatively high incidence of embryologically determined associated genital malformations. The diagnosis of these conditions will be difficult after the 1st months of life due to the regression of the uterus und ovaries, whereas a reliable diagnosis can easily be achieved by US, supplemented by sonographic genitography in the neonatal period (potentially supplemented by fluoroscopy using the same catheters see Fig. 89.1h). Furthermore, in certain conditions, such as syndromal disease, adequate additional assessment of other body areas (heart, brain, spinal canal) must be initiated, with many of these queries being sufficiently answered by US. Voiding cystourethrography (VCUG): In spite of US advances, VCUG still remains an important investigation for VUR, particularly in countries where UCA is not approved for paediatric use or pre-operatively. Furthermore, VCUG enables a detailed anatomic assessment of the urethra and is superior to US in depicting intermittently posing diverticula, in demonstrating intrarenal reflux and by offering a panoramic overview of the entire (refluxing) UT. However, VCUG poses a significant radiation burden, particularly to girls—thus indications must be very strict (and constantly reassessed, potentially investigations can be replaced by ce-VUS) and a proper fluoroscopy technique is essential:

M. Riccabona

• The radiopaque contrast media (CM) should always be instilled into a (nearly) empty bladder, with CM instillation under physiologic pressure to avoid overextension and non-­ physiological conditions. The way varies depending on how the various centres, both transurethral as well as suprapubic punctures, are used (the latter particularly for psychological and socio-cultural reasons, as well as in cases where a catheterisation is impossible such as in PUV or urethral trauma). Note that a balloon of a urethral catheter must be deflated to avoid incorrect findings. • Cyclic studies are recommended in neonates and infants to increase VUR detection rate—this is more difficult with a suprapubic puncture. • Adequate fluoroscopy units adapted for paediatric needs are mandatory—performing VCUG with blind-spot films or using a C-arm device is inadequate. • A VCUG should always include a fluoroscopic assessment during the early filling phase in an anterior-posterior projection, an oblique view of the distal ureteral region, a lateral or oblique view of the urethra during voiding, an assessment of the upper collecting system during and after voiding, as well as a post-void assessment for residual urine and of potentially refluxed CM drainage dynamics. Detailed guidelines on how to properly perform VCUG can be found in the paediatric ­radiology literature (e.g. Riccabona et al. 2008); VCUG-based VUR grading has internationally been standardised. Plain film still has a role in assessment of UT calculi, as well as in assessment of various drains, catheters and devices; it, furthermore, is helpful for the assessment of associated skeletal changes, such as spina bifida. It should be focused on the kidney, the ureter and the bladder (“KUB”) by using shutters (not retrospective collimation in digital devices!); an age-adapted technique (dose, grids, resolution of film-foil combination or digital detectors) is essential. Intravenous urography (IVU) was one of the mainstays in uroradiology decades ago that allows

89  Imaging of the Paediatric Urogenital Tract

for a reliable assessment of the collecting system anatomy and reveals information on renal function and urinary drainage. However, today it has lost most of its importance in the paediatric setting, as US, scintigraphy and MR-Urography (MRU) enable a superior assessment with less radiation in most paediatric queries. Still, in areas with restricted access to these modalities, for a focused pre- or post-operative assessment, in some conditions where maximal spatial resolution is indispensable for early diagnosis (e.g. for medullary sponge kidney), and in suspected urolithiasis that cannot be sufficiently assessed using the other means respectively or for planning lithotripsy, a modified and adapted IVU may still be an option. If IVU is performed, the number of films should be restricted, and targeting (shutters—not retrospective grabbing or collimating on digital devices, gonad protectors) as well as timing should be individually adapted to the diagnostically relevant moments and areas. X-ray dose, film-foil combination, digital detectors, grids and CM dose need to be adapted according to the patient’s age and necessary resolution. With this approach, a significant reduction of the film number (usually 2–4 films are sufficient for diagnosis) and of the delivered radiation dose can be achieved. Scintigraphy is usually performed by paediatric nuclear medicine specialists according to the standardised protocols issued by the paediatric nuclear medicine groups and societies/associations. It uses tracers labelled with radioactive material (usually Technetium 99m) that are specifically handled by the renal parenchyma and/or secreted into the urine. The change in radioactivity over time is measured by a gamma camera, for functional assessment. Static renal scintigraphy uses DMSA (dimercaptosuccinic acid). This tracer is specifically taken up by the tubuli, and it allows for an exquisite assessment of the functioning renal parenchyma. It is used for split renal size assessment and assessment of focal lesions, such as acute pyelonephritis (aPN) or scars. Spatial resolution is restricted and can be improved using SPECT; however, due to an often-increased radiation bur-

1233

den, this latter technique is rather reluctantly used in children. Dynamic (diuretic) renography uses the glomerularly and tubularly handled MAG3 (Mercapto-acetyltriglycine) allowing for differentiation of an arterial phase, a parenchymal phase, an excretion phase and a drainage phase. Standardised hydration is essential as well as diuresis provocation using Furosemid to produce reliable results; a bladder catheter should be inserted in infants and particularly in patients with (high degree) VUR also to avoid misdiagnosis by bladder filling induced “pseudo-­ obstruction”. It is mainly used for the assessment of obstructive uropathy, such as pelvi-ureteral junction obstruction (PUJ(O)) or megaureter (MU). Radionuclide cystography (RNC) uses either the late phase of a dynamic renogram with the bladder filled by the renally excreted MAG3 (only applicable in toilet-trained and cooperative patients—without catheterisation), or the isotope (generally Tc99m colloid) is installed directly into the bladder via suprapubic puncture or a transurethral catheter. Any activity increase in the upper collecting system proofs the VUR.  Some grading and functional assessment is achievable but anatomic resolution is restricted; RNC, therefore, is used mostly only for screening purposes and for follow-up examinations. Interventional radiologic procedures are less common in children—however, still PCN (usually US-guided, potentially complemented by fluoroscopy or increasingly intra-cavitarily applied UCA) is still performed (often by paediatric urologists), and intravascular diagnostic and therapeutic procedures may become necessary, such as embolisation of an acute (tumour, traumatic) haemorrhage or depiction and treatment of renal artery stenosis—the latter commonly performed only by interventional (paediatric) radiologists. CT and MRI are usually performed by paediatric radiologists and will not be covered in detail. CT has a significant radiation burden and, therefore, is very reluctantly used in children. If

1234

UT-CT is considered in neonates, infants or children, age-adapted low-dose protocols with adapted parameters, reconstruction algorithms, as well as CM amount and timing are mandatory; normal protocols used in adults are not acceptable for routine paediatric use. The generally accepted indications for CT in children’s UGT are: severe abdominal trauma, tumour assessment, assessment of renal abscesses (particularly if MRI is not available), complex malformations that involve the pelvic skeleton (if this information is relevant for planning surgery, e.g. bladder exstrophy), and assessment of complicated urolithiasis and infections, such as renal tuberculosis (Tbc) or xanthogranulomatous pyelonephritis (XPN). Note that increasingly, most of these queries are investigated by MRI if available. MRI is becoming the ideal one-stop-shop imaging modality, particularly for congenital UGT anomalies, as MRU constitutes a comprehensive investigation without a radiation burden that allows for both anatomic and (if diuretic stress and CM are applied) functional assessments. Its setbacks are the restricted availability, the sedation needs particularly in infants and young children, some spatial and temporal resolution issues, as well as artefacts derived from patient motion. Nevertheless, MRI has taken over, not only as the imaging assessment of complex UG anomalies, but also the imaging assessment and staging of most UG tumours in childhood; thus, increasingly replacing CT. And finally, MRI is promoted for assessment of equivocal or complicated UTI (additionally applying diffusion-weighted sequences), as well as assessment of genital or complex anorectal malformations and anomalies.

89.2.1 Typical Imaging Findings in Common Paediatric Urological Conditions One of the most common queries is pre- or neonatally detected UTD/PCD. The task of early imaging is to depict those neonates and infants who benefit from early treatment and appropriate management with consequent monitoring to pre-

M. Riccabona

vent future harm to the kidney. Basically, a neonatal US study in the well-hydrated patient at the end of the 1st week of life enables differentiation of the patients who do not need any further imaging, those patients who need follow-up or additional studies and patients who need urgent treatment, such as for severe bilateral obstruction or posterior urethral valve (PUV). The latter might often need an earlier study (within the first days of life), if indicated by prenatal US findings and clinical presentation. The most common diagnosis in this patient group is a “physiological” dilatation without any need for further studies or follow-up (Fig.  89.2). The other relevant group is those with (dilating) VUR and obstructive uropathy. The most important sonographic signs (constituting most of the “extended US criteria” that improve US diagnosis and differentiation, Table  89.3) that allow differentiation of obstructive versus non-obstructive dilatation are: thickening of the renal pelvic wall and the configuration of the calices, which remain normally shaped in low-pressure pelvi-ectasia, whereas they become distorted or rounded with more or less parenchymal narrowing in patients with increased intrapelvic pressure. Additionally, bladder wall changes or dilatation of the ureter— even if intermittent—may hint towards urine transport and drainage problems, particularly VUR.  Note that already neonatally renoparenchymal changes due to hypodysplasia, such as increased echogenicity, reduced cortico-­ medullary differentiation or renal cysts may be present. These findings usually indicate a relatively restricted prognosis with high probability of renal insufficiency that sometimes manifests in infancy. Creative new sonographic approaches, such as perineal US (Fig.  89.3) or ce-VUS/ US-urethrography can help in the further specification of the condition, either by directly showing a urethral valve or enabling a reliable assessment of VUR. As these UCA-based methods are increasingly advocated for radiation safety issues, the paediatric urologist should make him-/herself familiar with the respective imaging appearance. In high-grade neonatal PCD, early US may underestimate its severity as the physiologically

89  Imaging of the Paediatric Urogenital Tract

a

Fig. 89.2  US of a normal neonatal kidney. Longitudinal (a, no pelvi-calyceal distension) and axial (b, slight physiologic distension of the renal pelvis) section with a

1235

b

14 MHz linear transducer: Note some echogenic papillae, with a physiologically pronounced cortico-medullary differentiation and echogenic cortex

Table 89.3  Extended ultrasound criteria List of extended ultrasound criteria These sonographic features may indicate UT pathology and might prompt further investigations (individual personalised decision based on all information available):  Urethra: valve-like shape of open bladder neck, suspected stenosis or valve on perineal US during voiding, utriculus cysts or suspected duplications etc.  Bladder: atypical shape and size, bladder wall thickening and trabeculation, diverticula, ostium pathology, significant residual urine, persistent urachus  Ureter: dilatation, tortuosity, ureteral wall thickening, ectopic orifice, duplex ureter  Kidney: position (ectopic? malrotation? fusion such as Horseshoe kidney …), caliceal dilatation and shape (e.g. PCD III or higher, clubbing, or obvious variation of dilatation during investigation/during and after voiding), pelvic wall thickening (“urothelial sign”), abnormal/asymmetric kidney size, parenchymal abnormalities (echogenicity and cortico-medullary differentiation, narrowing, cysts ...), vascular anomalies (regional vessel rarefaction, accessory renal artery crossing area of pelvi-ureteric junction, abnormal course of renal vein etc.)

reduced renal function during the first days of life will prevent the collecting system from reaching its maximum dilatation. Therefore, a repeated investigation after 2–4 (6) weeks is usually necessary. The amount of dilatation does not necessarily equal the degree of obstruction or renal function; kidneys with significantly reduced function may exhibit less dilatation than systems that function well. Therefore, US is not sufficient for assessment and grading of obstruction. Earlier, IVU was generally used for the assessment of urine drainage dynamics; today, this has been replaced by diuretic renography and functional

MRU. Note that these functional studies (i.e. scintigraphy and MRU) should only be performed after kidney maturation, which is at the earliest after the 6th week, better the 3rd month of life. Particularly in patients with a dilated ureter, differentiation of obstructive uropathy from VUR is mandatory. In these cases—particularly in neonates and boys—ce-VUS or VCUG is recommended for comprehensive assessment of VUR, including the bladder and the urethra. If VUR is detected, a base assessment of renal function by DMSA scintigraphy should be performed, usually at approximately 3–6 months of age. VCUG

1236

Fig. 89.3  Perineal US during voiding in a neonate showing the urine-filled and nicely distended male urethra (arrow) and a cystiform urethral diverticulum (arrowhead). Bl: bladder, A: anal canal, S:symphysis. Note that urine in itself may act as a sufficient contrast medium on US so that not always UCA are needed

is mandatory before surgery, whereas VUR follow-­up or VUR screening (if indicated at all) can be performed by RNC or ce-VUS. In obstructive uropathy, differentiation of the level of obstruction is important. Besides the rare but serious infravesical obstruction secondary to a tumour or a PUV, the obstruction can be either at the level of the ureterovesical junction (UVJ) or pelvi-ureteric junction (PUJ). In UVJ obstruction (UVJO), a “megaureter” is present that can be usually and finely visualised by US, as well as a potential ureterocele and an often associated duplex system (Fig. 89.4a). US can additionally assess and document (by motion-mode or using cine loop clips) ureteral peristalsis, particularly useful for follow-up (Fig 89.4b). PUJO is sonographically characterised by significant dilatation of the renal pelvi-caliceal system (PCD III–V), with more or less parenchymal narrowing without a visible ureter (Fig. 89.5a, b), unless a combined stenosis or a combination with dilating VUR is present. US examinations should always measure the maximum axial distension of the renal pelvis and the calices, as well as the maximal narrowing of the parenchyma, and assess renal vascularity and perfusion. CDS may depict additional renal arteries compromising the PUJ, and Doppler trace analysis may exhibit asymmetrically elevated resistance indexes in severe acute obstruction, particularly under diuretic stimulation with Furosemid (= diuresis sonogra-

