Clinical Procedures in Emergency Medicine [5 ed.] 9781416036234, 2008041760

Citation preview

clinical procedures in

emergency medicine

A S S O C I A T E

E D I T O R S

Catherine B. Custalow,

MD, PhD Associate Professor, Retired Department of Emergency Medicine University of Virginia School of Medicine Charlottesville, Virginia

Arjun S. Chanmugam, MD, MBA Associate Professor and Director of Education Department of Emergency Medicine The Johns Hopkins School of Medicine Baltimore, Maryland

Carl R. Chudnofsky, MD Chairman Department of Emergency Medicine Albert Einstein Medical Center Associate Professor Jefferson Medical College Philadelphia, Pennsylvania

LTC John McManus, MD, FACEP Adjunct Assistant Professor Department of Emergency Medicine Oregon Health and Science University Portland, Oregon Clinical Research Physician Army Institute of Surgical Research Brooke Army Medical Center Fort Sam Houston, Texas

clinical procedures in

Emergency Medicine F I F T H

E D I T I O N

E D I T O R S

James R. Roberts,

MD, FACEP, FAAEM, FACMT Professor of Emergency Medicine Vice Chair, Department of Emergency Medicine Senior Consultant, Division of Toxicology Drexel University College of Medicine Hahnemann University Medical Center Chairman Department of Emergency Medicine Director, Division of Toxicology Mercy Catholic Medical Center Philadelphia, Pennsylvania

Jerris R. Hedges,

MD, MS Professor and Dean John A. Burns School of Medicine University of Hawaii—Manoa Honolulu, Hawaii Professor and Vice-Dean, Emeritus Department of Emergency Medicine Oregon Health and Science University School of Medicine Portland, Oregon

1600 John F. Kennedy Blvd. Ste 1800 Philadelphia, PA 19103-2899

CLINICAL PROCEDURES IN EMERGENCY MEDICINE Copyright © 2010 by Saunders, an imprint of Elsevier Inc.

ISBN: 978-1-4160-3623-4

All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permissions may be sought directly from Elsevier’s Rights Department: phone: (+1) 215 239 3804 (US) or (+44) 1865 843830 (UK); fax: (+44) 1865 853333; e-mail: [email protected]. You may also complete your request on-line via the Elsevier website at http://www.elsevier.com/permissions.

Notice Knowledge and best practice in this field are constantly changing. As new research and experience broaden our knowledge, changes in practice, treatment and drug therapy may become necessary or appropriate. Readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindications. It is the responsibility of the practitioner, relying on their own experience and knowledge of the patient, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the Editors assumes any liability for any injury and/or damage to persons or property arising out of or related to any use of the material contained in this book. The Publisher Previous editions copyrighted 2004, 1998, 1991, 1985. Library of Congress Cataloging-in-Publication Data   Clinical procedures in emergency medicine / editors, James R. Roberts, Jerris R. Hedges ; associate editors, Catherine B. Custalow … [et al.].—5th ed.     p. ; cm.   Includes bibliographical references and index.   ISBN 978-1-4160-3623-4   1.  Emergency medicine.  I.  Roberts, James R., 1946-  II.  Hedges, Jerris R.  III.  Title: Emergency medicine.   [DNLM:  1.  Emergency Treatment.  2.  Emergencies.  WB 105 C641 2010]   RC86.7.C55 2010   616.02′5—dc22 2008041760

Acquisitions Editor: Stefanie Jewell-Thomas Developmental Editor: Dee Simpson Design Direction: Steven Stave Illustrations Manager: Kari Wszolek Marketing Manager: Courtney Ingram

Working together to grow libraries in developing countries Printed in United States of America Last digit is the print number:  9  8  7  6  5  4  3  2

www.elsevier.com | www.bookaid.org | www.sabre.org

To Lydia, my eternally supportive wife; to Martha, a newly minted RN with unbridled enthusiasm and passion for emergency medicine; and to Matt, always an inspiration for everything. You all make my life worthwhile. To Cathy Custalow, MD who picked up the slack everywhere. To Gary Setnik, MD and Todd Thomsen, MD for their illustrative input. To Rose Bacher who kept us on target and productive. To the folks at Elsevier who made this edition a reality: Dee Simpson for gargantuan help with all aspects of this project; to Judy Fletcher, supportive yet again; Berta Steiner who made it all come together. J.R.R.

As the Fifth edition comes together, I wish to acknowledge the support that Catherine Custalow has brought to this edition by enhancing the overall text and stepping up as I ventured into new areas of responsibility. Without her efforts, this edition would not be possible. As always, I am appreciative of my family and the years of sacrifice they have made such that this knowledge could be made available to aid future generations of emergency physicians and other practitioners—who themselves seek to alleviate suffering and extend lives. J.R.H.

To my son, Nick and to the memory of my daughter, Lauren. To James and Doris Bomberger who first were parents and now are friends. To Dr. Harold F. Young for reasons that are obvious to anyone who knows him. C.B.C.

To my wonderful wife Karen, thank you for your inspiration, support and remarkable partnership; to Sydney, William and Nathan who will always be reminders of what is truly important; to my father who helped to foster the passion of medical education as a public service. Finally to those who practice and teach emergency medicine, may this book serve you well. A.S.C.

The Fifth edition of the book is dedicated to my past, current, and future residents at Albert Einstein Medical Center… your passion for learning, your devotion to healing, and your zeal for life are truly inspirational. C.R.C.

My dedication is first in memory of my Grandfather William “Big Daddy” McDonald for his love, wisdom, and guidance. Also, for my family Laura and our two children for their current and future love and support. Finally, for our military personnel who continually sacrifice for our freedom. J.G.M.

C O N T R I B U T O R S

Benjamin S. Abella, MD, MPhil

Assistant Professor, Department of Emergency Medicine and Associate Director, Center for Resuscitation Science, University of Pennsylvania, Philadelphia, Pennsylvania Artificial Perfusion during Cardiac Arrest

Bruce D. Adams, MD, FACEP

Clinical Professor of Emergency Medicine, Medical College of Georgia, Augusta, Georgia; Chief, Department of Clinical Investigation, and Chief, Department of Emergency Medicine, William Beaumont Army Medical Center, Fort Bliss, Texas; Colonel, Medical Corps, United States Army

Central Venous Catheterization and Central Venous Pressure Monitoring

James T. Amsterdam, DMD, MD, MMM, FACEP, FACPE

Professor of Clinical Emergency Medicine, Penn State University College of Medicine, Hershey; Adjunct Professor of Emergency Medicine, Drexel University College of Medicine, Philadelphia; Chair/Service Line Director, Department of Emergency Medicine, York Hospital, York, Pennsylvania Regional Anesthesia of the Head and Neck

Heatherlee Bailey, MD, FAAEM, FCCM

Assistant Professor of Emergency Medicine, Director of Critical Care Education, and Medical Student Critical Care Clerkship Director, Department of Emergency Medicine, Drexel University College of Medicine, Philadelphia, Pennsylvania Mechanical Ventilation

J. Dave Barry, MD

Clinical Assistant Adjunct Professor, Department of Surgery and Emergency Medicine, University of Texas Health Science Center at San Antonio, San Antonio, Texas; Adjunct Assistant Professor, Department of Military and Emergency Medicine, Uniformed Services University of the Health Sciences, Bethesda, Maryland; Faculty Emergency Physician, Department of Emergency Medicine, and Associate Program Director, San Antonio Uniformed Services Health Education Consortium (SAUSHEC) Emergency Medicine Residency Program, Brooke Army Medical Center, Fort Sam Houston, Texas Vital Signs Measurement

Steven J. Bauer, MD

Clinical Assistant Professor of Emergency Medicine, West Virginia University Hospitals-East; Jefferson Memorial Hospital, Ranson, West Virginia Alternative Methods of Drug Administration

Lance B. Becker, MD, FAHA

Professor, Department of Emergency Medicine, and Director, Center for Resuscitation Science, University of Pennsylvania, Philadelphia, Pennsylvania Artificial Perfusion during Cardiac Arrest

Kip Benko, MD, MS

Assistant Clinical Professor of Emergency Medicine, University of Pittsburgh School of Medicine; Faculty Staff, Presbyterian University Hospital, Pittsburgh, Pennsylvania Emergency Dental Procedures

Edward S. Bessman, MD

Assistant Professor, Department of Emergency Medicine, The Johns Hopkins University; Chairman, Department of Emergency Medicine, Johns Hopkins Bayview Medical Center, Baltimore, Maryland Emergency Cardiac Pacing

Courtney A. Bethel, MD

Clinical Assistant Professor of Emergency Medicine, Drexel University College of Medicine—Medical College of Pennsylvania and Hahnemann Medical School; Staff Physician, Mercy Catholic Medical Center, Philadelphia, Pennsylvania Burn Care Procedures

Barbara K. Blok, MD

Assistant Professor, Division of Emergency Medicine, University of Colorado Denver School of Medicine, Aurora; Assistant Residency Director, Denver Health Residency in Emergency Medicine, Denver Health Medical Center, Denver, Colorado Thoracentesis

Michael E. Boczar, DO

Assistant Clinical Professor, Department of Emergency Medicine, University of Michigan, Ann Arbor, and Hurley Hospital, Flint, Michigan Resuscitative Thoracotomy

vii

Contributors

Sudip Bose, MD, FACEP, FAAEM

Assistant Clinical Professor, Department of Emergency Medicine, University of Illinois, Chicago; Attending Emergency Physician, Department of Emergency Medicine, Advocate Christ Medical Center, Oak Lawn, Illinois; Former Major, United States Army Cricothyrotomy and Transtracheal Jet Ventilation

Thomas A. Brabson, DO, MBA, FACOEP, FACEP

Clinical Associate Professor, Department of Emergency Medicine, Philadelphia College of Osteopathic Medicine, Philadelphia, Pennsylvania; Medical Director, Department of Emergency Medicine, City Campus, and Department of Emergency Medical Services, AtlantiCare Regional Medical Center, Atlantic City, New Jersey Prehospital Immobilization

William J. Brady, MD

Professor of Emergency Medicine and Medicine, University of Virginia; Vice Chair, Department of Emergency Medicine, University of Virginia Hospitals, Charlottesville, Virginia Basic Electrocardiographic Techniques

G. Richard Braen, MD viii

Professor and Chairman, Department of Emergency Medicine, and Assistant Dean of Graduate Medical Education, University at Buffalo School of Medicine, Buffalo, New York Culdocentesis

N. Adam Brown, MD

Clinical Instructor, Robert Wood Johnson, New Brunswick; Attending Physician, Robert Wood Johnson at Rahway, Rahway, New Jersey Otolaryngologic Procedures

James H. Bryan, MD, PhD

Assistant Professor of Emergency Medicine, Oregon Health and Science University; Assistant Chief, Emergency Medicine, Veterans Affairs Medical Center, Portland, Oregon Alternative Methods of Drug Administration

Kenneth H. Butler, DO

Associate Professor of Emergency Medicine, Department of Emergency Medicine, University of Maryland School of Medicine; Associate Director, Emergency Medicine Residency, University of Maryland Medical Center, Baltimore, Maryland Incision and Drainage

Stacie E. Byers, DO, FACEP

Attending Physician, Emergency Medicine, St. Anthony Medical Center, Denver, Colorado Splinting Techniques

Sharon K. Carney, MD

Clinical Assistant Professor of Emergency Medicine, Department of Emergency Medicine, Drexel University College of Medicine, Philadelphia; Director, Department of Emergency Medicine, Mercy Fitzgerald Hospital, Darby, Pennsylvania Intravenous Regional Anesthesia

Merle A. Carter, MD, FACEP

Director, Emergency Medicine Residency Program, Albert Einstein Medical Center; Assistant Professor, Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania Compartment Syndrome Evaluation

Theodore C. Chan, MD

Professor of Clinical Medicine, University of California, San Diego; Medical Director, Department of Emergency Medicine, University of California, San Diego, Hospitals, San Diego, California Basic Electrocardiographic Techniques

Theodore A. Christopher, MD

Professor and Chairman, Department of Emergency Medicine, Thomas Jefferson University, Philadelphia, Pennsylvania Injection Therapy of Bursitis and Tendinitis

Carl R. Chudnofsky, MD

Chairman, Department of Emergency Medicine, Albert Einstein Medical Center; Associate Professor, Jefferson Medical College, Philadelphia, Pennsylvania Splinting Techniques

Joseph E. Clinton, MD

Professor and Head, Department of Emergency Medicine, University of Minnesota School of Medicine; Chief, Department of Emergency Medicine, Hennepin County Medical Center, Minneapolis, Minnesota Basic Airway Management and Decision-Making

Wendy C. Coates, MD

Professor of Medicine, and Chair, Acute Care College, David Geffen School of Medicine at UCLA, Los Angeles; Director, Medical Education, Department of Emergency Medicine, Harbor-UCLA Medical Center, Torrance, California Anorectal Procedures

Catherine B. Custalow, MD, PhD

Associate Professor, Retired, Department of Emergency Medicine, University of Virginia School of Medicine, Charlottesville, Virginia Educational Aspects of Emergency Department Procedures

Anthony J. Dean, MD

Assistant Professor of Emergency Medicine and Assistant Professor of Emergency Medicine in Radiology, Director, Division of Emergency Ultrasonography, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania Bedside Laboratory and Microbiologic Procedures

Adjunct Assistant Professor, Department of Military and Emergency Medicine, Uniformed Services University of the Health Sciences, Bethesda, Maryland; Staff Emergency Physician, Wilford Hall Medical Center, Lackland Air Force Base, Texas Central Venous Catheterization and Central Venous Pressure Monitoring

Kenneth Deitch, DO

Assistant Professor, Jefferson Medical College, Thomas Jefferson University; Associate Research Director, and Attending Physician, Department of Emergency Medicine, Albert Einstein Medical Center, Philadelphia, Pennsylvania Intraosseous Infusion

William R. Dennis, MD, MPH

Head of EMS for Navy Medicine East and National Capital Area, Emergency Medicine, Naval Medical Center Portsmouth, Portsmouth, Virginia Ophthalmologic Procedures

Denis J. Dollard, MD

Clinical Assistant Professor, Department of Emergency Medicine, Drexel University College of Medicine; Director, Emergency Medicine, Mercy Hospital of Philadelphia, Philadelphia, Pennsylvania Radiation in Pregnancy and Clinical Issues of Radiocontrast Agents

Timothy B. Erickson, MD, FACEP, FAACT, FACMT

Professor, Emergency Medicine and Clinical Toxicology, Associate Head, Department of Emergency Medicine, and Director, Division of Clinical Toxicology, University of Illinois at Chicago, Chicago, Illinois Procedures Pertaining to Hypothermia and Hyperthermia

Brian D. Euerle, MD

Associate Professor, University of Maryland School of Medicine; Attending Physician, Emergency Department, University of Maryland Medical Center, Baltimore, Maryland Spinal Puncture and Cerebrospinal Fluid Examination

Charles J. Fasano, DO

Assistant Professor, Department of Emergency Medicine, Jefferson Medical College; Attending Physician, Department of Emergency Medicine, Albert Einstein Medical Center, Philadelphia, Pennsylvania Physical and Chemical Restraint

Lisa Mackowiak Filippone, MD

Assistant Professor, Emergency Medicine, UMDNJ Robert Wood Johnson Medical School; Attending Physician, Emergency Medicine, Cooper University Hospital, Camden, New Jersey Ultrasound-Guided Procedures

Michael T. Fitch, MD, PhD

Associate Professor of Emergency Medicine, Wake Forest University School of Medicine, WinstonSalem, North Carolina Abdominal Hernia Reduction

Contributors

Paul T. DeFlorio, MD

Brenda A. Foley, MD, FACEP

Clinical Assistant Professor, Department of Emergency Medicine, Thomas Jefferson University, Philadelphia; Attending Physician, Department of Emergency Medicine, Doylestown Hospital, Doylestown, Pennsylvania Injection Therapy of Bursitis and Tendinitis

Robert T. Gerhardt, MD, MPH, FACEP, FAAEM

Associate Professor, Department of Military and Emergency Medicine, Uniformed Services University of the Health Sciences, Bethesda, Maryland; Staff Physician, San Antonio Uniformed Services Health Education Consortium (SAUSHEC) Emergency Medicine Residency Program, Brooke Army Medical Center, Fort Sam Houston, Texas Assessment of Implantable Devices

Diane L. Gorgas, MD

Associate Professor and Residency Director, Department of Emergency Medicine, The Ohio State University; The Ohio State University Medical Center, Columbus, Ohio Vital Signs Measurement; Transfusion Therapy: Blood and Blood Products and Reversal of Warfarin-Induced Coagulopathy

Steven M. Green, MD, FACEP

Professor of Emergency Medicine and Pediatrics, Loma Linda University, Loma Linda, California Systemic Analgesia and Sedation for Procedures

Brett S. Greenfield, DO, FACOEP/CAQ-EMS

Medical Director, MidAtlantic MedEvac, and Attending Physician, Department of Emergency Medicine, AtlantiCare Regional Medical Center, Atlantic City, New Jersey Prehospital Immobilization

Maria Halluska-Handy, MD

Assistant Director, Emergency Medicine Residency Program, Albert Einstein Medical Center; Department of Emergency Medicine, Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania Management of Amputations

Richard J. Harper, MS, MD

Associate Professor, Emergency Medicine, Oregon Health and Science University; Chief, Emergency Medicine Service, Portland Veterans Affairs Medical Center, Portland, Oregon Pericardiocentesis

ix

Contributors

Richard A. Harrigan, MD

Professor of Emergency Medicine, Temple University; Attending Physician, Department of Emergency Medicine, Temple University Hospitals, Philadelphia, Pennsylvania Basic Electrocardiographic Techniques

Randy B. Hebert, MD

Attending Physician, Advocate Illinois Masonic Medical Center, Chicago, Illinois Cricothyrotomy and Transtracheal Jet Ventilation

Jerris R. Hedges, MD, MS

Professor and Dean, John A. Burns School of Medicine, University of Hawaii—Manoa, Honolulu, Hawaii; Professor and Vice-Dean, Emeritus, Department of Emergency Medicine, Oregon Health and Science University School of Medicine, Portland, Oregon Ophthalmologic Procedures

Christopher P. Holstege, MD

Associate Professor, Departments of Emergency Medicine and Pediatrics, and Chief, Division of Medical Toxicology, University of Virginia School of Medicine; Medical Director, Blue Ridge Poison Center, University of Virginia Health System, Charlottesville, Virginia Decontamination of the Poisoned Patient

x

Laura R. Hopson, MD

Assistant Professor, Clinical Track, University of Michigan Health System, Ann Arbor, Michigan Pharmacologic Adjuncts to Intubation

J. Stephen Huff, MD

Associate Professor of Emergency Medicine and Neurology, University of Virginia Health System, Charlottesville, Virginia Special Neurologic Tests and Procedures

Charlene Babcock Irvin, MD, FACEP

Research Director, Emergency Medicine Department, St. John Hospital and Medical Center; Associate Clinical Professor, Wayne State University School of Medicine, Detroit, Michigan Autotransfusion

Tim Janchar, MD

Emergency Physician, Legacy Emmanuel Hospital, Portland, Oregon Arterial Puncture and Cannulation

Lewis J. Kaplan, MD, FACS

Associate Professor of Surgery, Director, Emergency General Surgery, Yale University School of Medicine, New Haven, Connecticut Mechanical Ventilation

Eric D. Katz, MD

Program Director and Vice-Chair for Education, Maricopa Medical Center, Phoenix, Arizona Commonly Used Formulas and Calculations

John J. Kelly, DO, FACEP, FAAEM

Associate Chair, Department of Emergency Medicine, Albert Einstein Medical Center; Associate Professor of Emergency Medicine, Jefferson Medical College, Philadelphia, Pennsylvania Nerve Blocks of the Thorax and Extremities

Kevin P. Kilgore, MD

Assistant Clinical Professor, University of Minnesota School of Medicine, Minneapolis; Senior Staff Physician, Department of Emergency Medicine, Regions Hospital, St. Paul, Minnesota Regional Anesthesia of the Head and Neck

Thomas D. Kirsch, MD, MPH

Associate Professor, and Director of Operations, Department of Emergency Medicine, The Johns Hopkins School of Medicine, Baltimore, Maryland Tube Thoracostomy

Kevin J. Knoop, MD, MS, Capt, MC, USN

Assistant Professor of Military and Emergency Medicine, Department of Military and Emergency Medicine, Uniformed Services University of the Health Sciences, Bethesda, Maryland; Executive Officer, US Naval Hospital, Yokosuka, Japan Ophthalmologic Procedures

Frederick K. Korley, MD

The Robert E. Meyerhoff Assistant Professor, Department of Emergency Medicine, The Johns Hopkins School of Medicine; Attending Physician, Department of Emergency Medicine, The Johns Hopkins Hospital, Baltimore, Maryland Management of Increased Intracranial Pressure and Intracranial Shunts

Baruch Krauss, MD, EdM, FAAP, FACEP

Associate Professor of Pediatrics, Harvard Medical School; Senior Associate Physician in Medicine, Division of Emergency Medicine, Children’s Hospital Boston, Boston, Massachusetts Devices for Assessing Oxygenation and Ventilation; Systemic Analgesia and Sedation for Procedures

Diann M. Krywko, MD

Assistant Professor and Emergency Medicine Residency Curriculum Director, Medical University of South Carolina, Charleston, South Carolina; Graduate Lecturer, University of Michigan School of Health Professions, Flint, Michigan Indwelling Vascular Devices: Emergency Access and Management

Richard L. Lammers, MD

Professor of Emergency Medicine, Michigan State University College of Human Medicine, East Lansing; Research Director, Kalamazoo Center for Medical Studies, Kalamazoo, Michigan Principles of Wound Management; Methods of Wound Closure

Assistant Professor, Emergency Medicine, Dartmouth Medical School, Hanover; Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire Venous Cutdown

John A. Marx, MD

Adjunct Professor of Emergency Medicine, University of North Carolina, Chapel Hill; Chair, Department of Emergency Medicine, Carolinas Medical Center, Charlotte, North Carolina

Contributors

Patricia L. Lanter, MD

Peritoneal Procedures

David C. Lee, MD

Clinical Associate Professor, Emergency Medicine, New York School of Medicine, New York; Director of Research, Emergency Medicine, North Shore University Hospital, Manhasset, New York Bedside Laboratory and Microbiologic Procedures

Matthew R. Levine, MD

Assistant Professor, Department of Emergency Medicine, Northwestern University Feinberg School of Medicine; Attending Physician, Northwestern Memorial Hospital, Chicago, Illinois Foreign Body Removal

Shan W. Liu, MD, MPH

Instructor, Harvard Medical School; Faculty, Massachusetts General Hospital, Boston, Massachusetts Peripheral Intravenous Access

Marie M. Lozon, MD

Associate Professor of Emergency Medicine and Pediatrics, University of Michigan Medical School; Director of Children’s Emergency Services, University of Michigan Health System, Ann Arbor, Michigan Pediatric Vascular Access and Blood Sampling Techniques

Matthew L. Lyon, MD

Associate Professor of Emergency Medicine, and Director of Emergency Ultrasound, Department of Emergency Medicine, Medical College of Georgia, Augusta, Georgia

Central Venous Catheterization and Central Venous Pressure Monitoring

Sharon E. Mace, MD, FACEP, FAAP

Professor of Medicine, Department of Medicine, Cleveland Clinic Lerner College of Medicine of Case Western University; Director, Observation Unit, and Director, Pediatric Education/Quality Improvement, Department of Emergency Medicine, Cleveland Clinic; Faculty, Department of Emergency Medicine, Metrohealth Medical Center/Cleveland Clinic Emergency Medicine Residency Program, Cleveland, Ohio Cricothyrotomy and Transtracheal Jet Ventilation

David E. Manthey, MD

Director, Undergraduate Medical Education, Wake Forest University School of Medicine, WinstonSalem, North Carolina Abdominal Hernia Reduction

Phillip E. Mason, MD, Maj, USAF, MC

Faculty Physician, Department of Emergency Medicine, Wilford Hall Medical Center, San Antonio, Texas

Devices for Assessing Oxygenation and Ventilation; Basic Airway Management and Decision-Making

Anthony S. Mazzeo, MD, FACEP, FAAEM

Clinical Assistant Professor, Emergency Medicine, Drexel University College of Medicine, Philadelphia; Attending Physician and Associate Director, Emergency Medicine, Mercy Fitzgerald Hospital, Darby, Pennsylvania Burn Care Procedures

Douglas L. McGee, DO

Assistant Dean, Jefferson Medical College; Chief Academic Officer, and Emergency Medicine Residency Program Director, Department of Emergency Medicine, Albert Einstein Medical Center, Philadelphia, Pennsylvania Local and Topical Anesthesia; Podiatric Procedures

John W. McGill, MD

Assistant Professor, University of Minnesota School of Medicine; Senior Associate Faculty, Hennepin County Medical Center, Minneapolis, Minnesota Tracheal Intubation

Robert M. McNamara, MD

Professor and Chair, Department of Emergency Medicine, Temple University School of Medicine, Philadelphia, Pennsylvania Management of Common Dislocations

Pauline E. Meekins, MD

Clinical Instructor, Division of Emergency Medicine, Medical University of South Carolina, Charleston, South Carolina Decontamination of the Poisoned Patient

Dave Milzman, MD, FACEP

Professor of Emergency Medicine and Adjunct Professor of Physiology, Georgetown University School of Medicine; Research Director, and Senior Faculty, Georgetown University/Washington Hospital Center Emergency Medicine Residency, Washington Hospital Center, Washington, DC Arterial Puncture and Cannulation

xi

Contributors

Bohdan M. Minczak, MS, MD, PhD

Assistant Professor, and Interim Vice Chairman, Department of Emergency Medicine, Methodist Hospital Division, Thomas Jefferson University Hospital; Associate Professor of Physiology, Philadelphia College of Osteopathic Medicine, Philadelphia, Pennsylvania

Techniques for Supraventricular Tachycardias; Defibrillation and Cardioversion

Daniel S. Morrison, MD, RDMS, FACEP

Assistant Residency Director, Wayne State University/ Sinai-Grace Hospital Emergency Medicine Residency Program, and Assistant Professor of Emergency Medicine, Wayne State University; Director of Emergency Medicine Ultrasound, Detroit Medical Center, Detroit, Michigan Arthrocentesis

David W. Munter, MD, MBA

Assistant Clinical Professor, Emergency Medicine, Eastern Virginia Medical School, Norfolk; Director, Emergency Medicine Department, Sentara Obici Hospital, Suffolk, Virginia Esophageal Foreign Bodies

Kathleen A. Neacy, MD xii

Associate Professor, Loyola University Medical Center, Maywood, Illinois Tracheostomy Care

Edward A. Panacek, MD, MPH

Professor of Emergency Medicine, University of California, Davis; Director, Office of Clinical Trials, University of California, Davis, Medical Center, Sacramento, California Balloon Tamponade of Gastroesophageal Varices; Arthrocentesis

Steven J. Parrillo, DO, FACOEP, FACEP

Associate Professor, Philadelphia College of Osteopathic Medicine, and Jefferson Medical College; Attending Physician, Department of Emergency Medicine, Faculty, Emergency Medicine Residency, and Medical Director, Emergency Department, Einstein Elkins Park Hospital, Albert Einstein Medical Center, Philadelphia, Pennsylvania Arthrocentesis

Margarita E. Pena, MD, FACEP

Associate Residency Director, Emergency Medicine Department, and Medical Director, Clinical Decision Unit, St. John Hospital and Medical Center; Associate Clinical Professor, Wayne State University School of Medicine, Detroit, Michigan Autotransfusion

James A. Pfaff, MD

Assistant Professor, Department of Military and Emergency Medicine, Uniformed Services University of the Health Sciences, Bethesda, Maryland; Staff Physician, San Antonio Uniformed Services Health Education Consortium (SAUSHEC) Emergency Medicine Residency Program, Brooke Army Medical Center, Fort Sam Houston, Texas Assessment of Implantable Devices

David Lee Pierce, MD

Assistant Professor of Clinical Emergency Medicine, and Clerkship Director, Emergency Medicine, University at Buffalo School of Medicine, Buffalo, New York Culdocentesis

Heather M. Prendergast, MD, FACEP

Associate Professor, Emergency Medicine, and Residency Research Director, University of Illinois at Chicago, Chicago, Illinois Procedures Pertaining to Hypothermia and Hyperthermia

Beatrice D. Probst, MD, FACEP

Associate Professor of Medicine and Surgery, Loyola University Stritch School of Medicine; Assistant Director, Emergency Department, Loyola University Health System, Foster G. McGaw Hospital, Maywood, Illinois Emergency Childbirth

Robert F. Reardon, MD

Assistant Professor, Department of Emergency, University of Minnesota School of Medicine; Faculty Physician, Department of Emergency Medicine, Hennepin County Medical Center, Minneapolis, Minnesota Basic Airway Management and Decision-Making; Tracheal Intubation

Emanuel Rivers, MD, MPH

Vice Chairman and Research Director, Department of Emergency Medicine, Henry Ford Hospital, Detroit, Michigan Resuscitative Thoracotomy

Ralph J. Riviello, MD, MS

Associate Professor, Department of Emergency Medicine, Drexel University College of Medicine; Attending Physician, Hahnemann University Hospital, Philadelphia, Pennsylvania Otolaryngologic Procedures

James R. Roberts, MD, FACEP, FAAEM, FACMT

Professor of Emergency Medicine, Vice Chair, Department of Emergency Medicine, Drexel University College of Medicine; Chairman, Department of Emergency Medicine, Director, Division of Toxicology, Mercy Catholic Medical Center, Philadelphia, Pennsylvania Intravenous Regional Anesthesia

Adjunct Assistant Professor of Emergency Medicine, University of North Carolina, Chapel Hill; Assistant Residency Director, Carolinas Medical Center, Charlotte, North Carolina Peritoneal Procedures

Brent E. Ruoff, MD

Chief, Emergency Medicine Division, Washington University School of Medicine, St. Louis, Missouri Commonly Used Formulas and Calculations

Carolyn Sachs, MD, MPH

Associate Clinical Professor, David Geffen School of Medicine at UCLA; Emergency Medicine Center, UCLA Medical Center, Los Angeles, California Examination of the Sexual Assault Victim

Leonard E. Samuels, MD

Assistant Professor, Emergency Medicine, Drexel University College of Medicine; Hahnemann University Hospital, Philadelphia, Pennsylvania Nasogastric and Feeding Tube Placement

Gregory Schneider, MD

Attending Physician, Department of Emergency Medicine, Holy Cross Hospital, Fort Lauderdale, Florida Physical and Chemical Restraint

Robert E. Schneider, MD

Senior Medical Advisor for Workforce Protection, Office of Health Affairs, United States Department of Homeland Security, Washington, DC Urologic Procedures

Richard B. Schwartz, MD

Chairman, Department of Emergency Medicine, Medical College of Georgia, Augusta, Georgia Pharmacologic Adjuncts to Intubation

Michael A. Silverman, MD

Instructor of Emergency Medicine, The Johns Hopkins School of Medicine; Chairman, Department of Emergency Medicine, Harbor Hospital, Baltimore, Maryland Urologic Procedures

Peter Erik Sokolove, MD

Professor, Vice Chair for Education, and Program Director, Department of Emergency Medicine, University of California, Davis, Health System, Sacramento, California

Extensor and Flexor Tendon Injuries in the Hand, Wrist, and Foot; Standard Precautions and Infectious Exposure Management

Cemal B. Sozener, MD

Clinical Lecturer, Department of Emergency Medicine, University of Michigan Medical School, Ann Arbor, Michigan Indwelling Vascular Devices: Emergency Access and Management

Contributors

Michael S. Runyon, MD

Mark Spektor, DO, MBA

Assistant Professor, SUNY-Downstate; Medical Director, Maimonides Medical Center, Brooklyn, New York Nerve Blocks of the Thorax and Extremities

Sarah A. Stahmer, MD

Associate Professor of Surgery, and Emergency Medicine Program Director, Division of Emergency Medicine, Department of Surgery, Duke University School of Medicine, Durham, North Carolina Ultrasound-Guided Procedures

Daniel B. Stone, MD, MBA, FACEP

Assistant Professor, Emergency Medicine, Northwestern University, Chicago, Illinois; Director, Emergency Medicine, Coral Springs Medical Center, Coral Springs, Florida Foreign Body Removal

Amita Sudhir, MD

Assistant Professor of Emergency Medicine, Department of Emergency Medicine, University of Virginia, Charlottesville, Virginia Educational Aspects of Emergency Department Procedures

Jacob W. Ufberg, MD

Associate Professor and Residency Director, Department of Emergency Medicine, Temple University School of Medicine, Philadelphia, Pennsylvania Management of Common Dislocations

Malinda Wheeler, MN, FNP

Director, Forensic Nurse Specialists, Inc., Los Alamitos, California Examination of the Sexual Assault Victim

Justin Williams, MD

Brooke Army Medical Center, Fort Sam Houston, Texas Venous Cutdown

Richard Zane, MD

Assistant Professor, Harvard Medical School; Vice Chairman, Brigham and Women’s Hospital, Boston, Massachusetts Peripheral Intravenous Access

xiii

How This Textbook Should be Viewed

xiv

HOW THIS MEDICAL TEXTBOOK SHOULD BE VIEWED BY THE PRACTICING CLINICIAN AND THE JUDICIAL SYSTEM The editors and authors of this textbook strongly believe that the complex practice of medicine, the vagaries of human diseases, the unpredictability of pathologic conditions, and the functions, dysfunctions, and responses of the human body cannot be defined, explained, or rigidly categorized by any written document. Therefore, it is neither the purpose nor intent of our textbook to serve as an authoritative source on any medical condition, treatment plan, or clinical intervention, nor should our textbook be used to rigorously define a standard of care that should be practiced by all clinicians. Our written word provides the physician with a literature-referenced database, and a reasonable clinical guide which is combined with practical suggestions from individual experienced practitioners. We offer a general reference source and clinical roadmap on a variety of conditions and proce-

dures that may confront clinicians who are experienced in emergency medicine practice. This text cannot replace physician judgment, cannot describe every possible aberration, nuance, clinical scenario or presentation, and cannot define rigid standards for clinical actions or procedures. Every medical encounter must be individualized and every patient must be approached on a case-by-case basis. No complex medical interaction can possibly be reduced to the written word. The treatments, procedures, and medical conditions described in this textbook do not constitute the total expertise or knowledge base expected to be possessed by all clinicians. Finally, many of the described complications and adverse outcomes associated with implementing or withholding complex medical and surgical interventions may occur, even when every aspect of the intervention has been standard or performed correctly. The editors and authors of Clinical Procedures in Emergency Medicine Fifth Edition

P R E F A C E

The Fifth edition of Clinical Procedures in Emergency Medicine continues the original concept of providing a complete, detailed, and up to date description of many common, and some uncommon, procedures encountered during emergency medical practice. The novice may find the discussions and figures devoted to many procedures somewhat daunting or overwhelming at first; but, hopefully most will eventually appreciate the details and verbiage contained in the text. The goal is to describe an intervention as though it were the nascent clinician’s first exposure to the concept, but with a depth that the seasoned operator would also deem helpful. It was difficult to find figures or photographs that convey the details, or elucidate the vagaries, to the extent one might want. The newly added color and additional figures were a much needed update, and morphed into an obvious improvement over previous editions. Many of the photographs were taken by me over 35 years of ED shifts, some were borrowed from other sources, such as the wonderful text by Cathy Custalow. Yet others were originally illustrated by Todd Thomsen with the help of Gary Setnik. No doubt Dr. Thomsen has found his calling, blending amazing art with equally impressive medical expertise. This edition is now available and fully searchable online at expertconsult.com, and MD Consult, with additional figures that will further educate the clinician to the nuances of emergency medicine procedures. There is, of course, more than one way to approach any patient, or any procedure, so this text is not a dictum. This book does not attempt to define standard of care. It is a compendium of self proclaimed tried and true, but occasionally prospectively tested, techniques, practical hints, and successful tactics gleaned from years of practice, all blended with a modicum of newer modalities (such as ultrasound). As with prior editions, this version also significantly incorporates the personal opinions of the authors and editors. This book is intended to help the clinician and the patients who rely upon them. It’s simply a clinical guide, not a legal document. Don’t

reference this book if you testify in court, for either the defense or the plaintiff. Today’s dogma too often becomes tomorrow’s heresy, and physician hubris is worse than incompetence. Simply stated, Emergency Medicine, and the human body, too often readily defy the written word, personal opinion, or local custom, and humble even the venerable and universally praised gray haired professor. Many new authors have been added, as well as a number of new concepts and approaches. My personal thanks are hereby conveyed to those who contributed to previous editions. The updated chapters often merely refine or further manipulate the scholarly work of others who originally assisted us. The current contributors include an enviable blend of friends and colleagues, former students of mine, up and coming stars in their own right, and my prior mentors and role models—all are accomplished physicians, and leaders in their own milieu. Most know more than I know, and most are likely infinitely more capable and facile with procedures. All are capable of writing a text themselves, but are now enlightened and eschew that primal urge since they now know how difficult it is to write even a single chapter. My able and erudite associate editors, Arjun Chanmugam, Cathy Custalow, Carl Chudnofsky, and John McManus provided the bulk of the original editing; but, in the end, my personal bias likely prevailed. If any of our editing changed, altered, or misinterpreted the original thoughts of the contributors, we apologize; but, hard decisions had to be made, and waffling was rarely an option. We attempted to squarely address such omnipresent vague topics as prophylactic antibiotics, and accepted the fact that not all foreign bodies or tendon lacerations will be identified in the heat of the moment. The prescient and sagacious clinician knows that the ability to practice medicine from a book is limited, and one learns best from past experiences; and, for certain, the most instructive past experience is one that was not textbook perfect.