M. Riccabona

phy). Furthermore, echoes and sedimentations may be present within a dilated system and may hint towards intraluminal infection (but also other phenomena may cause those echoes—therefore, US is not proof of such an infection, but can also not “rule out” an UTI!). This condition is relatively frequent in MU patients under antibiotic prophylaxis and may be clinically silent; urine samples may produce equivocal results, as a greater amount of urine will come from the normal functioning non-­ affected contra-lateral healthy system. Once the initial diagnosis is established, management varies: patients with non-obstructive dilatation or megacalycosis will usually be followed by repeated US examinations. Patients with obstructive uropathy or deterioration of mild (“non-obstructive”) dilatation need additional functional imaging, particularly for deciding on surgery and for preoperative anatomic assessment. US offers a non-invasive method for frequent follow-ups, but remains restricted in terms of functional assessment—therefore, diuretic scintigraphy, at present, remains the standard investigation for grading and follow-up of obstruction; usually, the decision for surgery is based on these findings (Fig. 89.5c). Increasingly, functional diuretic dynamic MRU is used for a combined approach to assess anatomy and function (Fig. 89.5d–h). Another common paediatric condition is urinary tract infection (UTI). Today, imaging increasingly focuses on differentiation of upper versus lower UTI, which clinically—particularly in infants—can be difficult. Potentially associated UT malformations need to be evaluated, but often these conditions have already been diagnosed by various screening programs. The importance of bladder function disturbances is increasingly recognised; therefore—particularly in older girls—bladder function assessment constitutes an essential part of the work-up after UTI. Again, US forms the mainstay of initial imaging allowing for assessment of UT anomalies, and using a supplementing aCDS enables a quite reliable assessment of renal involvement in UTI. Typical US criteria for (acute) pyelonephritis (aPN) are:

89  Imaging of the Paediatric Urogenital Tract

a

b

Fig. 89.4  US in megaureter. (a) US image of an ureterocoele (using an 8 MHz curved linear array and harmonic imaging) protruding into the bladder at the ostium, with its corresponding distal megaureter (parasagittal section, slightly oblique). (b) Cross-section of a slightly dilated

a

d

1237

b

e

distal left ureter behind the bladder in an infant using an 8 MHz curved linear array and harmonic imaging, with a motion-mode trace documenting lack of ureteral peristalsis

c

f

Fig. 89.5  Imaging in obstructive uropathy. (a, b) Typical US image of a pelvi-ureteric junction obstruction with gross pelvic dilatation, dilated calices (+ + 1) and parenchymal narrowing (+ + 2), in longitudinal (a) and axial (b) section using an 8 MHz curved linear array with harmonic imaging in an infant. (c) MAG3 dynamic diuretic renography showing the asymmetrically deteriorated drainage of the affected kidney (red line) in a decompensated pelviureteric junction obstruction. (d–h) MRU in pelvi-ureteric

g

h

junction obstruction: the T2-weighted “water image” shows the grossly dilated pelvi-calyceal system with malrotation on the left side (d). The serial T1 weighted gadolinium-enhanced images show asymmetric uptake (e), asymmetric excretion (f), asymmetric diuretic response (f, 3d-reconstructed image) and delayed excretion of contrast urine into the dilated left collecting system without sufficient urinary drainage (h, 3d-reconstruction)

M. Riccabona

1238

a

b

c

Fig. 89.6  Imaging in acute UTI. (a) Cross-section with an 8-MHz curved array transducer in a child with febrile UTI demonstrating an area of increased parenchymal echogenicity and reduced cortico-medullary differentiation (+ +), consistent with focal renal involvement. (b) Amplitude-coded colour Doppler sonography (= power

Doppler) in the same child depicts a focal perfusion defect at the site of the parenchymal abnormality, confirming renal segmental involvement in acute pyelonephritis. (c) Static renal DMSA scintigraphy in UTI, posterior acquisition, showing a photopenic defect in the upper pole (asterisk) in a different child with acute pyelonephritis

• • • •

constituting a nephrologic problem, such as glomerulonephritis or UTI including haemorrhagic cystitis. However, other severe UT conditions may also manifest with macroscopic haematuria such as haemolytic uraemic syndrome, renal vein thrombosis, or bladder and kidney tumours. Furthermore, conditions such as stress haematuria, retro-aortal left renal vein with nutcracker syndrome, idiopathic familiar microscopic haematuria, and haematuria associated with refluxing or even obstructing uropathy may also cause (microscopic) haematuria; renal infarction is an extremely rare event in children. In all these conditions, a comprehensive abdominal US including (a)CDS is indicated as the first imaging study to supplement urine (erythrocyte) microscopy and laboratory findings and often helps in the differential diagnosis. If (macroscopic) haematuria is associated with flank pain or dysuria or even abdominal colics, urolithiasis must be considered. US is the first imaging study allowing assessment of the collecting system and the renal parenchyma that may exhibit signs of nephrocalcinosis and other underlying nephropathies, or show different entities that need to be considered for differential diagnosis, for example, bleeding renal or bladder tumours, such as large angiomyolipoma or rhabdomyosarcoma. Particularly, for detection of distal ureteral calculi a sufficiently filled bladder and a detailed study of the distal ureters are mandatory. CDS may be helpful by demonstrating the

Enlarged kidney Focal or diffuse parenchymal changes Increased echogenicity of the perirenal sinus Thickening of the pelvic-, ureteral- and bladder wall (Fig. 89.6a)

There may be (but must not be) echoes within the urine. Furthermore, focal or diffuse, particularly asymmetric perfusion or vascularity disturbances caused by inflammatory oedema, vascular compression or necrosis, as well as scars can be reliably visualised by aCDS (Fig.  89.6b). In patients with ambiguous, equivocal or inconsistent findings, as well as in situations where aCDS is not available or of suboptimal quality, static DMSA scintigraphy or increasingly MRI using diffusion-weighted sequences are considered the gold standard for assessment of suspected aPN (however, only indicated if there is a therapeutic implication) (Fig. 89.6c). Increasingly, MRI (and in children, much rarely CT) are used for evaluation of complications, such as abscess formation, XPN, renal Tbc, as well as for differential diagnosis in sonographically unclear lesions such as complicated, infected cysts or tumours. The latter can be also better addressed by i.v. ce-US. VUR assessment is considered indicated in all infants and all patients with renal involvement or renal scarring after the infection—best at approximal 4 weeks after the onset of an upper UTI. Haematuria is a relatively rare event in childhood. There are a number of causes, most of them

89  Imaging of the Paediatric Urogenital Tract

a

b

1239

c

Fig. 89.7  Imaging in haematuria. (a) Colour Doppler sonography demonstrates the asymmetric urine inflow into the urinary bladder from the right ostium, and the twinkling colour signals at the area of the left ostium/left transmural ureter deriving from an ostial calculus (arrow). (b) Longitudinal section of the distal ureter (curved array,

harmonic imaging, 8 MHz) in an infant with haematuria and hypercalcuria demonstrates a distal ureteral calculus (+ +), with dorsal shadowing (arrow). (c) Same child, at maximum of peristaltic wave, with better dilated ureter, where the calculus exhibits twinkling signals on CDS (= “twinkling sign”) (arrow)

ureteric bladder inflow, enabling a differentiation of asymmetric urine inflow or complete obstruction (Fig. 89.7a, b). Furthermore, the “twinkling sign” may improve the detection of even small calculi in poorly distended distal ureters (Fig. 89.7c). Note that in the acutely obstructed kidney, US usually exhibits only little dilatation; the parenchyma is echogenic and the kidney itself appears enlarged and swollen. Doppler flow profiles may demonstrate an asymmetrically elevated resistance index. The calculi themselves are echogenic, though blood clots and mixed concrements with low calcium content may also appear less echogenic and produce less shadowing, whereas intraluminal fungus may appear similar to a calculus (i.e. very echogenic, with some dorsal shadowing). In cases with ureteral obstruction, the PUJ is often wide open, the proximal ureter is dilated and thus can be followed to the level of the obstruction (Fig.  89.8a–c). Sometimes, differentiation of papillary or wall calcifications from intraluminal concrements may be difficult—in these cases, positioning manoeuvres that demonstrate floating of concrements may be helpful. Note that other sedimentations in the papillae or the distal tubules may also cause similar appearances and twinkling artefacts such as in nephrocalcinosis or the physiologic hyperechoic medullae and papillae of neonates.

In many cases, US (possibly supplemented by a KUB film) is sufficient for diagnosis (Fig.  89.8d). In some instances, particularly for planning lithotripsy, an adapted IVU may become necessary, for example for differentiation of a duplex system or assessment of multiple ureteral calculi (Fig.  89.8e, f). Low dose stone CT has become the major imaging method in adults with urolithiasis, but, at present, is still used reluctantly in children due to its radiation burden—it, however, can be used as a problem-solving tool using a dedicated unenhanced low dose paediatric age- and size-/weight-adapted CT-protocol. Otherwise, only some rare conditions, such as prolonged complicated stone disease with infection, XPN or renal tumours with calcifications, as well as gross or complicated nephrolithiasis in an underlying problem as well as acute or traumatic haemorrhage, contrast-enhanced spiral CT may become indicated (Fig.  89.9a–c). At present, MRI plays no role in imaging urolithiasis. Kidney and bladder tumours are rare but important entities that only manifest clinically in part. They are often detected incidentally when doing an US for various unspecific symptoms. such as abdominal pain, haematuria, increased blood pressure or follow-up in various syndromes that are associated with a higher risk for renal tumours. The task of imaging is to detect and

M. Riccabona

1240

a

b

c

d

e

f

Fig. 89.8  Imaging in infants and children with urolithiasis. (a–c) US images pelvi-calyceal dilatation (a), an echogenic pelvic wall with indirect signs for a duplex system such as a central “parenchymal bridge” (b) and a calculus in the mid-ureter ( + + ) that exhibits the twinkling

sign on CDS. (c–f) KUB film (d) and adapted IVU using just two films for assessment of anatomy and drainage in the same girl as a–c with ureteral urolithiasis and suspected duplex system for planning shock wave lithotripsy (e = 15 min after contrast, f = 60 min after i.v. contrast)

confirm the tumour, to try to give some information on the tumour entity, to properly stage the tumour and search for metastasis, to give the preoperatively necessary anatomic-topographic information and eventually, to monitor the patients for follow-up during and after treatment. US usually is the initial investigation, and it detects or confirms the tumour. MRI, at present, is propagated for complete anatomic assessment

as well as characterisation and potentially staging of paediatric UG tumours. CT may become necessary for a thorough work-up, particularly concerning lung metastases or suspected calcifications (differential diagnoses versus, e.g. neuroblastoma), as well as in situations where (paediatric) MRI is not available. In children, every renal cyst or renal polycystic disease should prompt a thorough family assess-

89  Imaging of the Paediatric Urogenital Tract

a

1241

b

c

Fig. 89.9  Low dose unenhanced spiral multi-slice CT (adapted paediatric protocol). (a–c) Multi-slice CT (a) with sagittal reconstruction (b) and volume rendering (c) in an infant with a large concrement that filled the entire

collecting system, performed for therapy planning (i.e. additional parenchymal calcifications that would obviate lithotripsy or hinder percutaneous lithoapraxy)

ment and at least one follow-up examination. Basically, US is the primary investigation of choice in renal cysts, which are usually diagnosed and followed sonographically. Rarely, for example, for a complete diagnostic work-up in suspicion of cystic tumours, a range of modalities may become necessary, including plain film, CT or MRI (e.g. also for DD against tertiary calices or caliceal diverticula) and scintigraphy, depending on the underlying setting, the clinical suspicion and method availability. Note that microcysts may remain invisible to imaging, just altering the (US) appearance of the renal parenchyma. Every complicated cyst or every undefined cystic disease (particularly if cyst size or number increases over time) must be investigated by additional sectional imaging, such as CM-enhanced spiral CT or MRU; in complicated cysts i.v. ce-US may improve differential diagnosis by, for example, detecting enhancement in nodular areas or within thickened septae, then confirming the suspicion of a tumorous condition. International consensus statements have been issued to streamline and recommend imaging algorithms in cystic kidney disease based on the new insights into pathophysiology (e.g. model of ciliopathies, uromodelin pathology) both in the initial diagnosis (that very much relies on genetics and family history) and for follow-up, also respecting new treatment options with the respective imaging needs (e.g. MRI) for research purposes or to study the treatment effects.

Abdominal and urogenital trauma exists in children, too. Depending on the severity of trauma, as well as the trauma mechanism and impact, different imaging approaches are advocated. In mild trauma, as well as after renal biopsy, a comprehensive US study may suffice—with a mandatory follow-up after 12–24 h (Fig. 89.10a). In severe or multiple traumas, the role of US is restricted to the 1-min “FAST” examination in the emergency room used to check for free fluid; all other imaging is then achieved by contrast-enhanced multi-detector spiral CT. This approach grants a quick and reliable imaging of all necessary aspects (not only the kidney and ureter, but also other abdominal parenchymal organs, the spine, the major vessels and thorax etc.) as necessary for proper patient management and decision making, particularly as a conservative approach is increasingly advocated. It also allows for detailed grading of the injury. In cases with suspected bladder injuries or ureteral as well as urethral tears, dedicated additional imaging should be performed, such as fluoroscopicor CT-cystography (complete filling of bladder essential) and retrograde or antegrade urethraand ureterography. But note that again, only age-/size-adapted paediatric CT protocols must be used and have to be established beforehand— as there is little time to adapt and optimise CT protocols in emergency situations.