James R. Roberts

xv

F O R E W O R D

The emergency physician has the unique responsibility of offering his or her skills at all times to all people (young and old, friendly and hostile, rich and poor). No other health providers are always collectively there at the entrance to the hospital. As emergency physicians, our responsibilities have grown and our horizons have been expanded because of our commitment to people. We have built a system that creates a caring environment from the home to the street and to the hospital, and a system that also integrates firefighters, police officers, paramedics, nurses, clerks, students, pharmacists, and physicians into this caring service. Each new clinical problem and each creative intervention has led to innovations in thought and technical advances. The Fifth Edition of Roberts and Hedges’ text, Clinical Procedures in Emergency Medicine, takes another step in the pursuit of excellence in the provision of that care. One of the newest chapters—the use of chemical and physical restraints demonstrates the skill of the authors in integrating the physical (the device) and the chemical (the pharmaceutical). The editors’ and the authors’ effective integration of the discussion of all the aspects of our care has been the hallmark of this text since the first edition. Each task is described and each goal is defined allowing for a rigorous approach as to how, why, when and with what device and pharmacologic agent each procedure is most safely and effectively performed. The past 40 years in the history of emergency medicine have seen a remarkably rapid evolution in care. Organized medicine has often been criticized for its inability to change thought patterns and approaches to care, but the ability to change current patterns is the recognized strength of emergency physicians. We have undertaken our responsibilities, created new relationships, and developed new perspectives on clinical medicine in an area where previously no one dared to serve. This text exemplifies and describes the tremendous progress in thought, new techniques, new technology, new pharmaceuticals and their integration for the improved and safer care of the emergency patient. The rapid growth of prehospital care, the ever-increasing roles of emergency care, and the diversity of clinical issues and research dilemmas in emergency medicine have led to the development of a new type of physician in the emergency department. This text defines the breadth of investigative and clinical emergency medicine and the enormous technical skill and intellectual responsibility required of each emergency physician. These chapters are written by emergency physicians and other physicians working closely with emergency patients

who have highly specialized knowledge in particular aspects of emergency medicine. Almost a third of these authors are new contributors to this edition. The further integration of the clinical, investigative and educational roles of the emergency physician has led to the refinement of this Fifth Edition. As the basic science and clinical practice of emergency medicine have further developed, this book has grown to represent a complete view of our specialty. This text offers a balanced analysis of the entirety of the armamentarium at our disposal in the emergency department for the care of those with urgent and emergent problems. The authors attempt to simplify and clarify while focusing on knowledge and process in the environment where we practice. This text permits any practitioner the opportunity to perform his or her first emergency procedures with a foundation that emphasizes evidence and limits bias and ignorance. This text has filled a void in medical practice. Procedural interventions in the emergency department had previously been largely undefined and certainly inadequately analyzed. The emergency physician who is trained to address the airway, an obstetrical or cardiac emergency for example will be able to utilize this text to review, better understand and develop the requisite cognitive and technical skills. Knowledge of these skills and their indications, as well as the risks and benefits of practice, will permit emergency physicians to achieve the highest level of service and will foster their potential to initiate quality research. This book is also about motivating physicians to appreciate the clinical norms and expectations in our field. The editors have recognized for years many of the problems defined in the report To Err is Human released by the Institute of Medicine of the National Academy of Sciences in 1999. This text has moved the physician from anecdote to a rigorous analysis. The reader will not only feel more secure while performing an essential procedure with specific technical and pharmaceutical adjuncts but he or she will also become more confident about making the decision not to perform a procedure that entails more risk than benefit to an individual patient. The editors and authors have attempted to enhance education and limit the errors of commission as well as omission while improving the safety and occupational health of the emergency physician. Recognizing that the emergency department environment is by definition unpredictable and often chaotic, these authors have prepared us to change the human response in an attempt to make errors more difficult to commit. Understanding the remarkable spectrum of responsibility of the emer-

xvii

Foreword

xviii

gency physician is our essential task. We shall succeed as health providers if we understand our patients and their needs, the pathophysiology of emergency medicine and its therapeutics, and our procedures and their pitfalls. The Fifth Edition of Roberts and Hedges’ Clinical Procedures in Emergency Medicine provides enough thought-provoking information about the task, the appropriate technology and the appropriate pharmaceutical at the appropriate dose at the right time to prepare the emergency physician to care for the emergency department patient in a humane and intellectually sound manner. Although few physicians other than emergency physicians will use all the techniques, technology and pharmaceuticals detailed in this text, many other physicians can and will profit immensely from the use of this text. The techniques are well defined, well illustrated, and well referenced by clinicians who obviously use them daily. In this edition, there are many more helpful photographs and graphics with an esthetically pleasing and visually useful commitment to colors. This text remains unique with respect to the depth and breadth with which the editors and authors critically evaluate the tools of

our trade. The two leaders of our field, Roberts and Hedges, are once again assisted by four associate editors in this edition. These respected emergency physicians expand the excellent foundation of editorial contributions and ensure continued successful presentation of this complex material to help guide our clinical care. The understanding and application of the principles defined in this edition should be considered essential for each emergency physician in his or her attempt to continuously improve the delivery of the best possible health care to our patients.

Lewis R. Goldfrank, MD

Director, Emergency Medicine, Bellevue Hospital Center New York University Langone Medical Center Professor and Chair, Emergency Medicine New York University School of Medicine Medical Director, New York City Poison Center New York, New York

S E C T I O N  

I

Vital Signs And Patient Monitoring Techniques

C H A P T E R

1

Vital Signs Measurement Diane L. Gorgas and J. Dave Barry

Measuring the temperature, pulse, respiratory rate (RR), blood pressure, and pulse oximetry is generally recommended for all emergency department (ED) patients in addition to an assessment of pain in the appropriate patient population. Vital signs may not only indicate the severity of illness but also dictate the urgency of intervention. Although a single set of values suggests pathology, triage or initial vital signs may be spurious and simply related to stress, anxiety, and fear. Therefore, the greatest utility of vital signs is their observation over time. Deteriorating vital signs are an important indicator of a similarly deteriorating physiologic condition, whereas improving values provide reassurance that the patient is responding to therapy. Hence, when a patient undergoes treatment over an extended period of time, remember to repeat the vital signs as appropriate, particularly those that were previously abnormal. In some circumstances, it is advisable to monitor certain vital signs continuously. Vital signs should be measured and recorded at intervals dictated by clinical judgment and the scenario and patient’s clinical state or with any significant change in these parameters. Adhering to protocols or disease categories may not be useful or productive. An abnormal vital sign may constitute the patient’s entire complaint, as in the febrile infant, or may be the only indication of the potential for serious illness, as in the patient with resting tachycardia. Emergency medical service (EMS) begins the assessment of the patient’s status and vital signs in the prehospital setting. Surges of epinephrine and norepinephrine commonly occur during transport by EMS and these are known to alter the vital signs, leading to increases in heart rates of greater than 10%.1 Prehospital vital signs should always be interpreted with the entire clinical scenario in perspective. Blood pressure and pulse are frequently evaluated together, as a measure of the blood volume. Although body

temperature is usually the last vital sign measured during resuscitation, it has special importance for patients suffering from thermal regulatory failure. With these considerations in mind, the current chapter is organized according to the priorities of patient resuscitation and evaluation. Additional “vital signs” recently introduced into the practice of emergency medicine are pulse oximetry, capillary refill, and the visual analog or pain scale. Capillary refill is discussed in the blood pressure section of this chapter as an assessment of overall perfusion, circulatory volume, and blood pressure. Assessment of pain as a vital sign is gaining acceptance and is discussed briefly at the end of this chapter. Mental status can be viewed as a summation of measurable vital signs (blood pressure, heart rate, RR, and temperature) with the understanding that significant vital sign abnormalities can cause mental status changes. BACKGROUND can be found on

E x p e rt C o n s u lt

NORMAL VALUES The range of normal, resting vital signs for specific age groups must be recognized by the clinician to enable identification of abnormal values and their clinical significance. The normal ranges for vital signs are also influenced by gender, race, pregnancy, and residence in an industrialized nation. These ranges have not been validated in ED patients, who have many reasons for vital sign abnormalities, including anxiety, pain, and other forms of distress, in addition to altered physiology from their disease states. Importantly, ranges of normal vital signs commonly quoted as normal or abnormal in other settings serve only as a guide, and not an absolute criterion, for diagnosis, treatment, further observation, or intervention in the ED. Published vital sign norms for children are not as well accepted as for adult patients. Table 1–1 and Table 1–2 report normal vital signs for children by age group as means and standard deviations. In Table 1–1, the values for pulse and blood pressure for 0- to 2-month-olds are adapted from studies of newborn populations (i.e., 60 mm Hg Titrate to Po2 > 60 mm Hg and FIO2 ≤ 0.6 Target V T spont = V T set Use square wave in COPD only Not efficacious in rewarming or cooling

Note: 1. Use inspiratory time-cycled PCV to precisely control I/E ratio. IRV is used to manage hypoxia but may lead to hemodynamic instability owing to decreased venous return and decreased cardiac output. 2. Best to titrate ventilator settings to the shape of the pressure-volume curve (need ventilator with a graphics package, a.k.a. “open lung model”). Special Circumstances 1. Severe acute lung injury—Consider permissive hypercapnia. If able to achieve Po2 > 60 mm Hg on FIO2 ≤ 0.6, the Pco2 may be allowed to be greater than 40 mm Hg if pH > 7.25. Further attempts to raise VE to decrease Pco2 may induce additional lung injury. 2. Asthma—Defect is decreased gas flow. In conventional ventilation, use higher flow rate and lower respiratory rate to allow more time for exhalation. 3. Traumatic brain injury—Do not lower Pco2 < 35 mm Hg, because it may induce severe cerebral vasoconstriction and lead to cerebral ischemia. The goal is normal Pco2: 35–40 mm Hg. Acceptable to hyperventilate for a patient with an acute herniation syndrome as a bridging maneuver for definitive therapy. 4. PEEP—Used to raise alveolar recruitment and increase Po2 in patients with hypoxemic respiratory failure. Caution: Excessive PEEP can lead to hypotension from diminished venous return. Initial treatment of this hemodynamic instability is with volume replacement and lowered PEEP if possible. AC, assist control; ARDS, acute respiratory distress syndrome; COPD, chronic olstructive pulmonary disease; FIO2 , percent of inspired oxygen; I/E, inspiratory-to-expiratory ratio; IMV, intermittent mechanical ventilation; IRV, inverse ratio ventilation; Paw, airway pressure; Paw-peak, peak airway pressure; PC, pressure control (setting); Pco2, pressure of carbon dioxide; PEEP, positive end-expiratory pressure (5 cm H2O is considered physiologic); Po2, pressure of oxygen; PSV, pressure-support  flow rate; SIMV, synchronized intermittent mechanical ventilation; T , time in inspiration; V , minute ventilation; V , tidal volume. ventilation; Q, i E T

Decelerating (Ramp).  Once the maximal inspiratory flow is reached, the rate of gas delivery immediately begins to slow in a preprogrammed fashion. Therefore, relative to the square waveform, longer time is spent in inhalation to deliver  T or achieve the target pressure, which allows for the set V improved oxygenation. This waveform also achieves a lower Paw-peak and a higher Paw-mean. Sine Wave.  This is not useful in critically ill patients. Accelerating.  This is the flow pattern for neonatal gas intake and is generally not used in adult ventilation unless one

must use the Siemens Servo 900 ventilator, which has only two options: square or accelerating. Sighs.  This is a large single breath or large multiple breaths, both designed to help maintain alveolar recruitment by ventilating a patient on a periodic basis at close to vital capacity. Controversy exists regarding the utility of the sigh option with regard to alveolar overdistention.17 Pause.  Pause is a variable used on the Siemens Servo 900 ventilator to alter the I/E ratio. This is technically complex for anyone not working with ventilator setup on a daily basis.

L/min.

60

Shorter

Longer

r [bpm]

Paw

High CPAP

1 2 3

Release

23

40

Longer

20

Shorter

0

Ramp Time Equal VT but different Ti and Te

Figure 8–5  Effect of flow rate on inspiratory and expiratory times.  Note that as the flow rate changes, there are corresponding alterations in the effective times for inspiration and exhalation. Deflections above the x-axis (time) indicate inspiration, and those below indicate exhalation. The delivered tidal volume for each cycle is the same, but the inspiratory and expiratory times are different.

It is much easier to directly adjust the Ti/Te on more modern ventilators. A pause is useful to determine the plateau pressure on the Puritan Bennett 7200 or the Infrasonics Adult Star ventilator (as a point measurement). If a short pause is used to measure plateau pressure, the authors suggest no longer than 0.5 second as the pause duration; remove the pause when the measurement is completed.

NEW MODES Airway Pressure Release Ventilation Airway pressure release ventilation (APRV) is essentially a high level CPAP mode that is terminated for a very brief period. The CPAP level may be as high as 40 or more cm H2O pressure. The long time during which the high-level CPAP is maintained achieves oxygenation whereas the short release period achieves CO2 clearance (Fig. 8–6). The long time during which the high-level CPAP is present results in substantial recruitment of alveoli of markedly different regional time constants at rather low gas flow rates and lower Paws (by comparison with conventional ventilation strategies). The establishment of intrinsic PEEP by the short release time enhances oxygenation. CO2 clearance is aided by recruitment of the patient’s lung at close to total lung capacity (TLC); elastic recoil creates large-volume gas flow during the release period. This is a fundamentally different mode from cyclic ventilation. This mode allows the patient to spontaneously breathe during all phases of the cycle. This mode is enabled to succeed by having a floating valve that is responsive to the patient’s needs regardless of the location within the respiratory cycle. In other words, the patient is allowed to breathe in or out during the high-level CPAP phase as well as during the release phase. Accordingly, the sequence is called a phase cycle; there is no set inspiratory or expiratory time, and no readily identifiable respiratory rate in the traditional sense. During the high-CPAP phase, a patient may exhale 50 to 200 or more mL of gas as his or her lung volume becomes full of gas; this is not a full exhalation, and the release of excess gas should not be counted as a breath. APRV has been successfully used in neonatal, pediatric, and adult forms of respiratory failure. It is considered an alternative open lung model approach to mechanical ventilation.18

11.0

MV [L/min]

0 –10 2

4

L/min

6

8

10

7.4

s

Flow

50

Ppeak [m bar]

100 Phase cycle

50

4.9

Mechanical ventilation

Increased flow rate

A

m bar

● 

80 60 40 20 0 –20 –40 –60 –80

APRV

8 

FLOW RATE: IMPACT ON Ti AND Te

30

0

VTe [L]

–50 –100 0

2

4

6

8

10

s

.120

Figure 8–6  Airway pressure release ventilation (APRV)—airway pressure-time and flow-time traces.  Note that the peak airway pressure (Paw-high) is maintained for a long period. This phase establishes oxygenation (Thigh). There is a short period of release when most CO2 is cleared (Tlow). The bottom trace indicates flow over time. The combined time for the Thigh and Tlow is known as a phase cycle. Note that the number of phase cycles is not the respiratory rate because patients breathe within the entirety of the Thigh. As the release phase is initiated, the flow rate is identified as negative and is of a high rate (here ~7.5 L/min), consistent with significant alveolar recruitment. During the high continuous positive airway pressure (CPAP) phase, the patient is allowed to exhale (negative deflections on the flow-time trace). Thus, APRV is quite dissimilar from traditional cyclic ventilation. This unique mode is made possible by a floating valve system.

Given the spontaneous nature of the mode, there should be virtually no need for continuous infusions of neuromuscular blocking agents in patients placed on this mode of venti­ lation;19 exceptions to this observation do occur for the management of intracranial pressure (ICP) but not for oxygenation or clearance of CO2. This may result in a shorter length of ICU stay and a reduced incidence of prolonged neuromuscular blockade syndrome. Furthermore, because patients may be ventilated at lower Paws than using cyclic ventilation, there is a reduced need for pressor support of  O 2.19 Moreover, there is a reduced hemodynamics to ensure D sedative need because patients are more comfortable on this spontaneous mode than on cyclic ventilation.19 Hemodynamic assessment using a pulmonary artery catheter in patients on APRV has been investigated. The pulmonary artery occlusion pressure (PAOP) must be read at the middle or end of the release phase to maintain the fidelity of the reading. Reading the PAOP at any other point in the cycle will give a significantly different value by comparison with the end-expiratory reading obtained using PCV.20 Transport of patients on APRV with a Paw-high (sustained peak Paw) greater than 20 cm H2O pressure should be with the patient attached to the ventilator instead of being hand ventilated.21 Hand ventilation is unable to match the manner of gas delivery and pressure dynamics that the patient requires. Attempts at hand ventilation, even with an appropriately set PEEP valve, are frequently complicated by unexpected hypoxemia and hemodynamic instability.

143

RESPIRATORY PROCEDURES ● 

II 

Proportional Assist Ventilation Ventilators that are capable of performing in the proportional assist ventilation (PAV) mode will be able to assess on a breath-by-breath basis how much work of breathing support the patient needs to achieve the targets and goals that the clinician sets.22 The unique features of this type of ventilation promise to reduce inadvertent airway injury and, in many ways, serve as a self-weaning ventilator mode. As the patient requires less support, the ventilator delivers less support. Current data are lacking to determine whether this will realize a shortened length of ventilator support for those with acute respiratory failure. A small study showed no difference in cardiopulmonary function using PAV with automatic tube compensation compared with PSV.23 However, another study showed improved quality of sleep using PAV versus PSV.24

Permissive Hypercapnia

144

As stated, excessive Paw-peak may be quite detrimental. One means of limiting Paw-peak, and thereby offering protection from the trauma of ventilation, is to decrease the delivered  T until an acceptable and less deleterious Paw-peak is achieved V (≤35 cm H2O). However, changes in Paw-peak may alter the pH-Pco2 balance. If the pH is 7.25, and the patient can tolerate the elevated Pco2 while still remaining well oxygenated,  E is not increased. Alternatively, the f may be then the V decreased in similar fashion, but usually not less than 8 breaths/min. This paradigm is known as permissive hypercapnia, and the concept represents a major departure from previously accepted tenets of MV, which mandated that MV should always achieve a normal Pco2.25 Clearly, many patients can safely tolerate Pco2 elevations that have in the past been thought to be harmful. A slower respiratory rate reduces the shear forces active across the alveolar common walls by allowing for fewer openings and closings per minute.26 The side effect of such a rate reduction is greater time for exhalation that may lead to alveolar collapse and increased shear stress at the junction of open and closed alveoli. Nonetheless, a greater Te helps prevent auto-PEEP, alveolar overdistention, T and hemodynamic embarrassment. Likewise, a reduced V also helps prevent alveolar overdistention. Although these features appear very attractive, permissive hypercapnia is not entirely benign. An elevated Pco2 triggers cerebral vascular vasodilatation, which leads to increased cerebral blood flow and possible elevated ICP.27 Increased ICP greater than 20 mm Hg can be detrimental in patients suffering from head injury or cerebral ischemia.28 In patients with ALI complicating traumatic brain injury or stroke, such a management strategy is optimally accompanied by a measure of cerebral perfusion to evaluate for hyperemia or ICP monitoring to assess for intracranial hypertension from increased CO2 tension.29 Hypercapnia also shifts the oxyhemoglobin dissociation curve to the right, leading to increased early unloading of O2 at the tissue level. Hypercapnia also creates an acidosis that may initiate myocardial depression, dysfunction of pH-dependent enzyme kinetics, and distorted cellular metabolism.30 Severe acidosis, pH less than 7.2, may be effectively countered by using an NaHCO3 infusion. This proper intravenous infusion can be created by mixing 1 L of D5W and 150 mEq of NaHCO3, creating a sodium content similar to that of lactated Ringer’s solution (130.4 mEq/L). Alternatively, if

150 mL of D5W is removed before adding the 150 mL of NaHCO3, the resultant Na+ concentration is 150 mEq/L and approximates 0.9% normal saline solution. A common additional repair of the increased Pco2 if the pH is less than 7.2  E while is to increase the respiratory rate to increase the V maintaining the “lung protective low tidal volume.” This strategy is quite similar to the ARDSNet protocol commonly used for volume ventilation of patients with ALI or acute respiratory distress syndrome (ARDS; see later).31

Prone Positioning Multiple authors have proposed gravity as an effective aid in lung recruitment for those patients with severe or refractory hypoxemia.32,33 Several studies have identified benefits in terms of increased Po2 after pronation.32,33 Prone positioning has also been shown to improve recruitment of edematous lung and reverse overinflation as compared with recruitment maneuvers, thus making aeration of the lung more homogeneous.34 Several areas remain unresolved. It is unclear who will maximally benefit (up to one third have no benefit), how many proning cycles per day and for what duration are most beneficial, and how long to continue pronation once it has been initiated. What is clear is that chest geometry is critical in successful pronation32 (Fig. 8–7). Patients with ovoid chests have little benefit from pronation because there are relatively equal lung volumes that exchange the superior and inferior positions when the patient is placed prone. Patients whose thoracic cage is more triangular have a greater volume of lung posteriorly (while in the supine position) that is able to PRONE POSITIONING Triangular thoracic cage Ovoid thoracic cage Sup

Superior

Inferior

Inferior

More lung recruitment

No ∆ lung recruitment

Inferior

Inferior

Sup

Superior

PRONE POSITIONING AND CARDIAC SIZE

Pre-Injury Normal cardiac size Significant lung to recruit Pronation benefit

Post-Injury Increased cardiac size Limited lung recruitment Pronation less helpful

Figure 8–7  Chest geometry and prone position.  Note that ovoid chest geometry results in equal volumes above and below the transverse axis, leading to no change in recruited lung volume with pronation. Triangular chest geometry leads to a significant increase in lung volume with prone positioning. This is by virtue of the large volume of lung that lies posterior to the transverse plane with the patient in the supine position. Increases in cardiac volume reduce the available retrocardiac lung volume and impair the effectiveness of prone positioning with regard to pulmonary recruitment.

THE “BIRD’S BEAK” PROFILE

Tidal volume

0

5

10

15

20

25

30

35

Mechanical ventilation

40

Airway pressure Figure 8–8  Alveolar overdistention is reflected in the increase in airway pressure without any concomitant increase in tidal volume. This PV curve pattern approximates a “bird’s beak” profile.

PAW AND Ti: IMPACT ON AUC Increased Paw and Ti Increased AUC

40 30 cm H2O

a minute. It is calculated by multiplying the patient’s respira T ) for patients without tory rate by the tidal volume (f x V spontaneous breaths. It is conveniently determined by the ventilator and can be read directly for those with and without  E . The need for V  E varies a spontaneous component to their V with the patient’s condition, body mass, comorbidities, and acid-base status. For example, a patient who has a metabolic E acidosis from diabetic ketoacidosis needs an elevated V to decrease Pco2 and acutely buffer the acidosis. Thus, the  E range the emergency provider should determine what V  T , or PCV/Ti to compensate patient will need and set the f, V  E is 7 to for the increased metabolic acid load. A normal V 10 L/min. Spontaneous VE .  Spontaneous V E is the V E derived from spontaneous breathing. During weaning, progressive increases in this parameter are expected as mandatory breaths are decreased or eliminated. Spontaneous VT .  See “Synchronized Intermittent MV” and “Pressure-Support Ventilation,” earlier. Paw-peak.  Paw-peak is the maximum amount of reflected pressure in the patient’s airway. This peak occurs during inspiration—an important concern because of the welldocumented relationship of elevated Paw (and volume) causing biotrauma.4 Excessive Paw-peak (>35 cm H2O) commonly leads to alveolar overdistention and injury, causing release of inflammatory mediators and complications including pneuomothoraces, pneumatoceles, pulmonary interstitial emphysema, pneumomediastinum, ALI, and ARDS. The Paw-peak and alveolar overdistention are best evaluated using the PV curve, looking to abrogate any increases in Paw that are not accompanied by increases in delivered volume (Fig. 8–8).  T lead Increases in Paw without accompanying increases in V to a plateau of the PV curve, known as the “bird’s beak” profile. This profile is a reasonable indicator of alveolar overdistention and airway injury. Plateau Pressure.  Plateau pressure is the pressure reflected from the airways once the full set volume or targeted

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

● 

Overdistension

VENTILATOR ORDER GOALS VE .  V E is the amount of gas delivered to a patient over

8 

exchange with a smaller volume of anterior lung. This patient population benefits from pronation. Similarly, patients with small cardiac volumes benefit more than those with large volumes because there is more lung behind the heart to recruit with the pronation maneuver (see Fig. 8–7). It is quite clear that pronation is a challenging and potentially dangerous maneuver in a patient with invasive lines and an ETT in place, although no increase in unintentional extubations was noted in a study by Gattinoni and coworkers.32 This risk of tube or catheter dislodgment or malposition is compounded by the altered hemodynamics in a patient who is not yet volume-replete, especially after trauma. A recent study of trauma and surgical ICU patients with ALI did show improvement of PaO 2 FIO2 ratios, fewer ventilator days, and shorter hospital lengths of stay using a specialized bed to prone patients.35 Given the potential for complications and the level of expertise and close monitoring required, pronation cannot be recommended as first-line therapy for hypoxemia in the ED. Instead, it is best reserved for the more controlled environment found in the ICU. Moreover, it is the author’s experience that pronation use has been virtually eliminated since using APRV as the rescue mode of choice for refractory hypoxemia.

20

Higher Shorter

Longer

10 0 Time Figure 8–9  Mean airway pressure and the pressure-time trace.  Note that the greater the maximum airway pressure and the longer the Ti, the greater the area under the curve (AUC) described by the positivepressure (inspiratory) limb of the respiratory cycle. The increase in mean airway pressure (AUC) is the principal correlate of oxygenation in volume- or pressure-cycled ventilation.

pressure change has been achieved. It is a reflection of pulmonary compliance, airway resistance, and elastance. It is not  T , Q,  PCV, Ti, directly manipulable, but may be affected by V and PEEP. It does provide a basis for the initiation of other modes of ventilation and is quite useful in that regard (see “Airway Pressure Release Volume,” earlier). Paw-mean.  The area under the pressure-over-time curve (Fig. 8–9) may be calculated and represents the Paw-mean. The Paw-mean correlates most closely with the achieved Po2 in VCV or PSV modes. The longer the Ti, the greater the Paw-mean. When a patient has hypoxemia and the clinician wants to change the ventilator orders, it is important to not reduce the Paw-mean as a result of the change in therapy because a decreased Paw-mean consistently leads to a decrease in Po2.

COMPLICATIONS OF MV Pneumothorax.  Pneumothorax that is unassociated with trauma in a mechanically ventilated patient typically stems from alveolar overdistention (continuous or episodic), leading to alveolar rupture and escape of gas into the pleural space.36 For patients who are on PPV, it is wise to drain the

145

RESPIRATORY PROCEDURES ● 

II 

146

pleural space in order to prevent a simple pneumothorax from progressing to a tension pneumothorax with hemodynamic compromise. Loculated pneumothoraces may be successfully drained percutaneously under ultrasound or computed tomography (CT) guidance. Successful drainage of airspace disease leads to enchanced liberation from MV.37 Pneumothorax or tension pneumothorax may also result from aggressive bagvalve-mask ventilation. Patients with intrinsic lung disease such as COPD or asthma are more prone to developing pneumothorax than the average patient owing to the abnormal structural integrity of their alveolar air spaces.38 A simple pneumothorax can be drained by surgical tube thoracostomy with a small-bore tube (24 Fr), a commercially available pneumothorax kit (Arrow), or a pigtail catheter placed into the pleural space using the Seldinger technique (see Chapter 10). Each of these catheters should be placed to a chest drainage collection unit that incorporates a water seal chamber as well as a variable suction control. Treat persistent air leaks initially with continuous suction (usually 20 cm H2O suction) to evacuate the pleural space and promote coaptation of the visceral and parietal pleurae. Reduce suction and place the chest tube on water seal only after the resolution of the air leak. Remove the chest tube directly from water seal if there is no pneumothorax on chest film or after a test period of tube clamping and subsequent radiographic evaluation. The authors favor a 4-hour period of clamping because a recurrent pneumothorax is easier to treat by unclamping a tube than by placing a new one. Not all patients with a pneumothorax require invasive techniques to evacuate air from the pleural space. It is important to recognize that small pneumothoraces occurring in spontaneously breathing patients (i.e., negative-pressure ventilation) may be reevaluated in 4 to 6 hours with a repeat chest x-ray and drained only if they are expanding. This option is not advised for patients who are on any form of PPV because a simple pneumothorax can rapidly become a tension pneumothorax with subsequent hypotension and death. Tension pneumothoraces may be recognized by tachycardia, hypotension, elevated Paw-peak (if mechanically ventilated, tachypnea if not), jugular venous distention (if not intravascularly depleted), thoracic resonance by percussion on the affected side, diminished or absent breath sounds on the affected side, and tracheal deviation away from the affected side. Clearly, not all signs or symptoms are present in all patients and treatment should be dictated by the patient’s clinical condition. Certain patients develop loculated pneumothoraces or fluid collections. If the collections are either single or immediately adjacent to one another and readily identified, they may be drained using ultrasound guidance at the bedside.39 However, the loculations are frequently in inaccessible areas or are difficult to image with ultrasound. Therefore, CT scanning of the thorax can provide precise anatomic definition of the presence and number of loculated collections as well as a guide for the interventional radiologist. The authors have successfully used CT-guided drainage of loculated pleural collections (air and fluid) to assist weaning of head-injured patients from mechanical ventilator support.37 Biotrauma.  Biotrauma refers to the self-sustaining process of lung injury from MV that follows alveolar overdistention or rupture, alveolar hypoperfusion, and repetitive shear stresses across alveolar walls. Originally, this problem was thought to be from too much pressure (barotrauma).40 Current principles hold that elevated Paws are a straightforward reflection of excess volume delivered to a lung that

cannot accept that much gas (i.e., volutrauma: excess volume is delivered).12 When this process is active in a patient on MV, it is termed ventilator-induced lung injury. Lung injury is an inhomogeneous process with areas of normal lung immediately adjacent to diseased and injured segments.41 Thus, the healthy and compliant segments with shorter regional time constants will readily accept gas, whereas their neighbors with reduced compliance and longer regional time constants will not. The end result is overdistention of the compliant segments, alveolar injury, and the liberation of inflammatory cytokines, chemokines, and activation of endothelin and arachidonic acid pathways, as well as the expression of adhesion molecules along the vascular endothelium.4 This leads to infiltration of inflammatory cells, their destructive lysosomal enzymes, and the induction of toxic oxygen metabolites. Avoiding this inflammatory cascade is an intelligent means of protecting a patient’s lungs from volume-induced lung injury. Such a notion has given rise to lung-protective ventilator  T ventilation (6–7 mL/kg body strategies based on the low V weight).31 Several studies have reported the development of ventilator-induced lung injury in patients with normal lungs  S (12 ml/kg). Lung injury that were ventilated with larger V T can develop within hours and has been linked to ventilation  T .42–44 Current recommendations are for all MV with large V  T than the once-standard 12 to 15 mL/kg. to be with lower V Patients with abnormal lungs (interstitial lung disease, lung resection, severe pneumonia, edema) and/or the presence of an ALI risk factor (sepsis, aspiration, transfusion) should be  T of 6 mL/kg body weight. Those with normal started on V  T less lungs and no ALI risk factors should be started with V 44 than 10 mL/kg body weight. Hemodynamic Compromise.  In all circumstances, the volume of venous return exactly matches the cardiac output volume. Any process that impedes venous return will decrease the available volume that establishes cardiac output. For patients on PPV, each gas delivery increases the intrathoracic pressure while exhalation decreases that pressure. Therefore, venous return principally occurs during exhalation. If the ventilator orders are constructed in such a way as to lead to increased intrathoracic pressure during exhalation, venous return will be reduced. Variables that can lead to this circumstance are increased PEEP, auto-PEEP, and IRV. Recall that venous return not only depends on a relatively negative pressure within the thoracic cavity but also relies on a sufficient amount of time for flow into the thoracic vasculature and right side of the heart. Thus, significantly high respiratory rates may compromise venous return as well. An additional untoward side effect of impaired venous return is cerebral venous hypertension from impeded venous drainage. Because there are no valves between the cerebral parenchyma and the right atrium, increased pressure on the right atrium reduces cerebral venous flow and may contribute to cerebral ischemia in patients with traumatic brain injury or stroke, especially in those with compromised systemic hemodynamics. Such patients are prone to watershed infarction; cerebral venous hypertension may increase this risk. Ventilator-Associated Pneumonia.  The association between the duration of endotracheal intubation and the promotion of pneumonia is quite clear. In fact, the likelihood of developing pneumonia is four times greater for patients in a surgical ICU than those in a medical ICU.45 Endotracheal intubation for more than 12 hours increases the risk threefold.45 A recent study showed that increased ED length of stay in emergently intubated blunt trauma patients was an

Pulmonary hemorrhage Chemical aspiration

ALI/ARDS INFILTRATE Pneumonia Atelectasis Blood transfusion Pulmonary Pleural effusion Extrapulmonary embolism Congestive heart failure inflammation Tumor FEVER

Figure 8–10  Confounders in the diagnosis of pneumonia.  Fever, leukocytosis, radiographic infiltrate, and sputum production do not necessarily indicate the diagnosis of pneumonia. Multiple other causes should be considered as well so that one does not apply antibiotics when there is no infectious agent to address. ALI, acute lung injury; ARDS, acute respiratory distress syndrome.

β2-Agonists.  These agents stimulate β-adrenergic receptors in bronchial smooth muscle, and induce muscle relaxation. This reduces airway resistance and improves gas flow through the conducting airways.53 The β2-agonists also inhibit mast cell degranulation, leading to ameliorated immune stimulation of the reactive airway. The most widely used agent in ICUs in the United States is albuterol. This agent may be administered via a side port of the ventilator circuit using a metered-dose inhaler (MDI; cost-effective). Alternatively, albuterol may be delivered by placing an in-line nebulizer device between the circuit and the ETT or on a side port on the ventilator tubing’s inspiratory limb (ventilator tubing– dependent). Many patients may develop bronchoconstriction and wheezing when mechanically ventilated without a preexisting history of reactive airways disease. β2-Agonists should be administered to patients who have poor air movement, wheezing, or both. A physiologically appropriate means of detecting and following bronchospasm is the peak-plateau gradient. A normal gradient is less than 4 cm H2O pressure; increased values indicate increased airway resistance. The efficacy of treatment with β2-agonists, intravenous magnesium, or diuresis may be assessed by following the changes in this gradient. Acetylcholine Antagonists.  The main utility of acetylcholine agents is to dehydrate secretions, although these agents may also block cholinergic-mediated bronchospasm. The most common agent is ipatropium bromide (Atrovent), an atropine derivative that has twice its biopotency on an equimolar basis. Ipatropium is commonly prescribed in combination with β2-agonists every 4 to 8 hours. As with albuterol, ipatropium may be administered by MDI or nebulizer. Ipatropium is rarely used in isolation for the therapy of bronchospasm. Mucolytics.  The prototype for the mucolytic class is Nacetylcysteine (NAC). NAC is believed to reduce the adhesion of mucous strands to each other as well as to the luminal surface of the alveoli and larger airways. NAC may be administered by nebulizer or lavage routes but no MDI equivalent is available. NAC has the unfortunate side effect of inducing mucosal inflammation in an unpredictable fashion when used for longer than 24 hours. However, for the first 24 hours, NAC may provide significant benefit in liberating densely inspissated secretions from dependent portions of the airways. Recruitment Maneuvers.  Recruitment maneuvers are designed to apply consistent but well-regulated pressure to partly or completely closed alveoli to reintroduce gas into those segments.54 The targeted segment(s) are those with poor compliance and long regional time constants. The area of interest is placed in a nondependent position (e.g., for left lower lobe benefit, place the left side of the patient up and the right side down), and the patient is hand ventilated using a bag-valve device attached to the ETT. An in-line pressure monitor is needed. The patient will usually require sedation to comply with the maneuver—fentanyl and midazolam are ideal. The clinician then applies pressure to the bag to achieve 35 cm H2O pressure and holds it for 4 to 6 seconds, after which the patient is allowed to exhale. This cycle is repeated for up to 5 minutes. A second examiner listens to the area of

Mechanical ventilation

Drug reaction Extrapulmonary infection

ADJUNCTIVE THERAPIES

● 

FEVER PLUS INFILTRATE DOES NOT NECESSARILY EQUAL PNEUMONIA

is carbapenem.52 Carbapenems consistently demonstrate excellent efficacy in eradicating ESBL-producing microbes.