M. Riccabona

1242

a

d

b

e

c

f

g

Fig. 89.10  Post-operative imaging. (a, b) Longitudinal US image of a perirenal/subcapsular haematoma at the lower pole of the left kidney in a patient after renal biopsy, much clearly visible using aCDS ( + + ) than just on plain grey-­scale US (↔). (c, d) US after macroblast® injection for cystoscopic VUR treatment: note the echogenic deposit ( + + ) at the left transmural distal ureter/ostium (c), with slight ureteral dilatation (d). (e–g) Adapted IVU

after ureteral re-­implantation and closure of a cutaneous ureterostoma in an infant with obstructive megaureter who sonographically exhibited gross dilatation with asymmetrically elevated resistance indices post-operatively. A series of three films demonstrates the partial obstruction after 20 min (e) and after Furosemide application (f), but sufficient clearance as response to gravitation after upright positioning (g) obviating intervention

Preoperative imaging is outlined by the diagnostic imaging work-up of the various conditions discussed above. In order to enable an efficient use of imaging, a profound communication should be established between all physicians and specialties involved. For example, a PUJO diagnosed by US.  Diuretic scintigraphy has shown decompensated obstruction with asymmetrically reduced renal function of the affected kidney. If any additional imaging is needed, the important questions to answer are:

In this situation, preoperative imaging consists of VCUG, MRU with an included MR-Angiography sequence for the delineation of the exact anatomy of the collecting system, assessment of a potential duplex system and renal vascular supply; IVU is not suitable to answer all these questions, and ce-VUS may miss short-­ lasting, low-grade VUR into a non-dilated ureter (however, probably this condition is of less importance). Usually, US is sufficient for post-operative and post-interventional monitoring (Fig. 89.10a– d). However, more intense post-operative imaging may become necessary in cases with a complicated post-operative course, as US may be restricted in reliably answering all management relevant function-related questions. Therefore,

• Is there a duplex system? • Is there VUR? • Is there any vascular anomaly that may be associated with the UPJO or might pose a peri-operative risk?

89  Imaging of the Paediatric Urogenital Tract

antegrade pyelography (e.g. after pyeloplasty, using the peri-operatively placed drain) may be helpful; some centres even regularly perform an antegrade pyelogram as part of the assessment before drain removal (this can also be achieved using UCA and US). For questions concerning drain position and function, UCA may be applied intraluminally to reduce the need for fluoroscopy and can depict drain malposition with extravasation or assess ureteric patency. If no drain is in place (such as in patients after anti-reflux procedures), an adapted IVU (potentially only consisting of two delayed films) or scintigraphy may answer the question (Fig.  89.10e–g). Seldom may other imaging, such as MRI or VCUG, become necessary in the post-operative setting, except for rare cases after tumour or trauma surgery. Note that the value of post-operative imaging is restricted; for example, pre-existing chronic dilatation will not quickly resolve and may physiologically even increase temporarily post-operatively—thus, dilatation cannot be used to assess re-obstruction. Therefore, don’t overstress your imaging department with too many unnecessary US exams just done for “reassurance” without real clinical necessity, as these results may be even misleading.

89.3 Imaging Algorithms In order to standardise the imaging procedures, to allow for comparison during follow-up, to enable a reliable work-up even after working hours and to grant some basic quality at reasonable costs, imaging algorithms and recommendations are have become important; these should be applied for imaging work-up of the various conditions discussed above. They try to enable an effective use of the various modalities without increasing the risk of missing important conditions that would need urgent treatment. The European Society of Uroradiology, as well as The European Society of Paediatric Radiology, are engaged in developing and have issued numerous specific paediatric imaging recommendations for common queries, such as, for

1243

work-up of neonatal UTD/PCD, obstructive uropathy in childhood, or imaging in UTI. Furthermore, these societies have been and are issuing procedural guidelines that define a basic minimum standard for the various modalities used in paediatric UG radiology. These open access recommendations (mostly published in Pediatr Radiol) try to take specific paediatric considerations and growth variations into account (as the probability of various diseases as well as management options may differ depending on sex and age), and address radiation protection issues, as children are far more sensitive to ionising radiation than adults and need utmost radiation protection. For example, the imaging algorithm in UTI in childhood shows slight differences between girls and boys (VCUG versus ce-VUS) and for infants versus older patients, where bladder function studies become very important, while the need for VUR assessment constantly decreases (Table 89.4). The imaging algorithm for neonatal PCD is strategised according to the amount of dilatation on prenatal US as well as the findings on the first postnatal examination (Table 89.5). Note that all imaging recommendations and algorithms constitute a basic general minimum consensus; individual adaptation to the specific patient’s condition (in terms of a more “personalised” approach) as well as adjustment of the algorithms according to local facilities and availability and expertise remain essential. Furthermore, these recommendations will need constant updating because not only changes in management strategies and advances in imaging techniques will affect the recommendations and induce potential changes, but also new insights into pathophysiology and natural history of certain conditions may lead to altered imaging requirement. As a general message—not all that can be imaged needs to be imaged! If there is no therapeutic impact, or some studies may even carry the risk of misinterpretation, one should abstain from studies that could possibly create confusion and the need for further (follow-up) studies.

M. Riccabona

1244 Table 89.4  Flowchart “Imaging algorithm in UTI in infants and children”

UTI (critria & diagnosis = clincal & laboratory: catheter urine sample /culture and blood count) *1

US incl. (power) Doppler

Normal Clinically cystitis

Normal or equivocal*2 - e.g., poor quality, DDx … - poor Doppler …

Pyonephrosis nephrostomy

(Pyelo)nephritis*2 - aPN, upper UTI, scar …

Clinically upper UTI STOP possibly follow-up

DMSA or MRI (*2) acutely - if therapeutically or diagnostically necessary

*2 Follow-up US

Normal

* 1 Urine: lecucyturia, posoiutve notrite, positive caulture, urine calictonin … Blood : leucocytosis, C-reactive protein, NOTE: reliable clinical diagnosis essential = most important entry criteria for imaging US cannot relaible diagnose or rule out UTI * 2 for DDx and complications (e.g., Tu or cysts versus abscess) for other complicated situations such as XPN, Tbc … or for infected stone, inflammatory pseudotumor etc:MRI (or CT) CT Indiactions: complicated stone disease, MRI not available

- timing depends on clinical course and initial findings - assess scars & renal growth

VUR evaluation - always in neonates / infants - possibly in age< 5 years -in recurrent UTI & scaring /RNP - ce-VUS particularly in girls & follow-up - VCUG (boys, pre-operatively) - RNC follow-up Late DMSA or MRI - after 4-6 (-12) months - for scars Bladder function studies - in older, toilet trained patients

Adapted from Riccabona et al., Pediatr Radiol 2008, 38: 138–145 Suggested imaging algorithm in children with UTI aPN acute pyelonephritis, DMSA static renal scintigraphy, ee-US echo-enhanced uro-sonography, RNC radionuclide cystography, Tbc tuberculosis, Tu tumour, US ultrasound, UTI urinary tract infection, VCUG voiding cystourethrography, VUR vesico-ureteral reflux, XPN xanthogranulomatous pyelonephritis

89  Imaging of the Paediatric Urogenital Tract

1245

Table 89.5  Flowchart “Imaging in neonates with antenatal diagnosis of PCD” Prenatal US Normal (grade 0/I, plevis < 5/8 mm)

mild dilatation (grade II, plevis 5-10 mm)

suffiicent hydration

moderate dilatation grade III, pelvis 8/10 mm unilateral grade IV + no other changes) +

sufficient bladder filling,

no US at all?

severe dilatation (grade IV: bilateral, dysplastic parenchyma, supected PUV) use extended US criteria

Delayed US (day 7–14)

early US (day 3–10) If suspicion of VUR + VCUG / ce-VUS

Urgent US (day 1-3) + ce-VUS / VCUG

PCD I-II No other pathology

PCD III-IV other pathology complex malformation …

Early DMSA / MRU? US-guided intervention?

stop potentially US follow-up after some months

VCUG / ce-VUS? MRU? Scintigraphy? Depends on individual findings e.g. US-follow-up intervalls … + DMSA in VUR, …

If signs of obstruction or increasing dilatation etc … + MAG3 scintigraphy + (functional) ce-MRU (+ adapted paediatric IVU)* BUT wait for mature renal function (> 8-12 weeks)

Adapted from Riccabona M, Fotter R, 2004, Reorientation and future trends in paediatric uroradiology: Minutes of a symposium. Pediatr Radiol 34, 295–301 Suggested imaging algorithm in neonates with prenatally diagnosed PCD: DMSA static renal scintigraphy, ce-VUS contrast-enhanced voiding uro-­sonography, IVU intravenous urography, MAG3 dynamic renography, MRU MR-urography, PCD pelvi-calyceal distension, PUV posterior urethral valve, US ultrasound, UT urinary tract, VCUG voiding cystourethrography, VUR vesico-­ureteral reflux *Adapted paediatric IVU: only if no diuretic functional scintigraphy/ce-MRU available

1246

89.4 Conclusion Many paediatric UG(T) conditions and queries need dedicated and tailored expert imaging, best provided by a paediatric radiology unit. Due to the different conditions one encounters in childhood imaging, algorithms differ from adults, and the imaging needs and protocols have to be adapted individually according to the patient age, gender, query and therapeutic consequence. Usually, imaging starts with a dedicated US, complemented by either a VCUG or MRI, depending on the underlying query. Existing imaging and procedural recommendations help to optimally tailor and perform imaging in paediatric uroradiology. And finally—also respecting economic aspects and capacities, as well as local variances and availabilities—imaging has always to be questioned in terms of efficacy: studies that do not serve the patient and impact on management or prognosis and only serve for falsely ­pacifying referring physicians or parents should be avoided.

Further Reading Arthurs O, Easty M, Riccabona M (2020) Imaging of the kidneys, urinary tract and pelvis in children. In: Adam A, Dixon AK, Gillard JH, Schaefer-Prokop CM (eds) Grainger and Allison’s diagnostic radiology, Volume II, Section G pediatric imaging. Elsevier, New York, pp 1846–1885 Fotter R (2001) Pediatric uroradiology. Springer, Berlin Gordon I, Riccabona M (2003) Investigating the newborn kidney update on imaging techniques. Semin Neonatol 8:269–278 Kao SC (2005) The urinary tract. In: Carty H, Brunelle F, Stringer D, Kao SC (eds) Imaging children, 2nd edn. Elsevier, New York, pp 537–882 Ključevšek D, Ključevšek T (2020) Efficacy of contrast-­ enhanced percutaneous nephrosonography to evaluate urinary tract patency in children. J Clin Ultrasound 48:410–415. (Online June 16th) Ključevšek D, Riccabona M, Ording Müller LS et  al (2020) Intracavitary contrast-enhanced ultrasonography in children: review with procedural recommendations and clinical applications from the European Society of Paediatric Radiology Abdominal Imaging Task Force. Pediatr Radiol 50:596–606. https://doi. org/10.1007/s00247-­019-­04611-­1 Nguyen HT, Benson CB, Bromley B et  al (2014) Multidisciplinary consensus on the classification of

M. Riccabona prenatal and postnatal urinary tract dilation (UTD classification system). J Pediatr Urol 10:982–998 Puri P, Höllwarth M (2006) Pediatric surgery. Springer, Berlin Riccabona M (ed) (2002) Paediatric uroradiology. Eur J Radiol 43 (special issue) Riccabona M (2011) The pediatric kidney. In: Quaia E (ed) Radiological imaging of the kidney. Springer, Berlin, pp 675–714. ISBN 978-3-54087596-3 Riccabona M (ed) (2019) Pediatric urogenital radiology, 3rd edn. Springer, New-York Riccabona M, Fotter R (2006) Radiographic studies in children with kidney disorders: What to do and when. In: Hogg R (ed) Kidney disorders in children and adolescents. Taylor & Francis, Birmingham, pp 15–34 Riccabona M, Avni FE, Blickman JG et al (2008) Imaging recommendations in paediatric uroradiology: Minutes of the ESPR Workgroup session on urinary tract infection, fetal hydronephrosis, urinary tract ultrasonography and voiding cysto-­urethrography ESPR-meeting, Barcelona/Spain, June 2007. ESUR Paediatric Guideline Subcommittee and ESPR Paediatric Uroradiology Work Group. Pediatr Radiol 38:138–145 Riccabona M, Avni FE, Blickman JG et al (2009) Imaging recommendations in paediatric uroradiology, part II: urolithiasis and haematuria in children, paediatric obstructive uropathy, and postnatal work-up of foetally diagnosed high grade hydronephrosis. Minutes of a mini-symposium at the ESPR annual meeting, Edinburg, June. Pediatr Radiol 39:891–898 Riccabona M, Avni FE, Dacher JN et  al (2010) ESPR Uroradiology Task Force and ESUR Paediatric Working Group: Imaging and procedural recommendations in paediatric uroradiology, Part III. Minutes of the ESPR Uroradiology Task Force mini-symposium on intravenous urography, uro-CT and MR-urography in childhood. Pediatr Radiol 40:1315–1320 Riccabona M, Avni F, Dacher JN et  al (2011) ESPR Uroradiology Task Force and ESUR Paediatric Working Group: imaging recommendations in paediatric uroradiology, part IV Minutes of the ESPR Uroradiology Task Force mini-symposium on imaging in childhood renal hypertension and imaging of renal trauma in children. Pediatr Radiol 41:939–944 Riccabona M, Avni F, Damasio B et  al (2012) ESPR Uroradiology Task Force and ESUR Paediatric Working Group—Imaging recommendations in paediatric uroradiology, part V: childhood cystic kidney disease, childhood renal transplantation, and contrast-­ enhanced ultrasound in children. Pediatr Radiol 42:1275–1283 Riccabona M, Lobo ML, Willi U et  al (2014a) ESPR Uroradiology Task Force and ESUR Paediatric Working Group—imaging recommendations in paediatric uroradiology, part VI: childhood renal biopsy and imaging of neonatal and infant genital tract. Pediatr Radiol 44:496–502 Riccabona M, Vivier HP, Ntoulj A et  al (2014b) ESPR Uroradiology Task Force—Imaging recommendations in paediatric uroradiology—part VII:

89  Imaging of the Paediatric Urogenital Tract standardized terminology, impact of existing recommendations, and update on contrast-enhanced ultrasound of the paediatric urogenital tract. Pediatr Radiol 44:1478–1484 Riccabona M, Darge K, Lobo ML et  al (2015) ESPR Uroradiology Task Force—imaging recommendations in paediatric uroradiology—part VIII: retrograde urethrography, imaging in disorders of sexual development, and imaging in childhood testicular torsion. Report on the mini-symposium at the ESPR

1247 meeting in Amsterdam, June 2014. Pediatr Radiol 45:2023–2028 Riccabona M, Lobo ML, Ording-Muller LS et al (2017) ESPR Abdominal (GU and GI) Imaging Task Force— imaging recommendations in paediatric uroradiology, part IX: imaging in anorectal and cloacal malformation, imaging in childhood ovarian torsion, and efforts in standardising pediatric uroradiology terminology. Report on the mini-symposium at the ESPR meeting in Graz, June 2015. Pediatr Radiol 47:1369–1380

Management of Antenatal Hydronephrosis

90

Jack S. Elder

90.1 Introduction An abnormality involving the genitourinary tract is detected in 1–2% of pregnancies, depending on the sonographic criteria. The goal of management is to recognize and treat congenital anomalies that may adversely affect renal function or cause urinary tract infection (UTI) or sepsis. Many structural abnormalities of the urinary tract are characterized by hydronephrosis, which is frequently assumed to be obstructive. However, often antenatal hydronephrosis (ANH) results from nonobstructive causes, including vesicoureteral reflux (VUR), multicystic dysplastic kidney, and certain abnormalities of the ureteropelvic and ureterovesical junction.

90.2 Development of the Kidney and Renal Function The kidney is derived from the ureteral bud and the metanephric blastema. During the 5th week of gestation, the ureteral bud arises from the mesonephric (Wolffian) duct and penetrates the metanephric blastema, which is an area of undifferentiated mesenchyme on the nephrogenic ridge. The ureteral bud undergoes a series of approximately 15 generaJ. S. Elder (*) Division of Pediatric Urology, Massachusetts General Hospital, Boston, MA, USA

tions of divisions, and by 20 weeks’ gestation, it forms the entire collecting system, that is, the ureter, renal pelvis, calyces, papillary ducts, and collecting tubules. Under the inductive influence of the ureteral bud, nephron differentiation begins during the 7th week of gestation. By 20 weeks’ gestation, when the collecting system is completely developed, approximately one-third of the nephrons are present. Nephrogenesis continues at a nearly exponential rate and is complete at 36 weeks’ gestation. Throughout gestation, the placenta functions as the fetal hemodialyzer, and the fetal kidneys play a minor role in the maintenance of fetal salt and water homeostasis. Formation of urine begins between the 5th and 9th weeks of gestation. The rate of urine production increases throughout gestation, and at term, volumes have been reported to be 51 mL/h. The glomerular filtration rate (GFR) has been measured at 6 mL/ min/1.73  m2 at 28 weeks’ gestation, increasing to 25  mL/min/1.73  m2 at term, and thereafter triples by 3 months of age. The main factors responsible for this increase in GFR after birth include an increase in the capillary surface area available for filtration, changes in intrarenal vascular resistance, and redistribution of renal blood flow to the cortical nephrons, which are much more numerous than the medullary nephrons. A congenital obstructive lesion of the urinary tract may have a deleterious effect on renal function. Severe early obstructive uropathy disrupts nephrogenesis and results in renal dysplasia.

© Springer Nature Switzerland AG 2023 P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_90

1249

J. S. Elder

1250

90.3 The Fetus with Antenatal Hydronephrosis When a fetus is identified with a suspected urinary tract abnormality, the goals of management include determining the differential diagnosis, assessment of associated anomalies, and determining the fetal and postnatal risk of the anomaly. Hydronephrosis is recognized by demonstrating a dilated renal pelvis and calyces. The ureter and bladder may be dilated also. The likelihood of having a significant urinary tract abnormality is directly proportional to the severity of hydronephrosis. If the renal pelvic diameter is more than 2 cm, 94% have a significant abnormality of the urinary tract requiring surgery or long-term urologic follow-up. If the fetal renal pelvic diameter is between 1.0 and 1.5 cm, 50% have an abnormality, and if the dilated renal pelvis is less than 1  cm, only 3% have a significant abnormality. A renal pelvic diameter of at least 4  mm before 27 weeks’ gestation and at least 7 mm after 27 weeks’ gestation has been considered significant. The later the sonogram is performed, the more likely an existing abnormality will be detected, because the obstructed renal pelvis gradually enlarges throughout gestation. In addition, in utero the fetus is usually upside down in the uterus and urine is draining uphill. For example, Fugelseth and colleagues reported that only one-third of a series of women carrying babies with a urologic anomaly had an abnormal ultrasound (US) study at 15–21 weeks’ gestation. The differential diagnosis of ANH is provided in Table 90.1. Virtually all of these conditions can cause bilateral hydronephrosis. A distended bladder and bilateral hydronephrosis are suggestive of bladder outlet obstruction, such as posterior urethral valves or a large ectopic ureterocele obstructing the bladder neck, but fetuses with nonobstructive conditions, such as high-grade VUR or prune belly syndrome, also may have bilateral hydroureteronephrosis and a distended bladder. In fetuses with a urologic anomaly, associated abnormalities are common. For example, in an early series of fetuses with bilateral hydronephrosis and oligohydramnios, 16 of 31 (55%)

Table 90.1  Causes of antenatal hydronephrosis Anomalous UPJ/UPJ obstructiona Multicystic kidneya Retrocaval ureter Primary obstructive megauretera Nonrefluxing nonobstructed megauretera Vesicoureteral refluxa,b Midureteral stricturea Ectopic ureterocelea,b Ectopic uretera Posterior urethral valvesa,b Prune belly syndromea,b Urethral atresiab Hydrocolposa,b Pelvic tumora,b Cloacal abnormalitya,b May be unilateral or bilateral Bladder may be distended

a

b

had an associated structural or chromosomal abnormality (Reuss et al. 1988). Congenital heart disease and neurologic deformities can often be detected, if they are present. In contrast, large bowel abnormalities, such as imperforate anus, are more difficult to detect by prenatal sonography, whereas recognition of small bowel anomalies, such as atresia, are usually straightforward. The considerations in determining fetal management include overall fetal well-being, gestational age, whether the hydronephrosis is unilateral or bilateral, and the volume of amniotic fluid. Until recently, there were no guidelines for determining how frequently to image the fetus or whether a specific intervention was necessary. If hydronephrosis is unilateral, usually no specific fetal therapy is necessary. For example, if the hydronephrosis is secondary to a ureteropelvic junction (UPJ) obstruction, even if function is poor, the kidney has a significant capacity for improvement in function following neonatal pyeloplasty. Even with bilateral UPJ obstruction (characterized by bilateral hydronephrosis and a normal bladder), the amniotic fluid volume and pulmonary development typically are normal. Consequently, specific interventions, such as percutaneous drainage of the fetal kidney or early delivery to allow immediate urologic surgery, are unwarranted. These same principles apply to primary obstructive megaureter.

90  Management of Antenatal Hydronephrosis

The primary life-threatening congenital urologic anomalies include posterior urethral valves, urethral atresia, and prune belly syndrome, which are usually characterized by bilateral hydroureteronephrosis and a distended bladder that does not empty in a male fetus, and many have oligohydramnios. Approximately one-third of infants with urethral valves eventually develop renal insufficiency or end-stage renal disease. Although prune belly syndrome is considered nonobstructive, neonates with this condition frequently have renal insufficiency, in large part because of congenital renal insufficiency and also from renal deterioration in children with repeated episodes of pyelonephritis. Urethral atresia is nearly always fatal, because the kidneys are usually dysplastic. A significant adverse prognostic factor is oligohydramnios, which prevents normal pulmonary development. In fetuses with severe obstructive uropathy and renal dysplasia, neonatal demise usually results from pulmonary hypoplasia rather than chronic renal failure. Intuitively, it would seem that treatment of the obstructed fetal urinary tract by diverting the urine into the empty amniotic space might allow normal renal development to occur and restore amniotic fluid dynamics, stimulating lung development. Experimental procedures have been performed, including percutaneous placement of a vesicoamniotic shunt, creation of a fetal vesicostomy or pyelostomy, and even percutaneous urethral valve ablation through a miniscope. The complication rate is high, including shunt migration, urinary ascites, stimulation of preterm labor, and chorioamnionitis. Furthermore, in many cases, irreversible renal dysplasia has already occurred, and although the procedure may be successful technically, often the baby is stillborn, dies of pulmonary hypoplasia, or is alive with end-stage renal disease. Nevertheless, some fetuses may benefit from aggressive intervention if the kidneys do not have irreversible dysplasia. Unfavorable prognostic factors include: –– Prolonged oligohydramnios –– Renal cortical cysts –– Urinary Na >100 mEq/L, Cl > 90 mEq/L, and osmolarity>210 mOsm/L

1251

–– Beta2-microglobulin > 6 mg/L –– Reduced lung area and thoracic or abdominal circumference Unfavorable urinary electrolytes may reflect stale urine in the fetal urinary tract. Consequently, perinatal centers typically obtain two or three sequential samples, as subsequent samples yield fetal urine that is more reflective of true fetal renal function.

90.4 Guidelines on Antenatal Hydronephrosis In 2010, the Society for Fetal Urology (SFU) published a consensus statement regarding the evaluation and management of ANH. This document was a comprehensive attempt to provide a guide to antenatal and postnatal management of ANH and identify areas of controversy and prioritize research endeavors. The SFU did not provide guidance on when a voiding cystourethrogram (VCUG) should be performed. With regard to antibiotic prophylaxis, the authors recommended that neonates with an increased risk of UTI (girls, uncircumcised boys, moderate to severe hydronephrosis, and familial VUR) should receive prophylaxis until the initial evaluation is completed and management is planned with the family. In 2012, the American Urological Association (AUA) was asked to establish formal guidelines for ANH, but they thought that the SFU was thorough and did not need updating. The European Society for Paediatric Urology has published recommendations on the management of various congenital urinary tract malformations, but has not addressed ANH specifically. In 2013, the Indian Society of Pediatric Nephrology published updated guidelines on ANH (Sinha et  al. 2013). These recommendations were based on systematic reviews of the literature from 1990 to 2012. Level 1 recommendations included those applicable to most subjects based on consistent information. Level 2 recommendations were those that were based on equivocal or insufficient information. Highlights of the recommendations include:

1252

1. It is recommended that ANH be diagnosed and its severity graded based on anteroposterior diameter (APD) of the fetal renal pelvis (1B). ANH is present if the APD is ≥4 mm in second trimester and ≥7  mm in the third trimester. 2. Termination of pregnancy is not recommended in fetuses with unilateral or bilateral ANH, except in presence of extrarenal life-­ threatening abnormality (1D). 3. It is recommended that all newborns with a history of ANH should have postnatal US examination within the first week of life (1B). In neonates with suspected posterior urethral valves, oligohydramnios or severe bilateral hydronephrosis, ultrasonography should be performed within 24–48 h of birth (1C). In all other cases, the US should be performed preferably within 3–7 days, or before hospital discharge (1C). 4. It is recommended that assessment of the severity of postnatal hydronephrosis be based on the classification proposed by SFU or anteroposterior diameter of the renal pelvis (1B). 5. It is recommended that neonates with normal US examination in the first week of life should undergo a repeat study at 4–6 weeks (1C). Infants with isolated mild unilateral or bilateral hydronephrosis (APD 10 mm, SFU grade 3–4, or ureteric dilatation (1B). MCU should be performed early, within 24–72 h of life, in patients with suspected lower urinary tract obstruction (1D). In other cases, the procedure should be done at 4–6 weeks of age. MCU is recommended for infants with antenatally detected hydronephrosis who develop a urinary tract infection (1C). 7. It is recommended that infants with moderate to severe unilateral or bilateral hydronephrosis (SFU grade 3–4, APD >10  mm) who do not show VUR should undergo diuretic renography (1C). Infants with hydronephrosis and dilated ureter(s) and no evidence of VUR

J. S. Elder

should undergo diuretic renography (2C). The preferred radiopharmaceuticals are 99mTc-­ mercaptoacetyltriglycine (99mTc-MAG3), 99m 99m Tc-ethylenedicysteine ( Tc-EC) or 99mTc-­ diethylenetriaminepentaacetic acid (DTPA) (2D). The differential function is estimated and renogram curve is inspected for a pattern of drainage. Diuretic renography should be performed after 6–8 weeks of age (2D). The procedure may be repeated after 3–6 months in infants where US shows worsening of pelvicalyceal dilatation (2D). 8. It is suggested that surgery be considered in patients with obstructed hydronephrosis, and either reduced differential renal function or its worsening on repeat evaluation (2C). It is suggested that surgery be considered in patients with bilateral hydronephrosis or hydronephrosis in a solitary kidney showing worsening dilatation and deterioration of function (2D). 9. It is recommended that parents of all infants with antenatal or postnatal hydronephrosis be counseled regarding the risk of urinary tract infections and the need for prompt management (1B). It is recommended that infants with postnatally confirmed moderate or severe hydronephrosis (SFU 3–4; renal APD > 10  mm) or dilated ureter receive antibiotic prophylaxis while awaiting evaluation (1C). It is recommended that all patients detected to have VUR receive antibiotic prophylaxis through the first year of life (1B). In 2014, a multidisciplinary consensus conference involving eight societies was held and included pediatric urologists, pediatric nephrologists, pediatric radiologists, and maternal-fetal medicine specialists. The purpose was to try to standardize the fetal evaluation and early postnatal management of babies with ANH. For example, several different classification schemes had been published for significant and insignificant pelvocaliceal dilation. In addition, although the SFU 4-point grading scale is used by most pediatric urologists, most pediatric radiologists classify hydronephrosis into mild, moderate, and severe. The US parameters include anterior-posterior renal pelvic diameter (APRPD), calyceal dilation, whether it involves the major and/or minor