8 

independent risk factor for developing pneumonia. Each hour spent in the ED increased the risk of developing pneumonia by 20%.46 Early pneumonias (72 hr) typically stem from nosocomial pathogens that may be resistant to community antibiotics.47 Such patients need to have empirical coverage for Pseudomonas, methicillin-resistant Staphylococcus aureus (MRSA), and the other SPACE microbes (Serratia, Pseudomonas, Acinetobacter, Citrobacter, and Enterobactericiae). Empirical coverage for fungi is not warranted except in special circumstances (recrudescent pneumonia in a patient already on broad-spectrum antibiotics for >7 days with negative cultures; a solid organ transplant patient after implantation for >4 mo; poly-site–positive fungal cultures or fungemia). Unequivocally, clinical estimation of pneumonia is correct at best 33% of the time.48 The most sensitive and specific test to diagnose pneumonia in a patient with a radiographic infiltrate, fever, leukocytosis, and purulent secretions is bronchoscopy and bronchoalveolar lavage (BAL) with quantitative cultures.49 This strategy provides strong evidence of the exact pathogen(s), eliminates treating nonpathogenic microbes that are upper airway colonizers, and provides confidence in withholding antibiotic for the diagnosis of “no pneumonia,” because many other diagnoses can present with a similar clinical picture (Fig. 8–10). Nosocomial pathogens commonly have multiple resistance profiles, typically plasmid-mediated. Resistance pressure from the use of third-generation cephalosporins has led to the establishment of vancomycin-resistant enterococci (VRE) as well as extended-spectrum β-lactamase–producing (ESBL) organisms of which Klebsiella is the prototype.50 Plasmid-mediated resistance to fluoroquinolones parallels the rise of ESBL-producing organisms.51 Empirical antibiotic selection should be derived from each hospital’s local antibiogram based on likely pathogens. A β-lactamase inhibitor combination paired with an aminoglycoside and vancomycin are the authors’ empirical agents of choice based on their local antibiogram for ventilator-associated pneumonia. Should the reader’s microbiology laboratory identify an ESBLproducing pathogen, the appropriate antibiotic class of choice

147

RESPIRATORY PROCEDURES ● 

II 

148

interest for an increase in breath sounds. The recruitment may be terminated when there are good breath sounds on two consecutive maneuvers. Recruitment maneuvers may be combined with chest physiotherapy for added benefit; chest phy­ sical therapy should, in general, precede the recruitment maneuver. Bronchoscopy.  See the discussion of nosocomial pneumonia under “Ventilator-Associated Pneumonia,” earlier. Bronchoscopy/BAL.  With MV, the normal bacterial, viral, and secretion clearance mechanisms of the mucociliary elevator are compromised. Accordingly, excellent pulmonary toilet is required to prevent secretion impaction, atelectasis,   Q pneumonitis, pneumonia, intrapulmonary shunt, or V mismatch. Despite seemingly adequate nursing or respiratory therapist care of a patient’s airways, atelectasis, mucus plugging, and segmental or subsegmental airway obstruction and collapse may occur. Several initial maneuvers are indicated, including alveolar recruitment, chest physiotherapy, postural drainage, and aerosolized bronchodilator therapy. Frequently, these maneuvers resolve the elevated Paw-peak and hypoxemia that are the markers of complications. When these initial therapies fail, a more invasive approach is warranted. The traditional approach to clearance of inspissated secretions is therapeutic bronchoscopy. The adult flexible fiberoptic bronchoscope has an outer diameter of 3.3 mm and a working channel for suctioning of 2.5 mm. Therefore, an ETT of size 8.0 or larger is ideal because it will permit easy passage of the bronchoscope and allow for adequate MV of the patient. However, airway mucosal irritation is a powerful sympathetic stimulant. Tachycardia, systemic hypertension, and bronchospasm commonly complicate therapeutic bronchoscopy. In the setting of intracranial hypertension, the clinician must take steps to blunt any potential sympathetic stimulation. Mucosal irritation may be minimized by careful bronchoscopic technique that avoids impacting and suctioning the sidewalls. In addition, topical or systemic lidocaine will also blunt mucosal irritation. Preprocedure β-blockade with a relatively short-acting agent like esmolol will blunt the tachycardia and elevated dP/dt that accompanies heightened sympathetic tone. This will diminish any increase in cerebral blood flow that accompanies sympathetic discharge. Adjuvant therapy with narcotic analgesia, with a short-acting agent like fentanyl, will enhance sedation and ameliorate pain from mucosal injury. When these measures fail, significant sedation and cerebral protection may be achieved with cautious administration of barbiturates like sodium pentothal. Pentothal therapy may also be complicated by systemic hypotension. Alternatively, for very short procedures, etomidate is an excellent and powerful sedative that has the unique advantage of inducing diminished ICP. Thus, excellent intravenous access for fluid or inotrope administration is mandatory when using barbiturates. A postprocedure chest x-ray is indicated to assess the results of the bronchoscopy and to assess for complications such as a pneumothorax or ETT malposition. When a specimen is obtained, examine it by Gram stain as well as culturing it for bacteria, viruses, fungi, or acid-fast bacilli when indicated. Use these results to help guide initial antibiotic therapy if the patient’s clinical condition indicates infection (e.g., mucosal erythema, leukocytosis, fever, hypotension). The culture results are qualitative only and although they serve to identify which organism(s) is/are present, they do not indicate the bacterial burden. Accordingly, the utility of such results has been derided as being no more useful than an aspirate obtained by a closed-suction system. However, a

closed-suction system does not allow directed suctioning of a particular side of the airway. In fact, the right side is more frequently suctioned than the left, based on the straighter geometry of the right main stem bronchus. To combat the geometry, “steerable” suction catheters are available that allow for directed lavage and suctioning. Furthermore, the suction catheter is enclosed within a sleeve that protects it from contamination during passage through the ETT and upper airways. With both the “steerable” catheter system and therapeutic bronchoscopy, the techniques may be modified to allow for quantitative assessment of the bacterial/viral/fungal burden. The technique is called bronchoalveolar lavage, and relies on wedging the tip of the fluid instillation/suction catheter into a bronchopulmonary segment, instilling a known amount of fluid (usually 180 mL of normal saline in 60-mL aliquots), and recovering that fluid for analysis. An adequate recovery is greater than 50% of the instillate volume. Moreover, there are criteria for the diagnosis of infection (bacteria > 300 colony-forming units [CFU]/mL).55 The criteria are liberalized (>500 CFU/mL) if the lavage and aspirate were performed in a larger airway such as the bronchus intermedius instead of a segmental orifice such as the superior segment of the lower lobe. Another modality that may be useful in the diagnosis of pulmonary infection is the bronchoscopically directed “protected brush biopsy.”56 In this technique, the bronchoscope is advanced into the area of interest, and a sheathed brush is advanced through the working channel into the airway. Then the brush is extruded and worked back and forth against the airway to “biopsy” adherent microbes and airway mucosa. The brush is withdrawn into the sheath, and the entire assembly is withdrawn. The brush is then cut off and incubated in culture media. This technique has numerous advantages in that upper airway secretions may be suctioned without fear of contaminating the specimen and obtaining spurious results. By comparison, the standard BAL technique requires that the operator guide the bronchoscope into the affected region without suctioning so as to not contaminate the subsequently aspirated lavage sample. The downside to protected brush biopsy is that mucosal injury, bleeding, and pneumothorax occur more commonly than with bronchoscopically directed or steerable-catheter BAL. Regardless of the technique used, the clinician must match the sample results with the patient’s clinical picture. Gastric Content Aspiration and Pneumonia.  Additional consideration is needed to avoid gastric acid blockade in patients who require prolonged intubation and MV. Gastric acid inhibition has, in some studies, been associated with a higher rate of nosocomial pneumonia than in patients who received ulcer prophylaxis with sucralfate alone.57 It is believed that the gastric acid milieu destroys refluxed bacteria and that most nosocomial pneumonias occur because of aspiration of gastric contents. It is important to recall that aspiration may occur simply by “wicking” of gastric secretions along an indwelling nasogastric tube that stents open the upper and lower esophageal sphincters as well as by vomiting and passage of gastric contents along the sides of the cuffed ETT. Aspiration of gastric acid with intubation (the most common scenario) does not require antibiotic therapy, nor is it improved by the administration of glucocorticoids or the use of immediate bronchoscopy (unless there is large airway obstruction). The clinical syndrome of sterile gastric content aspiration is Mendelson syndrome.

easily managed with combination therapy such as β2-agonists, acetylcholine antagonists, and glucocorticoids. A true management challenge is the critically ill asthmatic. These patients are different from others with asthma exacerbation in that they require intubation and MV.60 Unlike patients with many other disease states, asthmatics are not immediately improved by PPV; asthmatics often become acutely worse before any improvement from intubation and ventilation is realized. After intubation, the asthmatic’s Paw-peak is usually elevated. This leads to various problems including, but not limited to, early termination of a volume-cycled breath (excessive Paw limiting the breath), impaired gas exchange (increased D V  T ), and the induction of “biotrauma” (see later). A V slower respiratory rate allows for a longer time in exhalation. A prolonged Te is essential for the patient with restrictive  T , as disease. In VCV, a lower respiratory rate with a low V in the ARDSNet protocol, may be used for the management of life-threatening asthma.31 With PCV (as discussed earlier),  T based on the the set pressure may generate an inadequate V restrictive component of the exacerbated asthma. It is essential that bronchodilator and anti-inflammatory therapy (e.g., glucocorticoids) be pursued in conjunction with PPV for an optimal outcome.61 A diligent search should be undertaken to discern any potential triggers (e.g., infection) that may be eliminated to hasten recovery and limit the duration of MV. Appropriate sedation is critical to ensure adequate gas exchange; it enables the patient to “synch” with the ventilator and not trigger early volume-cycled breath termination. If sedation alone is inadequate to reduce the restriction imposed by the chest wall or intra-abdominal contents, pharmacologic relaxation is then indicated (although uncommonly required). Heliox therapy has also been used with success for

A well-designed weaning protocol is an invaluable aid in reducing the length of stay in the ICU. An appropriate protocol will enable the respiratory therapist and bedside nurse to initiate the weaning process each day before clinician evaluation. Computer order entry may create an ICU admission data set that automatically activates such a protocol once the entry criteria are met (i.e., the cause of respiratory failure is improving or has been eliminated, FIO2 < 0.50, PEEP < 10 cm H2O, and no pressors other than dopamine at < 5 µg/kg per minute or epinephrine or norepinephrine at < 0.05 µg/kg per minute). A ventilator pathway to chart and modify the progress of each patient through her or his MV needs is a useful tool. Such a pathway allows clinicians to regularly review a patient’s progress along what would be considered a “usual course” for someone requiring MV. Deviation from this course should prompt an investigation into the cause(s). A pathway is also an excellent tool to use as a platform for quality assurance and improvement review.

Mechanical ventilation

Asthma.  Fortunately, most patients with asthma are

Ventilator-Weaning Protocols and Pathways

● 

SPECIAL TOPICS

the failing asthmatic as a means of avoiding intubation in select patients (see “Heliox Therapy,” earlier).62 APRV has been used for severe life-threatening asthma; insufficient data are currently available to recommend this as front-line therapy. Its role may be as a salvage mode for asthmatics with refractory hypoxemia.

8 

Heliox Therapy.  The interface of gas with airways creates a certain amount of friction. Heavier gases lead to greater amounts of friction than lighter gases. The more friction generated, the greater the work of breathing for a given gas. Severe asthma commonly entails significant work of breathing and may lead to respiratory failure from respiratory muscle fatigue. Altering the gas composition from N2 : O2 to He : O2 (Heliox) provides for a lighter gas that requires less work of breathing. Different percentage mixtures of He : O2 are prepared and commercially available (e.g., 70% O2 and 30% helium). Successful resolution of impending respiratory failure has been achieved using this strategy.58 Note that this is not a commonly used therapy, but it is an important adjunct to have available when the need arises. Negative-Pressure Ventilation.  This unique mode of ventilation is best achieved using the Hayak Oscillator, a device produced in Israel. Its appearance is quite similar to a Cuirasse vest, but the driving negative-pressure source is quite different. The Hayak Oscillator features independent controls for the application of negative pressure as well as positive pressure, frequency of cycling between negative and positive pressure (inspiration and expiration), a chest physiotherapy mode for sputum expectoration (useful for those with cystic fibrosis), and a cardiopulmonary resuscitation mode. It is not widely used in the United States, but has demonstrated use in the cystic fibrosis patient population and during upper airway surgery, when an indwelling ETT would be a significant obstruction.59

Neuromuscular Blockade Neuromuscular blocking agents are used to induce muscular paralysis for various reasons including, but not limited to, O , reducing Paw-peaks during MV, reducing total body V 2 protecting life-sustaining indwelling devices, and placing an artificial airway. Agent selection entails consideration of factors identical to those surrounding analgesic and sedative selection. Commonly used agents include pancuronium, vecuronium, and cisatracurium. All may be given by bolus or continuous infusion. Only pancuronium and vecuronium have active metabolites and reportedly result in prolonged neuromuscular blockade in some patients after cessation of drug therapy.63 Furthermore, aminoglycosides, for instance, may potentiate the effect of neuromuscular blocking agents, thus reducing the amount of drug necessary to achieve the desired paralysis.64 The authors prefer cisatracurium for neuromuscular blockade because it undergoes Hoffman elimination in the plasma and is therefore independent of renal or hepatic metabolism. However, the authors also rarely use neuromuscular blockade outside of the operating room except when placing an ETT. Paralysis is commonly titrated to an effect monitored by a peripheral nerve monitor applied over the ulnar or other peripheral nerve distribution.65 No blockade results in four twitches of the adductor pollicis muscle resulting from four supramaximal triggering stimuli; complete blockade yields no response. A common goal of blockade is use of enough agent to result in two twitches out of a “train of four.” Another goal of twitch monitoring is to avoid overparalysis, to diminish the risk of prolonged neuromuscular blockade after withdrawal of the agent. If, however, zero twitches are required to achieve the goals of therapy, the monitor cannot monitor overparalysis at all. In addition, if feasible, many clinicians allow patients to emerge from paralysis once during each 24-hour period to perform a neurologic assessment and help ensure return of

149

RESPIRATORY PROCEDURES ● 

II 

neuromuscular function after cessation of drug therapy. There are no data to support this practice as a preventive measure, but it seems to make intuitive sense. There is a growing trend to avoid chemical relaxation throughout the United States; chemical relaxation is rare in the European Union for patient management in the critical care arena. The myriad potential complications of neuromuscular blockade have been described in detail in standard anesthesia texts. However, two important complications deserve mention: prolonged paralysis syndrome63 and the polyneuropathy of critical illness.66 The postparalysis syndrome is diffuse motor weakness associated with elevated creatine kinase levels (MM fraction) and preserved sensory nerve function on electromyography and nerve conduction velocity testing. By comparison, critical illness polyneuropathy involves both sensory and motor nerves and is less frequently associated with neuromuscular blocking agents as an etiologic cause. Critical illness polyneuropathy is believed to be principally related to the underlying disease and carries a less favorable prognosis for recovery than postparalysis syndrome. Some data implicate the aminosteroid structure of vecuronium and pancuronium in the pathogenesis of either of the complications mentioned previously by drawing a parallel between the neuromuscular blockade polyneuromyopathies discussed earlier and those identified in patients on long-term steroid regimens. However, data are currently inconclusive as to the exact etiology of the syndromes discussed earlier.

Sedation 150

Patients on MV commonly require some sedation, which can be provided in an intermittent bolus fashion or by continuous infusion. Clinician monitoring of the depth and adequacy of sedation is feasible in an inactive patient. When patients require sedation for agitation control, are mechanically ventilated, or are chemically relaxed, the ability of the clinician to assess the depth and adequacy of sedation is severely impaired. In addition, because the use of pharmacologic paralysis presents the external appearance of a quiet, restful patient, it is important to have some means of titrating sedation to an appropriate level. The authors favor using a modified single-lead electroencephalogram montage known as the bispectral index (BIS; Fig. 8–11). This device integrates a power spectrum of the coherence of electrical activity of the monitored areas of the brain and translates the information into an analog value ranging from 0 to 100.67 Lower numbers indicate deeper levels of sedation. This device has been successfully used in the operating room to monitor and titrate the level of benzodiazepine or propofol sedation for surgical procedures.68 Further experience is being gained in the ICU titration of therapy in pharmacologically paralyzed patients as well as monitoring serial changes after central neuraxial injury or illness that results in brain death. The authors have used BIS monitoring to decrease sedative agent usage and cost in intubated, neuromuscularly blocked and sedated patients in the critical care setting.69 Protocol-driven sedation guidelines may decrease the duration of MV, complications, and ICU length of stay.70

Neonatal Ventilation It is clear that neonatal ventilation is not simply ventilation of small adults. The vast majority of neonatal ventilation is performed as PCV, albeit with smaller volume targets than in

Figure 8–11  The Bispectral Index monitor (Aspect Medical, Nantucket, MA).  This pole-mounted device attaches to the patient’s forehead and provides a modified single-lead electroencephalogram whose power spectrum undergoes a Fourier transformation to yield a numeric representation of the level of sedation.

VOLUME GUARANTEE Set Pinsp Airway pressure

insp Flow exp VT = 8 mL VT = 7 mL VT = 6 mL VT = 5 mL Figure 8–12  Volume guarantee ventilation.  Note that the ventilator uses the prior breath to determine how much support is required to achieve the desired tidal volume and remain within the set pressure limit.

adults (i.e., same milliliters per kilogram of body weight but smaller changes in pressure to achieve the smaller needed volumes). Note, however, that neonatal ventilation may also be performed using VCV. Not all ventilators can deliver the small volumes required for this kind of ventilation, and special ventilators dedicated to neonatal ventilation have been developed (e.g., Babylog by Drager, Bird Ventilator). Many adult ventilators are equipped with a software package that allows the microprocessor to control pressure, flow rates, and volume in this application (e.g., Drager E4 with Neoflow, Siemens Servo 300C). Furthermore, there are hybrid modes such as pressure-support ventilation volume guarantee (PSV-VG) that combines the best of both worlds.71 One simply sets a pressure-support limit as well as a desired volume for each breath. The ventilator then determines on a breath-by-breath basis how much pressure support is required to achieve the set target and remain within the pressure limit (Fig. 8–12). In this way, the mode is also self-weaning; as the patient’s need for support decreases, the support, in fact, decreases (Fig. 8–13). Because the equipment varies at each institution,

CPAP VT

Clung

Time Figure 8–13  Pressure support volume guarantee ventilation.  As the compliance of the lung (C lung) improves, the positive inspiratory pressure (PIP) decreases while maintaining a constant tidal volume ( V T ). When the pressure support volume is reduced to an acceptable minimum level, weaning has been achieved and the patient should be evaluated for liberation from mechanical ventilation.

clinicians are urged to familiarize themselves with the available ventilator and how to use it. Guidance from a neonatal intensivist in conjunction with a respiratory therapist is ideal.

Noninvasive Ventilation Intubation, MV, and its sequelae may be avoided in a select group of patients suffering from acute respiratory failure by using noninvasive ventilation (NIV) techniques (see Chapter 2). It is important to recall that if the patient’s physiology requires definitive airway control, NIV is not appropriate. This modality may be effectively used in patients who do not wish to be intubated (e.g., those with end-stage COPD) and in patients who need time for medical therapy to achieve its goals (e.g., those with congestive heart failure). In general, NIV is commonly used for temporary ventilatory and oxygenation support (e.g., 0.6 LDH, lactate dehydrogenase.

171

RESPIRATORY PROCEDURES ● 

II 

A

B

Figure 9–14  Pre- (A) and post-thoracentesis (B) chest radiographs. Postoperative films are not always routine but should be performed if air was aspirated, the patient has postprocedure chest pain or dyspnea, multiple attempts were made, or the patient is on a ventilator.  (A and B, From Thomsen T, Setnik G [eds]: Procedures Consult—Emergency Medicine Module.Copyright 2008 Elsevier Inc. All rights reserved.)

172

likely has a transudative effusion, and no further fluid analysis is necessary. Once a fluid is classified as transudative, it typically requires no further fluid analysis, and therapy is directed at the underlying cause of the effusion (e.g., CHF, cirrhosis). In the presence of an undiagnosed exudative effusion, however, more extensive fluid evaluation is required.

monary infarct. A predominance of lymphocytes is consistent with a more chronic pleural process, including malignancy, TB, PE, and viral pleuritis.10,39 Eosinophil counts of greater than 10% often have no clear etiology, but have traditionally been associated with blood or air in the pleural space.

Evaluation of Exudates

If infection is a concern, directly inoculate blood culture bottles at the bedside with 10 mL of fluid. TB cultures may be obtained from separate collection tubes.

All undiagnosed exudates, at a minimum, should have pleural fluid sent for cell count with differential, glucose, and cytology39 (Table 9–3). In regions with a high prevalence of tuberculosis (TB), adenosine deaminase (ADA) should be added to this list. Clinical suspicion for an underlying disease process should guide additional fluid evaluation.

Cell Count with Differential In general, the presence or absence of red blood cells (RBCs) is not useful in determining the etiology of the effusion because it takes a very small amount of blood to cause a blood-tinged appearance. A grossly bloody pleural effusion or RBC count of greater than 100,000 cells/mm3 is suggestive of trauma, malignancy, pneumonia, or pulmonary infarction,31,40 but a lack of RBCs does not exclude these diagnoses. Grossly bloody pleural fluid with a hematocrit of greater than 50% of the peripheral hematocrit often requires tube thoracostomy. Exudates typically have a pleural fluid white blood cell (WBC) count of greater than 1000 cells/mm3, and counts may reach levels of greater than 10,000 cells/mm3, most commonly with parapneumonic effusions.40 The differential cell count can be useful in identifying the cause of an exudative pleural effusion. A predominance of neutrophils indicates an acute process affecting the pleural surface, such as infection or pul-

Culture

Glucose The concentration of glucose in exudates is extremely variable and, in general, does not correlate with any specific disease process. Routine measurement of pleural fluid glucose for exudative effusion is recommended, in that a low glucose concentration (10,000 cells/mm3 >50% >50%

  Eosinophils Glucose

>10% 40 IU/L Presence of organism >100 U/L

Triglycerides Creatinine (with serum measurement) Albumin (with serum measurement)

>110 mg/dL Pleural fluid–to–serum creatinine ratio > 1 Serum–to–pleural fluid albumin gradient > 1.2 g/dL >140 pg/L58 Presence of TB DNA sequences, TB 50% of hemithorax Loculated effusion Pleural thickening by CT Aspiration of frank pus Pleural fluid pH < 7.2 Positive Gram stain or culture of pleural fluid CT, computed tomography.

Parapneumonic Effusions Patients with suspected parapneumonic effusions warrant rapid evaluation and outcomes risk assessment based on pleural anatomy, pleural fluid bacteriology, and pleural fluid chemistry.47 All parapneumonic effusions require at least diagnostic thoracentesis with the goal of identifying patients with complicated parapneumonic effusions. The indications for tube thoracostomy or other surgical management include large or loculated effusions, pleural thickening on CT scanning (the pleural peel), aspiration of frank pus, pleural fluid pH of less than 7.2, and positive Gram stain or culture (Table 9–4). Pleural pH measurement gives useful information regarding pleural inflammation. Normal pleural fluid pH is approximately 7.64. Pleural fluid pH less than 7.3 indicates pleural inflammation. The differential diagnosis of pleural fluid acidosis includes not only empyema and complicated parapneumonic effusion but also malignancy, TB, esophageal rupture, and collagen vascular disease.48 Measurement of pleural pH is essential in the evaluation of suspected parapneumonic effusions because pH is a key factor in the management algo-

rithm. The fluid must be collected anaerobically, but may be transferred from the initial 50 or 60 mL syringe into a heparinized blood gas syringe,49 and then left at room temperature for up to 1 hour prior to laboratory analysis50 without affecting the accuracy of the results. Because of these specifics regarding the collection and evaluation of pleural fluid pH, pleural fluid should routinely be transferred to a heparinized blood gas syringe and placed on ice while awaiting the decision for pH testing.

COMPLICATIONS Pneumothorax The most frequently reported complication of thoracentesis is pneumothorax, which has a reported incidence of 4% to 19% in some studies.23,28,51 Thoracostomy tubes were required in less than 50% of the post-thoracentesis pneumothoraces in each study. The mechanism for this complication is puncture of the lung or inadvertent air entry through the needle or

173

RESPIRATORY PROCEDURES ● 

II 

catheter during the procedure. Pneumothorax should be suspected if there is aspiration of air during fluid removal or if the patient develops new symptoms during or after the procedure. Procedure-related factors appearing to contribute to pneumothorax include inexperienced operator, therapeutic taps, and use of needles larger than 20 gauge.22,24,51,52 The risk of pneumothorax may be increased in patients with underlying chronic obstructive pulmonary disease.53

Cough Cough is another frequently encountered complication. Although typically considered a minor complication resulting in only patient discomfort, it may be associated with the creation of an iatrogenic pneumothorax.52,54 The procedure should be terminated if persistent patient coughing occurs.

Infection As with all procedures, there is a potential risk for infection which is estimated at 2%. The risk is kept low with proper attention to patient preparation and sterile technique.

Uncommon Serious Complications

174

Other serious complications have been reported, but occur in less than 1% of procedures. These include hemothorax, splenic rupture, abdominal hemorrhage, unilateral pulmonary edema, air embolism, and catheter fragment left in the pleural space.52 Hemothorax may be suspected by a rapid accumulation or reaccumulation of pleural fluid or by a change in patient vital signs after the procedure. Hemothorax may be due to

laceration of the lung or the diaphragmatic, intercostals, or internal mammary vessels. Careful attention to technique, such as avoiding the superior portion of the intercostal space, never puncturing medial to the midclavicular line, and not penetrating too deeply into the thorax during needle insertion, should be practiced. Hemothorax requires appropriate surgical consultation and drainage via thoracostomy tube. Puncture of the spleen or liver through the diaphragm may result in localized organ hematoma or hemoperitoneum.54 Clinically, this is suspected when the needle pass does not yield pleural fluid (dry tap) and is followed by a patient complaint of abdominal pain. If this diagnosis is suspected, appropriate resuscitation is the initial treatment, followed by a diagnostic imaging study, preferably CT scan. If the patient is hemodynamically unstable, bedside ultrasound and immediate surgical consultation should occur. Through-the-needle catheters are not commonly used because the catheter may be cut or sheared off while it traverses the needle or is withdrawn. Reexpansion pulmonary edema is a complication associated with rapid reexpansion of the lung. Symptoms include dyspnea, tachypnea, tachycardia, cough, and frothy sputum.55 It is believed that this can be avoided by monitoring the pleural pressures carefully after 1500 mL of fluid has been withdrawn and discontinuing the procedure when the pleural pressures are greater than −20 mm Hg.31,56 There is no proven way, however, to assure that reexpansion pulmonary edema will not occur.

REFERENCES  c a n

be found on

E x p e rt C o n s u lt

Thomas D. Kirsch

Tube thoracostomy (TT) to evacuate an abnormal accumula­ tion of fluid or air from the pleural space is a common elective, emergent, or urgent procedure. The procedure is performed by emergency clinicians or surgeons, usually dependent on the urgency of the scenario, or local protocols. Air or fluid in the pleural space can result from a spontaneous or traumatic pneumothorax (PTX), pleural fluid accumulations from blood, malignancy, infection (empyema), or lymph (chylothorax). The first modern methods to evacuate pleural contents were developed in the 19th century, but these techniques were not widespread until 1918, when they were used to treat postin­ fluenza empyema. Military experience demonstrated that thoracic drainage combined with antiseptics and antibiotics reduced the mortality from thoracic trauma from 62.5% during the Civil War, to 24.6% in World War I, to 12% in World War II.1 TT has since evolved to become a common and effective procedure.2

PATHOPHYSIOLOGY The pleural space that normally separates the visceral and parietal pluerae has a thin layer of lubricating fluid separating the layers. The parietal pleura lines the interior of the chest wall. The visceral pleura covers the lungs. Under normal cir­ cumstances, a small negative pressure in the pleural space keeps the lung inflated. With inspiration, the negative intra­ thoracic pressure increases, leading to the expansion of the lung from an influx of air from the environment. The addition of blood, fluid, or air in the pleural space disrupts the normal pressure gradient and interferes with normal inspiratoryinduced inflation, leading to the “collapse” of the lung. The degree of respiratory compromise depends on the volume of the fluid or air in the pleural space, the patient’s age and baseline pulmonary status, and the integrity of the chest wall. As the amount of fluid or air increases, respiratory function worsens and produces symptoms of dyspnea with exertion and then at rest, often with pleuritic chest pain, and anxiety. Large positive-pressure accumulations associated with a tension PTX lead to severe respiratory dysfunction and cardiovascular compromise.

PNEUMOTHORAX Because the lung remains inflated due to a negative pressure in the pleural space, a PTX results from the presence of air in the pleural space and loss of this negative pressure (Fig. 10–1). The air can enter the pleural space internally from a ruptured lung bleb, the trachea or from the outside due to a penetrating injury. A PTX often has an iatrogenic cause sec­ ondary to egress of air from puncture or rupture of the lung into the pleural space. Common causes are procedures such as subclavian venous cannulation, transthoracic biopsies, tho­ racentesis, positive-pressure ventilation (PPV), or cardiopul­ monary resuscitation (CPR).

A spontaneous PTX occurs from the rupture of a subpleural lung bleb with little or no trauma and the chest wall remains intact. Based on the presence of underlying lung disease and the approach to treatment, a spontaneous PTX can be catego­ rized as either primary or secondary. A primary spontaneous PTX occurs in a patient without underlying lung disease. The typical patient with a spontaneous PTX is a tall, thin, 20- to 40-year-old male smoker. Secondary spontaneous PTXs occur in patients with underlying lung or pleural disease, including emphysema, chronic bronchitis, asthma, Marfan’s syndrome, Pneumocyctis jiroveci pneumonia, other pneumonias, and neo­ plasm. The morbidity, mortality, and long-term complica­ tions of a PTX increase for patients with underlying lung disease. Whereas a primary PTX may be selectively observed or simply aspirated, a secondary PTX often requires a more aggressive approach to management. Common symptoms include the sudden onset of pleuritic chest pain and dyspnea with exertion or at rest. Occasionally, pain is absent and symptoms vary from mild dyspnea on excursion that the patient may ignore for days to severe dyspnea at rest. Signs and symptoms do not always correlate well with the size or cause of the lung collapse, often surprising the clinician when symptoms are investigated. Some individuals with a small spontaneous PTX may never seek medical attention, and the process will resolve spontane­ ously. TT is the most common treatment, but some experts recommend simple aspiration as first-line treatment for all primary PTXs requiring intervention.2,3 A spontaneous PTX may rarely progress to a tension PTX. The process can be bilateral.

Traumatic Open PTX Loss of the normal negative intrapleural pressure results in a lung collapse. An open PTX occurs when the chest wall is penetrated and the negative intrapleural pressure is lost as outside air enters the pleural space. A tension PTX may develop with even a small open chest wall defect when air enters the wound with each inspiration but is trapped during expiration by a flap that functions like a one-way valve. Each breath therefore increases the intrapleural pressure. If the diameter of the chest wound is greater than that of the trachea, then with each respiratory attempt, air moves preferentially through the chest wall opening rather than down the trachea. This is a more serious injury because, if left untreated, it can prevent any ventilation of the involved lung.

Traumatic Closed PTX A closed PTX may also result from an injury when the chest wall is not penetrated. This usually results from a rib fracture that penetrates the lung, but can also occur when an alveolus or bleb ruptures after blunt trauma as a result of an abrupt increase in intrathoracic pressure against a closed glottis. The air leak from a closed PTX is usually self-limited, but can rarely progress to a tension PTX.

Tension PTX A tension PTX is a life-threatening condition that requires immediate intervention. Tension PTX usually results from penetrating chest injuries. Other causes include fractures of

Tube thoracostomy

Tube Thoracostomy

Spontaneous (Closed) PTX

● 

10

10 

C H A P T E R

175

RESPIRATORY PROCEDURES ● 

II 

B A

176

C

R

D

E

Figure 10–1  A, Anteroposterior chest radiograph view of a right-sided, seemingly small, simple pneumothorax (PTX). Note absence of peripheral lung markings on the right side and the distinct line indicating the edge of the collapsed lung. Although this appears to be a small PTX, it produced significant dyspnea in this patient with chronic obstructive pulmonary disease and therefore required a chest tube. B and C, Computed tomography (CT) scans show the extent of the collapse, almost 100% in some places, not appreciated on a plain x-ray. Adhesions kept part of the lung expanded. D, Sometimes bullae can be seen on a chest x-ray and mistaken for a PTX because their thin wall can be visualized. E, Larger bullae are sometimes identified only by the fact that an area on the chest x-ray does not appear to have any pulmonary vessels (arrows), and at the periphery, crowding of the normal lung and vessels may appear. Skin folds and the border of the scapula or external wires can also be mistaken for a PTX. A CT scan can settle these issues.

the trachea or bronchi, a ruptured esophagus, the presence of an occlusive dressing over an open PTX, and PPV. Patients with chest or lung injuries who are undergoing PPV are at much greater risk to develop a tension PTX, either spontaneously from lung injury due to high-pressure ventilation or when a small PTX is expanded by PPV. Because of this, any patient with a penetrating thoracic injury (even without immediate evidence of a hemothorax [HTX] or PTX) may be a candidate for a

“prophylactic” chest tube before mechanical ventilation. A tension PTX may develop from a simple PTX caused by CPR, but be clinically evident only after PPV has been insti­ tuted. Asthmatics or patients with emphysema may also develop a PTX, followed by tension PTX, from the high pressure required for ventilation. A tension PTX occurs when a pulmonary or bronchial injury creates a “ball-valve” or “flap-valve” mechanism that

10  ● 

Tube thoracostomy

R

C A Inspiration

177

B Expiration

D

Figure 10–2  A and B, The pathophysiology of a tension PTX. During inspiration, air enters the pleural space through a one-way valve either from the outside or from the lung itself. Upon expiration, the injury/valve closes, trapping increasing amounts of air in the pleural space. Eventually, the mediastinum shifts and cardiac filling and eventually cardiac output are compromised. C and D, Tension PTX. C, On a posteroanterior chest x-ray, the left hemithorax is very dark or lucent because the left lung has collapsed completely (white arrows). The tension PTX can be identified because the mediastinal contents, including the heart, are shifted toward the right, and the left hemidiaphragm is flattened and depressed. D, A CT scan done on a different patient with a tension PTX shows a completely collapsed right lung (arrows) and shift of the mediastinal contents to the left. (A and B, From Vukich DJ, Markovchick VJ: Pulmonary and chest wall injuries. In Rosen P, Barkin RM, Braen CR, et al [eds]: Emergency Medicine: Concepts and Clinical Practice. St. Louis, Mosby-Year Book, 1988. Reproduced by permission.)

leads to the progressive accumulation of air in the pleural space. The one-way valve effect is thought to be due to the presence of a tissue flap that allows air into the pleural space but then closes with expiration and traps the air (Fig. 10–2). The increasing pressures lead to ipsilateral complete lung collapse, and if allowed to continue, impingement on the mediastinum with a shift of the heart towards the uninvolved side, restricting ventricular filling and subsequently decreas­ ing cardiac function. This severe disruption of both respira­ tory and cardiac function can lead to hypotension and reduced ventilation (both hypoxia and CO2 retention) and eventually to cardiopulmonary collapse.

Pneumomediastinum Air in the mediastinum is termed pneumomediastinum (PM)4–6 (Fig. 10–3). It is usually a benign condition that presents with nonspecific pleuritic chest or neck pain or symptoms similar to those of a pericarditis, pneumonia, pulmonary embolism, or spontaneous PTX. The most common cause is a sudden increase in intrapulmonary pressure, usually without an obvious lung alveolar rupture. Rarely, there is an associated PTX or esophageal rupture. Spontaneous PM is usually benign but Stiegmann and coworkers4 reported a rare tension PM in which air from an alveolar rupture near the pulmonary

RESPIRATORY PROCEDURES ● 

II 

A

B

178

C

Figure 10–3  A, Pneumomediastinum (PM). Chest x-ray reveals extensive gas in the subcutaneous tissue (arrows) and outlining the mediastinal structures (arrowheads). B, Pneumopericardium. On this chest x-ray, the pericardium (arrows) is outlined by air between the pericardium and the heart and air in the lungs. Note that the pericardium does not extend above the level of the pulmonary arteries. This helps distinguish pneumopericardium from PM. C, Barotrauma (note air in the neck, arrow) as a consequence of acute respiratory distress syndrome. Note the presence of a PTX, PM, pneumopericardium, and subcutaneous air (into the neck) in this patient receiving positive-pressure ventilation (PPV).  (A–C, Courtesy of Dr. Thomas E. Stewart.)

artery tracked to the subcutaneous mediastinal tissue where it became “trapped by scar tissue” and produced pressure on the great vessels and major airways. PM from barotrauma in a ventilated trauma patient may be very extensive and can be associated with pneumopericardium (see Fig. 10–3C ). The most common causes of benign spontaneous PM are inhalation drug abuse (crack cocaine or marijuana) when a Valsalva maneuver is performed to enhance drug absorption, acute asthma, blunt chest trauma, violent coughing or vomit­ ing, or barotrauma. Often, the condition is spontaneous and no cause is found. The most common symptoms are chest pain, usually pleuritic, and neck pain, sore throat, dyspnea, or persistent cough. Vital signs, including pulse oximetry, are usually normal. About half the time, subcutaneous emphysema, usually in the neck, can be palpated. Usually, the diagnosis is made by demonstrating air in the tissues on radiographs of the neck or chest or more readily by computed tomography (CT) scan. Lateral chest/neck radiographs are more diagnostic than anteroposterior views. This condition can be mistaken for pneumopericardium (see Fig. 10–3B). The classic Hamman crunch (crinkling or audible crepitance during chest ausculta­

tion) is present in only 50% to 80% of cases of PM, and is easily initially missed. The clinical course is almost always benign, especially in young male drug inhalers, and extensive testing, hospitaliza­ tion, follow-up radiographs, or antibiotics are not warranted.5,6 A few hours of observation in the emergency department (ED) seems prudent if the symptoms are of recent onset. For the rare tension PM, decompression may be undertaken. A needle is inserted into the second or third intercostal space just lateral to the sternum and directed toward the anterior mediastinum, suggested as initial therapy in infants. Alterna­ tively, a small incision 2 to 3 cm above the sternal notch can be made so finger dissection of the deep cervical and pre­ tracheal fascia can be performed. A chest tube is not therapeutic.

HEMOTHORAX A HTX is the accumulation of blood in the pleural space, caused by injuries to the heart, great vessels, or the vessels of the lungs, mediastinum, or chest wall. Bleeding from the lung parenchyma is low pressure, and is usually self-limited or

Chylothorax Chylothorax results from an injury to the thoracic duct during central line placement, operative injury, or chest trauma. The primary thoracic duct injury is usually asymptomatic because the chyle initially collects extrapleurally and may not begin to fill the pleural cavity for 2 to 10 days. As the fluid accumulates, the patient slowly develops respiratory symptoms. The chest radiograph demonstrates a pleural effusion, and the diagnosis is made when the thoracentesis reveals a milky fluid with a high fat and lymphocyte content and 4 to 5 g/dL of protein. The definitive treatment is by either repeated thoracentesis or TT combined with parenteral alimentation until the volume of chyle decreases.

DIAGNOSIS Symptoms The presentation of patients with abnormal collections of fluid or air in the pleural space can range from asymptomatic to cardiopulmonary arrest. The severity of symptoms depends somewhat on the size of the PTX and especially the rapidity of accumulation, age of the patient, and the presence of an underlying lung disease. Specific symptoms range from mild dyspnea with exercise and pleuritic chest pain for small dis­ ruptions to hypotension and severe dyspnea for those with a tension PTX. A cough may also be present. Severely injured patients may be unable to relate symptoms, rendering the physical examination and chest radiograph essential to diag­ nose a PTX and HTX. A tension PTX must be considered in any patient with sudden respiratory or cardiac deterioration and in intubated patients who become difficult to ventilate with increased airway pressure, hypotension, or elevated central venous and pulmonary artery pressure. Conscious patients with a tension PTX will rapidly develop severe dyspnea, restlessness, agitation, and a feeling of impending doom. They are usually tachycardic and tachypnic and can quickly become hypotensive. With a spontaneous PTX, 95% of patients complain of the sudden onset of sharp or pleuritic chest or shoulder pain, or both. Sixty percent of patients experience dyspnea, and 12% have a mild cough. Dyspnea and anxiety are more common in older patients. The symptoms of an HTX are often similar to those of a PTX, but may be accompanied by hypotension as blood accumulation in the pleural space increases. The onset of symptoms for effusions is usually much more gradual with increasing shortness of breath and dyspnea on exertion over days to weeks, a common scenario with malignant effusions.

Stable Patients For more stable patients (and those with smaller accumula­ tions), the physical examination is less sensitive and a chest radiograph is usually necessary for the definitive diagnosis. Physical findings may include unilaterally decreased breath sounds, tachypnea, tachycardia, decreased tactile fremitus, increased resonance with percussion, or subcutaneous emphy­ sema, but the examination may reveal little to no abnormali­ ties with a small PTX. A less than 20% PTX will often present with a completely normal chest examination, including equal breath sounds. Pleural fluid collections are difficult to detect by physical examination, particularly with less than 500 mL of fluid in the pleural space. Breath sounds may be decreased and percussion of the bases may be dull. Parapneumonic empyemas often present with fever, cough, chest pain, dyspnea, and purulent sputum. The physi­ cal examination will reveal diminished breath sounds, dullness

Tube thoracostomy

An empyema is an accumulation of pus in the pleural space, usually from parapneumonic infectious effusions. Empyemas are estimated to occur in 1% to 2% of hospitalized patients with pneumonia. The remainder of empyemas result from violations of the thoracic space by surgical procedures, trauma, or esophageal perforation. Staphylococcus aureus is the most common isolate.

Unstable Patients During the initial (“ABC”) phase of resuscitation, the vital signs often point to the presence of a tension PTX. This diagnosis must be considered for injured patients who are tachycardic, hypotensive, and dyspneic. Similar symptoms occur with a pulmonary embolus, pericardial tamponade, and severe pneumonia. No single examination will reliably diag­ nose a tension PTX, so multiple methods must be rapidly conducted. “Look, listen, and feel” is the rule. Observation of the chest wall may reveal asymmetrical chest expansion. Neck and forehead veins may be distended even if the patient is hypotensive, or the trachea may be deviated away from the side of the PTX. Auscultation may demonstrate diminished breath sounds on the injured side. In one prospective study, the sensitivity, specificity, and diagnostic accuracy of ausculta­ tion for HTX/PTX was 84%, 97%, and 89%, respectively.7 A false-negative auscultation is more likely than a falsepositive one.7 Finally, when percussing the chest wall, there may be hyperresonance on the affected side and subcutaneous emphysema may be present. Pulsus paradoxus (>12 mm Hg of normal inspiratory decrease in systemic blood pressure) may be evident. For intubated patients, an early sign of tension PTX is difficulty in bagging owing to increased airway pressures. In injured patients with apnea, hypotension, or cardio­ pulmonary arrest, the diagnosis and treatment of a tension PTX should be made by an immediate needle or catheter decompression thoracentesis, not by a radiograph. The diag­ nosis of a tension PTX is confirmed when there is a rapid improvement in the vital signs and possibly a rush of air through the needle. Post CPR PTX.  After vigorous CPR, it is not uncom­ mon to encounter an iatrogenic PTX, from either bleb rupture from compressions, lung trauma from a broken rib, or central line placement. Rib fractures are common during CPR in patients with severe osteoporosis. A small PTX is often incon­ sequential, but many CPR survivors undergo PPV, and a stable PTX may be converted into a tension PTX, with car­ diopulmonary deterioration. Such deterioration may be assumed to be secondary to a return of the original pathology, but a search for a treatable tension PTX must be undertaken. Subcutaneous air, decreased breath sounds, difficulty with providing ventilations, or a deviated trachea should raise sus­ picion for this event (Fig. 10–4).

● 

Empyema/Effusions

Physical Examination

10 

ceases when a chest tube is placed. Intercostal artery, pulmo­ nary artery, and internal mammary artery bleeding can be profuse and often requires surgical intervention.

179

RESPIRATORY PROCEDURES ● 

II 

A

180

B

C

Figure 10–4  A, After successful cardiopulmonary resuscitation and intubation, this patient began to deteriorate, with a precipitous drop in blood pressure and a decreasing oxygen saturation. It was believed that the cause of the initial cardiac arrest was returning. Marked subcutaneous air was noted in the scrotum and abdominal wall, but little air was noted in the chest wall tissue. The subcutaneous air had curiously tracked via tissue planes, a distinctly unusual place for air to accumulate. B, A chest tube (arrow) quickly reversed the decompensation. C, This patient had a respiratory arrest from heroin injected into a neck vein (note extensive scar). After resuscitation with a bag-mask, he had a return of respiratory depression, unresponsive to naloxone, and he was very difficult to ventilate. He had a small PTX from a nick in the lung from the neck injection, and PPV turned it into a tension PTX.

to percussion, egophony, and diminished tactile fremitus on the involved side. Patients with an indwelling chest tube with an empyema will develop a fever, the pleural fluid drainage may be excessive and become purulent, and respiratory symp­ toms may worsen.

Radiography Plain Radiographs A chest radiograph is essential to diagnose a PTX in the stable patient. Unstable patients with a potential tension PTX may be diagnosed clinically or receive a portable radiograph while monitored by a clinician. The best plain radiographs for an HTX or a PTX are an upright inspiratory posteroanterior and lateral chest. Contrary to common belief, an expiratory upright posteroanterior chest radiograph is no better at detecting a PTX than the traditional upright inspiratory view. Upright is preferable to a supine chest radiograph for an HTX because even with large amounts of blood only slight differences in the densities of the lung fields may be found. With an upright chest radiograph 300 to 500 mL of fluid is needed to cause costo­ phrenic angle blunting8 (Fig. 10–5). If there is significant clinical suspicion, then further studies are needed to rule out the presence of intrapleural

fluid or air. A thoracic CT scan is the “gold standard,” dem­ onstrating many PTXs that are never visible on a plain radio­ graph. If this is not available, then to look for an HTX, the best additional views are bilateral decubitus chest radiographs, with the PTX expected to be seen on the side away from the table as gravity pulls down the affected lung. On a chest radiograph, the partially collapsed lung of a PTX appears as a visceral pleural line with no pulmonary markings beyond it (see Fig. 10–1E). It is easy to initially mistake large blebs for a PTX or identify the scapular border, skin folds, or indwelling lines as a PTX, but a CT scan quickly resolves the issue. Other radiographic findings include hyperlucency of the affected HTX, a double diaphragm contour, increased visibility of the inferior cardiac border, better visualization of the pericardial fat at the cardiac apex, and possibly a depressed diaphragm. If subcutaneous air is noted on the chest x-ray of a patient with blunt chest trauma, it can be assumed that the air came from an injured lung and that a PTX exists. It is difficult to accurately predict the size of a PTX on plain radiographs, and such calculations have limited clinical applicability.9 The size is usually described as a percentage collapse, but the vagaries of the collapse can be correctly delineated only by CT. With a tension PTX, the chest radio­ graph reveals lung collapse with no lung markings, a depressed

Fluid

A

Ultrasound Ultrasound has been shown to be useful in diagnosing both PTXs and HTXs. At least two ultrasonographic signs are described that are used to identify a PTX. One is the absence of the “sliding lung” sign, which is the movement of the hyperechoic line between the chest wall and the aerated lung with each respiration. The presence of the sliding lung sign effectively rules out a PTX, and its absence suggests that a PTX may be present, but confirmation with other signs is necessary. Another sign is the absence of “comettail artifacts,” which are hyperechoic reverberation artifacts of the visceral pleural line that spread to the edge of the screen.13

Tube thoracostomy

No meniscus

● 

Straight line

10 

CT scans are not routine for the diagnosis of a PTX, but are more useful for HTXs and other fluid collections. They also offer invaluable information on the etiology of such abnormalities. A CT scan may be useful when the diagnosis is unclear or when looking for small amounts of pleural fluid. CT scans are particularly useful to determine whether an empyema is loculated or draining successfully.

INDICATIONS FOR TT PTX

B Figure 10–5  A and B, Stab wound to the chest. When accumulated fluid in the chest cavity is seen as a straight line on an x-ray (an airfluid level [see arrows]), with no meniscus up the side, air must be present, even if a PTX cannot be seen. This poor-quality radiograph failed to show a 20% PTX found on CT, but a better plain radiograph may have also been diagnostic. If the patient is supine, these findings may not be seen.

hemidiaphragm on the affected side and a shift of the medi­ astinum and trachea to the opposite side. With a bilateral PTX, no mediastinal shift may be seen. Thoracic CT Scan CT scans of the chest are much more sensitive than plain radiographs for detecting PTXs and HTXs and more accurate to estimate the size and other characteristics of a PTX (see Fig. 10–1A and C). About 10% of trauma patients with a normal chest x-ray will demonstrate a small HTX or PTX.10–12 The clinical significance of these small, previously undetected, occult injuries is likely not great, and it has been suggested that a small PTX seen only on CT scan may be left untreated, and simply observed, in the otherwise stable patient. Many patients with a PTX seen only on CT scan may also safely undergo PPV without the placement of a chest tube.12

TT is by far the most common treatment for all types of PTXs, but controversy exists over the treatment of a small traumatic and primary spontaneous PTX. However, the American College of Chest Physicians has developed useful guidelines for the management of primary and secondary spontaneous PTXs14 (Table 10–1). Chest tube placement as routine initial intervention is likely not necessary in healthy patients with small primary spontaneous or isolated small traumatic PTXs in the absence of respiratory compromise, concomitant injuries, or when PPV will not be required. For a primary spontaneous PTX, a recent review concluded that there is no significant difference between simple aspira­ tion and tube drainage with regard to immediate success rate, early failure rate, duration of hospitalization, and the 1-year success and pleurodesis.15 Needle aspiration is associated with reduced analgesia requirements and lower pain scores com­ pared with TT. Simple aspiration is associated with a reduc­ tion in the percentage of patients hospitalized when compared with tube drainage. There is growing evidence that “video-assisted thora­ scopic surgery” may be the optimum treatment method for patients with an uncomplicated spontaneous PTX.16 For patients with a simple PTX, prolonged suction is rarely required, and the tube can be simply attached to a Heimlich valve or underwater seal.14,17 Without any intervention or continuing air leak, a small PTX will resolve over days to weeks. Supplemental oxygen will speed the reexpansion process by increasing the rate of pleural air absorption. Most patients with a secondary spontaneous PTX have enough underlying disease to eventually require a chest tube. Patients with chronic obstructive pulmonary disease, malig­ nancies, cystic fibrosis, and acquired immunodeficiency syn­ drome (AIDS)–related Pneumocystis jiroveci pneumonia, and tuberculosis infections are usually symptomatic enough, and have the potential for recurrence, that a TT usually cannot be avoided. Underlying infected and necrotic tissue in the AIDS patients often portends poor response to pleurodesis

181

RESPIRATORY PROCEDURES ● 

II 

TABLE 10–1  Guidelines of the American College of Chest Physicians for the Management of Primary and Secondary Spontaneous Pneumothorax Primary Spontaneous Pneumothorax (No Underlying Lung Disease) A clinically stable patient must have all of the following present: respiratory rate, 60 beats/min or 90%; and can speak in whole sentences between breaths. Clinically Stable Patients with Small Pneumothoraces (2000 mL), because these are often asso­ ciated with continuing hemorrhage. Autotransfusion of the shed blood is desirable if the technique is available.

Empyema The treatment of patients with empyema depends on the severity of their infection and their underlying condition. Some patients with empyema can be treated with serial tho­ racenteses, but most will require continuous drainage with a T T. Thoracoscopic decortication represents definitive therapy for severe cases. Usually, a diagnostic thoracentesis is done first to assess the fluid for signs of infection. Thick pus on thoracentesis, a positive Gram stain fluid glucose less than 60 mg/dL, pH less than 7.20, or elevated lactate dehy­ drogenase is associated with effusions requiring chest tube drainage. Once an empyema is detected, therapy should not be delayed because the fluid can become loculated within hours. The tube is left in place until the volume of the pleural drainage becomes clear yellow and is less than 150 mL in 24 hours. An empyema that fails to resolve on the chest radiograph within 48 hours requires chest CT scan and a careful review of antibiotic choice. Multiloculated effusions are best managed with thoracoscopic decortication.

CONTRAINDICATIONS For unstable injured patients with a PTX or an HTX, there are no absolute contraindications to a TT. In critical patients,

Tube thoracostomy

This is meant to be a guide, and clinical judgment should always be used.

● 

Massive hemothorax, >1000–1500 mL initial drainage Continued bleeding >300–500 mL in 1st hr >200 mL/hr for first 3 or more hr Increasing size of hemothorax on chest film Persistent hemothorax after two functioning tubes placed Clotted hemothorax Large air leak preventing effective ventilation Persistent air leak after placement of second tube or inability to fully expand lung

the placement of a chest tube is often performed empirically, because procedures to confirm the presence of, assess the extent of, or prove the absence of pathology are prohibited by logistics of the resuscitation. In the stable patient, relative contraindications include anatomic problems such as the pres­ ence of multiple pleural adhesions, emphysematous blebs, or scarring. Coagulopathic patients should be evaluated for clot­ ting factor replacement before any invasive procedure.

10 

TABLE 10–2  Indications for Surgery after Tube Thoracostomy Based on the Results of the Thoracostomy

TREATMENT Treatment of a Tension PTX during a Resuscitation Immediate decompression of the chest must be considered in all injured patients who present in extremis with unexplained hypotension, particularly those with penetrating chest inju­ ries. The goal is to open the chest cavity quickly to allow the accumulated air to escape. This can be accomplished with a scalpel and forceps, as is done at the beginning stages of a thoracostomy, or by needle decompression (Fig. 10–6). Alter­ natively, a large-bore needle/angiocatheter (minimum 16 gauge) should be placed at the midclavicular second intercos­ tal space on the side with diminished breath sounds, or both sides if unclear (Fig. 10–7). The needle should be removed, but the angiocatheter left in place to create a simple PTX. Whether or not this is successful in improving the patient’s vital signs, an open TT is then needed. If the needle decompression is not effective, an open thoracostomy can be started even without the immediate availability of a chest tube to create an exit for the air to normalize the respiratory and cardiovascular function. The technique is the same as that for a TT (see later).

Prehospital Treatment Emergent needle decompression thoracostomy may be used in the prehospital setting or when a patient suspected of having a PTX rapidly deteriorates or presents in extremis. The needle (or catheter) may then be attached to a flutter valve (fashioned from the fingers of a surgical glove), under­ water seal, or commercially available one-way (Heimlich) valve so the air can continue to escape, but blocking its influx. A three-sided occlusive dressing is used to cover the wound of a stable patient with an open chest wound and PTX in the prehospital setting. Similar to the valve mechanisms listed previously, this dressing acts as a one-way (flap) valve but prevents the air from entering the chest cavity. In this case, the external three-sided occlusive dressings allow air to exit the pleural space while preventing air reentry through the wound. A sterile dressing, such as petrolatum-impregnated gauze that extends 6 to 8 cm beyond the wound in all direc­ tions, is used. Only three sides are taped down. Ideally, the patient is instructed to deeply inhale and then perform a Valsalva maneuver or to cough just as the dressing is placed.

ED Treatment Equipment Recommendations for standard instruments for a TT tray are listed in Table 10–3. The most basic needs are a scalpel, a large clamp (Kelly), and the chest tube. Because the contents of these trays vary among hospitals, emergency providers

183

II 

● 

RESPIRATORY PROCEDURES

Figure 10–6  During a resuscitation involving a tension PTX, there may be no time for a chest tube and needle decompression may not be effective rapidly enough. Under these circumstances, the pleural cavity can be vented in seconds. This assumes that the patient is intubated. A, A No. 10 scalpel is used to make a deep incision in the skin and subcutaneous tissue over the fourth or fifth rib in the anterior axillary line. B, A long closed Kelly clamp (or scissors) is inserted over the top of the rib and stabbed into the pleural space. A pop is usually felt. C, Once the instrument is in the pleural space, it is opened wide to create a rent in the parietal pleura. Air should immediately vent. If the patient is intubated, normal cardiorespiratory function can be maintained. A chest tube is then inserted.

A

B

184

C

Figure 10–7  A large-bore needle/catheter combination is used to puncture the parietal pleura and establish the presence of blood or air in the pleural space. The needle can be placed anywhere in the pleural space, but traditionally, the same sites used for tube thoracostomy (TT) are used: the anterior second intercostal space in the midclavicular line or the anterior axillary line in the fourth or fifth interspace is used. The needle is placed to enter over the rib to avoid neurovascular injury. The needle is then withdrawn, leaving the catheter behind to create a simple open PTX. The procedure can be done either with or without the syringe attached to the catheter. This is only a temporary therapeutic maneuver for a tension PTX and a chest tube must also be inserted. (Redrawn from Richards V: Tube thoracostomy. J Fam Pract 6:631, 1978.)

should familiarize themselves with the trays at their facility prior to an emergency. Chest tubes are open-ended clear plastic tubes of various diameters with a series of holes along the distal length. A radiopaque strip that is interrupted by the side ports (holes) runs along the length of the tube. This allows the provider to better visualize the tube on the postprocedure radiograph and to ensure that the side ports are within the pleural cavity. Adult tube sizes vary from 12 to 42 French, with smaller tubes used for a small PTX, and larger (a minimum of 36 Fr) for HTX and empyema. The largest possible tube should be used to drain suspected HTX. For pediatric patients, Nos. 14, 16, 20, and 24 French tubes are adequate. Before insertion, the beveled (extrathoracic) end of the tube is often cut squarely to better fit the commonly available connectors.

PROCEDURE Before any procedure, gown, glove, mask, and goggle precau­ tions must be used. When possible, consent should be obtained

Equipment

for

Tube

Tube thoracostomy

Procedure Sterile drapes 10- to 20-mL syringe and assorted needles (for local anesthesia) Local anesthetic (1%–2% lidocaine) Antiseptic solution No. 10 scalpel Large clamps (Kelly) Needle holder Chest tubes (size appropriate) No. 0 or 1-0 silk or similar suture Forceps Straight (suture) scissors Large, curved (Mayo) scissors Soft arm restraints

Standard insertion site for lateral placement

I.V. 30˚-60˚ angle

Drainage System and Tubing Drainage apparatus with sterile water for water seal Hard plastic serrated connectors Sterile tubing Dressing Petroleum gauze or similar occlusive dressing Gauze or similar pads Adhesive tape—cloth-backed Tincture of benzoin

B

A

Figure 10–8  Standard sites for TT.  A, The second intercostal space, midclavicular line is preferred for needle aspiration or catheter insertion. B, The fourth or fifth intercostal space, midaxillary line, lateral to the pectoralis muscle and breast tissue, is the preferred site for a chest tube, regardless of pathology. Note that placing the tube too far posteriorly will not allow the patient to lie down comfortably.

and the procedure should be conducted in the most sterile conditions that time allows.

Tube Insertion Site The most common location for a chest tube is the mid- to anterior axillary line, usually in the fourth or fifth intercostal space (Fig. 10–8). This approach is cosmetically preferable and better tolerated than placing the tube in the anterior chest wall in the second intercostal space in the midclavicular line.

● 

Oxygen

10 

TABLE 10–3  Recommended Thoracostomy

Figure 10–9  To insert a chest tube, the patient is placed semierect with the ipsilateral arm abducted as far as possible and preferably restrained. Supplemental oxygen and monitoring are recommended.

But a chest tube placed anywhere in the pleural cavity will drain blood, fluid, or air. The fifth intercostal space is approximately at the level of the nipple or the inferior scapular border in most patients, although the breast mass may lead to variance in females. The incision site should be lateral to the edge of the pectoralis major and breast tissue. To avoid penetrating the abdominal cavity, a more superior insertion site should usually be chosen because the external landmarks can be mis­ leading. The diaphragm of a supine patient who is not taking a deep breath is much higher than suspected. Before insertion, the tube should be held beside the chest wall with the tip of the tube at the level of the clavicle to estimate the distance the tube should be advanced from the incision site to the apex of the lung. The level of the insertion site must be sufficient to ensure that the last drainage hole on the tube will be within the pleural space. A clamp may be placed on the tube to mark the maximum length the tube is inserted to prevent the common problem of advancing the tube too far. In markedly obese patients, it is common to fail to advance the tube far enough, thereby not ensuring that the last hole is in the pleural space. There is no evidence in adults that tube location affects the ability to drain fluid collections. As the lung expands and the pleural space becomes smaller, air and fluid that is not loculated will follow the path of least resistance and enter a functioning drainage tube, regardless of the tube’s location.

Patient Preparation Patients should be started on oxygen and placed on continu­ ous pulse oximetry monitoring. When possible, the head of the bed should be elevated 30° to 60° (Fig. 10–9) to lower the diaphragm and decrease the risk of injury to the dia­ phragm, spleen, or liver. The arm on the affected side is placed over the patient’s head and restrained in that position. A semierect position helps lower the diaphragm. The skin should then be cleaned with a standard surgical scrub and draped sterilely.

185

RESPIRATORY PROCEDURES ● 

II 

Anesthesia The procedure can be extremely painful, so stable patients should be given parenteral analgesics or procedural sedation prior to the procedure. Unstable patients or those with severe sleep apnea should be monitored closely and consid­ ered for ketamine or propofol anesthesia rather than highdose narcotic/benzodiazepine analgesia. Generous local anesthesia should also be used—up to 5 mg/kg of locally injected 1% lidocaine with or without epinephrine. A wheal of anesthetic is made in the area of the incision over the rib. While slowly infiltrating with a longer and larger-bore needle (19- or 21-gauge), the needle is directed over the superior aspect of the rib through the muscle, periosteum, and to the parietal pleura along the entire anticipated track of the tube’s passage (Fig. 10–10). The needle may also be used to intermit­ tently aspirate for air or fluid to find the pleural cavity. If air or fluid is not found, the insertion site should be changed. A common problem is inadequate systemic analgesia and local anesthesia. Be prepared to give additional doses of each throughout the procedure. Once the tube is in place, local anesthetic may be admin­ istered through the chest tube into the pleural space to reduce the pain of the tube against the pleura. One approach for stable patients is to administer 10 mL of 0.5% bupivacaine through the chest tube while the patient is lying on the con­ tralateral side.18 After 5 minutes without drainage of the thorax, standard gravity or vacuum drainage is reinitiated. Parenteral analgesic agents should be used as needed to control the pain associated with the initial injury and the procedure.

186

Insertion A common problem during the procedure is that the skin incision is too short to create and maintain an adequate track to insert the thoracostomy tube. The incision should be no less than 3 to 5 cm long, and there is no harm, other than slightly more scarring, in making it longer (Fig. 10–11A).

Pleural fluid Lung Anesthetic

Figure 10–10  Local anesthesia is essential to reducing the pain from the insertion of a chest tube. Both the skin and the pleura should be infiltrated with a generous amount of local anesthetic. A, The anesthetic is first infiltrated over the rib at the site of the incision. B, The needle is then advanced slowly over the top of the rib while intermittently infiltrating and aspirating until the pleura is breeched and air is withdrawn. Anesthetic is then injected liberally (maximum 5 mg/kg) to cover the pleural lining.  (A and B, Redrawn from Hughes WT, Buescher ES: Pediatric Procedures, 2nd ed. Philadelphia, WB Saunders, 1980, p 234.)

Traditionally, the initial skin incision is made over a rib or two lower than the intercostal space that the tube will pass through. The tube is then “tunneled” under the skin up over the next rib and then through the intercostal space. This is done to prevent air leaks, but there is no good evidence to support this. A transverse incision through the skin and the subcutaneous tissues should be made with a No. 10 blade over the rib. A large Kelly clamp is used to push and spread the deeper tissues and bluntly dissect a track over the rib. The intercostal vessels and the nerve are located on the inferior margin of each rib and must be avoided. The bluntly dissected track should pass immediately over the superior surface of the lower rib in the chosen intercostal space (see Fig. 10–11B). Firm resistance will usually be felt when the tough parietal pleura is met. At this point with the clamp closed, firm pres­ sure must be made to penetrate the cavity. This often takes considerable force. Penetrating the pleura is usually the most painful portion of procedure, and extra anesthetic or analgesia may be needed at this point. To prevent penetrating too deeply, hold the clamp midshaft a few centimeters distal to the incision when resting the tip against the pleura before pushing through (see Fig. 10–11D and E). A palpable pop may be felt and a rush of air or fluid may occur when entering the pleural cavity. With only the clamp tips in the pleural cavity, the clamp is spread to make an adequate pleural entry and withdrawn (see Fig. 10–11C). The opening in the parietal pleura should be wide enough to comfortably insert a finger and the tube; however, an extensive pleural opening should be avoided because this opening provides an egress for air. Because the pleura cannot be closed, a gapping hole predisposes to subcutaneous emphysema after the tube is secured (Fig. 10–12). Another common problem occurs at this point, particu­ larly in obese patients: the dissected track and pleural opening are lost when the clamp is withdrawn. To prevent this, a gloved finger should be slid over the clamp and into the pleura prior to withdrawing the clamp (Fig. 10–13A). This is done to further define the tract and to verify that the pleura has been entered and that no solid organs are present. Whenever possible, the finger should always be left in the pleural space so the hole is not lost (see Fig. 10–13B and C) and the tube is passed over, under, or beside the finger into the pleural space. This step allows the clinician to feel the tube passing into the pleural cavity and avoids subcutaneous dissection with the tube. The tube can be passed alone or held in a large curved clamp, with the tube tip protruding beyond the tip of the clamp (Figs. 10–14 and 10–15). The tube should pass with little resistance; if it is hard to pass it may not be in the pleural cavity and may be passing subcutaneously (Fig. 10–16). The tube should be directed posteriorly, medially, and superiorly until the last hole of the tube is clearly intrathoracic, the marker clamp that was previously attached touches the chest wall, or resistance is felt. Ensure that all the holes in the tube are within the pleural space. Rotate the tube 360° to reduce the likelihood of kinking. The tube should be attached to the previously assembled water seal or suction before the clamp is released. Asking the patient to cough and thereafter observing bubbles in the water seal chamber is a good way to check system patency.

Confirmation of Tube Placement Multiple ways are available to confirm the location of the tube. The tube must be located in the pleural cavity such that adequate drainage can take place, without any undue bending

10  ● 

Tube thoracostomy

A

Parietal pleura

Egress of air/fluid Vessels and nerves Adequate skin incision here

187 Note: Too large of a hole in the pleura predisposes to subcutaneous air leaks.

B

D

C

E

Figure 10–11  A, The incision should be made with a No. 10 scalpel through the skin and subcutaneous tissue over the fourth or fifth rib in the anterior axillary line. B, A long Kelly clamp is used to bluntly dissect over the top of the rib and then pushed into the pleural space. A “pop” is usually felt when the pleura is penetrated. C, Bluntly dissect to the pleural lining by pushing the closed points of the clamp forward, then spreading the tips and pulling back slightly with the points spread. Remember to dissect immediately over the rib on the superior aspect. A rush of air or fluid signifies penetration into the pleural space. Considerable force may be required when pushing through the pleura. By holding the clamp in the midsection of the curve, (D) the fingers will stop at the chest wall and prevent deep penetration of the tips of the clamp. To protect against inadvertent puncture of the lung when the clamp penetrates the pleural cavity, place an index finger on the distal portion of the clamp (E) or use the thumb and fingers of the opposite hand on the clamp to serve as a stop once the desired depth is reached.  (B, From Millikan JS, Moore EE, Steiner E: Complications of tube thoracostomy for acute trauma. Am J Surg 140:739, 1980; C, from Bricker DL: Safe, effective tube thoracostomy. ER Reports 2:49, 1981.)

RESPIRATORY PROCEDURES ● 

II 

or kinking of the tube. Initially, if possible, a finger can be slid along the tube to verify that it enters the pleural cavity. Condensation on the inside of the tube and audible air move­ ment with respirations, the free flow of blood or fluid, and the ability of the operator to rotate the tube freely after inser­ tion are also indicators that the tube is in the pleural space. The ability to rotate the tube freely after insertion also sug­ gests that the tube is not kinked, which can happen during tube placement. The definitive assessment of tube placement is the chest radiograph. If the tube and most proximal hole are not completely in the pleural space, the tube should be advanced if the field has remained sterile. If the tube is kinked or dysfunctional or the sterile field has been lost and advance­ ment is required, a new tube should be placed in a sterile fashion through the same track. If the tube has been advanced too far, it may simply be withdrawn to the correct depth.

Intercostal muscles

Pleura

Securing the Tube Once the tube position is verified, it can be secured. It is best to await radiographic confirmation before extensive efforts to secure the tube are undertaken, because adjustments may be

Kelly forceps

A

188

B Figure 10–12  Right-sided subcutaneous emphysema after chest tube placement, secondary to making too large a hole in the pleura, with a subsequent air leak. It is usually benign and self-limited, but with PPV, it can be problematic. Because there is no way to close the pleura, making just the right-sized hole is the key to success.

Figure 10–13  A and B, After puncturing the pleural lining and spreading with the clamp, slide a gloved finger over the clamp to ensure that the pleural space has been reached and that no solid masses are present. Then, withdraw the clamp and use the finger as a guide for the chest tube to ensure entry into the pleural cavity. C, Once the pleural opening is found, do not remove the finger because the hole may be easily lost, especially in an obese patient.  (A and B, From Millikan JS, Moore EE, Steiner E: Complications of tube thoracostomy for acute trauma. Am J Surg 140:739, 1980.)

C

10 

Securely tied initial stay suture

● 

Tip of clamp grasps the chest tube

Left long

Tube thoracostomy

Long ends wrapped around tube and tightly tied

Left long

A

B

Figure 10–16  A, To secure the tube, first close the skin incision with a “stay” suture near the tube. B, Tie the knot securely and leave the suture ends long for wrapping around and tying the tube. Wrap the suture tightly at least twice around the tube, enough to indent the tube slightly, and tie securely.

A Initial stay suture

United horizontal matress suture Long ends wrapped loosely around tube and tied in bow Surgeon’s knot

Note: Horizontal mattress suture encircles tube

A B Figure 10–14  A and B, To reduce the risk of damage to the lung, the tube is grasped with the curved clamp, with the tube tip protruding from the jaws.

Figure 10–15  Subcutaneous placement of a chest tube (arrows) can occur because the tube can dissect through tissue planes with relative ease. If this tube had been directed posteriorly, the radiograph would erroneously “confirm” intrapleural placement despite the tube being subcutaneous throughout its entire course.

B

Figure 10–17  Another method to close the wound and secure the tube is with a horizontal mattress suture combined with a stay suture. A, A horizontal mattress suture is placed on either side (above and below) of the tube and is held only with a surgeon’s knot. B, The loose ends are also wrapped around the tube and are tied loosely in a bow to identify the suture. This suture will be untied and used to close the skin incision after tube removal.

required. There are numerous methods to secure a tube. The usual method is to sew the tube to the skin with large 0 or 1-0 silk or nylon sutures. Nylon sutures are acceptable but must be tied tightly or they may slip on the surface of the chest tube. One common method is to use a “stay” suture in which the same suture that closes the skin incision is used to hold the tube (see Fig. 10–16). After this suture is used to close the skin incision at the site of tube insertion, the ends are left long and then wrapped tightly and repeatedly around the chest tube and tied securely. The sutures must be tied tightly enough to indent the chest tube slightly to avoid slippage. Longer skin incisions may require additional simple sutures to close completely. Some clinicians use a suture technique that can both help close the skin around the tube and subsequently completely close the incision after the chest tube is removed. To do this, a horizontal mattress suture is placed approximately 1 cm across the incision on either side of the tube, essentially encircling it (Fig. 10–17). This is secured with a simple knot that can be easily untied so that it can be opened and retied to close the incision after the tube is removed. After suturing the tube in place, an occlusive dressing of petrolatum-impregnated gauze should be applied where the tube enters into the skin. This may help to reduce air leaks.

189

RESPIRATORY PROCEDURES ● 

Y-cut

II 

Gauze sponge Vaseline gauze next to wound Y-cut gauze at 90˚ angle Figure 10–18  To dress the wound and reduce the risk of air leaks, an occlusive dressing should be applied. First wrap the base of the tube at the skin incision with a petroleum-impregnated dressing. A twolayer dressing of gauze sponges with a Y-shaped cut centered at the tube is shown. Place the second layer at a 90° angle to the first.

To skin on one side of tube Wide tape

To wrap around chest tube To skin on other side of tube

A

190

Half length of tape torn into 3 pieces

Open to atmosphere or attach to suction

Heimlich chest drain valve

To patient

Flow direction One-way air flow Figure 10-21  A one-way Heimlich valve alone is often sufficient to treat a pneumothorax, but it cannot be used to treat a hemothorax.

Middle strip of torn tape Torn tape

B

Figure 10–20  The tube can be further secured with an additional anchor system further down on the tube. Wrap a 20- to 25-cm piece of tape or elastic, adhesive dressing around the tube and seal at least 3 cm of the tape together on the side of the tube nearest the chest wall. Spread the remaining tape against the dry skin of the chest wall and secure with additional tape.

Tape secures anchoring tape

Figure 10–19  One method to further secure the tube is to use wide, split cloth tape. A, The distal half of a 15- to 20-cm-long wide piece of tape is longitudinally split into three pieces. The two outside pieces are placed on the skin on either side of the tube, and the center strip is wrapped around the chest tube itself. B, This process may be repeated with a similar piece of tape placed at a 90° angle. The tape is securely anchored to the skin (benzoin is optional, but the skin must be clean and dry), and the torn tape is wrapped around the tube. Each anchoring piece is covered by another piece of tape.

The skin should then be covered with two or more gauze pads with a Y-shaped cut from the middle of one side to the center (Fig. 10–18). This dressing should be secured with wide (8to 9-cm) cloth or elastic adhesive tape with or without benzoin. Use approximately 10 to 12 cm of tape split into three pieces extending halfway along its length. The two outside pieces are placed on the skin on either side of the tube site, and the center section is wrapped tightly around the tube (Fig. 10–19). This is repeated with a second piece of tape placed at 180° to the first. Also securely tape the tube connections. The tube can be further secured by using tape to create a loop or stalk by wrapping it around the tube and then pressing the tape together for 1 to 2 cm before applying the tape to the chest wall (Fig. 10–20).

Drainage and Suction Systems A basic understanding of the functions of chest tube drainage systems is necessary to prevent life-threatening complications associated with their use. There are two essential components to all drainage systems: a one-way valve to allow air or fluid to drain out of the pleural space without allowing air back into the pleural space and a suction mechanism to increase the rate of drainage. The simplest drainage device is just a one-way valve without suction. This can be accomplished by either an underwater seal or with a flutter (Heimlich) valve attached to the end of the chest tube (Fig. 10–21). Normal respiration and coughing are often sufficient to create the pressure needed to remove the excess air from the pleural space, and the lung will then expand. The Heimlich valve does not require suction and has been used for outpatient therapy. With a one-bottle underwater seal system, the intrapleu­ ral fluid or air exits under a small amount of water and collects into the single reservoir mixing with the water (Fig. 10–22A). The water above the tube acts as a seal because it is too heavy to be drawn back into the chest. It is important to remember that the intrathoracic pressure must be greater than the water pressure at the distal immersed tube to allow air or fluid to drain into the bottle. This pressure is determined by the height of the water above the exit port of the tubing. When the height is too great (the tube is too deep in the water), even coughing may not raise the intrapleural pressure sufficiently to drain the chest. To prevent inspiration from generating enough negative pressure to pull the collection bottle con­ tents into the chest cavity, the collecting bottle must be below the patient, usually on the floor. Suction is used initially to treat patients with a PTX or an HTX, but should be replaced by a water seal once the

A

Water seal bottle

To patient Open to atmosphere

Water level

B

Trap bottle To chest tube

Trap chambers

Water seal bottle To suction source

Air inlet Float valve

Suction control chamber

Water seal chamber

C Figure 10–22  A, A single-bottle (water-seal) collection device. B, A two-bottle system. The trap reservoir proximal to the water seal keeps the accumulating drainage from affecting the water-seal pressure. C, This has now been replaced by a disposable system that mimics the two-bottle system (Thora Klex system, Davol, Inc.)

drainage and expansion are satisfactory and there are no per­ sistent air leaks. The suction device should have high suction flow (≤20 L/min) and be able to keep the suction constant. A wall suction of 10 to 20 cm H2O is normally used, but remem­ ber that the amount of suction in the chest tube is dependent on the depth of water in the water seal reservoir, not on the suction from the wall valve. When the negative pressure from the suction source exceeds the depth of the water in the chamber, air enters from the top of the third tube, causing continuous bubbling (see Fig. 10–22B). This prevents a further pressure increase in the chest tube. The wall suction dial can be turned down until only occasional bubbling can be detected. Vigor­ ous bubbling does not equate with more suction. Bottle combinations are rarely used now and many types of commercial, enclosed systems are available that essentially mimic the bottle system. Current commercial drainage

Prophylactic Antibiotics The use of prophylactic antibiotics after chest tube placement in the ED is common, but controversial, and no specific stan­ dards exist. Multicenter trials have demonstrated no benefit.20,21 Routine antibiotics have no proven value in the reduction in the incidence of chest tube–associated empyema or pneumo­ nia. If a chest tube is placed under less than ideal sterile con­ ditions or if there is significant lung damage, prophylactic antibiotics to cover S. aureus may be considered, but there is a possibility of selecting out resistant organisms.