90  Management of Antenatal Hydronephrosis

calyces, parenchymal thickness and appearance, normal or abnormal ureter, and normal or abnormal bladder (Nguyen et al. 2010, 2014). Normal values for urinary tract dilation (UTD) are APRPD: Antenatal Postnatal

16–27 weeks >28 weeks (>48 h)

48 h to 1 month of age, and a second renal US 6 months later. Genetic screening is not indicated unless there are associated congenital malformations. If the APRPD is >7  mm at 16–27 weeks or >10 mm at >28 weeks with any peripheral calyceal dilation or any other upper urinary tract abnormality, they are classified as UTD A2–3, or Increased Risk. The assigned risk is based on the most concerning feature. For UTD A2–3, the panel recommended a follow-up US in 4–6 weeks, although with suspected posterior urethral valves (PUV) or severe bilateral hydronephrosis, more frequent follow-up was recommended until delivery. Following delivery, a renal US after 48 h but before 1 month was suggested, again with more immediate evaluation if PUV is suspected or there is significant bilateral hydronephrosis. In addition, specialist consultation with pediatric urology or nephrology was recommended. For postnatal presentation, at >48  h, an APRPD 15  mm, there is peripheral calyceal dilation and/or abnormal ureters, it is classified UTD P2, Intermediate Risk. SFU hydronephrosis grade 3 corresponds to UTD P2. The panel recommends a follow-up renal US in 1–3 months. A VCUG, antibiotic prophylaxis, and a functional renal scan are optional, at the discretion of the clinician. If the APRPD is >15 mm and there is peripheral calyceal dilation, abnormal parenchymal thickness, abnormal parenchymal appearance, abnormal ureters, and/or abnormal bladder, it is classified UTD P3, High Risk. SFU hydronephrosis grade 4 corresponds to UTD P3. The consensus group recommends a follow-up renal US in 1 month. A VCUG and antibiotic prophylaxis are recommended. A functional renal scan is optional, at the discretion of the clinician (but is virtually always recommended). This classification scheme was validated at an NIH Consensus Conference by the participants. In addition, it was evaluated in a retrospective study of 490 patients (Hodhod et al. 2016). The authors found that the UTD classification appropriately identified babies that were likely to require surgical intervention, while the SFU hydronephrosis grading system was most likely to predict the likelihood of resolution of hydronephrosis.

90.5 Management of the Newborn with Antenatal Hydronephrosis 90.5.1 Management in the Nursery At birth, the abdomen is inspected to detect the presence of a mass, which most often is secondary to a multicystic dysplastic kidney or UPJ

1254

obstruction. In male newborns with posterior ­urethral valves, often a walnut-shaped mass, representing the bladder, is palpable just superior to the pubic symphysis. Newborns should also be evaluated for anomalies involving other organ systems, as urinary tract abnormalities often occur in babies with congenital heart disease, lung abnormalities, and anorectal malformations. Renal function should be monitored with serial serum creatinine levels, particularly in infants with bilateral hydronephrosis or a unilateral poorly functioning kidney. At birth, the serum creatinine level is identical to the mother’s, but by 1 week of age, the creatinine should decrease to 0.4 mg/dL. The exception is premature infants, in whom the creatinine may not decrease until these children reach 34–35 weeks’ conceptional age because of renal immaturity before that age.

90.5.2 Antibiotic Prophylaxis Neonates with hydronephrosis who are at risk for UTI should be placed on antibiotic prophylaxis with either amoxicillin 20  mg/kg daily or cephalexin 15 mg/kg daily. At 2 months beyond term, the prophylaxis is usually changed to trimethoprim-­sulfamethoxazole suspension, cephalexin, or nitrofurantoin because these medications provide broader antibacterial coverage. In addition, circumcision should be considered in male neonates to minimize the likelihood of UTI.  However, which babies are at risk and therefore who needs prophylaxis is controversial. The risk group UTD P3, High Risk deserves prophylaxis. However, there is significant variability among pediatric urologists and nephrologists in prescribing prophylaxis for children with hydronephrosis. For example, 30% of Canadian pediatric nephrologists recommended CAP for bilateral low-grade ANH compared with 11% of pediatric urologists (Braga et  al. 2014). With high-grade ANH, 73% of nephrologists and 38% of urologists recommended CAP.  Their recommendations for VCUG also were quite variable. Early studies suggested that children with VUR, ectopic ureter and ureterocele, and posterior urethral valves benefit from prophylaxis, but those with hydronephrosis secondary to an abnormality

J. S. Elder

of the UPJ or ureterovesical junction are not at increased risk. More recently, infants with highgrade hydronephrosis (OR 2.40), female gender (OR 3.16), and uncircumcised males (OR 3.63) were at the highest risk, but multivariate analysis suggested that prophylaxis was not beneficial (Zaraba et al. 2014). However, a subsequent prospective study from the same institution indicated that ‘lack of continuous antibiotic prophylaxis’, hydroureteronephosis, and VUR were significant risk factors for febrile UTI (Braga et  al. 2015). Another retrospective study found that the risk of febrile UTI was 7.9% in children receiving prophylaxis vs. 18.7% in those not receiving prophylaxis. Children with ureteral dilation >11  mm, ureterovesical obstruction, and high-grade VUR were at the greatest risk for a febrile UTI (Herz 2014). Most agree that children with low-grade hydronephrosis do not benefit from prophylaxis. Antibiotic resistance is a definite concern when prescribing prophylaxis unnecessarily.

90.5.3 Initial Radiologic Evaluation Radiologic evaluation should be performed to delineate the abnormality responsible for changes in prenatal sonography. Serial renal sonograms, a voiding cystourethrogram (VCUG), and a diuretic renogram usually provide the diagnostic information necessary to guide management, although all of these studies are unnecessary in many children.

90.5.3.1 Renal/Bladder Ultrasound A renal and bladder sonogram should be obtained first. Because neonates have transient oliguria, a dilated or obstructed collecting system may appear normal for the first 24–72  h of life (Fig.  90.1). Ideally, if unilateral hydronephrosis was present prenatally, the renal sonogram should not be obtained until 72 h to maximize its sensitivity. It is most appropriate to wait until 3–4 weeks of age unless the antenatal US showed severe unilateral or bilateral hydronephrosis. Renal length, degree of caliectasis, parenchymal thickness, and presence or absence of ureteral dilation should be documented. The ­ severity of hydronephrosis can be graded from 1

90  Management of Antenatal Hydronephrosis

a

1255

b

Fig. 90.1 (a) Ultrasound of left kidney at 12 h. Normal study. Echolucent areas (arrows) in renal cortex are renal pyramids (normal finding). (b) Ultrasound of same patient at 6 weeks shows grade 4 hydronephrosis

Table 90.2  Grading of hydronephrosis Renal Image Renal parenchymal Grade of Central renal thickness hydronephrosis complex 0 Intact Normal 1 Slight splitting Normal 2 Evident splitting, Normal complex confined within renal border Normal 3 Wide splitting pelvis dilated outside renal border; calyces uniformly dilated 4 Further dilatation Thin of pelvis and calyces (calyces may appearconvex)

to 4 using the Society for Fetal Urology (SFU) grading scale (Table  90.2). Inexperienced radiologists may misinterpret a normal neonatal kidney with hypoechoic pyramids for caliectasis. Most significant urologic anomalies that require surgical correction or long-term urological follow-up are associated with SFU grade 3 or 4 hydronephrosis. More sophisticated analyses, such as the renal resistive index, as well as urinary studies have been assessed, but efforts to demonstrate obstruction have been inconsistent. The degree of pelvocaliectasis correlates closely

with the likelihood that a significant urological condition is present. Lee et al performed a metaanalysis of reports of ANH and determined that the risk of finding postnatal pathology was 11.9% with mild ANH, 45.1% for moderate, and 88.3% for severe ANH. Their definitions of mild, moderate, and severe hydronephrosis depended on gestational age at the time of diagnosis. Similarly, Sidhu et al performed a meta-analysis and found that when postnatal hydronephrosis was SFU grade 1 or 2, there was stabilization or resolution of pelviectasis in 98%, whereas when there was SFU grade 3 or 4, there was stabilization or resolution in only 51%. The bladder should be imaged to detect a dilated posterior urethra (urethral valves), bladder wall thickening, ureteral dilation, inadequate bladder emptying, or a ureterocele. Perineal sonography may demonstrate a dilated prostatic urethra, which is consistent with posterior urethral valves.

90.5.3.2 Voiding Cystourethrogram In selected cases, a VCUG should be performed. This study may demonstrate VUR, posterior urethral valves, or a bladder diverticulum. A radiographic cystogram is preferred over a radionuclide cystogram because the latter does not provide sufficient delineation of bladder and urethral anatomy and because VUR, if present, cannot be graded. In an analysis by the AUA Pediatric Vesicoureteral Reflux Guidelines Committee, an

1256

overall VUR detection rate of 16.2% was found. The mean incidence of VUR into a non-dilated kidney was 4.1%. In cases with ANH and a normal postnatal sonogram, the incidence of VUR was 17%. The prevalence of VUR was significantly higher in girls than boys with ANH. The likelihood is highest if there is SFU grade 3 or 4 hydronephrosis or if a dilated ureter is identified. The chance of identifying VUR on a VCUG is less if there is only SFU grade 1 or 2.

90.5.3.3 What If the Initial Sonogram Is Normal? A common dilemma is whether a full evaluation is necessary if the initial renal sonogram is normal. Assuming a significant degree of fetal renal pelvic dilatation (i.e., >4  mm anteroposterior pelvic diameter before 33 weeks, 7 mm diameter after 33 weeks) was present, the child may have VUR. Blane and colleagues reported that 12% of children with grade V, 31% with grade IV, and 80% with grade III VUR had a normal renal sonogram. However, the AUA Reflux Guidelines determined that the mean incidence of VUR into a non-dilated kidney was 4.1%. Because VUR may cause intermittent renal pelvic dilation, theoretically babies with prenatal hydronephrosis and a normal postnatal sonogram may have VUR, and early diagnosis and medical treatment of VUR may reduce the likelihood of developing reflux nephropathy. On the other hand, others have advocated performing a VCUG only if the postnatal sonogram is abnormal; however, in these reports, neonates with a normal postnatal renal sonogram were not systematically evaluated to determine the real incidence of VUR in this group. Currently, most do not recommend a VCUG unless the postnatal renal sonogram demonstrates grade 3 or 4 hydronephrosis and/or ureteral dilation.

90.5.4 Follow-Up Evaluation and Treatment If bilateral hydronephrosis or unilateral hydronephrosis in a solitary kidney is present, then close monitoring of the serum creatinine and electrolytes is necessary. If the hydronephrosis is caused by posterior urethral valves, then valve ablation

J. S. Elder

should be performed before hospital discharge. If the hydronephrosis is secondary to VUR, the infant should be placed on prophylaxis and managed as described later in this chapter. If the hydronephrosis is SFU grade 3 or 4, the VCUG is negative and bilateral UPJ or ureterovesical junction obstruction is suspected, evaluation with diuretic renography is indicated. If unilateral hydronephrosis and a normal contralateral kidney are present, abnormalities in serum creatinine or electrolytes are uncommon. Nevertheless, these serum studies should be drawn to document that renal function is normal. Usually, follow-up functional radiographic studies can be delayed until 6–12 weeks of age, when renal function is more mature and studies of renal function and obstruction are more likely to be accurate. If the sonogram and VCUG are normal, then only a follow-up sonogram in 6–8 weeks is necessary. In general, if hydronephrosis is discovered on the initial postnatal sonogram, pediatric urologic or nephrologic consultation is advisable to direct subsequent radiologic evaluation and plan therapy.

90.5.4.1 Diuretic Renogram The diuretic renogram is used to determine whether upper urinary tract obstruction is present. It is used to assess the differential renal function and efficiency of drainage of the kidneys. Infants with grade 3 and 4, and occasionally grade 2, hydronephrosis should undergo this study. Mercaptoacetyl triglycine (MAG-3) is generally used and is secreted by the renal tubules. It provides excellent images with minimal background activity. During a diuretic renogram, a small dose of the radiopharmaceutical is injected intravenously. During the first 2–3 min, renal parenchymal uptake is analyzed and compared, allowing computation of differential renal function. Subsequently, excretion is evaluated. After 20–30  min, furosemide 0.5–1  mg/kg is injected intravenously, and the rapidity and pattern of drainage from the kidneys to the bladder are analyzed. If no upper urinary tract obstruction is present, then normally half of the radionuclide is cleared from the renal pelvis within 10–20 min, termed the t½. A t½ > 20 min is consistent with upper urinary tract obstruction, but it is not diagnostic of obstruction, because there are factors

90  Management of Antenatal Hydronephrosis

that can prolong the t½ in addition to an obstructive lesion (see below). A t½ between 15 and 20  min is indeterminate. The images generated usually provide an accurate assessment of the site of obstruction. With unilateral hydronephrosis, the normal kidney should be used as a control for interpretation, and ideally it has already ­accumulated and excreted most of the radionuclide before the furosemide is administered. Numerous variables affect the outcome of the diuretic renogram. For example, newborn kidneys are functionally immature, and in some cases, even normal kidneys may not demonstrate normal drainage following diuretic administration. Dehydration prolongs parenchymal transit and can blunt the diuretic response. Giving an insufficient dose of furosemide may result in slow or inadequate drainage. In addition, a full bladder may impede bladder drainage. Furthermore, if VUR is present, continuous catheter drainage is mandatory to prevent the radionuclide from refluxing from the bladder into the dilated upper tract, causing a prolonged washout phase. Consequently, a urethral catheter should be inserted and bladder drainage measured.