Tube thoracostomy

Water level

● 

Open to atmosphere

systems combine a two-bottle method that can be connected to suction, but with “air leak chambers” (see Fig. 10–22C). Bubbling in this chamber indicates the presence of an air leak, either in the drainage system itself (usually a loose tube con­ nection) or from a large hole in the lung parenchyma. If an air leak is found, first check the system and the tube and con­ nectors. If that does not correct the leak, check that all holes of the chest tube are within the thorax. If the air leak persists, it may be from the patient; this is usually seen only with expiration or with coughing. A continuous air leak or a leak seen during inspiration indicates a larger and possibly more significant lung injury.19 Surgical intervention is indicated if an air leak persists for longer than 72 hours or the lung is not completely reexpanded. When the drainage system is functioning properly, the height of the fluid level in the drainage tube fluctuates with inspiration and expiration. The absence of respiratory fluctua­ tion or a decrease in the drainage may indicate that the system is blocked or that the lung is fully expanded. If the tube is blocked, the chest tube or collecting tubing or both can be changed, “stripped” to dislodge clots. Although replacing the tube is a complicated process, the routine use of stripping should be used sparingly because of the potentially high pres­ sures generated. If the blockage is within the thorax, the tube can be cleared by forcing air or fluid back into the chest. The tube must be clamped distally and then compressed and stripped to force the contents proximally. Stripping is the opposite maneuver, in that the tube is clamped proximally and progressively compressed distally followed by a release to allow the tube to spring open. The sudden increase in nega­ tive pressure may extract clots and fluid from a more proximal location. The drainage reservoir must remain below the level of the chest to prevent the fluid in the collection system from reentering the chest. The reservoir is usually placed on the floor or hung from the edge of the bed. Simple respirations do not generate enough negative intrathoracic pressure to pull the water in the reservoir up to the height of the chest if the reservoir is kept on the floor. The length of the tubing must be sufficient to keep the reservoir below the level of the patient, but not long enough to cause it to form dependent loops of fluid or kinks. Dependent loops collect fluid and create an additional water seal that, if large enough, requires greater intrapleural pressure to drain. If these pressures become high enough (15–25 cm H2O), a tension PTX may result. Occlusive clamping of chest tubes should be performed only with close monitoring because it can lead to a tension PTX in rare cases. Patients with chest tubes in place are best transported with a Heimlich valve or water seal only, not with a clamped tube. Clamping the chest tube as a trial maneuver before removal of the tube is discouraged.

10 

To patient

191

RESPIRATORY PROCEDURES ● 

II 

192

Tube Removal Chest tubes are rarely removed by emergency clinicians. The usual indications for chest tube removal are after a chest radiograph demonstrates complete resolution of the PTX and there is no evidence of an ongoing air leak. Suction should be discontinued prior to removal, and the patient should be placed on a water seal. Clamping the tube is more controver­ sial. Finally, most experts recommend that a repeat radio­ graph be completed 5 to 12 hours after suction is discontinued before pulling the tube. For empyema, the removal depends on the clinical and radiographic resolution of the infection. To remove the chest tube, place the patient sitting upright at about 45° and remove the dressings. Prepare and drape the insertion site and follow sterile technique. If a pursestring suture was placed at the time of the insertion, only sterile scissors are needed to cut the suture. If there is no pursestring suture, suturing equipment will be needed to close the wound once the tube is removed. Additional equipment should be available to reinsert a chest tube if the lung collapses. A pet­ rolatum- or antibiotic-impregnated gauze dressing should be prepared for covering the wound. The pursestring suture that was previously placed should be loosened and readied for closing the wound. Then the skin loop of the suture holding the tube to the skin should be cut and removed from the skin. The tube should be clamped to prevent leakage of body fluids and then disconnected from the connecting tubing. The patient should inhale fully and perform a mild Valsalva maneuver. The tube is pulled out in one swift motion while the patient holds the breath. The pursestring suture is quickly tied and then covered with the occlusive dressing. The patient should be observed for 2 to 6 hours, with a chest radiograph obtained before discharge. Any increase in symptoms requires prompt reevaluation. After 48 hours, the dressing may be removed. Sutures may be removed in 7 to 10 days.

OTHER TECHNIQUES Minicatheter Insertion A less invasive alternative to traditional TT for patients with a simple PTX is treatment with a minicatheter. This tech­ nique and observation are widely used in Europe, but less so in the United States. Healthy patients with iatrogenic PTX (e.g., after central line attempts, intravenous drug injection), victims of minor nonpenetrating chest trauma, and patients with spontaneous primary PTXs are potential candidates for catheter aspiration of the PTX. Patients with underlying lung pathology, such as pneumonia, congestive heart failure, asthma, or emphysema, are generally not candidates for mini­ catheter use. Advantages of this technique include the ease of catheter insertion, decreased patient discomfort, less scarring, and decreased cost. The drawbacks include catheter kinking and the inability to perform video-assisted thoracoscopy through the site. After successful reexpansion of the lung, selected patients may be treated as outpatients with a Heimlich valve. Many protocols are available for using catheter aspiration as the first step in treating simple PTXs. In general, patients with successful aspiration are observed in the ED for 4 to 6 hours after the catheter is inserted, and if a repeat radiograph shows no reaccumulation of air, the catheter is removed. After 2 more hours, another chest radiograph is obtained, and the

patient is released if there is no recurrent PTX. Patients with continued residual PTX often receive a conventional TT. Minicatheters should not be used for patients on a ventilator, with continuing air leaks, or with an HTX. A common problem with catheters is that they occasionally clog and become nonfunctional in 24 to 48 hours because of the small lumen. A 14-gauge intravenous catheter or an 8.5-French trauma catheter can be used, but these catheters have only single distal holes, which can easily become obstructed or adhere. It is preferable to use catheters designed specifically for aspirat­ ing a PTX. These are made of flexible, thrombosis-resistant, radiopaque material with multiple distal side ports to reduce the risk of occlusion. A commercially available pigtail catheter system is ideal for this procedure.

Guidewire Technique for Catheter Aspiration The catheters are placed using a standard “over-the-wire” (Seldinger) technique. The most common insertion site is the second intercostal space in the midclavicular line, but either of the standard locations (the mid- to anterior axillary line, usually in the fourth or fifth intercostal space, or the midcla­ vicular second intercostal site) can be used. The patient is placed in a semi-upright position and the skin is cleaned with an antiseptic solution and the area draped. Lidocaine is infil­ trated locally for anesthesia. The guide needle is then advanced in a straight line at a 60° angle cephalad over the top of the rib (Figs. 10–23 to 10–25). Unless a straight track is created, it will be difficult to advance the floppy catheter; a tunneling approach cannot be used. When the pleural space is identified by intermittent aspiration, the advancement of the needle is halted. A guidewire is fed through the needle into the pleural space. Then the needle is removed while stabilizing the guide­ wire to keep it in the pleural space. A small incision is made in the skin with a No. 11 blade at the base of the wire to allow passage of the catheter through the skin. Some systems use a dilator over the wire to open the path through the soft tissues. The minicatheter is then threaded over the guidewire into the pleural space and the wire and dilator are removed, leaving the catheter in the pleural space. A twisting motion may be needed to advance the catheter through the subcutaneous tissues. The catheter should be secured to the skin with a suture and dressed. The catheter may be removed after a period of observation or the suction may be maintained for a few days. If used for a few days, the catheter will become clogged with mucus or blood, which may be cleared by inject­ ing sterile saline through the device. To aspirate the PTX, a three-way stopcock is attached to the catheter and the air is slowly aspirated with a 60-mL syringe until resistance is felt. Gentle wall suction can also be used, because a number of aspirations may be required until all air exits. A chest radiograph is taken to determine whether the lung is fully expanded. If residual PTX is present, further aspirations can be attempted. If air cannot be aspirated, the catheter may be kinked or blocked with soft tissue. To relieve the blockage, place the patient in the full upright position and have him or her cough or take a deep breath. Alternately, the catheter can be twisted or rotated gently.

TT IN PEDIATRIC PATIENTS PTXs can occur in the neonatal population. They are often associated with resuscitative measures (such as mechanical

10  ● 

Tube thoracostomy

A

B

C

D

193

E

F

Figure 10–23  Aspiration of a pneumothorax (caused by subclavian vein catheterization) with an Arrow 14-French Percutaneous Cavity Drainage Catheterization Kit. This 23-cm pigtail multihole catheter is ideal for such purposes. Air can be aspirated from the catheter with a syringe or the catheter can be attached to suction or a Heimlich valve. This catheter is not used for patients on a ventilator, those with continuing air leaks, or those with a hemothorax. It is ideal for stable patients who have a primary PTX or a collapse that can be expected to be stable if the lung is reexpanded (such as intravenous drug use–induced, minor blunt trauma, secondary to central venous catheter insertion). A, Seldinger-type catheter kit demonstrates the pigtail catheter and all necessary equipment, including local anesthesia, introducing needle and syringe, scalpel, guidewire, and dilator. B, After generous local anesthesia, the introducing syringe is advanced in a straight line over the top of the fifth rib until air is aspirated. Unless a straight track is created, it will be difficult to advance the floppy catheter, and a tunneling approach cannot be used. C, The guidewire is advanced into the pleural space and the introducing needle is removed. D, Puncture the skin at the site of wire insertion with a scalpel. E, A dilator is advanced over the wire to create a track for the catheter. F, The pigtail catheter is advanced over the wire through the dilated tract, assuming its pigtail configuration when it is in the pleural space. A twisting motion may be needed to advance the catheter through the subcutaneous tissues. The catheter is advanced to the hilt and secured to suction. This catheter may be removed after a period of observation or the suction may be maintained for a few days. If used for a few days, the catheter will become clogged with mucus or blood, which may be cleared by injecting sterile saline through the device.

ventilation) for meconium aspiration or prematurity. For the rest of the pediatric population, trauma is the most common cause. Approximately one third of children with thoracic trauma will develop a PTX. As with adults, the physical exam­ ination of newborns and infants with PTX can be highly variable, necessitating the use of a chest radiograph for diag­ nosis. Ideally both anteroposterior and cross-table lateral pro­ jections are used because small PTXs may be seen only on the lateral view.

In general, TT is the treatment of choice once a symp­ tomatic PTX is detected in infants. When signs of tension PTX are present, immediate aspiration with a plastic catheter over-the-needle device is recommended. Small PTXs ( 0.12 sec). The rhythms of these dysrhythmias can be regular or irregular. Examples of narrow-complex SVTs are sinus tachycardia; atrial fibrillation (AF); atrial flutter; AV nodal re-entry; atrial tachycardia, both ectopic and re-entrant; multifocal atrial tachycardia (MAT); junctional tachycardia; and accessory pathway–mediated tachycardia. The term widecomplex tachycardia describes rhythms such as VT, SVT with aberrancy, or a pre-excitation tachycardia facilitated by an accessory pathway between the atria and the ventricles. Tachycardias can be benign or can have significant physi­ cal effects on the patient. When the HR is 60 beats/min, approximately one cardiac cycle of contraction (systole) and relaxation (diastole) occurs per second. The excitation for the cardiac contraction typically originates in the SA node, the intrinsic “pacemaker” of the heart. The pacemaker impulse traverses across and depolarizes the atria causing atrial con­ traction or systole. Subsequently, this depolarization reaches the AV node. Upon initiating depo­larization of the AV node, the conduction velocity of this depolarizing impulse tran­ siently decreases (i.e., it undergoes “decremental conduc­ tion”), so that the ventricles can fill with blood from the antecedent atrial contraction. (Remember: The duration of diastole is roughly twice the duration of systole to allow for adequate ventricular filling.) The AV node also serves as a gate/ selective block to prevent an excessive number of depolarizing impulses from reaching the ventricles when the atrial rate is accelerated. Immediately thereafter, this depolarizing wave acceler­ ates as it travels down the bundle of His to the Purkinje fibers, causing ventricular depolarization and contraction systole. Subsequently, the ventricles begin to relax (i.e., enter diastole and begin to fill with blood prior to the next depolarization).

197

CARDIAC PROCEDURES ● 

III 

This describes the events of one cardiac cycle or heartbeat. The electrochemical voltage changes of these events are depicted on the electrocardiogram in the usual sequential PQRST (the P-wave indicates SA nodal depolarization, the P-R interval denotes atrial depolarization followed by AV nodal activation, and the QRS complex summarizes electrical activity during ventricular depolarization). The discharge rate of the SA node is usually modulated by a balance of input from the sympathetic and parasympa­ thetic nerves (i.e., the autonomic nervous system). The sym­ pathetic input to the heart is provided via the adrenergic nerves, which innervate the atria and ventricles, and by circu­ lating hormones such as epinephrine/norepinephrine, which are released from the adrenal gland and cause the HR to increase. The parasympathetic input to the heart is provided by the vagus nerve (cranial nerve [CN] X) fibers. These nerve fibers innervate the SA and AV nodes. Vagal output to the SA node causes slowing of the HR by decreasing the depolariza­ tion rate of the “intrinsic pacemaker” whereas vagal output to the AV node enhances nodal blockade of atrial depolarization impulses to the ventricles. Under normal physiologic circum­ stances, the HR is modulated to meet the metabolic needs of the body’s peripheral circulation. Changes in the AV electro­ chemical events (i.e., rates and rhythms) are manifested as

TABLE 11–1  Diagnostic and Therapeutic Approaches to Supraventricular Tachycardias Vagal Maneuvers Carotid sinus massage Pressure on the carotid sinus Valsalva technique Forced expiration of air against a closed glottis Apneic facial exposure to cold water (“cold water diving reflex”) Immersion of the face into cold water Oculocardiac reflex The trigeminovagal reflex initiated by pressure on the eyeball Pharmacologic Agents Adenosine Amiodarone Verapamil Diltiazem β-Blockers including esmolol Digoxin Procainamide Ibutilide Cardioversion Administering a synchronized shock

ECG AND MEMBRANE POTENTIAL OF VENTRICULAR CELLS Normal electrocardiogram PQRST: Summary of time dependent ionic fluxes/action potentials A in the myocytes

198

R

Mechanical force generated by myocyte contraction/shortening T

1mV P Overshoot +30 Myocardial cell 10 mM Na+ 140 135 " K+ 4 10+ mM Ca2+ 2

0

Extracellular fluid

mV

Q

ECG S

Phase 1: Transient efflux of K+ Phase 2: Influx of Ca2+ and Na+

−70 (Internal − external potential) = −90

Phase 0: Fast Na+ - influx

Phase 3: Efflux of K+ > influx of Ca2+ and Na+

Tension Threshold Phase 4: Na+- K+- pump Contraction

0 Absolute refractory period Relative Fast Na+- channels refractory are closed period

300 ms

Steep phase 0 means rapid depolarization

APhase

0 Upstroke - Increased sodium conductance into myocytes Phase 1 Early repolarization - Increased potassium conductance out of myocytes Phase 2 Plateau - Calcium influx into the myocytes/potassium efflux increasing Phase 3 Repolarization - Sodium influx decreased, calcium influx decreased/potassium efflux still present Phase 4 Steady state - Sodium, potassium, calcium conductance returns to resting membrane potential Figure 11–1  Electrocardiographic and membrane potential of ventricular cells.

2

3

Normal

1

2

3

Reentry

Figure 11–2  Cardiac conduction in supraventricular tachycardia (SVT).  Top, Normal depolarization down path 1 and 2 that will “extinguish” or “cancel out” at point 3 normal depolarization/repolarization and conductance. Bottom, Abnormal. 1: Normal conduction; 2: delayed/ slowed conduction with unidirectional block; 3: normal conduction pathway.

Techniques for supraventricular tachycardias

1

● 

initial point of entry into this pathway. This allows the depo­ larization wavefront to restimulate the myocytes and initiate another propagated depolarization through the same tract (Fig. 11–2). If this condition is allowed to persist and these impulses stimulate the atria effectively and traverse the AV node, an SVT may develop as a result of re-entry. Suppression of this dysrhythmia can occur by terminating the conditions favoring re-entry, and the hemodynamic consequences may be attenuated by enhancing AV nodal blockade of the ventri­ cles (e.g., through vagal stimulation, medication), thus slowing the ventricular response to this condition. Termination of the re-entry can be accomplished using either pharmacologic modification of the myocytes, rendering them refractory to depolarization impulses for a longer period of time in the stable patient, or synchronized cardioversion to uniformly depolarize the myocytes and terminate the conditions favor­ ing the SVT. Another situation to consider in the development and propagation of SVTs is the presence of pre-excitation or an accessory pathway between the atria and the ventricles. Arrhythmias secondary to these etiologies can be managed by the use of appropriate pharmacologic agents to either sup­ press the conduction through the accessory pathway or appro­ priately block the AV nodal transmission without enhancing conduction through the accessory pathway.

11 

changes in the electrocardiographic (ECG) intervals and waveforms. As noted earlier, supraventricular tachycardic rhythms can be either sinus (i.e., originating in the SA node; sinus tachycardia) or ectopic (i.e., originating in the atrial myocytes above the ventricles). The rate of SA node discharge often varies as a result of various physiologic and pharmacologic stimuli, including fever, hypovolemia, shock, anemia, hypoxia, pain, use of cocaine, and amphetamines. These conditions often require or precipitate an increased blood flow/cardiac output (CO) to the peripheral tissues. This increase in periph­ eral blood flow or CO is accomplished by an increase in the HR (Remember: CO = HR × SV [stroke volume]). These are usually normal, benign physiologic responses to various stimuli or triggers. Direct treatment of these rhythms is usually not necessary; however, determining and treating the cause of the sinus tachycardia usually eliminates the fast HR. However, when single or multiple ectopic, spontaneously dis­ charging foci develop in the atria or upper portions of the AV node, they can begin to “take over” or “override” the normal pacemaker activity in the heart (i.e., the SA node) and produce a rapid HR exceeding 100 beats/min. These foci may develop owing to an increased irritability/automaticity of the atrial myocytes secondary to electrolyte abnormalities, hypoxia, pharmacologic agents, or atrial stretch due to volumetric overload. If these foci are not treated/suppressed and the atrial depolarization rate proceeds to accelerate to rates greater than 150 beats/min (meaning that the heart is beating in excess of 2 beats/sec) with the impulses getting through the AV node to the ventricles, the time for diastolic filling of the ventricles will be compromised, causing a precipitous drop in SV. This will ultimately cause a drop in CO regardless of the increase in HR. Furthermore, as CO begins to drop, the mean arterial blood pressure (MABP) will drop, causing hypoperfu­ sion of the brain and other peripheral tissues (Remember: MABP is the product of CO times total peripheral resistance [TPR]: MABP = CO × TRP). Treatment of this tachycardia can be achieved by pharmacologically suppressing the auto­ maticity of the myocytes with medications (e.g., calcium channel blockers or β-blockers) and subsequently treating the underlying cause(s)—the hypoxia, electrolytes, and the like. Decreasing the hemodynamic consequences of this arrhyth­ mia requires increasing the “blocking” of these impulses from reaching the ventricles via the AV node. This can be done by enhancing vagal input to the AV node or by pharmacologic enhancement of AV blockade. Multiple rapid depolarizations of the atria, which are conducted to the ventricles, can ulti­ mately have a bimodal type of response; a modest increase in HR will cause an increase in CO whereas a massive increase in atrial rate with a concomitant increase in ventricular rate will cause a drop in CO. This can lead to an unstable patient with signs and symptoms such as confusion, altered mental status, or persistent chest pain. When the patient becomes unstable, immediate treatment is indicated. In addition to areas of increased automaticity that can precipitate SVTs, a condition described as re-entry can also cause an SVT. Re-entry describes a condition in which a depolarization impulse is being propagated down a pathway in which some of the myocytes are still in the effective refrac­ tory period and a “unidirectional block” is present, preventing the impulse from traveling normally down this pathway. However, as the impulse travels around the area of the “uni­ directional block,” the tissue allows the depolarization front to travel in the opposite (antidromic) direction, back to the

199

CARDIAC PROCEDURES

Glossopharyngeal nerve

III 

● 

Superior cervical ganglion Vagus nerve Internal carotid artery Contact point for digital massage/stimulation

External carotid artery

Carotid sinus nerve

Carotid body Carotid sinus

Carotid sinus

Common carotid artery

Figure 11–3  Carotid sinus.

200

To complete this discussion, we must also consider that there may be the possibility of an interventricular conduction delay being present prior to the development of an SVT. If this is the case, the SVT may appear as a wide-complex tachy­ cardia and can be confused with other dysrhythmias. However, an even more dangerous situation can occur if a wide-complex tachycardia of ventricular origin (VT) is present and is mis­ diagnosed as an SVT with aberrancy. As a result, the patient could be treated inappropriately, with the intervention causing suppression of ventricular activity and ultimately cardiac arrest. VT with a pulse is considered an unstable rhythm that often requires synchronized cardioversion and is dis­ cussed in more detail in Chapter 12, Defibrillation and Cardioversion. The clinician must have a means of slowing down and sorting out these physiologic events so that an appropriate diagnosis and treatment/intervention decisions can be made. With the application of vagal maneuvers, in some cases, the activity of the atria and ventricles may be isolated enough to facilitate a correct diagnosis. An understanding of the under­ lying pathophysiology will allow for appropriate treatment.

VAGAL MANEUVERS Anatomy and Physiology The physiologic effects of pressure on the carotid sinus have been known for centuries. They were first described in the medical literature in 1799 when Parry wrote a treatise entitled “An inquiry into symptoms and causes of syncope anginosa, commonly called angina pectoris.”1 He noted that pressure on the bifurcation of the carotid artery produced dizziness and slowing of the heart. The term carotid is derived from the Greek karos, meaning heavy sleep.

Figure 11–4  Stretch receptors of the carotid sinus.

The bifurcation of the common carotid artery possesses an abundant supply of sensory nerve endings located within the adventitia of the vessel wall (Figs. 11–3 and 11–4). These nerves have a characteristic spiral configuration, continually intertwining along their course and eventually uniting to form the carotid sinus nerve. The afferent impulses travel from the carotid sinus via Herring’s nerve or carotid sinus nerve to the glossopharyngeal nerve (CN IX) and then to the vasomotor center in the medullary area (nucleus tractus solitarius) of the brainstem (Fig. 11–5). The vasomotor center is composed of three distinct areas, each with a distinctive function. The vasomotor center is located bilaterally in the reticular sub­ stance of the medulla and in the lower third of the pons. The center transmits efferent impulses downward through the spinal cord and the vagus nerve. The efferent impulses, which originate in the medial portion of the vasomotor center, travel along the vagus nerve (CN X) to the sinus node and the AV node of the heart. The vasomotor center’s medial portion lies in immediate apposition to the dorsal motor nucleus of the vagus nerve (CN X). These medial portion vasomotor center impulses decrease HRs. Efferent impulses originating in the lateral areas of the vasomotor center travel along the sympa­ thetic chain to the heart and to the peripheral vasculature. These sympathetic impulses control either vasoconstriction or vasodilatation of the vascular system. A balance between the vasoconstriction and the vasodilatation maintains proper vasomotor tone.2,3 The afferent nerve endings in the carotid sinus are sensi­ tive to MABP and to the rate of change of pressure. Research

Medulla

Techniques for supraventricular tachycardias

NTS

Carotid

Vagus

+ Vagus Aortic arch

+ –

Vagal nuclei

RVLM

+

Heart

–

+ Vein

SNS

CVLM

● 

Figure 11–5  Schematic depicts the arterial baroreceptor reflex.

11 

Glossopharyngeal

+

+

Arteries Arterioles

NTS: Nucleus of Tractus Solitarius RVLM: Rostral Ventrolateral Medulla CVLM: Caudal Ventrolateral Medulla Vagal Nuclei: Dorsal motor nuclei, Nucleus Ambiguous

201

indicates pulsatile stimuli are more effective than sustained pressures in evoking a response. Elevated blood pressure stretches the baroreceptors, leading to increased firing of the afferent nerve endings.2 As for low blood pressure states, the carotid sinus baroreceptors are exquisitely sensitive to low blood pressure. Hypotension causes a drop in afferent firing.2 The parasympathetic and the sympathetic nervous systems play independent but coordinated roles in the carotid sinus reflex. Increased firing of the carotid sinus results in reflex stimulation of vagal activity and reflex inhibition of sympathetic output. The parasympathetic effect is almost immediate; it occurs within the first second and causes a drop in HR. The sympathetic effect, which causes a drop in blood pressure through vasodilatation, becomes manifest only after several seconds.4 The blood pressure changes may not take full effect until a minute has elapsed.5 The changes in blood pressure and HR are independent phenomena. Epinephrine blocks the reduction in blood pressure, whereas a fall in HR is blocked by the administration of atropine. A cerebral effect, characterized by a loss of consciousness, was once thought to be due to stimulation of the carotid sinus. However, it is seen only when sufficient pressure is exerted to occlude the more distal temporal artery pulsation and when contralateral carotid disease is present. This cerebral effect is now believed to be a result of decreased bilateral cortical perfusion. The parasympathetic branch of the carotid sinus reflex supplies the sinus node and the AV node. The effect of the parasympathetic stimulation is to slow the HR. The SA pace­

TABLE 11–2  Potential Observations with Vagal Maneuvers in the Management of Tachydysrhythmias 1.  Vagal maneuvers may slow the atrial rate in VT or complete heart block and may therefore demonstrate previously hidden P-waves or obvious (AV) dissociation. 2.  Abrupt changes in the heart rate without conversion are a result of increasing AV block. 3.  Gradual slowing of the ventricular rate suggests the presence of a sinus rhythm. Only rarely do vagal maneuvers decrease AV conduction in the presence of a sinus mechanism. 4.  The dysrhythmias most likely to convert to sinus rhythm are PAT and paroxysmal nodal tachycardia. 5.  Dysrhythmias that are associated with AV conduction defects (PAT with block, atrial flutter, and atrial fibrillation) infrequently convert to a sinus rhythm, but the ventricular rate slows. Rarely, atrial slowing will be sufficient to allow 1 : 1 AV conduction, which may actually increase the ventricular rate (Fig. 11–6). AV, atrioventricular; PAT, paroxysmal atrial tachycardia; VT, ventricular tachycardia.

maker is more likely to be affected than the AV node, except when digitalis has been administered.2,5,6

Indications for Vagal Maneuvers Vagal maneuvers are potentially useful in attempting to slow down or break an SVT. Vagal maneuvers are also indicated in settings in which slowing conduction in the SA or AV node could provide useful information (Table 11–2). These settings

CARDIAC PROCEDURES ● 

III 

include patients with wide-complex tachycardia in whom carotid sinus massage (CSM) aids in the distinction between SVT and VT. CSM can elucidate narrow-complex tachycar­ dia in which the P-waves are not visible or aid in detection of suspected rate-related bundle branch block or suspected pace­ maker malfunction. After CSM, a wide-complex SVT may be converted to normal sinus rhythm, P-waves may be revealed after increased AV node inhibition, or ventricular complexes may narrow as the ventricular rate slows. Because CSM slows atrial and not ventricular activity, AV dissociation may be more easily seen, indicating VT (Fig. 11–6). In rapid AF or atrial flutter with 2 : 1 block, either P-waves or irregular ven­ tricular activity with absent P-waves may be revealed. Sinus tachycardia may also be more apparent once P-waves are unmasked by slowing the SA node (Figs. 11–7 to 11–14). Adenosine may be used for the same diagnostic purpose in these situations as well.7 In order of decreasing frequency, the ECG changes seen with CSM and vagal maneuvers are pre­ sented in Table 11–3. Vagal maneuvers, and in particular CSM, may also be a useful aid to the diagnosis of syncope in the elderly. Some BEFORE C.S.M. P

P

ATRIAL RATE 102 P P P P

P

DURING C.S.M. P

202

P

Atrial Rate

166↓

Carotid Sinus Pressure↓ VI

Figure 11–7  Paroxysmal atrial tachycardia with variable block.  Carotid sinus pressure uncovers P-waves hidden in the ventricular complex. Upper strip resembles atrial flutter or atrial fibrillation with ventricular ectopic beats. Lower strip shows paroxysmal atrial tachycardia with variable block at an atrial rate of 166 beats/min.  (From Lown B, Levine SA: Carotid sinus—clinical value of its stimulation. Circulation 23:766, 1961. Reproduced by permission.) CSM

VENTRICULAR RATE 150 CRI

88 P

VI

150 P

P

P

CRI

Figure 11–6  Ventricular tachycardia.  Carotid sinus massage (CSM) slows atria but not ventricles, thus establishing the presence of AV dissociation, supporting the diagnosis of ventricular tachycardia. The QRS measures 0.16 sec. Note the atrial rate slowing from 102 to 88 beats/min while the ventricular rate is unaffected.  (From Lown B, Levine SA: Carotid sinus—clinical value of its stimulation. Circulation 23:766, 1961. Reproduced by permission.)

Figure 11–8  Sinus tachycardia.  The sinus P-wave is obscured within the descending limb of the T-wave. CSM transiently slows the sinus rate and exposes the P-wave. The rate then increases. The strips are continuous.  (From Silverman ME: Recognition and treatment of arrhythmias. In Schwartz GR, Safar P, Stone JH, et al [eds]: Principles and Practice of Emergency Medicine, vol 2. Philadelphia, WB Saunders, 1978. Reproduced by permission.) Figure 11–9  Sinus tachycardia with high-level block.  Arrows indicate sinus P-waves. Strips IIa to IId are continuous. The basic rhythm is sinus, but marked first-degree AV block is present. High-degree (advanced) AV block associated with transient slowing of sinus rate is produced by CSS.  (From Chung EK: Electrocardiography. 2nd ed. New York, Harper & Row, 1980. Reproduced by permission.)

II-a CSS II-b

II-c

II-d

J CSM

Figure 11–10  Paroxysmal atrial tachycardia.  CSM abolishes the dysrhythmia and results in a period of sinus suppression with a junctional (J) escape beat. Prolonged periods of asystole may produce anxiety in the physician who is waiting for the resumption of a sinus pacemaker.  (From Silverman ME: Recognition and treatment of arrhythmias. In Schwartz GR, Safar P, Stone JH, et al [eds]: Principles and Practice of Emergency Medicine, vol 2. Philadelphia, WB Saunders, 1978. Reproduced by permission.)

11  ● 

II-a

Ventricular standstill Figure 11–11  Atrial flutter.  CSS (downward arrow) produces marked slowing of the ventricular rate in atrial flutter. Note the obvious flutter waves with an atrial rate of 300 and a long period of ventricular standstill. The strips are continuous.  (From Chung EK: Electrocardiography, 2nd ed. New York, Harper & Row, 1980. Reproduced by permission.)

CSM

Techniques for supraventricular tachycardias

C S S (left) II-b

Figure 11–12  Atrial fibrillation.  CSM slows the ventricular response transiently, revealing the fibrillating baseline. The ventricular rate subsequently accelerates.  (From Silverman ME: Recognition and treatment of arrhythmias. In Schwartz GR, Safar P, Stone JH, et al [eds]: Principles and Practice of Emergency Medicine, vol 2. Philadelphia, WB Saunders, 1978. Reproduced by permission.)

C.S.S. II II

II Figure 11–13  Occult premature ventricular contractions.  CSM reveals ventricular extrasystoles, thereby explaining the cause of palpitation in this case.  (From Lown B, Levine SA: Carotid sinus—clinical value of its stimulation. Circulation 23:766, 1961. Reproduced by permission.)

Figure 11–14  A run of ventricular tachycardia is seen immediately after a supraventricular dysrhythmia is terminated by CSM. The patient remained asymptomatic, and a normal sinus rhythm was established spontaneously within a few seconds. If asystole is prolonged, ask the patient to vigorously cough (cough CPR) or apply a precordial thump.

TABLE 11–3  Order of Decreasing Frequency of Electrocardiographic Changes with Vagal Maneuvers 1.  Sinoatrial slowing, occurring in approximately 75% of cases and leading to sinus arrest approximately 3% of the time. 2.  Atrial conduction defects, manifested by an increase in width of the P-wave on the electrocardiogram 3.  Prolongation of the PR interval and higher degrees of atrioventricular block, seen in approximately 10% of cases. 4.  Nodal escape rhythms. 5.  Complete asystole, defined as sinus arrest without ventricular escape lasting > 3 sec, occurring in 4% of cases. 6.  Premature ventricular contractions.

14% to 45% of elderly patients referred for syncope are thought to have carotid sinus syndrome (CSS).6,8,9 CSS is defined as an asystolic pause greater than 3 seconds or a reduction of systolic blood pressure greater than 50 mm Hg in response to CSM. It shares many characteristics with sick sinus syndrome, suggesting that both are manifestations of the same disease. CSS causes cerebral hypoperfusion, leading to dizziness and syncope. Analysis of patients with the syndrome indicates that it results from a baroreflex-mediated bradycardia in 29% of patients, hypotension in 37%, or both in 34%.10,11 Therefore, syncope, near-syncope, or a fall of unclear etiology in the elderly is an important indication for diagnostic CSM.12

203

CARDIAC PROCEDURES ● 

III 

Although the use of digoxin has been overshadowed by the use of other potentially less toxic agents such as calcium channel blockers and β-blockers, the clinician can still pro­ spectively simulate the cardioinhibitory effects of digoxin on a patient by performing vagal maneuvers. This can guide use and dosage of the medication before initiating treatment/ therapy with digoxin. Significant slowing or block with CSM suggests a similar sensitivity to digoxin, and a smaller loading dose should be considered.

Equipment and Setup Prior to the initiation of any clinical intervention such as vagal maneuvers, administration of medication, or cardioversion, for SVT, if there is time, the patient should be placed on a cardiac monitor, intravenous (IV) access should be estab­ lished, and a slow, keep-vein-open (KVO; 60 mL/hr saline IV) solution should be infused. The patient should also be monitored with a pulse oximeter and an indirect blood pres­ sure monitor. Numerous antiarrhythmic medications should be readily available. A defibrillator/pacemaker should be at bedside in anticipation of a worsening dysrhythmia. The administration of oxygen is advised for the procedure, espe­ cially if conscious sedation is anticipated. The patient should be placed in the reverse Trendelenburg position if tolerated. Merely placing the patient in this position may terminate the SVT owing to increased pressure on the carotids, giving maximum carotid bulb stimulation. This position may also prevent syncope if there is a significant decrease in blood pressure or HR.

204

CAROTID SINUS MASSAGE CSM is a bedside vagal maneuver technique involving digital pressure on the richly innervated carotid sinus. It takes advantage of the accessible position of this baroreceptor for diagnostic and therapeutic purposes. Its main therapeutic application is for termination of SVTs owing to paroxysmal atrial tachycardia (PAT). It also has diagnostic utility in the assessment of tachydysrhythmias and rate-related bundle branch blocks. In addition, it can provide clues to latent digoxin toxicity, as described previously, by potentiating man­ ifestations of toxicity. It can also be used to sort out the dif­ ferential diagnosis of syncope. Returning to the use of CSM as a diagnostic technique for assessing digoxin toxicity, adverse effects/toxicity from digoxin depend more on the response of the host than on the actual digoxin level. In cases of suspected digoxin toxicity, before the level is available, or when the digoxin level is in the “normal range,” CSM may be a useful diagnostic adjunct. Significant inhibition of AV node conduction associated with ventricular ectopy, especially ventricular bigeminy, should lead to the suspicion of digoxin toxicity.1 Other therapeutic uses of CSM have been made obsolete by current medical therapy. In 1961, Lown and Levine1 described the dramatic effect CSM had in the 1920s on reliev­ ing acute pulmonary edema in a group of patients with hyper­ tension and coronary artery disease. They reported: “Relief is immediate and coincides with the onset of bradycardia. In the majority, it is associated with a drop in blood pressure. The patient is promptly able to lie flat. Fear, dyspnea and chest oppression disappear.” CSM also has been reported to relieve anginal pain. The technique may be useful when the diagnosis of angina is uncertain.13 The advantage of the CSM technique

over the use of nitroglycerin is unknown. Although CSM is no longer the first approach to either pulmonary edema or angina, it remains a therapeutic or adjunct diagnostic tool in some cases or when modern pharmacologic agents are unavail­ able. Because adenosine may not always be readily available and cannot be used to assess the sensitivity of the carotid sinus, CSM remains a useful bedside tool. Contraindications CSM is contraindicated in the very rare patient likely to suffer neurologic or cardiovascular complications from the proce­ dure. Patients with a carotid bruit should not have CSM because of the risk of carotid embolization or occlusion. A recent cerebral infarction is another contraindication, because even marginal reduction of cerebral blood flow may produce further infarction. The presence of diffuse, advanced coronary atheroscle­ rosis is associated with increased sensitivity of the carotid sinus reflex. This hypersensitivity is further augmented during an anginal attack or an acute myocardial infarction. Brown and coworkers14 found that the degree of carotid sinus hyper­ sensitivity was directly proportional to the severity of coro­ nary artery disease documented by cardiac catheterization. Patients with acute myocardial ischemia or with recent myo­ cardial infarction are already at higher risk of VT or ventricu­ lar fibrillation (VF). A CSM-induced prolonged asystole may further predispose them to these dysrhythmias. Therefore, CSM should be avoided in these patients. Both digoxin and CSM act through a vagal mechanism to inhibit the AV node. Patients on digoxin may experience a greater inhibition of the AV node with longer AV block as a result. Patients with apparent manifestations of digoxin toxic­ ity or known digoxin toxicity should not have CSM, becauses AV inhibition may be profound.15 Technique This technique can be performed with or without a concomi­ tant Valsalva maneuver. Alternatively, pressure can be applied to the abdomen by an assistant. Some clinicians prefer to place the patient supine or with the head of the bed tilted down­ ward. The clinician should begin CSM on the patient’s right carotid bulb because some investigators have found a greater cardioinhibitory effect on this side.12,16,17 However, scientific agreement on this issue is not unanimous. Simultaneous bilat­ eral CSM is absolutely contraindicated, because cerebral cir­ culation may be severely compromised. Before attempting CSM, the clinician should first auscultate for carotid bruits on both sides of the neck. The presence of a bruit is a con­ traindication to massage. Keeping the patient relaxed is helpful for two reasons: a tense platysma muscle makes palpation of the carotid sinus difficult, and an anxious patient will be less sensitive to CSM as a result of heightened sympathetic tone. With the patient’s head tilted backward and slightly to the opposite side, palpate the carotid artery just below the angle of the mandible at the upper level of the thyroid carti­ lage and anterior to the sternocleidomastoid muscle. Once the pulsation is identified, use the tips of the fingers to administer CSM for 5 seconds in a posteromedial direction, aiming toward the vertebral column. Although earlier practitioners used a longer duration of massage, a shorter period of massage minimizes the risk of complications and is adequate for diag­ nostic purposes in the majority of patients.18 Pressure on the carotid sinus may be steady or undulating in intensity;

11  ● 

II

the force, however, must not occlude the carotid artery. The temporal artery may be simultaneously palpated to ensure that the carotid remains patent throughout the procedure. If unsuccessful, CSM may be repeated after 1 minute. If the procedure is still unsuccessful, the opposite carotid sinus may be massaged in a similar fashion. Simultaneous Valsalva maneuvers and the head-down position to enhance carotid sinus sensitivity should be done before the technique is aban­ doned. Importantly, CSM should be repeated once antiarrhythmic medication has been given, and often the combination is more effective. Complications Neurologic complications of CSM are rare and usually tran­ sient. In a review of neurologic complications in elderly patients undergoing this procedure, Munro and associates19 found 7 complications from a total of 5000 massage episodes, for an incidence of 0.14%. Reported deficits included weak­ ness in 5 cases and visual field loss in 2 others. In 1 case, the visual field loss was permanent. Patients in this study were excluded from CSM if they had a carotid bruit, recent cerebral infarction, recent myocardial infarction, or a history of VT or VF. The duration of massage was 5 seconds. Lown and Levine1 described 1 patient with brief facial weakness during several thousand tests. Carotid emboli and hypotension have both been implicated as possible causes of the neurologic deficits. Unintentional occlusion of the carotid artery may also be responsible for some neurologic complications. Cardiac complications include asystole, VT, or VF. A normal pause of less than 3 seconds is part of the physiologic response to CSM; a longer pause may be diagnostic of CSS. In a review of reported cases of ventricular tachydysrhythmias, five cases were described.20 All five patients were receiving digoxin, and in several cases, VT or VF followed AV block. Digoxin is associated with more prolonged AV block resulting from CSM, perhaps leaving these patients more vulnerable.