1257

to establish a diagnosis and establish a plan of management. In particularly complicated cases, however, cystoscopy with retrograde pyelography, computed tomography (CT) scan, antegrade pyelography, or a Whitaker antegrade perfusion test is necessary.

90.6 Congenital Anomalies Causing ANH 90.6.1 UPJ Obstruction or Anomalous UPJ

The most common cause of severe hydronephrosis without a dilated ureter or bladder in newborn infants is UPJ obstruction, which results from an intrinsic fibrotic narrowing at the junction between the ureter and renal pelvis (Fig. 90.2). At times, an accessory artery to the lower pole of the kidney also causes extrinsic obstruction, but this finding is rare in newborns with hydronephrosis; it is much more likely to occur in older children and adults. In kidneys with a UPJ obstruction, renal function may be significantly impaired from pressure atrophy. 90.5.4.2 Magnetic Resonance The anomaly is corrected by performing a Urography pyeloplasty, in which the stenotic segment is Magnetic resonance urography (MRU) occasion- excised and the normal ureter and renal pelvis are ally is used to evaluate suspected upper urinary reattached. Success rates are 91–98%. Lesser tract pathology. The child is hydrated and given degrees of UPJ narrowing may cause mild hydrointravenous furosemide. Next, gadolinium-­ nephrosis, which is usually nonobstructive, and DTPA, which is filtered and excreted, is injected typically these kidneys function normally. intravenously and routine T1-weighted and fat-­ Another cause of mild hydronephrosis is fetal suppressed fast spin-echo T2-weighted imaging folds of the upper ureter (Fig. 90.3), which also is performed through the kidneys, ureters, and are nonobstructive. The spectrum of nonobstrucbladder. This study provides superb images of the tive UPJ abnormalities often is termed ‘anomapathology, and methodology has been developed lous UPJ’. to allow assessment of differential renal function Hydronephrosis in many newborns gradually and drainage. There is no radiation exposure, but diminishes or resolves over months to years. The younger children need sedation or general anes- goal of early evaluation is to determine whether a thesia. It is the procedure of choice for delineat- true anatomic obstruction is present that should be ing complex genitourinary pathology (e.g., corrected or whether it is safe to follow the infant cross-fused ectopia with hydronephrosis and/or non-operatively. Defining what constitutes segmental multicystic kidney, cloacal anomaly). obstructive and nonobstructive hydronephrosis is a constant source of debate in pediatric urology. 90.5.4.3 Ancillary Studies Cartwright and colleagues in 1992 studied 80 In most cases, a renal sonogram, VCUG, and neonates with suspected UPJ obstruction. Of 39 diuretic renogram provide sufficient information with unilateral hydronephrosis and at least 35%

J. S. Elder

1258

a

c

b

Fig. 90.2 (a) Ultrasound shows grade 4 left hydronephrosis without dilated ureter. VCUG was normal (b) MAG-3 diuretic renogram. Differential renal function 46% left, 54% right. Right kidney drains spontaneously

before furosemide is administered at 20 min. There is no drainage from the left kidney, consistent with left UPJ obstruction. (c) Follow-up renal US 4 months following left pyeloplasty showing improved hydronephrosis

differential renal function who were managed non-operatively, only six (15%) later underwent pyeloplasty, primarily because of deteriorating differential renal function on renal scintigraphy. Following pyeloplasty, the differential renal function returned to its initial level in these patients. One might question whether early pyeloplasty in these patients would have allowed the renal func-

tion to improve to 50% (normal). The remaining patients managed non-operatively maintained differential function greater than 40%. Onen et  al reported 104 consecutive neonates with unilateral hydronephrosis managed non-operatively, with follow-up as long as 5 years. In follow-up, only seven (7%) underwent pyeloplasty because of reduction in differential renal function of more

90  Management of Antenatal Hydronephrosis

Fig. 90.3  Excretory urogram in infant with bilateral grade 2 hydronephrosis shows bilateral fetal folds of upper ureter (arrows). Note sharp calyces (normal) [from previous chapter]

than 10% or progression of hydronephrosis. Pyeloplasty returned differential renal function to prepyeloplasty levels in all cases. Of 16 patients with significantly reduced renal function on initial scan and grade 4 hydronephrosis, rapid improvement was noted on follow-­up diuretic renography in 15 and the washout curve became nonobstructive in six. In addition, hydronephrosis disappeared in six, improved in six, remained stable in three, and deteriorated in one. The physiology of the resolution or reduction in hydronephrosis and the improvement in differential renal function in these babies is unknown. In another more recent report from the same institution, of 19 newborns with bilateral grade 3 or 4 hydronephrosis, a total of 13 kidneys were subjected to pyeloplasty. Of those managed non-operatively, 21 kidneys were grade 0–2 and grade 3 in two kidneys. The mean follow-up time to achieve maximum improvement in hydronephrosis was 10 months. Although these studies suggest that it is safe or appropriate to manage neonates with a sus-

1259

pected UPJ obstruction non-operatively, an infant’s kidney has greater capacity for improvement in differential renal function than an older child’s. In addition, all of these studies base differential renal function on the uptake during the first 2–3 min of the study, and there is variability in the way this percentage is calculated. Finally, these studies have not reported the pattern of washout on diuretic renography. In a review of renal biopsies obtained at pyeloplasty at the author’s previous institution, 63% showed minimal or no obstructive histologic changes; however, of those with differential function greater than 40%, 21% showed significant histopathologic changes, including reduced glomerular number, glomerular hyalinization, interstitial inflammation, and dysplastic glomeruli (Elder et  al. 1995). Of the kidneys with a differential function less than 40%, 33% showed minimal or no histologic changes. Overall, in 25% of the patients the findings on renal biopsy did not correlate with the computed differential renal function and reflect the need for more sensitive markers of obstruction. The approach of many pediatric urologists to neonates with a suspected UPJ obstruction is as follows. The hydronephrosis is graded from 1 to 4 using the SFU grading scale and a VCUG is obtained for grade 3 and 4 hydronephrosis. Nearly all infants requiring pyeloplasty have grade 3 or 4 hydronephrosis, and those with grade 1 or 2 hydronephrosis do not seem to be at significant long-term risk. If a neonate has an abdominal mass from a hydronephrotic kidney, bilateral severe hydronephrosis, or a solitary kidney, a prompt MAG-3 diuretic renogram is obtained. If signs of obstruction are apparent, prompt pyeloplasty is performed. Otherwise, the diuretic renogram is obtained at 6–12 weeks of age. If the study is obtained when the kidneys are immature, the diuretic response to furosemide might show delayed drainage even with an unobstructed UPJ. If diuretic renography shows at least >40% differential renal function and there is some drainage on the diuretic renogram, the child can be managed non-operatively, regardless of the drainage pattern. A follow-up renal sonogram is performed 3 months later. If the hydronephrosis is unchanged from baseline, a repeat MAG-3 diuretic renogram is obtained. If

1260

there is deterioration in differential renal function or worsening of the diuretic washout curve, pyeloplasty is recommended. However, if these parameters remain stable or improved and the child does not develop a UTI, follow up 3–6 months later with another renal sonogram or MAG-3 diuretic renogram is performed, and management is individualized. It is incumbent on clinicians caring for these infants to have a good understanding of the vagaries of the diuretic renogram and to monitor infants with a suspected UPJ obstruction closely. In addition, a review of the radiologic studies (not just the radiology report) by the pediatric urologist is important. There has been significant progress in the development of minimally invasive techniques in pediatric pyeloplasty, even in infants. Although infant pyeloplasty can be performed through a small incision (lumbotomy or flank muscle-­ splitting) in many centers, some pediatric urologists are performing the procedure with traditional laparoscopic techniques or laparoscopically with the da Vinci robot. Success rates have been compared to series of open surgical repair. Hospital stay and narcotic use are less with the minimally invasive approach. Generally, a transperitoneal approach has been used, with mobilization of the colon. However, on the left side, the pyeloplasty can be performed with a transmesenteric approach.

90.6.2 Multicystic Dysplastic Kidney A multicystic dysplastic kidney is composed of multiple noncommunicating cysts of varying sizes with a stromal component with dysplastic elements. These kidneys do not function. Although multicystic kidney is the most common cause of an abdominal mass in neonates, the vast majority of multicystic kidneys are detected by prenatal sonography. Some clinicians incorrectly assume that multicystic kidney and polycystic kidney are synonymous terms. Polycystic kidney disease is an inherited disorder and has an ‘adult form’ (autosomal dominant) and an ‘infantile form’ (autosomal recessive) and affects both kidneys. In contrast, a multicystic kidney is almost always

J. S. Elder

Fig. 90.4  Renal US in newborn with right multicystic dysplastic kidney, showing kidney replaced by multiple noncommunicating cysts of various sizes with minimal dysplastic parenchyma

unilateral and is usually not an inherited disorder. Sonography of multicystic kidneys is often diagnostic, demonstrating multiple echolucent cysts of varying sizes with no discernible cortex (Fig. 90.4). Occasionally, the cysts may resemble a severe UPJ obstruction with minimal parenchyma, termed the ‘hydronephrotic variant’. The contralateral kidney is abnormal in 5–10% of cases. Renal scintigraphy (MAG-3 or DMSA [dimercaptosuccinic acid] scan) shows non-function, but with current US techniques, renal scintigraphy generally is unnecessary for confirmation. On occasion, there is a segmental multicystic kidney, in which there is a complete duplication anomaly of the upper urinary tract, with the upper pole being multicystic. Previously, a VCUG often was ordered because as many as 15% have contralateral VUR, but currently, it seems unnecessary unless there is contralateral hydronephrosis. The management of a multicystic kidney is becoming less controversial. If an abdominal mass that is symptomatic is present, early nephrectomy is indicated. However, left untreated, most multicystic kidneys become smaller. Potential complications include malignancy and hypertension. In a review of 26 clinical series, no cases of Wilms’ tumor were reported among 1041 children, and the maximum estimated risk is 0.07% (Chang et al. 2018). Tumors arise from the stromal, not the cystic, component of multicystic kidneys. Consequently, even if the

90  Management of Antenatal Hydronephrosis

cysts regress completely, the likelihood that the kidney could develop a neoplasm is not altered. With regard to hypertension, the risk is 3.2%. Generally, a follow-up sonogram is recommended at 6 months of age. If the cysts enlarge, the stromal core increases in size, or hypertension develops, laparoscopic simple nephrectomy is recommended. However, further follow-up sonography is unnecessary unless there is concern regarding the contralateral kidney, because finding a Wilms’ tumor incidentally would be extremely rare. Because of the occult nature of hypertension, annual blood pressure measurement is recommended, and if hypertension occurs, nephrectomy should be considered.

90.6.3 Primary Megaureter (Non-refluxing) A megaureter refers to a wide ureter and may be (1) primary or secondary, (2) obstructive or nonobstructive, and (3) refluxing or non-refluxing. Non-refluxing megaureter results from an aperistaltic segment of the distal ureter that does not allow normal propulsion of urine. In this condition, sonography shows a dilated ureter and renal pelvis with variable renal parenchymal atrophy. VCUG shows no VUR in most cases. Before the antenatal sonography era, most patients with this condition presented with flank pain, flank mass, pyelonephritis, hematuria, or stone disease. Surgical correction consists of excision of the aperistaltic segment, tailoring (also known as a

b

Fig. 90.5  Female newborn with non-refluxing left megaureter discovered by antenatal US (a and b). Ultrasound shows grade 4 hydronephrosis and dilated ureter. Diuretic renogram, 8 weeks of age, showed 50% differential renal function in left kidney. Obstructive drainage pattern was

1261

tapering) of the ureter, and reimplantation of the ureter into the bladder. Although severe hydronephrosis may be present, often there is gradual reduction in hydronephrosis over a period of several years (Fig. 90.5). Braga et al. (2016) reported that the majority of their patients demonstrated resolution of hydronephrosis in a median of 17 months, although those with a mean ureteral diameter >17 mm required surgical intervention. This group found that continuous antibiotic prophylaxis and circumcision reduced the risk of UTI. DiRenzo et al. (2015) found that resolution or improvement of hydronephrosis occurs in all cases of mild postnatal dilation and 60% of those with moderate or severe upper urinary tract dilatation. The British Association of Paediatric Urologists recommends initial nonoperative management of primary obstructive megaureter. In follow-up, surgical intervention is recommended for UTI, pain, worsening hydronephrosis, differential renal function 90% of cases. However, there is a significant risk of post-operative VUR through the ureterocele into the upper pole moiety, which may require subsequent definitive treatment. If the ureterocele is orthotopic, approximately 30% show VUR following TUI, whereas if it is ectopic, 75% have VUR following TUI. TUI is often the only procedure necessary.

TUI of a ureterocele draining a nonfunctioning moiety will not result in the development of any significant degree of function. In recent years, many centers have been performing minimally invasive (laparoscopic) upper pole heminephrectomy. These procedures have been performed either with a retroperitoneal or transperitoneal approach. Common complications include perirenal urinoma and, in some cases, devascularization of the lower pole moiety. The latter complication is most common in infants. Another option is to perform transperitoneal laparoscopic partial nephrectomy with

J. S. Elder

1264

robotic assistance. Consequently, if the hydronephrotic upper pole does not function, this author maintains the infant on antibiotic prophylaxis and laparoscopic upper pole heminephrectomy (or nephrectomy, if the entire kidney does not function) with or without robotic assistance performed electively at 6 months of age, assuming the upper pole system remains hydronephrotic. If the renal scan shows significant upper pole function, however, ureteropyelostomy or ureteroureterostomy, in which the upper pole ureter is anastomosed to the lower pole renal pelvis or ureter, is recommended. This procedure can be performed either at the level of the kidney, with the removal of part of the redundant distal ureter, or low, through an inguinal incision. Total urinary tract reconstruction in neonates and infants is not recommended because of the high complication rate caused by the small size of the infant’s bladder.

90.6.5 Posterior Urethral Valves The most common cause of severe obstructive uropathy in children is posterior urethral valves (PUV), which are tissue leaflets fanning distally from the prostatic urethra to the external urinary sphincter (Fig.  90.7). The incidence of this abnormality is 1:8000. Typically, the leaflets are separated by a slit-like opening. Approximately one-third ultimately develop chronic renal failure or severe renal insufficiency. Prognosis is significantly better if the antenatal sonogram before 24 weeks’ gestation was normal. In 1 study, 9 of 17 patients with PUV whose hydronephrosis was discovered before 24 weeks’ gestation developed renal failure, whereas only one of 14 recognized after 24 weeks’ gestation developed end-stage renal disease. Favorable prognostic factors include a serum creatinine level of less than 0.8–1.0  mg% after bladder decompression, unilateral VUR into a nonfunctioning kidney (‘VURD syndrome’), ascites, and identification of the corticomedullary junction on renal sonography. Early delivery of infants with an antenatal diagnosis of suspected PUV is not recommended, unless there is oligohydram-

nios. If severe bilateral renal dysplasia is present, pulmonary hypoplasia is often also present, and problems with ventilation may result. Initially, a small feeding tube should be passed into the bladder for urinary drainage until electrolyte imbalances can be corrected. A Foley catheter is not recommended because the balloon may cause significant bladder spasm and impede upper tract drainage. Care should be taken ­passing the catheter, as the prostatic urethra is dilated and there is bladder neck hypertrophy; the feeding tube may coil in the prostatic urethra and not drain the bladder. In this setting, catheter irrigation typically results in the fluid coming out of the urethra next to the catheter. AVCUG should be obtained to confirm the diagnosis, and a renal scan should be performed to evaluate the upper tract differential renal function. In newborns, alternative treatments include transurethral endoscopic ablation of PUV, cutaneous vesicostomy, and high diversion (cutaneous pyelostomy). The ideal initial treatment is valve ablation with a small Bugbee electrode, as is used with TUI, or the holmium:YAG laser. In small neonates, the 8 or 9 Fr resectoscope may be too large for the urethra, and a temporary vesicostomy may be necessary. A vesicostomy also should be considered for those with a serum creatinine level that remains significantly elevated after bladder decompression. Cutaneous pyelostomy rarely affords better drainage compared with cutaneous vesicostomy and diverts urine away from the bladder, which may prevent normal bladder growth. However, in selected cases the Sober-­en-­T cutaneous ureterostomy is useful. In this procedure, the upper ureter is brought out to the abdomen and transected, and the distal segment is anastomosed to the renal pelvis; this option allows urine to drain both through the ureterostomy, as well as to the bladder.

90.6.6 Vesicoureteral Reflux Some neonates with medium- and high-grade VUR are detected following the finding of ANH (Figs.  90.8 and 90.9). Approximately 80% of such patients are boys. In the most severe cases of

90  Management of Antenatal Hydronephrosis

a

c

1265

b

d

Fig. 90.7  Male newborn with posterior urethral valves. (a, b) Right (a) and left (b) enlarged kidneys with bilateral grade 3–4 hydronephrosis with echogenic parenchyma. The distal ureters were dilated. (c, d) VCUG shows severe

trabeculated bladder with diverticula, dilated prostatic urethra and grade V left vesicoureteral reflux. Patient underwent transurethral ablation of posterior urethral valves

massive VUR, the bladder may also become distended from aberrant micturition into the upper urinary tract. In the AUA Reflux Guidelines analysis, reflux-related renal scarring was present in 47.9% of those with grades IV–V VUR, but only 6.2% of those with grades I–II VUR. Consequently, in neonates with grades III–V VUR, a DMSA scan is recommended to determine whether

reflux-related renal scarring is present. Initially, neonates with prenatally diagnosed VUR are managed medically. Most are placed on amoxicillin prophylaxis for 2 months, followed by nitrofurantoin or trimethoprim- sulfamethoxazole prophylaxis, and circumcision is recommended for male neonates to decrease the risk for UTIs.

J. S. Elder

1266 Fig. 90.8  Male newborn with antenatal and postnatal bilateral hydronephrosis. VCUG shows bilateral grade IV vesicoureteral reflux. The urethra was normal

a

b

c

Fig. 90.9  Male newborn with antenatal left hydronephrosis in a duplex kidney. No ectopic ureter or ureterocele. (a) Dilated left lower pole, normal upper pole. The distal left lower pole ureter was dilated. (b) VCUG shows

grade V reflux into lower pole, left kidney. (c) DMSA renal scan, left kidney on left. Differential function left 42% right 58%. Normal right kidney and left upper pole, reduced function left lower pole

90  Management of Antenatal Hydronephrosis

Neonates with VUR are more likely to show spontaneous resolution than are older children. Indeed, 20–35% of ureters with grade IV or V VUR have reflux resolution within 2 years; however, a significant proportion develop a breakthrough UTI, and anti-reflux surgery is recommended in these cases. VUR that occurs only when the bladder is quite full or during voiding is more likely to resolve spontaneously the reflux that appears during bladder filling. The success rate for open surgical correction of VUR in infants can be as high as in older children. Another option is subureteral injection of dextranomer microspheres/hyaluronic acid into the ureterovesical junction, in which the success rate is 70–80% with a single injection.

90.7 Conclusions Approximately 1–2% of newborns have an antenatal diagnosis of hydronephrosis or significant renal pelvic dilation. Hydronephrosis is often caused by nonobstructive conditions. The likelihood of significant urologic pathology is directly related to the size of the fetal renal pelvis, and 90% with an anteroposterior diameter >2 cm need surgical intervention or long-term urologic follow-up. Following delivery, antibiotic prophylaxis should be administered if there is severe hydronephrosis or hydroureteronephrosis. Circumcision should be offered to affected boys. A renal sonogram is generally obtained at 3–4 weeks. A VCUG should be obtained if VUR or posterior urethral valves is suspected. If there is grade 3 or 4 hydronephrosis and the VCUG is negative, a diuretic renogram is recommended. Pediatric urologic or pediatric nephrologic consultation is helpful in planning evaluation and treatment. Antenatal recognition of hydronephrosis allows postnatal diagnosis and treatment of urologic pathology, preventing complications of pyelonephritis and obstruction. In the past decade, significant progress has been made in the development of minimally invasive treatment options.

1267

References Braga LH, Ruzhynky V, Pemberton J et  al (2014) Evaluating practice patterns in postnatal management of antenatal hydronephrosis: a national survey of Canadian pediatric urologists and nephrologists. Urology 83:909–914 Braga LH, Farrokhyar F, D’Cruz J et al (2015) Risk factors for febrile urinary tract infection in children with prenatal hydronephrosis: a prospective study. J Urol 193:1766–1771 Braga LH, D’Cruz J, Rickard M et al (2016) The fate of primary nonrefluxing megaureter: a prospective outcome analysis of the rate of urinary tract infections, surgical indications and time to resolution. J Urol 195:1300–1305 Chang A, Sivananthan D, Nataraja RM et  al (2018) Evidence-based treatment of multicystic dysplastic kidney: a systematic review. J Pediatr Urol 14:510–519 DiRenzo D, Persico A, DiNocola M et  al (2015) Conservative management of primary non-refluxing megaureter during the first year of life: a longitudinal observational study. J Pediatr Urol 11:226.e1–6 Elder JS, Stansbrey R, Dahms BB et al (1995) Renal histologic changes secondary to ureteropelvic junction obstruction. J Urol 154:719–722 Herz D, Merguerian P, McQuiston L (2014) Continuous antibiotic prophylaxis reduces the risk of febrile UTI in children with asymptomatic antenatal hydronephrosis with either ureteral dilation, high-grade vesicoureteral reflux, or ureterovesical junction obstruction. J Pediatr Urol 10:650–654 Hodhod A, Capolicchio JP, Jednak R et  al (2016) Evaluation of urinary tract dilation classification system for grading postnatal hydronephrosis. J Urol 195:725–730 Nguyen HT, Herndon CD, Cooper C et  al (2010) The Society for Fetal Urology consensus statement on the evaluation and management of antenatal hydronephrosis. J Pediatr Urol 6:212–231 Nguyen HT, Benson CB, Bromley B et  al (2014) Multidisciplinary consensus on the classification of prenatal and postnatal urinary tract dilation (UTD classification system). J Pediatr Urol 10:982–998 Reuss A, Wladimiroff JW, Steward PA et  al (1988) Noninvasive management of fetal obstructive uropathy. Lancet 2:949 Sinha A, Bagga A, Krishna A et al (2013) Revised guidelines on management of antenatal hydronephrosis. Indian J Nephrol 23:83–97 Zaraba P, Lorenzo AJ, Braga LH (2014) Risk factors for febrile urinary tract infection in infants with prenatal hydronephrosis: comprehensive single center analysis. J Urol 191:1614–1618

Upper Urinary Tract Obstructions

91

Leon Chertin and Boris Chertin

91.1 Pelviureteric Junction Obstruction 91.1.1 Historical Overview The detection of renal abnormalities during prenatal ultrasonography was first reported in the beginning of the seventeenth century (Garrett et al. 1970). Since then, the routine use of ultrasonography for the detection of congenital anomalies has become a part of routine care during the antenatal period. Currently, it is estimated that genitourinary anomalies comprise nearly 20% of all prenatally detected fetal anomalies. Amongst these, hydronephrosis is one of the most commonly detected anomalies, seen in approximately 1–5% of all pregnancies and occurs due to various causes. Thus, we have an increasing number of patients who are presenting to the clinician with a presumptive diagnosis, rather than a symptom, and sometimes before they are even born. Although the initial reports of surgical outcome of the correction of the neonatal PUJ obstruction

L. Chertin Sakler School of Medicine, Tel Aviv University, Tel Aviv, Israel B. Chertin (*) Department of Urology, Hebrew University, Jerusalem, Israel e-mail: [email protected]

were excellent (King et  al. 1984), following observations regarding renal function preservation during conservative treatment have started a new era in the treatment of antenatal hydronephrosis (Ransley et  al. 1990). In our earlier reports, we clearly demonstrated that prenatal diagnosis of hydronephrosis with close follow-up after delivery is much superior, in terms of renal function preservation, compared to those children who were diagnosed to have PUJ obstruction due to clinical symptoms (Chertin et  al. 1999). Multiple reports demonstrated that approximately 30% of children will require surgery during surveillance, therefore expectant management will spare the majority of children from surgery (Koff and Campbell 1992; Koff 2000; Ulman et  al. 2000; Onen et  al. 2002; Chertin et al. 2002).

91.1.2 Incidence The overall incidence of neonatal hydronephrosis, which leads to the diagnosis of PUJ obstruction, approximates 1  in 500 births. The ratio of males to females is 2:1  in the neonatal period, with left-sided lesions occurring in 60% of cases. In the newborn period, a unilateral process is most common, but bilateral PUJ obstruction was found in 10–49% of neonates in some reported series (Gokce et al. 2012).

© Springer Nature Switzerland AG 2023 P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_91

1269

L. Chertin and B. Chertin

1270

91.1.3 Etiopathogenesis PUJ obstruction is classified as intrinsic, extrinsic, or secondary. Intrinsic obstruction results from the failure of transmission of the peristaltic waves across the PUJ, with the failure of urine to be propulsed from the renal pelvis into the ureter, which results in multiple ineffective peristaltic waves that eventually causes hydronephrosis by incompletely emptying the pelvic contents. Extrinsic mechanical factors include aberrant renal vessels, bands, adventitial tissues, and adhesions that cause angulation, kinking, or compression of the PUJ.  Extrinsic obstruction may occur alone but usually coexists with intrinsic ureteropelvic junction pathology. Secondary PUJ obstruction may develop as a consequence of concomitant severe vesicoureteric reflux (VUR), which occurs in 15–30% of children who have ipsilateral PUJ obstruction, in which a tortuous ureter may kink proximally.