Interpretation of Vagal Maneuvers

II P

P

P

P C.S.S.

II

II

Figure 11–16  Acceleration of ventricular rate by carotid sinus stimulation (CSS).  Continuous tracing. Upper strip shows 2 : 1 atrioventricular (AV) block: atrial rate = 102/min; ventricular rate = 51/ min. The second and third strips were recorded during and after CSS, when the atrial rate was reduced to 68/min; a 1 : 1 response occurs.  (From Lown B, Levine SA: Carotid sinus—clinical value of its stimulation. Circulation 23:766, 1961. Reproduced by permission.)

Valsalva maneuver (i.e., exhaling against a closed glottis or bearing down as if to defecate), intrathoracic pressures are increased, leading to increased arterial pressure, as a result of increased afterload. This increased pressure is transferred to the peripheral vascular system. Venous return to the heart is decreased, resulting in a decreased SVT. This is followed by increased venous pressure. All of these pressure changes lead to an initial increase in HR and carotid sinus pressure. As the maneuver is sustained, vagal tone is increased, leading to a compensatory decrease in SA and AV conduction. This is the expected/desired diagnostic/therapeutic response. Contraindications Patients must be able to cooperate with the clinician’s com­ mands. Dyspneic or tachypneic patients may not be able to hold their breath for the period of time needed to complete the maneuver.

A pause longer than 3 seconds, or a drop in systolic blood pressure greater than 50 mm Hg in patients to whom CSM is administered while they are in a supine position, is diag­ nostic of CSS (Figs. 11–15 and 11–16). Patients should be supine during testing to reduce the risk of cerebral hypoperfusion.12,21 Although CSM is one of the better-known vagal maneu­ vers, a variety of other physical modalities are available to the clinician to affect a change in HR. The anticipated rhythm response to different vagal maneuvers in the setting of differ­ ent underlying rhythms is shown in Table 11–4.

Technique With the patient supine; monitor in place; IV access secured; and atropine, lidocaine, and defibrillation available, have the patient take a deep breath and hold it. Instruct the patient to bear down and try to exhale without allowing the air to leave the lungs. Patient should try to hold this position for 10 to 20 seconds.23,24 An adjunct method is to have the patient take and hold a deep breath and try to push against the clinician’s hand with her or his abdomen while the clinician gently pushes on the anterior wall of the abdomen.

Valsalva Maneuver

viewed as a variation on the simple Valsalva maneuver. It has been found to be useful in children who may be unable to cooperate with or be capable of performing a Valsalva maneu­ ver. Classically, the technique consists of facial immersion,

In general, mean bradycardia changes are greatest for the Valsalva maneuver and the diving response.2,21,22 During the

Techniques for supraventricular tachycardias

Figure 11–15  Hyperreactive carotid sinus reflex.  Gentle pressure was applied to the carotid sinus for 3 seconds, resulting in a pause in sinus rhythm of approximately 7 seconds. This syndrome may be the cause of syncope.  (From Bigger JT Jr: Mechanisms and diagnosis of arrhythmias. In Braunwald E [ed]: Heart Disease, vol 1. Philadelphia, WB Saunders, 1980. Reproduced by permission.)

Apneic Facial Exposure to Cold (“Diving Response,” Diving Bradycardia): Technique.  This technique can be

205

CARDIAC PROCEDURES ● 

III 

TABLE 11–4  Ventricular Response to Carotid Sinus Massage and Other Vagal Maneuvers Type of Arrhythmia

Atrial Rate (bpm)

Response to CSM and Release

60–100 100–180

Slowing with return to former rate on release. Slowing with return to former rate on release Slowing with return to former rate on release; appearance of diagnostic P-waves Termination or no effect Slowing with return to former rate on release; increasing AV block: flutter persists Slowing with persistence of gross irregular rate on release; increasing AV block Abrupt slowing with return to normal sinus on release; tachycardia often persists None; ± slowing Abrupt slowing; termination or no effect; may unmask WPW None; ± slowing None; may unmask AV dissociation None None None Gradual slowing caused by sinus slowing; return to former rate on release Sinus slows with increase in block; return to former rate with release Slowing None Slowing with return to former rate with release Slowing with return to former rate with release Do not attempt CSM

Normal sinus rhythm Normal sinus bradycardia Normal sinus tachycardia AV nodal re-entry Atrial flutter

150–250 250–350

Atrial fibrillation

400–600

Atrial tachycardia with block

150–250

AV junctional rhythm Reciprocal tachycardia using accessory (WPW) pathways Nonparoxysmal AV junctional tachycardia Ventricular tachycardia Atrial idioventricular rhythm Ventricular flutter Ventricular fibrillation First-degree AV block

40–100 150–250 60–100 60–100 60–100 60–100 60–100 60–100

Second-degree AV block (I) Second-degree AV block (II) Third-degree AV block Right bundle branch block Left bundle branch block Digitalis toxicity–induced arrhythmias

60–100 60–100 60–100 60–100 60–100 Variable

AV, atrioventricular; CSM, carotid sinus massage; WPW, Wolff-Parkinson-White (syndrome). Adapted from Braunwald E (ed): Heart Disease: A Textbook of Cardiovascular Medicine, 6th ed. Philadelphia, WB Saunders, 2001, p 642.

206

without breathing, for 15 to 30 seconds in cold water (0°C to 15°C). Another variation of this technique is to drip ice water into the nostril of a small child. The procedure is based on the classic diving reflex of bradycardia. Slowing SVT to unmask the hidden, underlying rhythm is similar to the effects of CSM. The conversion of PAT to sinus rhythm should be observed in 15 to 35 seconds. The procedure is convenient and noninvasive and can be self-administered.25–30 Berk and colleagues16 have demonstrated in healthy volunteers that cold water face immersion and the Valsalva maneuver can produce a greater vagal response than CSM. Lim and associates in 199831 and Mehta and coworkers in 198817 also found that the Valsalva maneuver was more effec­ tive than CSM for conversion of induced SVT.

Oculocardiac Reflex (Trigeminovagal Reflex) This reflex is clinically significant during strabismus surgery in children, although the manifestations of this reflex are not consistent. The oculocardiac reflex (OCR) is induced by pres­ sure on the eyeball. The afferent pathway follows the long and short ciliary nerves to the ciliary ganglion. From there, it travels to the gasserian ganglion body along the ophthalmic division of the trigeminal nerve (CN V). The afferent pathway ends in the main trigeminal sensory nucleus in the floor of the fourth ventricle.3,22,23 Efferent impulses start at the vasomotor center and travel through the vagal nerve (CN X) and the sympathetic chain. Bradycardia is the result of increasing parasympathetic tone. Decreasing sympathetic tone causes vasodilatation. The

cardiac effect stops when eye pressure is relieved.22 Patients should be monitored to recognize ECG changes. Atropine should be available to reverse life-threatening bradycardia. Contraindications Common sense dictates that the physician should try to elicit the following histories: (1) Recent retinal or lens surgery; (2) glaucoma; (3) thrombotic-related eye conditions; (4) pene­ trating or recent blunt trauma to the eye, which all seem to be obvious possible contraindications. Care should be taken when pressing on the eye globe to prevent corneal or scleral injury. Technique With the eyelid closed, nonrotating pressure is applied to the eyeball for 10 to 20 seconds. There is no advantage to the use of either eye. Ventricular slowing and possible decrease in blood pressure should be observed almost immediately when the pressure is applied. The cardiac effect of bradycardia will cease when pressure is removed. As with all vagal maneu­ vers, monitoring, IV access, atropine, lidocaine, and defibril­ lation should be available during the procedure (see Table 11–4). Equipment and Setup As a precaution against hypotension, and life-threatening dys­ rhythmias, an IV line with normal saline should be started before attempting any vagal maneuver, the use of pharmaco­ logic agents, or electrical cardioversion. The patient should be placed on a cardiac monitor. Atropine and lidocaine, as well as a transvenous or transcutaneous pacemaker and defi­

Amiodarone has become one of the workhorses of dysrhyth­ mia treatment in the emergency department (ED). It is often considered a Vaughan-Williams class III drug, because it is a potassium channel blocker. However, this medication also blocks the sodium and calcium channels and also blocks αand β-adrenergic receptors. As a result of its potassium block­ ing properties, amiodarone prolongs the action potential duration and increased refractoriness of the atria and the ventricular tissues, the sinus and AV nodal tissues, and the Purkinje fibers. Amiodarone also blocks sodium channels in depolarized tissue. Amiodarone slows depolarization in the SA node and slows conduction through the AV node. Its calcium antagonist effect is minimal.15,33,34,38 Indications and Contraindications.  Amiodarone is used in the control of narrow-complex supraventricular and ventricular dysrhythmias. It is useful in the management of narrow-complex tachycardias that originate from a re-entry rhythm (SVT). It is effective in conversion of the stable widecomplex tachycardias, and it is useful in managing polymor­ phic VT with a normal Q-T interval. Amiodarone can be used for wide-complex tachycardias of undetermined origin. This drug can also be used for the management of AF and atrial flutter with aberrancy, SVT with accessory pathway conduc­ tion, and the rare adult junctional tachycardia. Another use for this medication is rapid ventricular rate control due to accessory pathway conduction in pre-excited atrial arrhyth­ mias. It is a strong second-line choice with procainamide for hemodynamically stable VT. Its use can precipitate heart failure, hypotension, and severe bradycardia. When used with β-blockers and calcium channel blockers, amiodarone can have the added risk of hypotension and bradycardia. Torsades de pointes has been reported after the use of amiodarone in conjunction with drugs that have increased the Q-T interval. Dosage.  The IV dosage is 150 mg IV over 10 minutes. This can be followed by a 1-mg/min infusion for 6 hours and then a 0.5-mg/min maintenance infusion over 18 hours. If a dysrhythmia is refractory or resistant, 150 mg can be repeated every 10 minutes to a maximum of 2.2 g/24 hr.39 The major adverse effects of amiodarone are bradycardia and hypotension.

Calcium Channel Blockers Diltiazem Diltiazem is a nondihydropyridine calcium channel blocker. Diltiazem controls the rate of calcium influx into the myo­ cytes during depolarization. This calcium channel blocker slows conduction of impulses through the AV node and pro­ longs the refractory period of the AV node. As a result, this drug is capable of terminating re-entry–based tachycardias that have not converted with the use of adenosine or vagal maneuvers, and it can be utilized to control the ventricular response rate in a variety of SVTs (AF, atrial flutter). In addi­ tion, diltiazem can be utilized for the treatment of stable,

Techniques for supraventricular tachycardias

The use of vagal maneuvers has been eclipsed in recent years by the use of adenosine, an endogenous, ultrashort-acting, vagal-stimulating purine nucleoside that is ubiquitous in all body cells. Its action is to slow conduction time through the AV node and depress the AV node. If vagal maneuvers have been attempted and failed to produce the desired response, use of adenosine is an appropriate subsequent intervention. Extracellular adenosine is rapidly cleared from the circu­ lation by the erythrocyte and vascular endothelium system that transports adenosine intracellularly. Here, rapid metabo­ lism via a phosphorylation or deamination cycle produces inosine or adenosine monophosphate. Adenosine produces a short-lived pharmaceutical response because it is rapidly metabolized by the described enzymatic degradation. The half-life of adenosine is less than 10 seconds, with the metabo­ lites becoming incorporated into the phosphate, high-energy pool.32–35 Indications and Contraindications.  (1) Most forms of paroxysmal supraventricular tachycardia (PSVT) affect a reentry pathway involving the AV node, and adenosine depresses AV node and sinus node activity. Adenosine is indicated for the conversion of PSVT associated with or without accessory tract bypass conduction (Wolff-Parkinson-White [WPW], Lown-Ganong-Levine [LGL]). (2) The other use of adenos­ ine is in diagnostic slowing of SVT to unmask AF, atrial flutter, or VT. The diagnostic and therapeutic effects of ade­ nosine on tachydysrhythmias are similar to those elicited by the vagal maneuvers. Adenosine’s safety is derived from its short duration of action—usually about 10 to 12 seconds. Adenosine should not be used in patients with a known history of second- or third-degree AV block or sick sinus syndrome, unless there is a functioning internal pacer. Also, if the patient has a known hypersensitivity to the adenosine, the drug should not be used. In addition, adenosine should not be used in patients with an underlying accessory tract bypass conduction (WPW, LGL) in the setting of AF. In this circumstance, the HR may increase when enhanced AV node blockade permits greater use of the bypass tract. Dosage.  The recommended initial dose is a 6-mg rapid bolus over 1 to 3 seconds. The dose should be followed by a 20-mL saline flush. If there is no response within 1 to 2 minutes, a 12-mg repeated dose should be administered in the same manner as the initial dose. This second dose should be followed by a 20-mL saline bolus. Side effects of adenosine are common and transient, and many patients experience an unsettling feeling that should be explained before drug administration. Common sensations include flushing, dyspnea, and chest pain. Important drug interactions include theophylline or related methylxanthines (caffeine and theobromine), which can block the adenosine

Amiodarone

● 

SELECTED PHARMACOLOGIC AGENTS Adenosine

receptor sites. If these medications are being taken by the patient, a larger dose of adenosine should be administered. If the patient is taking dipyridamole or carbamazepine, which may block adenosine uptake and potentiate its effects, a smaller dose of adenosine should be contemplated (3 mg IV).36 Adenosine is safe and effective in pregnancy.37

11 

brillator, should be readily available at the bedside. Airway management equipment should be readily available. The patient should be in the supine or slight reverse Trendelen­ burg position if it can be tolerated. Occasionally, PAT will convert merely by lowering the back of the bed, presumably because the supine position results in a stretching of the carotid bulb, giving maximum baroreceptor sensitivity. The supine position may also prevent syncope in the event of a significant drop in HR or blood pressure.

207

CARDIAC PROCEDURES ● 

III 

208

narrow-complex tachycardias that are driven by automaticity (e.g., ectopic, multifocal, or junctional tachycardias). Its effects on AV nodal tissue are selective in that it reduces AV conduction in tissue responsible for the tachydys­ rhythmia but spares normal conduction tissue.34,38,40,41 Indications and Contraindications.  Its beneficial effects are (1) ventricular slowing of rapid AF/atrial flutter without accessory bypass conduction (2) rapid conversion of narrow-complex PSVT to sinus rhythm.33,36,38,41,42 Diltiazem is contraindicated in the following settings: (1) sick sinus syndrome, second-degree block, third-degree block, except in the presence of an internal pacer; (2) severe hypo­ tension or cardiogenic shock; (3) diltiazem hypersensitivity; (4) use of IV β-blockade within a few hours of need to use diltiazem; (5) AF or atrial flutter with coexisting accessory bypass tract conduction (WPW, LGL); and (6) VT. Dosage.  An initial dose of 0.25 mg/kg can be followed by a repeat dose of 0.35 mg/kg. Maintenance infusion should be at 5 to 15 mg/hr.15,36,38,41 Verapamil Verapamil is also a calcium channel blocker. This medication blocks the slow channel for calcium entry into the myocytes. Verapamil blocks not only the calcium channels in the special­ ized conduction tissue of the myocardium but also the con­ tracting cells of the heart. As a result, verapamil prolongs the effective refractory period within the AV node and slows conduction.2,40 It also has a modest effect on myocardial contractility.2 Indications and Contraindications.  Verapamil is effec­ tive in (1) converting narrow-complex PSVT to normal sinus rhythm and (2) controlling the ventricular response in AF or atrial flutter, if the AF or atrial flutter is not complicated by the presence of an accessory bypass tract (WPW, LGL). Verapamil should not be used in the following settings: (1) PSVT with accessory bypass tract conduction, (2) AF/ atrial flutter with accessory bypass tract conduction, (3) coex­ istence of a sick sinus syndrome or second- or third-degree AV block unless an internal pacer is present; (4) severe left ventricular dysfunction (systolic blood pressure < 90 mm Hg), or cardiogenic shock; and (5) in patients with a known vera­ pamil hypersensitivity.7,33,34,38,40,41

β-Adrenergic Blockade β-Blockers are very useful agents for the control of ventricular response in PSVT, AF or atrial flutter, and atrial tachycardia. It is generally considered that no β-blocker offers a distinctive advantage over another because when used clinically, they all can be titrated for a desired effect on dysrhythmias, and hypertension. Examples of β blockers are atenolol, metopro­ lol, propranolol and esmolol. What separates the different drugs and their use is the various pharmacological character­ istics which control adverse reactions, speed of onset, dosage regimes, contraindications, and drug interactions. The electrophysiological effect of β-blockers results from the inhibition of catecholamine binding at β-receptor sites. These medications reduce the effects of circulating catechol­ amines and this is manifested in a decrease in HR, blood pressure, and myocardial contractility. The PR interval may be prolonged, but the QRS and the Q-T intervals are not affected. Their actions are most noted on cells that are most stimulated by adrenergic actions. Typically, these sites are the

sinus node, the Purkinje fibers, and ventricular tissue when it is stimulated by catecholamines.2,33,34,40 These medications also have various cardioprotective effects for patients suffering from acute coronary syndromes. They exert their cardiopro­ tective effects by decreasing myocardial workload, and hence, they decrease myocardial oxygen consumption and demand.2 β-Blockers are useful in the treatment of narrow-complex tachycardias that originate secondary to re-entry phenome­ non or an automatic focus (MAT, an ectopic pacemaker, or a junctional rhythm). These drugs can also be used to control rates in patients suffering from AF or atrial flutter, as long as ventricular function is nominal. Some representative doses of these β-blockers are (1) atenolol (β1) 5 mg IV slowly over 5 minutes; if no effect, repeat in 10 minutes; (2) metoprolol (β1) 5 mg IV slowly, may repeat up to 15 mg total; and (3) propranolol 0.1 mg/kg IV slow push divided into three equal doses at 2- to 3-minute intervals; may repeat total dose in 2 minutes. Administration rate of the drug should not exceed 1 mg/min. In general, β-blockers should not be used in patients with a history that includes diabetes, lung diseases, bradycardias or heart blocks, use of calcium channel blockers, hypotension, or the presence of a vasospasm condition. Propranol Propranol is the representative drug of the β-adrenergic blockade agents. It is nonselective. It has β1 and β2 effects on the heart that allow for its use in controlling rapid ventricular rates. Rate slowing is caused by (1) slowing SA node impulse formation and (2) depression of myocardial contractility. The usual effects on the electrocardiogram are rate reduction and prolongation of the PR interval. The QRS and Q-T intervals are not affected. Because it is relatively nonselective (has effects on both β1 and β2 receptors), its contraindications are somewhat extensive.33,40,43 Esmolol Esmolol is a rapid-action, short-acting β1-selective (car­ dioselective) β-blocker. At therapeutic doses, it inhibits β1receptors located in cardiac muscle. At higher doses, the selectivity is lost and it affects β2-receptors in the lung and vascular system. Esmolol is rapidly metabolized in erythro­ cytes and has a half-life of about 2 to 9 minutes. Its elimina­ tion half-life is approximately 9 minutes.34,38 Indications and Contraindications.  Esmolol is indi­ cated for the rapid conversion of SVT and the rapid control of ventricular rate in patients with non–pre-excited, AF or atrial flutter. It also has a function in rate control of noncom­ pensated sinus tachycardia when a clinician feels the tachy­ cardia requires slowing. It also has been proved to have a benefit as an adjunct therapy in the VT of torsades de pointes.34,36,38,40 Esmolol should not be used in patients with second- or third-degree heart block, or in frank heart failure. Like all βblockers, care should be exercised when used in patients with bronchospastic disease and diabetes. Dosage.  Esmolol has a complicated dose regimen. A loading dose of 0.5 mg/kg is given over the first 1 minute. This is followed by a maintenance infusion of 50 µg/kg per minute over 4 minutes. If this is not successful, a second bolus dose of 0.5 mg/kg followed by a maintenance infusion of 100 µg/kg over 4 minutes is started. This bolus/maintenance dosing can be repeated up to a maximum infusion rate of

Ibutilide Ibutilide is a short-acting, antiarrhythmic that prolongs the refractory period of the myocardium by prolonging the dura­ tion of the cardiac action potential. This drug is useful in the management conversion of AF and atrial flutter, when the arrhythmia is present for less than 48 hours. This medication can also be used to control HR in the face of AF or atrial flutter when calcium channel blockers and β-blockers have proved ineffective. The dose for ibutilide is 1 mg IV over 10 minutes in a 0.1-mg/mL dilution in an adult weighing more than 60 kg. This dose can be repeated in about 10 minutes after the first dose. In patients weighing less than 60 kg, the initial dose should be 0.01 mg/kg. This drug minimally affects blood pressure and HR. However, caution should be used when using this drug in patients who have either a high potassium or a low magnesium level. The patient should be on a cardiac monitor during and after administration of the medication, Do not use this drug if the patient has a corrected Q-T inter­ val longer than 440 msec.

Digoxin Digoxin is a time-honored drug used for treatment of AF and atrial flutter. It is the only antidysrhythmic with inotropic

SPECIAL CONSIDERATION OF ANTICOAGULATION IN AF:   EVALUATION AND TREATMENT The most common, sustained tachydysrhythmia that presents to the ED is AF. The incidence of AF in the general popula­ tion is 1% to 2 %. The incidence increases with age. Approxi­ mately 1% of the population under 50 years of age have AF whereas 8.8% of the population older than 80 years have AF.53 The connection between AF, structural heart disease, and antecedent coronary artery disease is strong. As a dysrhythmia in acute myocardial infarction, AF is relatively uncommon (11%). However, its presence is associated with a 40% mor­ tality.54 The long-recognized association between valvular heart disease and AF is well documented. Rheumatic valvular disease is the classic valvular disease associated with AF. However, other dysrhythmias have an association with AF: WPW, atrial tachycardia, sick sinus syndrome, and AV nodal reentrant tachycardias. Long-standing medical or cardiac conditions having strong associations with AF are hyperten­ sion, cardiac myxomas, diabetes, thyroid disease, left ventricu­ lar dysfunction, congestive heart failure, pulmonary edema, chronic obstructive pulmonary disease, and pulmonary embolism.47,54,55 AF was thought to be caused by abnormal pulse forma­ tion originating in the atria. In human beings, the atria as a

Techniques for supraventricular tachycardias

A time-honored antiarrhythmic, procainamide slows conduc­ tion and decreases automaticity and excitability of atrial, ven­ tricular, and Purkinje tissue. It also increases refractoriness in atrial and ventricular tissue. Procainamide prolongs the Q-T interval without having much effect on Purkinje fibers or ventricular tissue.36 Indications and Contraindications.  A long-established clinical application is in the rate management of SVT, SVT with aberrancy conduction (wide-complex SVT), AF/atrial flutter associated with WPW conduction, and VTs. The advan­ tage to using procainamide is the ability to convert to the oral form when rate control is achieved. The advanced cardiac life support (ACLS) recommended dose of procainamide is usually 20 mg/min, although in urgent situations, up to 50 mg/min can be used. Procainamide is generally used in clinical situations in which time is not a factor in patient care. Long-term management in the ED necessitates monitoring of the plasma concentrations of pro­ cainamide and the n-acetyl procainamide (NAPA) metabolite. Hypotension and conduction disturbances (torsades de pointes, heart blocks, and sinus node dysfunctions) are often signs of high plasma levels. Caution should be used in patients with histories of hypokalemia, long Q-T intervals, and tors­ ades de pointes. Hematologic and rheumatologic disturbances are factors in long-term use. The end point of administration of the drug is when the arrhythmia is suppressed, hypotension occurs, or the Q-T duration increases by 50% of baseline or a maximum of 17 mg/kg of the drug has been administered (1.2 g in a 70-kg adult).32

● 

Procainamide

characteristics. Digoxin is less useful for the emergency clini­ cian because of its long delay of onset. Digoxin is a cardiac glycoside found in a number of plants. Digoxin is extracted from the leaves of the Digitalis lanata plant. Digoxin increases intracellular Na+ and K+ by inhibiting Na-K-ATPase, the enzyme that regulates the quantity of Na+ and K+ inside the cell. An intracellular increase in Na+ stimulates Na+-Ca+ exchange, leading to increased intracellular Ca+. Digoxin effects are both direct action on cardiac muscles and indirect action on the cardiovascular system. The indirect effects are mediated by the autonomic nervous system. The results of these actions are vagomimetic effects on the SA node and the AV node. The consequences of these actions are (1) increased force and velocity of myocardial contraction (positive inotropic effect); (2) slowing of the HR and AV nodal conduction (vago­ mimetic effect); and (3) decrease in symptomatic nervous system effects (neurohormonal deactivating effect).32,33,44–47 Indications and Contraindications.  Although its use in rate control of the ventricular response in chronic AF is well established, it no longer is the mainstay of therapy for narrowcomplex tachycardias. Newer agents have replaced digoxin in narrow-complex tachycardias. Its inotropic character is still widely utilized in the setting of heart failure. Use of digoxin should be avoided in the clinical settings of sinus node disease and AV blockade. It may cause complete heart block or severe sinus bradycardia. Do not use in the presence of accessory bypass tract rhythms (WPW or LGL). It may cause a rapid ventricular response or VF. Patients with idiopathic hypertrophic subaortic stenosis, restrictive cardio­ myopathy, constrictive pericarditis, or amyloid heart disease are particularly susceptible to digoxin toxicity.48 Dosage.  The IV loading dose of 10 to 15 µg/kg, fol­ lowed by individual parental dosing until the desired rate is achieved.32–34,36,40,49–52

11 

300 µg/kg per minute for 4 minutes.34,38,40 Similar dosing has been recommended for children using a 100- to 200-µg/kg maintenance rate between 100-µg/kg increases in bolus doses.34

209

CARDIAC PROCEDURES ● 

III 

210

result of disease, drug toxicity, or excessive endogenous hor­ mones (e.g., catecholamines) trigger spontaneous automatic­ ity of a sufficient number of atrial cells in multiple atrial sites. These firings sustain the chaotic, simultaneously firing atrial impulses that travel to the ventricles over multiple irre­ gular routes. The transmission of these erratic low-amplitude atrial or fibrillatory f-waves through the AV node to the ventricles paints the classic ECG picture of an irregular rhythm.54,56 First introduced in 1959 by the Russian researchers Moe and Abildskov,57 the hypothesis that AF was a self-sustaining rhythm, independent of multiple firing focus, was verified in 1985. At that time, an animal model was constructed showing four to six waves or “wavelets” were needed to sustain AF in a multiple circulating wave of atrial re-entry, producing the classic ECG picture. These wavelets interact to maintain the optimum atrial conditions needed to maintain sustained AF.54 Any adverse effects from this tachydysrhythmia are related to the disruption of the normal filling and eject components of the cardiac cycle. Classically, patients with AF present with feelings of palpitations, exertional fatigue, dyspnea on exertion, and lightheadedness. Further along in their presentation, patients respond to the fast ventricular rate and develop fluid overload, congestive failure, frank pulmo­ nary edema, and myocardial ischemia.54,58,59 Treatment of symptomatic rapid AF is along three treat­ ment fronts: (1) slowing the rapid ventricular response, (2) conversion to normal sinus rhythm, and (3) prevention of thromboembolism. This section addresses the issue of anti­ coagulation and prevention of thromboembolism.38,54,59,60 Hyperthyroidism may produce AF, occasionally in the absence of obvious signs and symptoms of thyroid storm. It is not standard to test all patients for thyroid disease before proceeding with needed or aggressive interventions, but thyroid function is a common test ordered for a complete evaluation of patients with AF. Failure to respond as expected to antiarrhythmic therapy may be a tip-off to underlying hyperthyroidism, but in subclinical cases, it would be impos­ sible to predict this at the bedside.

CARDIOVERSION Urgent restoration of symptomatic new-onset AF is best achieved with direct cardioversion using either the monopha­ sic or the biphasic defibrillators. In life-threatening or unstable presentations, patients in AF are to be immediately cardioverted because the risk of continued AF outweighs the risk of throm­ boembolism38,54,59,61 (Table 11–5). Of course, the definition of unstable/life-threatening is a clinical judgment call that must often be made with little data. The current guidelines for treatment of symptomatic new-onset AF focus on the length of time the patient has been in AF or atrial flutter as the determining factor for the initia­ tion of anticoagulation when confronted by the need for cardioversion to sinus rhythm. Accordingly, 48 hours or less has been determined to be the time limit that a patient with new-onset AF can be cardioverted without the need for anti­ coagulation. Studies have shown that staying under the 48hour limit allows cardioversion to occur with the lowest risk for thromboembolism.38,54,59,62 For patients who have been determined to be in AF longer than 48 hours and are not in need of urgent care need to be anticoagulated to an Interna­ tional Normalized Ratio (INR) of 2.0 to 3.0 for a 3-week

TABLE 11–5  Guidelines for Anticoagulation in Atrial Fibrillation Atrial Fibrillation I. Duration < 48 hr Low risk for thromboembolism Immediate electrical cardioversion if unstable No anticoagulation necessary II. Duration > 48 hr or Undetermined High risk for thromboembolism Immediate electrical cardioversion if unstable Stable clinical situation Warfarin: INR 2.0–3.0 for 3 wk Cardioversion, then warfarin: INR 2.0–3.0 for 4 wk OR TEE and heparinization Left atrial appendage clot not present Cardioversion, then continue warfarin: INR 2.0–3.0 for 4 wk Left atrial appendage clot visualized Continue warfarin: INR 2.0–3.0 for 4 wk, then cardioversion INR, International Normalized Ratio; TEE, transesophageal echocardiography. Adapted from Pelosi F, Morady F: Evaluation and management of atrial fibrillation. Med Clin North Am 85:225, 2001.

duration before cardioversion.43 If this approach is not clini­ cally acceptable, the patient should have a transesophageal echocardiogram (TEE) and be heparinized. If no left atrial appendage clot is visualized on TEE, the heparinized patient should be immediately cardioverted and anticoagulated for the next 4 weeks. If a left atrial appendage clot is visualized, the patient should be anticoagulated to an INR or 2.0 to 3.0 for 3 weeks’ duration and cardioverted63–67 (see Table 11–5). A synchronized shock (cardioversion) from a monophasic or biphasic defibrillator should be delivered with the patient sedated. Dosages for cardioverting atrial flutter are 50 to 100 J for a monophasic defibrillator; 100 to 200 J for AF. As of this writing, no specific doses are recommended for biphasic car­ dioversion. This topic is covered in greater detail in Chapter 12, Defibrillation and Cardioversion. Success rates with the biphasic defibrillators is reported to be approximately 94% to 95%.68–70 An alternative treatment strategy with a reported success rate of 50% to 70% is the use of ibutilide in a bolus infusion or the use of amiodarone. Caution is recommended with the use of ibutilide in patients with prolonged Q-T intervals or severe left ventricular dysfunction. Ibutilde has a 4% risk of ventricular arrhythmia. Pretreatment of patients to be electrically cardioverted with ibutilde can increase their chances for successful conversion to nearly 100%.32,33,40,71–74 Amiodarone has the advantage of being effective for tachydysrhythmias when the mechanism is unclear and can be used for either wide-complex or narrow-complex tachycar­ dias. Central venous access is advised if concentrations greater than 2 mg/mL are to be used. Amiodarone should be given as an initial bolus of 5 mg/ml over 20 to 30 minutes, followed by a maintenance infusion of 1.0 g/24 hr for a total of 48 to 72 hours.32,34,45,49 Other drugs with good to excellent evidence in obtaining rate control in narrow-complex AF include verapamil, diltia­ zem, procainamide, and β-blockers.33,34,36,40,54

be found on

E x p e rt C o n s u lt

Techniques for supraventricular tachycardias

REFERENCES  c a n

● 

The advent of β-blockers, calcium channel blockers, adenos­ ine, amiodarone, and other effective medications to treat tachydysrhythmias—particularly the SVTs—has diminished the therapeutic use of the vagal maneuvers. However, the vagal maneuvers still remain an important diagnostic tool. These maneuvers are especially important in unmasking the underlying rhythms of narrow-complex tachydysrhythmias and in determining the presence of CSS in patients with syncope.

As for the advent of medications, which quickly and safely control the rate in tachydysrhythmias, their availability has given the emergency clinician a more varied and powerful armamentarium to be used in cardioverting these lifethreatening dysrhythmias to normal sinus rhythms.

11 

CONCLUSIONS

211

CARDIAC PROCEDURES

12

Defibrillation and Cardioversion Bohdan M. Minczak

III 

● 

C H A P T E R

212

Defibrillation is an emergency procedure performed to terminate ventricular fibrillation (VF). This procedure is also useful in the treatment of pulseless ventricular tachycardia (VT). VF and pulseless VT are potentially lethal dysrhythmias1–3 that require immediate assessment and treatment. When VF or pulseless VT occurs, the myocytes of the ventricles become “irritable” or easily excitable. Multiple foci in the walls and septum of the heart begin to produce random, chaotic electrical discharges (Fig. 12–1). As a result, numerous cardiac cells enter into various stages of depolarization and repolarization. This activity disrupts the normal sequence of depolarization and repolarization within the myocardium. The rhythmic, sequential atrioventricular (AV) depolarization of the myocardium that originated in the sinoatrial (SA) node, the “pacemaker” of the heart, is distorted by the disorganized electrical activity from the ventricles. The surface electrocardiographic (ECG) waveforms become distorted from the normal PQRST. The electrical cardiac rhythm that is depicted on the cardiac monitor becomes a “squiggly” chaotic line (Fig. 12–2), with variable amplitude, frequency, and changing polarity. Subsequently, the mechanical activity of the ventricles becomes disorganized and the ventricles begin to contract randomly or “twitch,” producing many local, disorganized weak contractions. This renders the heart incapable of ejecting blood effectively into the peripheral circulation. As a result, the victim becomes unconscious, pulseless, and apneic. This condition describes sudden cardiac arrest (SCA). Failure to treat VF/VT/SCA within the first moments after collapse renders treatment of VF/VT/SCA more difficult.4 If defibrillation and termination of VF/VT/SCA is not accomplished within several minutes of VF/pulseless VT onset (≤4–5 min),2 the patient is unlikely to survive neurologically intact and is very likely to suffer a fatal outcome. SCA is a major cause of death in North America, with deaths from SCA conservatively estimated to exceed 330,000/ yr.5 SCA can strike at any time and in any place. Many of these deaths occur outside the hospital in the prehospital setting. However, a significant number of SCA events occur within the confines of the hospital; in the emergency department (ED), critical care units, non–critical care floors of the hospital, and outpatient areas. In most of these cases, VF is the underlying rhythm. Although the majority of SCA events occur in adults, SCA can also occur in the pediatric population, resulting from hypoxia, trauma, and other causes discussed later. Several rhythms are associated with SCA: pulseless VT, VF, pulseless electrical activity (PEA), and asystole. This chapter is limited to a discussion of treatment of VF and pulseless VT. SCA/VF/VT in adults has many potential causes: SCA independent of myocardial infarction (MI), myocardial ischemia, MI, undiagnosed coronary artery disease, and electrical injuries. The use of medications such as tricyclic antidepres-

sants, digitalis, quinidine, and other proarrhythmics that cause Q-T segment prolongation and changes in the refractory period of the cardiac cycle are capable of precipitating VF. Furthermore, chest trauma, hypothermia, cardiomyopathy, profound hyperkalemia or hypokalemia, hypocalcemia, hypercalcemia, electrolyte disturbances, and various toxidromes can induce conditions favoring the development of VF. Hypoxia is another culprit that frequently precipitates VF in adults and the pediatric population. Congenital malformations of the heart and great vessels have also been associated with an increased incidence of VF in young children. Therefore, based on the multiple, potential causes of VF, the probability of a health care provider, especially a clinician charged with the care of critically ill patients, encountering SCA/VF/pulseless VT is quite high. Prompt defibrillation is a well-documented, effective treatment for VF.6 Defibrillation entails providing a brief “burst” of therapeutic current through the chest wall, across the myocardium, for the purpose of terminating VF. Defibrillation uniformly depolarizes a critical mass of myocardium and “stuns” the fibrillating myocardial fibers, rendering the myocytes refractory to the various chaotic stimuli. Immediately after application of the defibrillatory current, all electrical and mechanical activity is reset. As the myocytes regain their excitability, the SA node will presumptively begin to depolarize rhythmically and reinitiate pacing the heart. Subsequently, the normal, organized, sequential electrical activity of the heart resumes. Shortly thereafter, depending on the metabolic conditions of the myocardium, the mechanical pumping activity of the heart begins to return to baseline levels and circulation of blood to the vital organs slowly resumes. As a result, SCA is terminated and there is a gradual return of spontaneous circulation (ROSC). In the last decade, much has been learned and implemented to improve survival from SCA. Defibrillators have been redesigned in order to increase the efficacy of first-shock defibrillation. New biphasic defibrillators are now touted as having a first-shock efficacy of greater than 90%.7 Defibrillator waveforms have been modified in order to provide a configuration that defibrillates the myocardium with the least amount of energy imparted to the myocardium. This endeavor will presumptively lead to a decrease in the incidence of myocardial damage secondary to defibrillation current and decrease the incidence of nonperfusing dysrhythmias such as PEA and asystole after shock. The characteristics of VF have been explored, yielding information regarding salient features that may serve as indicators for the timing of defibrillation. Currently available data indicate that VF has multiple phases and that defibrillation is most effective in the initial phase of VF.8 If VF persists for several minutes, metabolic byproducts produced by VF may be building up in the myocardial tissue, rendering defibrillation more difficult. Furthermore, some of these myocardial depressants may be interfering with the contractility of the heart in the period immediately after the shock. As a result, modifications have been made in the priorities and sequence of interventions provided during the resuscitative effort of SCA/VF/pulseless VT (i.e., cardiopulmonary resuscitation [CPR] first vs. defibrillation first and immediate CPR after shock) in order to enhance the potential for successful termination of VF with subsequent increased survivability from SCA.9 Recent advances in electronics and computer software have further expanded the arena for defibrillation. Defibrilla-

aVR

Defibrillation and cardioversion

I

● 

Figure 12–1  Fibrillating heart.

“first responders” who are participants in public access defibrillation (PAD) programs are providing CPR and early defibrillation to victims of SCA. As a result, the time to first shock has decreased in many venues. However, although a decrease in time to first shock has decreased and there have been many advances in technology, statistics reporting survival from SCA/pulseless VT/VF and successful defibrillation have not achieved anticipated levels.10 A review of data from the prehospital arena has suggested that a return to the basics (e.g., good, effective CPR) may be an important therapeutic intervention that will enhance efficacy of defibrillation. This speculation occurred because it was observed that when EMS personnel responded to unwitnessed SCA calls and response times exceeded 4 to 5 minutes, victims who received several minutes (±2 min) of CPR prior to initial defibrillation demonstrated an increased rate of initial resuscitation versus those victims who were immediately defibrillated upon arrival of EMS, without prior CPR. This led to studies addressing the effect of CPR on VF. The effects of CPR on VF have been evaluated and reported to increase the amplitude and duration of VF. In addition, performance of CPR prior to defibrillation of protracted VF (>4 min) has been suggested to increase success of VF termination by providing a nominal amount of blood flow through the coronary arteries during SCA.11 It is currently hypothesized that this blood flow may provide substrates to the myocytes that are needed to facilitate defibrillation and

12 

tors manufactured today are lighter, smaller, and more portable. Integration of computer microchips and software has added sophisticated rhythm recognition to the capability of these devices. Many prehospital emergency medical services (EMS), fire departments, police departments, and even lay

V4

V1

213

II

aVL

V2

V5

III

aVF

V3

V6

RHYTHM STRIP II 25 mm/sec; 1 cm/mY

A

00000-0000

F

40

78421

Ventricular fibrillation

B

Heart rate

Rhythm

P wave

300–600

Extremely irregular

Absent

PR interval

QRS

N/A

Fibrillatory baseline

(in seconds) (in seconds)

Figure 12–2  A, Ventricular fibrillation. B, Ventricular fibrillation.

CARDIAC PROCEDURES ● 

III 

that these substrates may enhance resumption of normal electromechanical activity of the ventricles after shock. In addition, byproducts of fibrillation that may be cardiac depressants may be washed away from the myocytes. As a result, defi­brillation is achieved more easily. Furthermore, it has been observed that performance of CPR immediately after defi­brillation facilitates the transition from SCA to ROSC by enhancing ejection of blood from the ventricle into the vasculature. Based on these observations, the steps, sequence, and priorities of CPR have been modified to improve the efficacy of and decrease delays in providing chest compressions. Current emphasis is on increasing the number and frequency of chest compressions. This has been recommended to improve blood flow and circulation during SCA and immediately after defibrillation. Airway management and use of medications have been revised and reprioritized. A comprehensive, in-depth review of resuscitation data has been conducted to provide an evidenced-based guide for the practice of SCA resuscitation. As a result, the American Heart Association has incorporated these new findings and released new guidelines and recommendations for resuscitation of SCA/VF/pulseless VT.12 These new recommendations are presented later in the text. The first part of this chapter is dedicated to describing the procedure for defibrillation of VF and pulseless VT. The second part of the chapter presents current recommendations and guidelines for the treatment of VT and other supraventricular arrhythmias via cardioversion. Use of medication in the resuscitation sequences is presented where appropriate. Defibrillation and cardioversion in the pediatric population are also covered.

214 BACKGROUND  c a n

be found on

E x p e rt C o n s u lt

INDICATIONS AND CONTRAINDICATIONS FOR DEFIBRILLATION (AND CPR) As stated earlier in the text, it is well established that prompt electrical defibrillation is the most effective treatment of acute SCA/VF.6 Starting with the onset of collapse, the survival rate for SCA/VF drops 7% to 10% for every minute of down time without defibrillation. If CPR is initiated, the survival rate declines less rapidly (i.e., 3%–4%/min of down time). If an SCA is witnessed and immediate CPR is provided, coupled with immediate defibrillation, the survival from this event has been reported to increase threefold.22 Therefore, defibrillation is indicated whenever a patient is diagnosed with VF or pulseless VT. At times, mitigating circumstances may exist surrounding the SCA event, causing the clinician to reevaluate the sequence of interventions; however, few absolute specific contraindications to early defibrillation exist other than the presence of a pulse, absence of SCA, medical futility for the procedure, or a valid do-not-resuscitate (DNR) order. Several scenarios are provided to illustrate pertinent points. If a patient suddenly becomes unresponsive, pulseless, and apneic (e.g., a potential victim of SCA), it is reasonable to assume that the underlying cardiac rhythm of the patient is most likely VF. Therefore, immediate action must be taken to prepare for defibrillation. A defibrillator with “quicklook” paddles or an AED should be promptly brought to the victim’s side. Application of either of these devices to the

victim will provide immediate monitoring and assessment of the patient’s rhythm. If there is any delay in getting a defibrillator to the victim’s side, initiation of CPR is indicated. If VF or pulseless VT is diagnosed by looking at the ECG rhythm via the quick-look paddles or if the AED indicates a shockable rhythm, defibrillation must be expediently performed. Thus, witnessed SCA/VF is an indication for defi­ brillation and CPR. The American Heart Association has published a scientific advisory on emphasizing the importance of continuous chest compressions in SCA. The advisory recommends that bystanders who are not confident of their ability to provide conventional CPR should use hands-only (compression-only) CPR until the arrival of an AED or health care provider.23 If a patient is found unresponsive, pulseless, and apneic and the “down time” is unknown, it is suggested that goodquality CPR be performed while preparation is made for defibrillation. As the CPR is performed, preparation for rhythm analysis and possible defibrillation, as noted earlier, should be initiated. After performing CPR for 2 minutes, rhythm analysis should be performed. If VF or pulseless VT is diagnosed, defibrillation should be performed promptly. Data from prehospital resuscitations in which time to first shock was delayed owing to prolonged response times demonstrated that the rate of successful defibrillation increased if patients received bystander CPR prior to defibrillation.22 A scientific evaluation of this information has proposed that CPR may enhance the defibrillation threshold by restoring substrates to the myocytes for the facilitation or resumption of normal excitation-contraction coupling. Furthermore, CPR may wash out myocardial depressants that may have built up during prolonged VF. Hence, the potential for first defibrillation shock success may increase with the performance of about 2 minutes (5 cycles of 30:2 compressions to ventilations) of CPR.22 Therefore, administration of CPR prior to defibrillation is recommended in the prehospital setting. Data to substantiate this sequence for in-hospital resuscitation have not been presented. Thus, the issue of unknown down time, although not a definitive contraindication to immediate defibrillation, may be a factor in the clinician’s decision making process regarding the resuscitation sequence. The question still remains for in-hospital resuscitation as to whether to defibrillate first or administer CPR prior to defibrillation when the down time is unknown. Other relevant questions being asked are how long the CPR should be performed prior to defibrillation and how long VF should be present before prioritizing CPR over defibrillation? Nonetheless, if the down time is unknown, initiate CPR and prepare to defibrillate as soon as possible. Consider performing CPR for a brief period before defibrillation, if deemed clinically appropriate. Victims of SCA due to traumatic injuries usually do not survive.24 The heart, aorta, and pulmonary arteries may have sustained injury that will prevent the resumption of normal cardiovascular function. There is a high probability that underlying hypovolemia and organ damage may preclude success of the resuscitation. However, the cause of the trauma may have been SCA with subsequent loss of consciousness. Therefore, if SCA/VF is present in the trauma patient, treatment with CPR and defibrillation should be attempted. However, if unsuccessful, then a search and treatment of the underlying cause of the trauma and the SCA should be pursued. Nonetheless, trauma is not a contraindication to defibrillation, although the resuscitative effort may be futile.

● 

Defibrillation and cardioversion

adjuncts) present on their chest. Presence of these patches is NOT a contraindication to defibrillation. However, placement of the electrodes or paddles used for defibrillation should be modified to avoid contact with these items. If needed, these items should be removed prior to defibrillation to avoid current diversion from the myocardium, current arcing, sparks, and other problems. Developments in defibrillation and computer electronics have led to the availability of implantable defibrillators (automatic implantable cardiac defibrillators [AICDs]; pacemakers) in the chest of patients who have known coronary artery disease. These patients are prone to dysrhythmias and may have episodes of VT and VF that are automatically detected and defibrillated or cardioverted. However, these devices can malfunction. If these patients present in SCA/VF, defibrillation should be performed as indicated. Presence of an AICD or pacemaker is NOT a contraindication for defibrillation. The only caveat is to avoid placement of the defibrillation paddles over the AICD or pacemaker because the current for defibrillation may be redirected away from the fibrillating myocardium and compromise termination of VF. In addition, because current from the defibrillation could enter the AICD or pacemaker, the device could be prone to future malfunction. These devices should be reevaluated after the patient has been defibrillated. Current trends in fashion sometimes include piercing of the body in various locations. In addition, certain items of clothing and jewelry may require modification of electrode or paddle placement. The presence of metals in locations proximal to the heart or in locations on the chest should be avoided to minimize the potential for diverting the defibrillating current from the myocardium. Also, if the metal object provides a potential short circuit from the patient or leads to “ground,” this object should be removed, if feasible, to avoid current diversion from the myocardium or the possibility of arcing and burns across the chest. However, the presence of these materials is NOT a contraindication to defibrillation. In this part of the chapter on defibrillation the recommendations are intended for application to adult (>8 yr or weighing > 25 kg [55 lb]) victims of SCA/VF. If a patient is a child (e.g., 1–8 yr of age or weighing < 25 kg [55 lb]), modifications in the sequence, defibrillation energy, energy attenuation equipment, and size of defibrillation paddles must be addressed. Pediatric defibrillation details are discussed later in this chapter. If a defibrillator or AED and equipment suitable for use in children are not available, the health care provider can resort to use of a standard AED or defibrillator. Use of AEDs or defibrillation in infants less than 1 year of age is neither indicated nor contraindicated. Therefore, the age of the patient is not a contraindication, but certain modifications in the resuscitation must be considered. Pediatric defibrillation is addressed later. As obvious from the text, defibrillation may have to be performed in various areas. Defibrillation can be an ignition source for explosion if any arcing occurs or if there are any stray or aberrant electrical discharges that occur as a result of the paddle or electrode discharge. Therefore, care should be taken to avoid defibrillation or to ensure that electrical conductivity through the patient’s chest is optimal during defibrillation in an environment in which volatile explosive materials are present, such as the operating room or other areas of critical care. Some of the materials to avoid are anesthetic agents and oxygen. Therefore, a potentially explosive environment is a relative contraindication to defibrillation.26

12 

If the victim of VF/pulseless VT is a pregnant female, treatment of the mother is critical. Therefore, prompt defibrillation is indicated, using the same guidelines and sequencing as for the nonpregnant patient.25 No harm to the fetus has been reported as a result of defibrillation. Therefore, pregnancy is not a contraindication to defibrillation. New guidelines suggest that only one shock be administered after identification of SCA/VF.6 After defibrillation, CPR should be resumed immediately and continued for about 2 minutes or 5 cycles, without interruptions for rhythm evaluation or pulse checks. Hence, the pulse and rhythm checks are delayed, and subsequent shocks are not recommended, but contraindicated. Administering additional shocks may indeed precipitate PEA or asystole. The administration of CPR immediately after defibrillation is indicated to enhance the mechanical function of the “stunned” heart in the immediate postdefibrillation period. This step may be modified at the discretion of the clinician in the hospital setting because of the presence of mitigating circumstances or the availability of hemodynamic monitoring devices such as Doppler devices, arterial pressure lines, or central venous pressure (CVP) monitoring. Prior recommendations suggested delivering a “stacked” sequence of up to three shocks without interposed chest compressions if the first shock was unsuccessful in terminating VF. This was done to decrease transthoracic impedance with the monophasic damped sinusoidal (MDS) defibrillators in use (discussed later) and to deliver more current to the myocardium. However, this recommendation has been rescinded owing to lack of supporting evidence. Now with the higher first-shock efficacy (90%) in successfully terminating VF (termination of VF for 5 sec) through the use of biphasic defibrillators,7 the recommendation to repeat a shock if the first treatment was unsuccessful is harder to justify. Furthermore, as previously mentioned, the post shock rhythms often observed after defibrillation were either asystole or PEA. These rhythms require immediate CPR as part of their treatment. Therefore, additional or multiple shocks are not recommended and are contraindicated. However, immediate CPR is recommended. Defibrillation is also an effective treatment modality used to terminate pulseless VT. If the patient has a pulse, is stable, and has a perfusing rhythm while in VT, defibrillation is contraindicated. However, if the patient in VT becomes unstable or develops signs of poor perfusion, change in mental status, or persistent chest pain with pulmonary edema and hypotension and subsequent shock, then synchronized cardioversion is recommended. This procedure is addressed later. If the patient becomes unstable as a result of polymorphic VT, or becomes pulseless during the episode of VT, an unsynchronized shock (i.e., defibrillation) is indicated. Patients found “down” or who have just become unresponsive can have other “rhythms present” beside VF or pulseless VT (e.g., PEA or asystole). Defibrillation is contraindicated in PEA. True asystole is NOT a shockable rhythm, and current evidence suggests that defibrillating “occult” or false asystole is not beneficial and may actually be harmful. Patients who are found unresponsive, but who have a pulse, should obviously have their cardiac rhythm determined. However, any patient who has a pulse, fast or slow, should NOT receive defibrillation. Some patients who succumb to SCA may have various medication-releasing patches (e.g., nitroglycerin, contraceptive hormones, antihypertensive agents, smoking-cessation

215

CARDIAC PROCEDURES ● 

III 

216

When performing defibrillation, care should be taken to avoid excessive moisture on the chest or around the patient. It is unlikely that there will be any significant or dangerous current leaks from the patient onto a wet floor; however, care should be taken to avoid creating an electrical hazard. Try to ensure that the area is not wet. Thus, a wet surface is not an absolute contraindication to defibrillation. Defibrillation can be performed on ice and wet pavements. With a high index of suspicion for the presence of VF in an unresponsive patient in early SCA, “blind” defibrillation can be life saving or it can make things worse, especially if the underlying rhythm is not determined prior to defibrillation. If the patient is indeed in VF, providing prompt early defibrillation may indeed help. If the underlying rhythm is NOT a shockable rhythm such as PEA/asystole, the resuscitation could be compromised. Shocking a patient in PEA can bring about asystole. If the patient is in asystole, things may be made worse by damaging the myocardium with electrical current, which will increase the possibility of an intraventricular conduction delay. Therefore, blind defibrillation is not recommended.27 However, data to support or refute this are sparse. Lastly, the defibrillation of “occult” or “false” asystole or very fine VF not detectable owing to paddle or electrode position may be considered, but is not recommended.27 Fine VF can occasionally masquerade as ventricular standstill or asystole. This may be a function of perpendicular electrode orientation with respect to the wavefront of depolarization. When evaluating the rhythm of a patient if there is any doubt or confusion regarding the type of rhythm present, the operator should make sure that several leads are checked or that the paddles are rotated 90° from their original position to ensure that asystole is indeed present before abandoning the possibility of defibrillation. If fine VF is unmasked, provision of aggressive CPR prior to defibrillation should be considered. Also, the controls on the ECG monitor should be placed on maximal gain to ensure adequate amplification of weak signals.

TABLE 12–1  Defibrillation Equipment List of Materials for Defibrillation • Defibrillator/ECG monitor • Handheld defibrillation electrodes “quick-look” paddles • Patient interface cables; multifunctional for ECG monitoring and defibrillation • Electrodes and pads for ECG signal acquisition and defibrillation • Conductive gel (not ultrasound gel) Additional “Equipment” (Pertinent to VF/VT)* ACLS Medications • Epinephrine • Vasopressin • Amiodarone • Lidocaine • Magnesium sulfate • Procainamide • Atropine Miscellaneous • IV access equipment, central line kits, and the like *List of suggested equipment and medications for a code cart. ACLS, advanced cardiac life support; ECG, electrocardiographic; VF, ventricular fibrillation; VT, ventricular tachycardia.

DEFIBRILLATION EQUIPMENT Preparation Sudden cardiac death usually occurs without advanced warning and requires prompt action. To ensure an expedient response and timely intervention, it is advisable to have all of the equipment listed in Table 12–1 easily accessible, on a prearranged tray, and preferably placed on a mobile, cartlike device (Fig. 12–3) in the anticipated order of use. This equipment should always be kept in a constant state of readiness. Wherever possible, the defibrillator (Fig. 12–4), patient cables, quick-look electrode paddles (Fig. 12–5), and ECG and defibrillation electrodes and pads (Fig. 12–6) should be preconnected and labeled to facilitate application to the patient. This will provide easy access to and prompt utilization of the equipment during the resuscitation. Furthermore, members of the designated resuscitation team should check the equipment at the beginning of their clinical shifts to ensure that the equipment is fully operational and that all of the components of the cart are present and/or restocked from any prior usage. Having a list of required equipment and a log of when the cart and equipment are checked is helpful. Intrinsic to the variability of the workplace is the potential of the health care team to encounter different types of

Figure 12–3  “Code cart” with defibrillation equipment.

defibrillation equipment. Input from biomedical research into the development and design of better, more effective defibrillators, capable of achieving better success with the first shock, along with competition from various manufacturers trying to increase their percentage of the health care market share by making defibrillator/monitor units more user-friendly and capable of more functions has provided a significant number of variations in defibrillator design and configuration.

Defibrillation and cardioversion

Defibrillators Central to the procedure of defibrillation is the defibrillator. To better understand the relevance and function of the various types of defibrillators and the significance of the controls, options, and settings, a brief discussion describing the basic components of a defibrillator is provided. Defibrillator units come in various configurations, depending on the manufacturer’s design (Figs. 12–7 and 12–8; see also Fig. 12–4). Most defibrillators manufactured today are also cardiac monitors capable of displaying several ECG leads (see Figs. 12–4 and 12–7). Some units can display and provide a hard copy of a 12-lead electrocardiogram (Fig. 12–9). In addition, many defibrillator/monitors have rate alarms, computer microchips, and software to detect certain arrhythmias and the possibility of acute MI. Most defibrillator/monitors also have the capability of performing car­ dioversion, external pacing, pulse oximetry, telemetry, and sphygmomanometry. A comprehensive review of the various options and varieties of defibrillator monitors is beyond the scope of this chapter. Therefore, the discussion provided

Figure 12–5  “Quick-look” paddles.

Figure 12–6  Multifunctional defibrillation pads/electrocardiographic (ECG) electrodes.

● 

Figure 12–4  Defibrillator with paddles preconnected/paper ready.

12 

Although the basic functions and operation of all defibrillators currently in clinical use are similar, and all of these defibrillators are capable of being used for defibrillation, the layout of the controls and connections may vary significantly and cause unnecessary delays in time to first shock as the team determines how to operate the defibrillator during the resuscitation. Perceived malfunctions of equipment are often due to the operator’s misuse of, or inappropriate operation of, the defibrillator controls and connections. Furthermore, differences in the type of defibrillator (e.g., waveform characteristics, discussed later) may affect the actual amount of energy delivered to the myocardium, causing unnecessary damage to the heart or causing a nonperfusing, postshock dysrhythmia if the operator is not fully aware of the equipment characteristics. Therefore, the clinician and resuscitation team or code team should become thoroughly familiar with the operation and type of equipment available in their designated patient care area. Choices and decisions made during the resuscitation may be affected by the type of equipment being used at the time. This should be done before the need to use the equipment suddenly arises. Remember, the longer VF persists, the harder it is to defibrillate.

Figure 12–7  Multifunction defibrillator/monitor.

217

CARDIAC PROCEDURES

Charging switch

● 

Battery (or means of power) provides the energy to charge the capacitor

Capacitorfor rapid discharge (as opposed to a battery)

Charging switch

III 

Inductor ensures that the electric pulse has an optimal shape and duration for rhythm restoration

Discharge switch Figure 12–10  Defibrillator/patient circuit.

Figure 12–8  Automatic external defibrillator (AED).

218

Figure 12–9  Defibrillator monitor capable of 12-lead ECG/ cardioversion/pacing/limited ECG interpretation.

herein is limited to elements relevant to defibrillation and cardioversion (later in the chapter).

Basics of Defibrillator Function and Operation (Basic Controls/Switches/Components) Defibrillation As stated previously, defibrillation entails delivering a brief burst of current across the myocardium for the purpose of terminating VF. To perform this function, a source of energy or current and a means of producing a controlled release of this current are needed. The defibrillator is the device that serves this purpose. The defibrillator serves as the energy source for the defibrillation/cardioversion. A diagram of the patient-defibrillator circuit (Fig. 12–10) is shown to provide a visual reference regarding the components under discussion. An AC power source provides electricity to an internal transformer and power supply. The transformer and power supply convert the incoming energy to DC and store the energy in a battery. The battery holds power within the unit, even when the defibrillator is unplugged from the AC source, rendering the device portable. Some devices have a backup battery and a selector switch to provide the option of switching between alternate batteries, if the primary battery fails. When defibril-

Chest impedence/ DC resistance

lation is to be performed, controls are set by the operator, which initiates a chain of events. First, the defibrillator monitor unit must be turned on, using the appropriate selector switch or rotating dial. The electrode pads (handheld quick-look or multifunctional electrode pads) must be appropriately placed onto the chest wall of the patient, thus interfacing the patient into the defibrillator-patient circuit. When the decision is made to defibrillate based on an assessment of the electrocardiogram, an energy level (in joules) for defibrillation is selected by the operator. This switch selects how much energy will be released during the shock. A charge switch is then activated to initiate current flow from the battery onto the capacitor. This action causes the capacitor to become charged. A capacitor stores a large amount of energy in the form of separated charges. When the shock controls are activated, electronic components and/or computer software select the appropriate pathway (e.g., a combination of resistors, inductors, capacitors, circuits, and switches) that allows release of a therapeutic amount of current into the paddles or electrodes that are interfaced with the patient. These internal controls affect the amount of energy (volts/joules/amplitude), the duration of current release (time span over which current flows through the electrodes in milliseconds), and the polarity (monophasic or biphasic) or direction that the current travels through the chest wall across the myocardium and between the electrodes or paddles. Cardioversion Basics Most defibrillator/monitors also have ECG detection and display devices incorporated into the device. This permits the operator to analyze the patient’s rhythm and decide whether a shockable rhythm requiring intervention is present. Furthermore, the ECG device is interfaced with the defibrillator so that when shock delivery needs to be performed or synchronized at a specific time point or phase of the ECG event (e.g., during the absolute refractory period of the cardiac cycle [Fig. 12–11], discussed later), as in cardioversion, software and electronic components can be preset to discharge energy to the electrodes or paddles at the appropriate time. A selector switch designed to select the mode for current discharge is present on the defibrillator/monitor unit that allows the operator to either engage or disengage the ECG device from the defibrillator output circuit. This switch allows control of when the current will be delivered. Thus, when the health care provider has decided to perform defibrillation, the device is placed in the unsynchronized mode, so that the current is discharged independent of the ECG device. The current is then discharged when the operator depresses the “shock/

Defibrillators currently in use are classified by the type of shock waveforms that they produce. There are two general types of waveforms, monophasic and biphasic. These waveforms can be further classified by the rate of current drop to baseline or the actual shape/polarity of the waveform: MDS (Fig. 12–12 left), monophasic truncated exponential (MTE) (see Fig. 12–12 right), biphasic truncated exponential (BTE) (Fig. 12–13), or biphasic rectilinear (see Fig. 12–13). Several figures have been included for comparison of the waveforms (Fig. 12–14). Although no specific waveform has been proved to be superior to another regarding survival from SCA or for the return of spontaneous circulation, biphasic waveforms have been shown to be more efficient in achieving first-shock termination of VF than monophasic waveforms.7 Of note, the amount of energy actually delivered to the myocardium by a specific waveform can differ from the actual selected

Absolute Relative Figure 12–11  Absolute refractory period of the cardiac cycle. MONOPHASIC DAMPED SINUSOIDAL WAVEFORM (MDS)

MONOPHASIC TRUNCATED EXPONENTIAL WAVEFORM (MTE)

40

Current (amps)

Current (amps)

30 20 10 0 –10 –20

0 1 2 3 4 5 6 7 8 9 10 11 12

25 20 15 10 5 0 –5 –10 –15

Figure 12–12  Monophasic damped sinusoidal (MDS; left) and monophasic truncated exponential (MTE; right) waveforms.

219

0

5

10

15

20

25

30

Time (msec)

Time (msec) BIPHASIC TRUNCATED EXPONENTIAL

50 40 30 20 10 0 –10 –20

0 4 150 Joules at 50 Ohms

0

4

8

Current (amps)

Current (amps)

RECTILINEAR BIPHASIC 50 40 30 20 10 0 –10 –20

12

Time (msec)

Figure 12–13  Biphasic waveforms.

150 Joules at 50 Ohms

0

4

8

12

Time (msec)

WAVEFORMS FOR EXTERNAL DEFIBRILLATION 2 kV

0 –2

6 msec

Edmark

Lown

Sinusoidal

Monophasic Biphasic Monophasic Biphasic Triphasic Rectangular

Defibrillation and cardioversion

Defibrillator Types

● 

peak of the R-wave of the QRS. This synchronization is done by the defibrillator/cardioversion circuitry. This procedure is discussed in more detail in the “Cardioversion” section of the chapter.

12 

discharge” buttons on either the paddles or the defibrillator unit itself. If cardioversion is to be performed, the operator will select the synchronized mode or “sync” setting for current discharge, placing the ECG component into the discharge circuit. This will result in current being released during the

Truncated exponential

Figure 12–14  Waveforms for external defibrillation.

CARDIAC PROCEDURES ● 

III 

220

energy level. Therefore, acquiring familiarity with the type of waveform that a particular defibrillator produces may be useful when it comes time to make the energy selection for defibrillation. A brief overview of the various waveforms is provided. Defibrillators can also be categorized or described by their operational characteristics. Defibrillators can be described as manual, semiautomated, and fully automatic, based on their operational characteristics. Biphasic Defibrillators/Energy Selection Defibrillators manufactured today are primarily biphasic. These devices produce an output waveform that flows back and forth between the electrodes, sending the current sequentially in both directions. These devices produce either a BTE (see Fig. 12–13) waveform or a biphasic rectilinear waveform (see Fig. 12–13). A brief description of the various types of waveforms and the ascribed significance are provided for the curious reader interested in further elaboration of defibrillator wave types. These devices have been found to deliver a more successful defibrillation shock using less energy and have been found to have a better first-shock defibrillation success.7 Although an optimal energy level for first-shock biphasic waveform defibrillation that will provide the most effective termination rate for VF has not been established, several studies have demonstrated that using relatively low energy of 200 J or less is not harmful and is as effective, if not more so, than monophasic waveform shocks using higher energy (360 J).28 Recommendations extrapolated from these studies suggest that it is reasonable to use the lowest energy range for the termination of VF that has been shown to be effective for a given type of defibrillator. This is termed the device-specific effective waveform dose range. Therefore, when using a defibrillator that produces a biphasic rectilinear waveform, energy levels of 120 J are appropriate; if the defibrillator produces a BTE waveform, energy ranges of 150 to 200 J are recommended. If the operator is unaware of the devicespecific effective waveform dose range, a consensus default of 200 J should be used for the initial shock. Use of the least amount of energy possible to accomplish termination of VF is currently speculated to cause less myocardial damage and precipitate fewer nonperfusing rhythms after shock delivery. These biphasic defibrillators are replacing the older monophasic defibrillators. Monophasic Defibrillators/Energy Selection Monophasic defibrillators were the first type introduced (generally). These devices are still available in various patient care settings, but are being phased out of production by manufacturers. Hospitals and EMS systems are replacing these defibrillators with the newer biphasic type. These defibrillators were designed to produce an MDS (see Fig. 12–12) waveform for defibrillation. When using this type of defibrillator, expert consensus suggests that an energy level of 360 J be used for the first shock.6 These defibrillators are still capable of performing a successful defibrillation and have not been deemed inferior.

Waveforms Until recently, modern defibrillators put out what was termed a damped, monophasic, half-sinusoidal waveform or a trapezoidal truncated exponential decay (voltage falling instantaneously) waveform, further explained later. The trapezoidal waveform

was modified to resemble a square waveform. The more square the waveform, the more effective it was for experimental defibrillation.29 In a comparison of square waveforms and damped half-sinusoidal waveforms (voltage falling to zero gradually) for animal defibrillation, it was found that less peak current per kilogram was needed with the square waveforms, although the average current levels were equivalent.30 This work was extended further, leading to the development of multiple new biphasic waveforms. These waveforms are achieved by manipulating the current (amperes), amplitude, duration, voltage, and ultimately, the energy delivered to the myocardium. In light of the current research into the utility of the various energy waveforms that modern-day defibrillators deliver, it would be relatively useful to briefly describe the terminology. Current output from a defibrillator is graphed with respect to time on an x-y cartesian plot. The form of the wave can be either monophasic or biphasic. A characteristic monophasic wave is described as a rapid positive unidirectional increase in current flow to a predetermined peak with a return to baseline. If the return of the current to baseline is gradual, the waveform is termed a damped waveform (MDS waveform). These waves often resemble a sine wave (see Fig. 12–12). If the return of the current level to baseline is paroxysmal or sudden, the wave is an MTE waveform (see Fig. 12–12). A rapid rise in current with respect to time with a slight plateau and then a subsequent paroxysmal or sudden reversal in current flow at a predetermined time until all of the energy is delivered with a return to baseline is termed a biphasic waveform because of the two phases—positive and negative—in current flow. Essentially, for a biphasic waveform to occur, current travels from one pad or paddle to the other, then a reversal occurs so that current now flows from the second pad or paddle to the first. If the polarity or direction of the current flow is gradual, the wave is termed a damped waveform. If the current reversal is abrupt, the waveform is deemed a truncated exponential waveform; hence, the term biphasic truncated exponential waveform (BTE) (see Fig. 12–13). The highest current flow attained is termed the peak energy delivered. This, however, is not synonymous with the total amount of energy delivered. Energy is delivered throughout the duration or period of the wave. The current thinking is that if there is less peak energy and a smaller amount of energy delivered to the fibrillating myocytes, there will be less damage to the heart tissue. In addition, this may decrease the perpetuation of conditions favoring VF. This suggests that these waveforms will enhance defibrillation efficacy, decrease myocardial damage, and decrease postdefibrillation arrhythmias.31,32 The first biphasic AED approved by the U.S. Food and Drug Administration utilized a BTE waveform.33 Additional experiments are being done to further explore various biphasic rectilinear first-pulse waveforms. The motive behind these modifications is to find an optimal waveform that will deliver the least amount of energy to the myocardium, thus decreasing the structural damage to the myocytes34 while achieving successful defibrillation.24,35,36 Injury to myocardial tissue has been associated with the peak current, not the amount of energy actually delivered to the myocytes.37 With the biphasic defibrillators, lower energy levels (150–175 J) can be used without escalating the energy up to 300 or 360 J. Experimental findings suggest that the clinical outcomes of these defibrillations are equivalent to those that used the escalating

Fully automatic defibrillators require only that the operator turn the device on and connect the electrodes to the patient. Subsequently, the defibrillator analyzes and shocks the patient, if indicated, without any further operator input, based on preprogrammed or hard-wired programs. These devices are usually provided for families of patients with known unstable cardiac arrhythmias. It is doubtful that the health care professional will encounter this type of device in a clinical or patient care setting. The only other fully automatic”defibrillators are the implantable defibrillator/pacer units or AICDs. These devices are discussed in Chapter 13 of this text.

Manual defibrillator/monitors are the most likely type to be found in many clinical settings. However, some of the manual defibrillators recently manufactured now have rhythmrecognition capabilities and can be configured to function as an AED with the turn of a switch or dial. This is especially helpful when there is nobody qualified present to interpret the rhythm when SCA happens in a noncritical area of the hospital. This facilitates the performance of defibrillation in various noncritical patient care areas without the presence of a clinician. Manual defibrillators require that the operator turn on the device, select the input (e.g., the quick-look paddles or the patient ECG electrodes), place the quick-look paddles or the multifunctional ECG or defibrillation pads onto the patient’s chest, and determine the type of rhythm present (e.g., VF, pulseless VT). Subsequently, the operator must select the appropriate energy level for the particular patient, based on the type of defibrillator available (monophasic or biphasic), charge the capacitor, and deliver the shock by activating the appropriate controls. If the required action is cardioversion, the operator will need to execute the steps as described previously. However, the SYNC switch needs to be activated (MODE/SYNC) in order to perform a synchronized shock. Thus, the manual defibrillator requires the setting of several controls—the input selector, the energy level, the charge button— and checking of the mode switch, followed by delivery of the shock by depressing the “Shock” controls.

Fully Automatic Defibrillators

Pads/Electrodes The patient is placed into the defibrillation circuit or interfaced with the defibrillator/monitor via the placement of handheld paddle electrodes or self-adhesive pad electrodes onto the chest (Fig. 12–15; see also Figs. 12–5 and 12–6) and/or back (Fig. 12–16). Most, if not all, defibrillators currently have multifunctional, handheld or insulated paddles with several controls located on the handles, which can acquire the electrocardiogram and be used to deliver the defibrillation shock. The controls usually provided are for setting energy levels for the shock, a charge button for charging the capacitor remotely from the patient’s side, and two “shock” buttons,

Semiautomated Defibrillators Semiautomated defibrillators or automated AEDs require that the operator turn the device on, and follow the voice and/or visual prompts provided by the device (e.g., “attach electrodes to the patient’s bare chest,” to “press the analyze button” on the defibrillator to initiate analysis of the rhythm, and then to either “press the shock button” or “initiate CPR” as directed by the defibrillator). Actual operation of the AED requires fewer steps and decisions by the operator. Once the AED is turned on, either the operator must place the electrodes in the appropriate position on the victim’s chest and then allow the unit to analyze the underlying ECG rhythm, which was triggered by completing the patient-defibrillator circuit when the electrodes were applied to the patient’s chest or the operator will be required to or prompted to depress the “analyze” switch. The AED will then determine whether there is a shockable rhythm present. If so the operator will be prompted to “clear” the patient for defibrillation and then she

Defibrillation and cardioversion

Manual Defibrillators

● 

or he will depress the shock button, delivering the defibrillatory shock to the patient. The rhythm analysis and choice of energy are done by the preprogrammed electronics intrinsic to the device. In summary, the operator need only depress the analyze switch, if specific to this type of AED, and if applicable, the operator presses the “shock” button if defibrillation is indicated (i.e., a shockable rhythm is present). Some of these devices have an optional “override” control in the module to allow a clinician or other healthcare provider to change the sequence of operation, energy levels, and so on. (These details are device-specific and can be found in the operator’s manual of the particular defibrillator in use.)

12 

monophasic shocks.38 However, the actual delivered energy is dependent on thoracic resistance/transthoracic impedance. Currently manufactured defibrillators are capable of determining transthoracic impedance/resistance and can actually modify the waveform and thus the amount of energy delivered across the myocardium. A more detailed discussion of transthoracic impedance is presented later in this chapter. In current clinical practice, there is little clinical difference in the effectiveness of the currently available waveforms. The trend of the future will probably be to use biphasic, impedance compensating defibrillators. Currently, monophasic defibrillators are still in use in some locations. There is no evidence to exclusively support the use of one defibrillator waveform over the other at this time.

Figure 12–15  Use of quick-look paddle electrodes for rhythm (ECG) determination and defibrillation.

221

III 

● 

CARDIAC PROCEDURES

Figure 12–16  Front/back position of electrodes on patient (alternate position).