91.1.4 Pathophysiology Renal morphogenesis is a complex, temporally and spatially regulated process by which precursor cells develop into a structurally and functionally normal kidney. Abnormal or dysregulated renal development results in a wide range of renal abnormalities, collectively known as congenital anomalies of the kidney and urinary tract (CAKUT), which compose the most common cause of end-stage renal disease (ESRD) in children. In humans, kidney and urinary tract development begins at approximately 3 weeks gestation, with the formation of the initial urinary excretory precursor, the pronephros, which undergoes complete involution (Little and McMahon 2012; Combes et al. 2015; Short and Smyth 2016). By 34–36 weeks gestation, nephrogenesis is complete, and the structural and functional relationship of each nephron segment is fully developed. On average, 1 million (range 0.2–2.7 million) individual nephrons in each kidney arise from the embryonic precursor cells. Concurrent with the morphogenesis of the kidney, functional development of the fetal kidney

also progresses with increasing gestational age. Prenatally, the placenta controls fluid and electrolyte homeostasis, and the primary function of the fetal kidney is the production of urine to maintain amniotic fluid volume. In the later stages of gestation, urine output, tubular function, and glomerular filtration increase with gestational age. After birth, the neonatal kidney undergoes physiologic changes to adapt to the extrauterine environment; rapid changes occur over the first several weeks and continue until reaching adult levels at 1–2 years of life. Changes in hydrostatic pressure in the renal pelvis and individual nephrons are critical in determining the effects of obstruction on Glomerular Filtration Rate (GFR). Normal proximal intratubular pressure of 12 mmHg increases in direct correlation with rising intrapelvic pressure with a maximum of approximately 40  mmHg. Higher intrapelvic pressure of 50–70 mmHg is not transmitted to the proximal tubules, perhaps because of compression of the renal papilla. Continuation of obstruction increases in preglomerular vascular resistance (afferent arteriolar vasoconstriction), which in turn could lead to the increased accumulation of fibrotic tissue and appearance of renal dysplasia. Furthermore, chronic partial obstruction leads to the gradual decrease in GFR and is accompanied by an increase in the fractional excretion of filtered sodium, indicating decreased tubular reabsorption. Leaving the obstruction untreated might lead to the irreversible tubular and glomerular injury with eventual renal loss.

91.1.5 Diagnosis 91.1.5.1 Prenatal Diagnosis The bladder is visualized by 14 weeks gestation. The ureters are usually not seen in the absence of distal obstruction or reflux. The fetal kidney may be visualized at the same time as the bladder. If not, they are always visualized by 16th weeks gestation. However, it is not until 20–24 weeks gestation, when the fetal kidney is surrounded by fat, that the internal renal structures appear distinct. Renal growth can then be assessed easily. Beyond 20 weeks, fetal urine production is the

91  Upper Urinary Tract Obstructions

main source of amniotic fluid. Therefore, major abnormalities of the urinary tract may result in oligohydramnios. Because of the distinct urine tissue interface, hydronephrosis can be detected as early as 16 weeks gestation. An obstructive anomaly is recognized by demonstrating dilated renal calyces and pelvis. A multitude of measurements and different gestational age cut-off points have been recommended in the assessment of fetal obstructive uropathy. Routine estimation of anteroposterior (AP) diameter of renal pelvis in the fetus with ­hydronephrosis is considered a useful marker for classification of renal dilatation and possible obstruction. AP renal pelvis threshold values ranged between 2.3 and 10 mm. Positive predictive values for pathological dilatation confirmed in the neonate ranged between 2.3 and >40% for AP renal measurements of 2–3 mm and 10 mm, respectively. One study, which included more than 46,000 screening patients, published the standards regarding renal pelvic measurement. This study clearly demonstrated that only fetuses exhibiting third-trimester AP renal pelvis dilatations >10 mm would merit postnatal assessment. In order to standardize postnatal evaluation of prenatal hydronephrosis a grading system of postnatal hydronephrosis was implemented in 1993 by the Society for Fetal Urology (SFU). In SFU system, the status of calices is paramount while the size of the pelvis is less important. In SFU grading of hydronephrosis, there is no hydronephrosis in Grade 0. At Grade 1, the renal pelvis is only visualized. Grade 2 of hydronephrosis is diagnosed when a few (but not all) renal calices are identified in addition to the renal pelvis. Grade 3 hydronephrosis requires that virtually all calices are depicted. Grade 4 hydronephrotic kidneys will exhibit similar caliceal status with the involved kidney exhibiting parenchymal thinning. Often this classification is applied also on prenatal hydronephrosis. We have published our data regarding prenatal findings with the special emphasis on the natural history of hydronephrosis during the postnatal period. Our data shows that SFU grade of prenatal hydronephrosis is not a significant predictive factor for surgery in unilateral hydronephrosis. However, SFU Grades

1271

3–4 prenatal bilateral hydronephrosis indicates that the majority of the children will require surgical correction during the postnatal period. In 2014, the “urinary tract dilation (UTD)” classification system was introduced to replace the SFU system and other grading systems. It consists of 6 parameters, namely APD of the renal pelvis, urinary tract dilation, parenchymal thickness, parenchymal appearance, ureteral status, and bladder status, furthermore distinguishing whether these parameters are antenatal (“A”) or postnatal (“P”). There are two antenatal and three postnatal categories of risk: A1 (low risk) or A2–3 (intermediate/high risk) for antenatal UTD; and P1 (low), P2 (intermediate), P3 (high risk) for postnatal UTD.  Persistent UTD A1 or UTD A2 to A3 warrants postnatal evaluation (Nguyen et al. 2014). In the case of severe prenatal bilateral hydronephrosis, severe hydroureteronephrosis, or severe impairment of the solitary kidney, fetal bladder aspiration for urinary proteins and electrolytes is recommended from 17 weeks of gestation in some reports in order to predict the renal injury secondary to obstructive uropathy. Fetal urinary sodium level less than 100 mmol/L, chloride level of less than 90 mmol/L and an osmolality of less than 210 mOsm/kg are considered as prognostic features for good renal function.

91.1.5.2 Clinical Presentation The most common presentation is abdominal flank mass. Fifty percent of abdominal masses in newborns are of renal origin with 40% being secondary to PUJ obstruction. Other clinical presentations include urinary tract infection irritability, vomiting, and failure to thrive. Ten to 35% of PUJ obstructions are bilateral and associated abnormalities of the urinary tract are seen in about 30%. PUJ problems are often associated with other congenital anomalies, including imperforated anus, contralateral dysplastic kidney, congenital heart disease, VATER syndrome, and esophageal atresia. 91.1.5.3 Differential Diagnosis With the increasing number of antenatally diagnosed hydronephrosis it is difficult to interpret

L. Chertin and B. Chertin

1272

the underlying pathology and its significance. Severe obstructive uropathies are detrimental to renal function. However, on the other hand, hydronephrosis without ureteral or lower tract anomaly is common. The important aspect of postnatal investigations is to identify the group of patients who will benefit from early intervention and those who need to be carefully followed. Ultrasound: Follow-up ultrasound examination is necessary in the postnatal period in antenatally detected hydronephrosis. If bilateral hydronephrosis is diagnosed in utero in a male Fig. 91.1 (a) A sagittal plane scan through the obstructed right kidney confirms obstruction at the level of the pelviureteric junction. (b) 99Tc MAG3 scan in the above patient. Clearance curve for right kidney confirming the high-grade obstruction on this side

a

b

infant, postnatal evaluation should be carried out within 24 h primarily because of the possibility of posterior urethral valves. If the ultrasound scan is negative in the first 24–48 h in any patient with unilateral or bilateral hydronephrosis, a repeat scan should be performed after 5–10 days, recognizing that neonatal oliguria may mask a moderately obstructive lesion. If hydronephrosis is confirmed on the postnatal scan, further careful scan of the kidney, ureter, bladder, and in boys, the posterior urethra is essential (Fig. 91.1a).

91  Upper Urinary Tract Obstructions

Radionuclide Scans: DTPA is completely filtered by the kidneys at maximum concentration of 5% being reached in 5  min, falling to 2% at 15 min. Recently, it has been reported that use of tracers that rely on tubular extraction such as 123 I-Hippuran and 99Tc MAG3 (Fig.  91.1b) may improve diagnostic accuracy. The kidney of the young infant is immature; renal clearance, even when corrected for body surface, progressively increases until approximately 2 years of age. Therefore, the renal uptake of tracer is particularly low in infants, and there is a high background activity. Thus, the traces with a high extraction rate, such as 123 I-Hippuran and 99Tc MAG3, provide reasonable images enabling the estimation of the differential kidney function during the first few weeks of life. It is also helpful in assessing the size, shape, location, and function of the kidney. Diuretic augmented renogram is a provocative test and is intended to demonstrate or exclude obstructive hydronephrosis by stressing an upper urinary tract with a high urine flow. Obstruction usually is defined as a failure of tracer washout after diuretic stimulation. If unequivocal, it eliminates the need for further investigations. In equivocal cases, F15 in which furosemide is given 15  min before the test provides a better assessment of the drainage of upper urinary tract. Forced hydration prior to a scan increases THE predictive value of non-obstructed patterns by up to 94%. Since glomerular filtration and glomerular blood flow are still low in the newborn, the handling of isotype is unpredictable and can be misleading. Functional magnetic resonance urography (MRU) has been recently proposed by many study groups as an alternative technique to evaluate the drainage curve and split renal function (SRF) in obstructive uropathy. This method allows the precise understanding of the kidney anatomy while providing information regarding renal functioning without radiation exposure obviating the need to use contrast media. When surgical correction is planned, MRU aids in clearly identifying anatomically crossing vessels and obstructive pathology. Pressure-Flow Study: In the equivocal cases and in the presence of impaired function, the

1273

pressure-flow study (Whitaker Test) and antegrade pyelography may be necessary to confirm or exclude obstruction. Whitaker Test is based on the hypothesis that if the dilated upper urinary tract can transport 10 ml/min without an inordinate increase in pressure, the hydrostatic pressure under physiological conditions should not cause impairment of renal function and the degree of obstruction if present is insignificant. However, it is an invasive test and is seldom required. Antegrade pyelography may be performed with ultrasound guidance in patients where diagnosis is difficult. Retrograde pyelography is seldom required to determine the status of ureters. The disadvantages include difficulty in ureteral catheterization in neonates, trauma, and edema that may change partial obstruction to the complete one. In patients where diagnosis is equivocal, serial examinations may be necessary. Routine use of micturating cystourethrogram (MCUG) in patients with antenatal unilateral hydronephrosis is controversial. Some authors advocate regular use of MCUG as a part of postnatal evaluation citing 15–30% of incidence of concomitant VUR either uni- or contralateral. Others recommend only performing MCUG in patients with SFU Gr III and IV hydronephrosis. We have abounded to perform MCUG routinely in children with unilateral antenatal hydronephrosis based on the fact that even if the reflux exists usually it is of low grade and does not require any treatment. We reserve MCUG only for patients with bilateral hydronephrosis or for those whose ureter was seen at any stage of antenatal or postnatal follow-up.

91.1.6 Management A considerable controversy exists regarding the management of newborn urinary tract obstructions. Some authors advocate early surgical intervention to prevent damage to maturing nephrons, while others feel that early surgery carries no specific benefit. During late prenatal and early postnatal life, there is progressive increase in glomerular filtration rate. Additionally, this transition is associated with an

1274

abrupt decline in urine output from what appears to be a quite high in utero output to a rather low early neonatal level of urine production. These physiological observations may explain the common observation of hydronephrosis detected antenatally, which on postnatal follow-up reverts to an unobstructed pattern. Surgery is usually undertaken in infants whose renal function deteriorates during the observation period. We have analyzed our database of 343 children (260 males and 83 females) with an antenatal diagnosis of hydronephrosis, which led to the postnatal diagnosis of PUJ obstruction, who were deliberately followed up conservatively at our department over a 16-year period, in order to define which factors lead to surgery (Chertin et  al. 2006). One hundred and seventy-nine children (52.2%) required surgical correction in the course of conservative management. Average age at surgery was 10.6 months (range 1 month–7 years). Of these, 50% underwent surgery during the first 2 years of life and majority of the remaining patients underwent surgery between the age of 2 and 4. Only two patients required surgery later on. Univariate analysis revealed that child sex, side of hydronephrosis are not significant predictive factors for surgery. However, SFU Grades 3–4 of postnatal hydronephrosis (p  6, >7, and 8  mm has been described in the first trimester and >10, 12, and 15  mm in the early second trimester (Taghavi et  al. 2017). Fetal megacystitis is also seen in urethral stenosis or atresia (7.4%), Prune Belly Syndrome (3.8%), Megacystis-Microcolon Intestinal Hypoperistalsis Syndrome (MMIHS) (1.1%), and Cloacal anomalies (0.7%). Additional sonographic features reported in b association with a distended bladder include bladder wall thickness (>3 mm) and the “keyhole sign”, which represent a dilatation of the posterior urethra in patients with posterior urethral obstruction. The keyhole sign is not a reliable predictor of the diagnosis of PUV, it has a high sensitivity (94%) but low specificity (43%) for the prenatal diagnosis of PUV (Bernardes et al. 2009). It is also present in approximately a third of patients with other pathologies, such as VUR, primary megaureter, ectopic obstructive ureter Fig. 94.5  Postnatal ultrasound showing renal dysplasia (Bernardes et al. 2009), and in a minority of cases (small hyperechoic kidneys) in a newborn with prenatal of Prune Belly Syndrome and MMIHS (Osborne diagnosis of bilateral hydronephrosis and oligohydramnios et al. 2011). A neurogenic bladder may also have a keyhole appearance, most likely due to the hypotonic bladder neck and dyssynergic vesical Table 94.1  Prenatal ultrasonographic features suggestive of PUV contraction. Other prenatal sonographic variables have Bilateral uretero-hydronephrosis Oligohydramnios been shown to be poor prognostic factors and Abnormal renal parenchyma/cystic appearance/renal being discriminative between PUV and VUR: hyperechogenicity Distended bladder/fetal megacystis oligohydramnios, bilateral uretero-­ hydronephrosis, and abnormal renal parenchyma Dilated posterior urethra/keyhole sign Thick-walled bladder (>3 mm) being more common in PUV patients (Fig. 94.5). Urinary ascites or perirenal urinoma Moreover, anhydramnios and ascites or perirenal urinoma are exclusively described in fetuses later diagnosed with PUV and in none of the fetuses calcium