222

Figure 12–17  Pediatric paddles/adapters for use over adult quick-look paddles.

one on each paddle, to actually deliver the shock. (Both buttons usually have to be depressed to deliver the shock.) (Some defibrillators also have “event-marker buttons,” which allow the operator to mark an event such as administration of medication during the resuscitation.) The paddles may have adapters on them (Fig. 12–17) for the purpose of pediatric defibrillation and the electrodes may vary in size to allow use in the pediatric patient39 (discussed in the “Pediatric Equipment” section). When using the handheld paddle electrodes for adult defibrillation, use of the 12cm-diameter paddles, if available, is recommended.40 However, pad electrodes or electrodes ranging in size from 8 to 12 cm in diameter have been shown to perform well. The goal is to use the larger paddles for the purpose of decreasing resistance or impedance at the chest wall and for providing an optimal current density across the myocardium that will be successful in terminating VF while causing minimal to no significant myocardial damage from excess current.41 To this end, use the largest available pads or paddles that will fit on the chest wall without overlap.41 Try to avoid using paddles that are too small in diameter, because they may cause myocardial necrosis.42

Figure 12–18  A 250-g tube of conductive gel for defibrillation.

Conductive Materials Use of conductive materials (Fig. 12–18) is important to lower the impedance or resistance to current flow at the electrode– chest wall interface. High impedance or resistance to current flow can compromise the amount of current actually delivered to the myocardium, leading to a failed first shock. Inappropriate use of conductive material can lead to current bridging or a short circuit and arcing of electrical current. This can cause spark production and unnecessary burns on the patient’s skin. In addition, the arcing of electricity can become a possible explosion hazard, depending on the circumstances. These conductive materials need to be used when using the handheld electrodes. There are various electrode gels on the market; these should be kept in the proximity of the defibrillator, on the prearranged cart ready to use (see Fig. 12–3) (Do not use ultrasound gel!!). The mean range of impedance across the human chest wall varies from 70 to 80 ohms,43–45 with bare skin contact yielding an impedance value of approximately 91 ± 20 ohms. With the use of saline-soaked pads, the impedance decreases to about 71 ± 11 ohms and with the conductive gel to 64 ±

CPR The victim’s airway should be opened by using the head-tilt/ chin-lift method. If cervical spine injury is suspected, the jawthrust maneuver should be used. Then, using the “look, listen, and feel” maneuver, the patient’s airway should be evaluated by the EC/HCP. This is done by the rescuer placing her or his ear close to the victim’s mouth and nose while looking down at the patient’s chest to assess whether the patient is breathing. If there is no chest wall movement and no airflow is perceived to be coming through the mouth or nose, the patient is deemed to be apneic. Auscultating the patient with a stethoscope may be attempted to verify lack of air movement in the airways. If there is no evidence of breathing, ventilation must be initiated. The patient is then slowly ventilated, delivering two breaths using either a bag-valve mask (BVM) or some type of barrier device. The ventilations are interposed over 2 seconds, providing one breath/sec. The EC/HCP doing the ventilation should proceed carefully, observing for modest chest wall rise and fall, so as not to overinflate the thorax. Hyperinflation of the chest can lead to inadvertent pressurization of the esophagus, causing the lower esophageal sphincter pressures to be exceeded. This can lead to retro-

Defibrillation and cardioversion

When confronted with a patient who has just become unresponsive, the emergency clinician (EC) or health care provider (HCP) should summon assistance from the ED team and follow guidelines for universal precautions. The defibrillation equipment should expediently be brought to the patient’s side and preparation for immediate defibrillation should begin. As soon as the defibrillator is available and the patient is connected to the monitor, rhythm assessment should begin. In the interim, as the defibrillation equipment is being turned on and the paddles or electrodes are being readied for placement on the chest, the EC/HCP should begin assessment of the patient and initiate the steps of CPR by applying the ABCs. If more than one person/HCP is present, several tasks can be performed simultaneously. The victim can be assessed by one rescuer as the equipment is readied by others and preparations are made to initiate resuscitation. If enough staff members/resuscitation team members are available, one staff member can prepare to ventilate the patient, while another positions himself or herself for possible chest compressions.

Rhythm Assessment Once the defibrillator is at the bedside, the defibrillator/ monitor should be turned on and the electrodes should be placed on the patient’s chest using either the quick-look paddles or the multifunctional electrode pads that can acquire ECG signals and be used concomitantly to defibrillate the patient. The handheld electrode paddles should have a conductive material such as a gel applied to the contact surface of the electrodes to decrease chest wall impedance. Care should be taken to avoid streaking the electrode gel or comparable material across the chest because this could cause electrical arcing, sparks, burns, and an electrical short circuit. This would be unfavorable for the patient and also compromise the efficacy of the defibrillation current across the myocardium, leading to an unsuccessful attempt at defibrillation. The gel is not indicated when using the multifunctional electrode pads on the chest. The correct position for the placement of either the handheld quick-look paddle electrodes (see Fig. 12–15) or the self-adhesive pads (see Fig. 12–16) is illustrated in the depicted figures. Often, the pads are labeled with a diagram to help in placing the electrodes onto the chest wall (see Fig. 12–6). Using the patient’s right side for orientation, the sternal electrode is placed just below the clavicle, just to the right of the sternum. The apical electrode is placed in the midaxillary line around the fifth or sixth intercostal space (Fig. 12–20). This approximates the lead 2 position for ECG acquisition. Once the electrodes or pads are in position, the selector dial or switch on the defibrillator monitor should be set to the appropriate position to acquire the ECG signal from the correct input source—either the handheld quick-look paddles or the multifunctional electrode pads. Errors sometimes occur when the selector switch is in the position for the patient cable and electrode pads while the operator is attempting to use the handheld paddles. (This could lead to misinterpretation of the rhythm, in which the operator perceives that the patient is in asystole, while in reality VF, pulseless VT, or some other rhythm is actually present.) (Be familiar with the operation of the switches!) In addition, the controls for gain of ECG signal should be adjusted to increase the sensitivity or gain of the ECG amplifier to ensure that fine VF is not interpreted as asystole. As the ECG rhythm appears on the monitor, a diagnosis of the type of rhythm, or lack thereof, should be made. If there is a shockable rhythm (i.e., VF or pulseless VT), defibrillation is indicated and the operator should proceed to select the appropriate energy level for the anticipated defibrillation.

● 

PROCEDURE Witnessed SCA (Fig. 12–19)

grade flow of gastric contents into the esophagus, with the potential for subsequent aspiration of acid and debris into the trachea, if the airway is not adequately protected. Next circulatory status must be determined. A quick attempt (≤10 sec) to check for a carotid pulse in the unresponsive, pulseless victim should be made. If the provider does not definitively feel a pulse, CPR should be initiated. If a pulse is absolutely present, then ventilation at 10 to 12 breaths/min or 1 breath every 5 to 6 seconds should be provided. If there is no pulse, start compressions immediately. The compression to ventilation ratio should be 30 compressions for every 2 ventilations. The rate of chest compressions should be 100 compressions/min or more. The rescuer performing the compressions should push hard and push fast. Every attempt should be made to minimize interruptions of compressions. Continue until the defibrillator is available.

12 

15 ohms. Multiple factors affect this range of impedance (e.g., body weight, chest size, chest hair, moisture on the skin surface of the patient, paddle size [diameter], paddle contact pressure, phase of respiration, and type of conductive material used). The number of serial shocks delivered and the interval between shocks were believed to affect chest wall impedance. However, with the use of and application of biphasic defibrillation waveforms, the clinical significance of this issue has been abandoned.46 Self-adhesive pad electrodes now have a resistancereducing, conductive material incorporated into the adhesive, rendering the use of a gel or other conductive material unnecessary. Firmly applying the self-adhesive electrode pads to the skin will usually be sufficient enough to minimize the impedance and allow adequate ECG acquisition and, if indicated, defibrillation.

223

CARDIAC PROCEDURES ● 

III 

UNRESPONSIVE PATIENT/WITNESSED SCA/VF/PULSELESS VT ADULT Determine Unresponsiveness Call for Help/Assistance Get Defibrillator/AED* to Patient’s side Start ABC’s

Open Airway Determine Breathlessness Ventilate Patient × 2 (1 breath/second) slowly

Check pulse ( 0.14 sec • Left axis deviation • AV dissociation • Certain configurational characteristics of the QRS morphology From Wellens HJ, Bär FW, Lie KI: The value of the electrocardiogram in the differential diagnosis of a tachycardia with a widened QRS complex. Am J Med 64:27, 1978; and Zipes DP: 1997.

Special Considerations: Wide-QRS-Complex Tachycardias Wide-complex tachycardias (wide-complex SVT) are diagnostic challenges in clinical medicine. The criterion often used to define wide-complex supraventricular tachycardia (WCSVT) is a tachycardia with a QRS duration of greater than 0.12 second. It is important to differentiate the rhythm as one of the following: VT, SVT with aberrancy (left bundle branch block [LBBB] or right bundle branch block [RBBB]), or an accessory AV pathway (“pre-excitation”). The need for a proper diagnosis is obvious. Incorrect diagnosis and inappropriate treatment can be life threatening. This is especially true in misdiagnosing VT as SVT. Studies have shown that VT is often misdiagnosed, even with the ready availability of clinical and ECG criteria.82 236

Etiology Normally, ventricular depolarization is initiated when the His bundle depolarizes both ventricles simultaneously through the bundle branches and Purkinje fibers. Normal depolarization takes place within 80 to 120 msec. Prolongation of the QRS duration happens (1) if the ventricles are activated sequentially rather than simultaneously as is the case in VT, bundle branch blocks, or accessory pathway ventricular activation (WPW, Lown-Gagong-Levine [LGL]); or (2) when His-Purkinje-myocardium conduction is slowed from ischemia, drugs, or electrolyte disturbances. Some dysrhythmias occur as a result of re-entry. There is an area of delayed conduction, due to metabolic changes in the tissue causing the depolarization front to travel at a slower speed through the myocardium. If an area proximal to the myocardium with the delayed conduction becomes reactivated, especially if there is a unidirectional block, a circular rhythm becomes established, causing the continuous firing of new ectopic focus. Classification Wide-complex tachycardias (QRS ≥ 0.12 sec) fall into three classifications based on mechanism: (1) VT, (2) SVT with aberrancy, and (3) pre-excited tachycardia. VT is the most common cause of wide-QRS-complex tachycardias. It is defined as three or more consecutive ventricular beats at a rate of 100 beats/min. VT is further classified as nonsustained (tachycardia lasting < 30 sec) or sustained (tachycardia that lasts > 30 sec). Sustained tachycardia usually results in hypotension or syncope and requires termination intervention. SVT is a tachydysrthymia using the normal AV conduction system for ventricular activation. This tachycardia originates in the SA or AV node. To sustain propagation, the AV

node is recruited. Aberrancy refers to the existence of an aberrant or nontraditional conduction mechanism resulting in a longer depolarization phase. SVTs with aberrancy by definition must result in wide-QRS-complex tachycardias. The two forms of aberrancy are fixed: a permanent bundle branch block or a functional block, which is a rate-dependent bundle branch block. The most common areas of the His-Purkinje functional block are in the left or right bundle branches. Sudden acceleration is often the initiating cause of the SVT with aberrancy. The aberrancy is maintained by a continuous, concealed, retrograde conduction pathway, which leads back into the blocked area. For pre-excitation wide-QRS-complex tachycardias, AV conduction occurs over two circuits: (1) normal AV nodal conduction or (2) through an accessory pathway. These two pathways create the needed circuits for a re-entry circuit: the circus movement tachycardias. Aberrancy appears because of the presence of intraventricular conduction block. Pre-excited tachycardia is any tachycardia in which the ventricles are antegradely activated over an accessory pathway. The most common pre-excited tachycardia is atrial fibrillation with ventricular activation over an accessory pathway. Clinical Diagnosis The directed history on presentation may provide the most valuable clues to the diagnosis of tachydysrhythmia. The onand-off occurrence of tachydysrhythmia in the past, the age of the patient, and the age of past occurrence all are important indicators of pre-excitation rhythms usually found in young patients. Sudden onset of tachydysrhythmia in the older coronary-prone patient or a patient with structural heart disease points more toward VTs. Symptoms associated with the tachydysrhythmia are important clues in diagnosis. The young patient often has few if any symptoms when experiencing the wide-complex or narrow-complex SVTs. The older patient may experience the entire range of cardiac symptoms. Tachydysrhythmia present for long periods often defines SVT. Patients with SVT often have recurrent tachycardias from their childhood or early adulthood. Attention must be paid to the medications the patient is using. Antiarrhythmic medications have a use-dependency property. Conduction velocity is slowed as rates increase.83 ECG Criteria for Differentiating VT from Wide-Complex SVT The clinician should not attempt the differential diagnosis of wide-QRS-complex tachycardias without the use of the 12lead electrocardiogram and extended rhythm strip. A common error is to attempt to determine the cause of a tachycardia based only on the rhythm strip. A comparison of past electrocardiograms is often helpful. Examination of the electrocardiogram should focus on the following areas: rate, regularity, AV dissociation, QRS axis, QRS duration, QRS concordance, and QRS morphology84 (Table 12–5). Treatment Therapy is dictated by the specific wide-complex tachycardia and the patient’s clinical presentation. The EC’s initial approach must always be led, and modified if necessary, by the patient’s presentation and subsequent changes. It is recommended in all cases of wide-complex tachycardias, and narrow-complex tachycardias that are producing hemodynamic instability, such that the clinician should immediately consider utilization of cardiovascular electrical cardioversion.

Comments

0.15 mg/kg

Most commonly used Induction occurs in about 2 min Small drop in blood pressure Flumazenil, antagonist available Quicker onset than midazolam Shorter duration than midazolam Small drop in blood pressure Rare complication of laryngospasms No drop in blood pressure Painful IV infusion Small drop in blood pressure Painful IV infusion Painful IV infusion An opiate Added for more sedation Can cause respiratory depression

Methohexital

1 mg/kg

Etomidate

0.15 mg/kg

Propofol

1.5 mg/kg

Thiopental Fentanyl*

3 mg/kg 1.5 µg/kg

*IV medications for sedation during cardioversion.

Synchronized monophasic or biphasic cardioversion is the appropriate first choice of treatment for these cases.78 In patients with wide-complex tachycardias who are cardiovascularly stable, the therapeutic options are more diverse. Stable, wide-complex tachycardia can always be considered VT and treated according to current VT treatment.85 Verapamil should never be used in unknown etiology wide-complex tachycardia. A reasonable treatment protocol for stable patients may be the use of adenosine, procainamide, lidocaine, and finally, cardioversion. Amiodarone is effective for most SVTs and its use in stable unknown wide-complex SVT is both appropriate and safe.86 Despite the criteria listed earlier, when in doubt, the clinician should assume that the dysrhythmia is a wide-complex tachycardia and treat it as such. If it is determined not to be, the treatment should obviously be modified.

Equipment and Setup The critical components of preparation for cardioversion are IV access, airway management equipment, drugs for sedation, and monitoring and DC delivery equipment (cardioverter). Secure IV access is essential for delivery of sedatives, antidysrhythmics, fluids, and possibly, paralytic agents. Although many of these drugs are not used routinely, if they are needed, timing is likely to be critical. A large-bore IV catheter should be inserted and firmly taped to the patient’s skin. A significant and preventable complication of procedures involving sedation is hypoventilation leading to hypoxia. Airway management equipment includes the secure IV catheter discussed previously, working suction with a tonsil-tipped device attached, BVM apparatus, oxygen, and appropriatesized laryngoscope and endotracheal tube. A pulse oximeter is generally recommended for patients undergoing conscious sedation. Another adjunct is continuous carbon dioxide pressure (Pco2) monitoring. A rising Pco2 level will be an earlier clue to hypoventilation due to sedation, because the oxygen saturation may remain normal for several minutes, especially if the patient has been preoxygenated.

Technique If time permits, metabolic abnormalities such as hypokalemia and hypomagnesemia should be corrected before attemp­­ ting cardioversion. At a minimum, hypoxia should be corrected with supplemental oxygen. If a patient has metabolic acidosis, compensatory hyperventilation after endotracheal intubation may be indicated prior to cardioversion. Respiratory acidosis should always be treated prior to the use of sedative drugs. Sedation Cardioversion may be extremely painful or terrifying, and patients must be adequately sedated prior to its use. Patients who are not adequately sedated may experience extreme anxieties and fear.87 Several IV medications are available for sedation of patients prior to cardioversion, including etomidate (0.15 mg/kg), midazolam (0.15 mg/kg), methohexital (1 mg/kg), propofol (1.5 mg/kg), and thiopental (3 mg/kg). In addition, fentanyl (1.5 µg/kg), a synthetic opioid analgesic, is sometimes administered 3 minutes prior to induction. Midazolam (Versed) is probably the most commonly used agent, with induction occurring about 2 minutes after a dose of about 0.15 mg/kg, or at least 5 mg for an averagesized adult. Although induction with midazolam takes slightly longer than with the other medications, it has the advantage that a commercial antagonist, flumazenil, is available for reversal if necessary. Small additional doses of fentanyl (1–1.5 µg/kg) may be added for more profound sedation. Fentanyl can cause respiratory depression, but can be reversed with naloxone. Methohexital has the advantage of quick onset and somewhat shorter duration than midazolam, but it has a rare association with laryngospasm. All the drugs except etomidate cause a small drop in blood pressure, and infusion of propofol and etomidate is painful. In elderly patients, the pharmacodynamics and kinetics are altered by coexisting illness and polypharmacy, rather than by any intrinsic effect of old age.88 Older patients with medical conditions such as congestive heart failure, renal failure, cancer, or malnutrition will therefore experience deeper, prolonged sedation with increased respiratory depression. Drug dose should be reduced in these patients. Administer the anesthetic agent(s) IV over about 30 seconds and wait until the patient is unable to follow simple commands and loss of the eyelash reflex is noted. Pushing the agent too quickly may result in hypotension; pushing the agent too slowly may not allow blood levels to reach a therapeutic range, if the agent has a rapid rate of metabolism. Cardioverter Use Selection of synchronized or nonsynchronized mode is the next critical step. In the synchronized mode, the cardioverter searches for a large positive or negative deflection, which it interprets as the R- or S-wave. It then automatically discharges an electric current that lasts less than 4 msec, avoiding the vulnerable period during repolarization when VF can be easily induced. When the cardioverter is set to synchronize,

Defibrillation and cardioversion

Midazolam

Dose

● 

Drug

Sedative medications should be ready for use in labeled syringes, with a prefilled saline syringe available for flushing the catheter. Antidysrhythmic medications for ventricular dysrhythmias (e.g., amiodarone, lidocaine) and for unexpected bradycardia (e.g., atropine) should be readily accessible.

12 

TABLE 12–5  Commonly Available Intravenous Medications Used for Sedation in Cardioversion

237

CARDIAC PROCEDURES ● 

III 

238

a brief delay will occur after the buttons are pushed for discharge, as the machine searches for an R-wave. This delay may be disconcerting to the unaware operator. If concern exists about whether the R-wave is large enough to trigger the electrical discharge, the clinician can place the lubricated paddles together and press the discharge button. Firing should occur after a brief delay. When the Ror S-wave deflection is too small to trigger firing, change the lead that the monitor is reading or move the arm leads closer to the chest. If there is no R- or S-wave to sense, as in VF, the cardioverter will not fire. Always turn off “synchronization” if VF is noted. Electrode Position Electrode paddles may be positioned in two ways on the chest wall: (1) the anterolateral (or base and apex) position, with one paddle placed in the left fourth to sixth intercostal space, midaxillary line, and the other just to the right of the sternal margin in the second to third intercostal space (see Fig. 12– 19), or (2) the anteroposterior position, with one paddle placed anteriorly over the sternum and the other on the back between the scapulae (see Fig. 12–16). The anterolateral position is used for emergent cardioversion, when placement of an electrode on the patient’s back may not be feasible. Paddles should be pressed firmly against the skin to avoid arcing or skin burns. Safety is a key concern in the performance of cardioversion. Any staff member acting as a ground for the electrical discharge can be seriously injured. The operator must announce “all clear” and give staff a chance to move away from the bed before discharging the paddles. Care must be taken to clean up spills of saline or water, because they may create a conductive path to a staff person at the bedside. Energy Requirements The amount of energy required for cardioversion varies with the type of dysrhythmia, the degree of metabolic derangement, and the configuration and thickness of the chest wall. Obese patients may require a higher energy level for cardioversion; the anteroposterior paddle position is sometimes more effective in these patients. If patients are shocked while in the expiratory phase of their respiratory cycle, energy requirements may also be lower. VT in a hemodynamically stable patient should be treated with amiodarone 150 mg IV, and this can be repeated as needed up to a dose of 2.2 g/24 hr. If unsuccessful, cardioversion is then used. Cardioversion with 10 to 20 J is successful in converting VT in more than 80% of cases. Cardioversion will be accomplished with 50 J in 90% of cases, and conversion should be initially attempted at this energy level.89 Cardioversion should be synchronized unless the T-wave is large and could be misread as the R-wave by the cardioverter. If the initial attempts at electrical cardioversion are unsuccessful, the energy level should be doubled, and doubled again if necessary, until a perfusing rhythm is restored. Immediately after conversion of VT, antidysrhythmic medication should be given to prevent recurrence. Patients with pulseless VT should be initially shocked with 200 J, followed by 300 J if the first shock is not successful. Re-entrant SVTs generally respond to low energy levels. Atrial flutter, for example, usually requires less than 50 J for conversion.17 Cardioversion of atrial flutter in the ED is indicated when the ventricular rate is not slowing in response to

pharmacologically enhanced AV node blockade or if the patient is unable to tolerate the aberrant rhythm. The majority of patients with paroxysmal atrial tachycardia respond to adenosine. If they do not, or if urgent conversion is needed owing to a high ventricular rate, electric countershock should be administered in the synchronized mode at 50 J, and doubled if necessary. In atrial fibrillation, the response to cardioversion is dependent on the duration of atrial fibrillation and the underlying cause. Cardioversion is successful in 90% of cases secondary to hyperthyroidism but in only 25% of cases secondary to severe mitral regurgitation.90 However, 50% of cases revert within 6 months, especially those with long-standing atrial fibrillation.91,92 Most patients with atrial fibrillation do not require cardioversion in the ED unless their ventricular response is high owing to a bypass tract, as in WPW syndrome. They may also require cardioversion when sequelae of rapid ventricular contraction are present or anticipated and the ventricular rate is not responding to drug therapy aimed at slowing AV node conduction. Conversion of atrial fibrillation generally requires more energy than the re-entrant SVTs (~100 J in most cases).93

Complications Complications of cardioversion may affect the patient, particularly the patient with a cardiac pacemaker, as well as health care personnel at the bedside. Patient complications are dose related and may involve the airway, heart, or chest wall, or they may be psychological. Injuries to health care personnel with cardioversion/defibrillation include mild shock and burns. Hypoxia may result if sedation is excessive or if the airway becomes compromised. With proper preparations and precautions, airway complications can be minimized. Respirations may also be depressed by any of the anesthetic agents, and the adequacy of tidal volume must be continually assessed by either direct observation or end-tidal CO2 monitoring. If another clinician is available, he or she should be placed in charge of monitoring the patient’s airway. Routine supplemental oxygen is suggested for all patients undergoing sedation. Chest wall burns resulting from electrical arcing are generally superficial partial-thickness burns, although deep partial-thickness burns have occurred.94 These are preventable by adequate application of conductive gel and firm pressure on the paddles. Paddles should not be placed over medication patches or ointments, especially those containing nitroglycerin, because electrical discharge may cause ignition, resulting in chest burns.95 Cardiac complications after cardioversion are proportionate to the energy dose delivered. In the moderate energy levels used most commonly, the hemodynamic effects are small. At higher energy levels, however, complications include dysrhythmias, hypotension, and rarely, pulmonary edema, which may occur several hours after the countershock. A transient failure of myocardial oxygen extraction due to a direct effect on cellular mitochondria has been proposed as an explanation for some of these cardiac complications.96 The dysrhythmias after high-dose (~200 J) DC shocks include VT and VF, bradycardia, and AV block, in addition to transient and sustained asystole. Sustained VT or VF was reported following 7 of 99 shocks in a study of patients undergoing electrophysiologic study and requiring cardioversion

Conclusions Cardioversion is a safe and effective method of quickly terminating re-entrant tachycardia. Complications related to psychological trauma, respiratory depression, and unin­ tentional health worker shock can be avoided with proper precautions. Adequate sedation is essential. Synchronized shock should be administered after close scrutiny of the lead used for sensing, to be sure that the R- or S-wave is significantly larger than the T-wave. Be prepared for postshock VT or VF, and if VF occurs, switch the cardioverter to “nonsynchronized” and defibrillate. Atropine and temporary pacing equipment should be available to treat postshock bradycardia, especially in patients with myocardial ischemia or MI.

Defibrillation and cardioversion

In summary, cardioversion is performed on perfusing arrhythmias. The goal is not to cause VF. Therefore, during cardioversion, the shock is administered at the peak of the R-wave, during the absolute refractory period. Delivery of the shock during the relative refractory period can cause the development of a nonperfusing arrhythmia such as VF. Cardioversion is performed to treat unstable SVT due to re-entry, unstable atrial fibrillation, and unstable atrial flutter. These arrhythmias are caused by re-entry. Delivering the shock interrupts the re-entrant focus causing these rhythms. In addition, cardioversion is recommended for the conversion or treatment of unstable monomorphic regular VT. The recommended energy level for the treatment of atrial fibrillation is 100 to 200 J with a monophasic waveform.

● 

Summary

If a biphasic waveform defibrillator is used, 100 to 200 J is reasonable. As of this writing, there are no specific data to change prior energy recommendations for cardioversion. However, this may change in the near future, owing to the increasing number of biphasic defibrillators in use today. Conversion of atrial flutter and other SVTs requires less energy. Usually, a dose of 50 to 100 J MDS is ofter sufficient. If the lower dose of 50 J is unsuccessful, increasing the dose in a stepwise fashion, using 50 to 100-J increments, is suggested. Cardioversion may not work for multifocal atrial tachycardia or junctional tachycardias, owing to the presence of a spontaneously depolarizing automatic focus. Treatment of VT has several variables that must be considered. The rate and morphologic characteristics must be considered. If the VT is monomorphic with a regular rate and form, but the patient is unstable, utilize 100 J with a monophasic defibrillator. If this intervention is unsuccessful, increase the energy dosage: 100 J, 200 J, 300 J, 360 J. If the patient deteriorates and becomes pulseless or synchronization is not possible, use high-energy defibrillation in the unsynchronized mode. There are two general types of tachycardias, narrow and wide complex. Management of these tachycardias requires several steps. First, determine the patient’s stability. If the patient is unstable, proceed to immediate cardioversion, with sedation if possible. If the patient is stable, obtain a 12-lead ECG reading to determine the underlying rhythm. A widecomplex tachycardia (>0.12 sec) may be VT, SVT with aberrancy, or a pre-excitation accessory pathway. Next determine whether the rhythm is regular or irregular. Treat patients that are unstable (e.g., with hypotension, chest pain, change in mental status) with immediate cardioversion. If the patient is stable, chemical cardioversion using amiodarone 150 mg over 10 minutes may be an option. If a narrow-complex tachycardia is present and the patient is stable, attempt to determine the underlying cause (e.g., fever, dehydration, hypovolemia, shock, anemia). If possible, attempt vagal maneuvers (see Chapter 11, Techniques for Supraventricular Tachycardias) or pharmacologic cardioversion. If the patient becomes unstable, proceed to immediate cardioversion. Use of medications can be attempted if the patient is stable and has good peripheral circulation. Medications that can be considered are amiodarone, calcium channel blockers (diltiazem, verapamil). In some cases, digitalis glycoside may be used.

12 

for VT, VF, or atrial fibrillation.97 These episodes occurred only in the patients with prior VT or VF. Patients with ischemia or known coronary artery disease appear to be at much higher risk for significant postshock bradycardia, with rate-support pacing required after 13 of 99 shocks in the study. Asystole requiring pacing occurred only once in 99 countershocks. Therefore, the proclivity for dysrhythmias is greater in high-dose cardioversion of an ischemic heart. Two types of VF after cardioversion have been described. The first occurs immediately after countershock and is easily reversed by a second, nonsynchronized shock. This type of VF results from improper synchronization, with discharge of current occurring during the vulnerable period. The second variety, which is more ominous, occurs approximately 30 seconds to a few minutes after attempted cardioversion. This dysrhythmia is characteristically preceded by the development of paroxysmal atrial tachycardia with block or a junctional rhythm. In affected patients, it may be very difficult to convert the dysrhythmia to a sinus rhythm. This phenomenon occurs in patients who have been taking digitalis glycosides and is presumably a manifestation of digitalis toxicity. In the event of VF after cardioversion, the equipment and manpower should be present for immediate defibrillation. If postcardioversion VF occurs, switch the cardioverter to “nonsynchronized” before attempting defibrillation. Electrical discharge will not occur in the “synchronized” mode, because the machine will be searching for a nonexistent R-wave. VF is much more likely to result if depolarization occurs on the T-wave. If a patient has large T-waves in the lead selected for cardioverter sensing, the electric shock may discharge during the vulnerable period of the cardiac cycle, resulting in VF.97 Always examine the complexes on the cardioverter monitor carefully for large T-waves and, if necessary, change the sensing lead. A randomly firing pacemaker can also be sensed by the cardioverter, resulting in countershock during the vulnerable period.98 Transient and intermittent ST-segment elevation has also been reported to occur (although rarely) after cardioversion, with myocardial injury or coronary vasospasm offered as possible explanations.99 An increase in serum enzyme levels (creatine kinase, lactate dehydrogenase, aspartate aminotransferase) may also occur after cardioversion, and the incidence has been reported to be between 10% and 70%. The enzyme rise is usually a consequence of skeletal muscle injury rather than myocardial damage. Cardioversion appears not to alter the enzyme profile of patients with MI100; however, some controversies exist.

239

CARDIAC PROCEDURES ● 

III 

BOX 12–2 Caveats regarding Cardioversion* • Electrical cardioversion is much less effective in treating arrhythmias caused by increased automaticity (e.g., digitalis-induced tachycardia, catecholamine-induced arrhythmia, multifocal AT).5 • Patients presenting with AF or atrial flutter lasting longer than 36– 48 hr are at risk for stroke from embolized thrombus originating in the left atrium. Studies have shown that patients are often unaware of the onset of AF, and thus, patients with stable AF should undergo either a 3- to 4-week course of anticoagulation treatment or TEE to rule out a clot in the left atrium before cardioversion. In the absence of contraindications, acute cardioversion of AF should be accompanied by anticoagulation therapy. • Younger patients, patients with AF of short duration, and those with AF secondary to hyperthyroidism are more likely to convert with cardioversion. Older patients, patients with prolonged AF (>1 yr), and those with structural heart disease are less likely to convert with electrical cardioversion.6

• Remember, using synchronization avoids energy delivery in the early phase of repolarization when the ventricular myocardium is susceptible to VF. This is also referred to as the “R on T phenomenon.” • If there is no R-wave (e.g., in the presence of VF), the cardioverter will not discharge in synchronized mode. • Place the paddles at least 10 cm from each other and from any internal pacemaker/defibrillator and 5 cm from the monitor electrodes. Avoid placement over a pacemaker or medicine patch because these may interfere with conduction. Nitroglycerin can pose a fire hazard if electrical arcing occurs from the paddles and so should be removed. • Defibrillator/cardioverters default to unsynchronized mode after each electrical discharge. You must press the SYNCH button after each synchronized shock if an additional shock is indicated.

AF, atrial fibrillation; AT atrial tachycardia; TEE, transesophageal echocardiography. From Thomsen T, Setnik G (eds): Procedures Consult—Emergency Medicine Module. Philadelphia: Saunders, 2008. Copyright 2008 Elsevier Inc. All rights reserved.

Pediatric Cardioversion

240

Pediatric cardioversion is similar to adult cardioversion. As previously described, the purpose of the procedure is to depolarize the myocytes completely at the most opportune time, during the peak of the R-wave, so as not to precipitate VF, and allow a slower perfusing rhythm to resume. However, the energy levels for pediatric cardioversion are different from those for the adult. In the pediatric procedure, the initial recommended energy dose is 0.5 to 1 J/kg, while the defi­ brillator is in the synchronized mode. If needed, a repeated cardioversion may be attempted at 2 J/kg, again while the defibrillator is in the synchronized mode! Remember to resynchronize the defibrillator after each cardioversion attempt and look for the appropriate markers on the monitor to ensure that the current is delivered at the appropriate phase

of the cardiac cycle! If medication is needed, amiodarone at a dose of 5 mg/kg IV over 20 minutes or procainamide at a dose of 15 mg/kg over 60 minutes can be used. (Do not give these drugs together!) Box 12–2 reviews clinical caveats regarding cardioversion.

Acknowledgment The editors and author would like to acknowledge the significant contributions to this chapter in previous editions by Steven Gazak, MD, William Burdick, MD, Jerris R. Hedges, MD, Michael Greenberg, MD, and John Krimm, DO. REFERENCES  c a n

be found on

E x p e rt C o n s u lt

James A. Pfaff and Robert T. Gerhardt

Patients with implanted pacemakers or automatic implantable cardioverter-defibrillators (AICDs) are commonly seen in the emergency department (ED). Fortunately, the increased reliability of these devices has prevented a marked increase in patients presenting with true emergencies related to device malfunction, but such patients clearly have serious underlying medical problems that must be considered. Pacemaker complications are not uncommon, with rates ranging from 2.7% to 5%.1 Many pacemakers fail within the 1st year.2 AICD complication rates, including inadvertent shocks, occur in up to 34% of patients with the device.3 The basic evaluation and treatment of patients with cardiac complaints who have pacemakers and AICDs are not substantially different from those of patients without the devices. However, a general knowledge of the range of problems, complications, and techniques for evaluating or inactivating pacemakers or AICDs is important for emergency clinicians. These devices are complicated so appropriate consultation, depending on the clinical situation, may be necessary.

HISTORY AND CLINICAL BACKGROUND can be found on

E x p e rt C o n s u lt

PACEMAKER CHARACTERISTICS In essence, a pacemaker consists of an electrical pulse– generating device and a lead system that senses intrinsic cardiac signals and then delivers a pulse. The pulse generator is hermetically sealed with a lithium-based battery device weighing about 30 g with an anticipated lifetime of 7 to 12 years. A semiconductor chip serves as the device’s central processing unit. The generator is connected to sensing and pacing electrodes that are placed in various locations in the heart, depending on the configuration of the pacemaker. Newer models are programmable for rate, output, sensitivity, refractory period, and modes of response,12 and they can be reprogrammed radiotelemetrically after implantation. Pacemakers are classified according to a standard fiveletter code developed by the North American Society of Pacing and Electrophysiology/British Pacing and Electrophysiology Group (Table 13–1). Known as the NBG code, it consists of five positions or digits. The first letter designates the chamber that receives the pacing current; the second, the sensing chamber; and the third, the pacemaker’s response to sensing. The fourth letter refers to the pacemaker’s rate modulation and programmability, and the fifth describes the pacemaker’s ability to provide an antitachycardia function. Whereas standard pacemakers generally do not have an antitachycardia function, AICDs do have this capability and overdrive pacing is the device’s first response to tachycardia. In

Assessment of implantable devices

Assessment of Implantable Devices

● 

13

13 

C H A P T E R

normal practice, only the first three letters are used to describe the pacemaker (e.g., VVI or DDD).13 Pacemaker wires are embedded in plastic catheters. The terminal electrodes, which may be unipolar or bipolar, travel from the generator unit to the heart via the venous system. In a unipolar system, the lead electrode functions as the negatively charged cathode, and the pulse generator case acts as the positively charged anode, into which electrons flow to complete the circuit. The pulse generator casing must remain in contact with tissue and be uninsulated for pacing to occur. In the case of bipolar systems, both of the electrodes are located within the heart. The cathode is at the tip of the lead, and the anode is a ring electrode roughly 2 cm proximal to the tip. Bipolar leads are thicker and draw more current than unipolar leads, and are commonly preferred owing to several advantages. These include a decreased likelihood of pacer inhibition due to extraneous signals and less susceptibility to electromagnetic field interference.14 The typical entry point for inserting the leads is the central venous system, which is typically accessed by the subclavian or the cephalic vein. The terminal electrodes are placed either in the right ventricle or in both the right ventricle and the atrium, under fluoroscopic guidance. Proper lead placement is checked by electrocardiograms (ECGs) checking sensing and pacing thresholds.15 The typical radiographic appearance of an implanted pacemaker is seen in Figure 13–1. The pacemaker rate is typically programmed to pace between 60 and 80 beats/min. A significantly different rate usually indicates malfunction. When the battery is low, the rate usually begins to drop, getting slower as the battery fades. Sensing of intracardiac electrical activity is a combination of recognizing the characteristic waveforms of P-waves or QRS complexes while discriminating these from T-waves or external interfering signals, such as muscle activity or movement. The pacing electrical stimulus is a triphasic wave consisting of an intrinsic deflection, far-field potential, and an injury current, which typically delivers a current of 0.1 to 20.0 mA for 2 msec at 15 V.16 Pacemakers have a reed switch, which may be closed by placing a magnet over the generator externally on the chest wall; this inactivates the sensing mechanism of the pacemaker, which then reverts to an asynchronous rate termed the magnet rate. Essentially, the magnet turns the demand pacemaker into a fixed-rate pacemaker. The magnet rate is usually, but not always, the same as the programmed rate. Several new innovations in rate regulation have been incorporated into some pacemakers. When present, the hysteresis feature causes pacing to be triggered at a rate greater than the intrinsic heart rate. When the hysteresis feature is employed in a single-chamber ventricular pacemaker, it is designed to maintain atrioventricular (AV) synchrony at rates that are lower than what would be normal for a ventricularpaced rhythm alone. To illustrate, were the hysteresis feature of the pacemaker set at 50 beats/min, an intrinsic rate under 50 beats/min would trigger ventricular pacing. Unlike a standard ventricular pacemaker, the hysteresis feature might be set to offer a ventricular pacing rate at 70 beats/min or greater once the pacer is triggered. Rate modulation by sensor-mediated methods is an additional feature triggered and mediated by a sensed response to various physiologic stimuli. The primary application for this rate modulation feature is in the case of pacemaker patients who continue to engage in vigorous physical activity. When

241

CARDIAC PROCEDURES ● 

III 

TABLE 13–1  North American Society of Pacing and Electrophysiology/British Pacing and Electrophysiology Group Generic Pacemaker Code (NBG Code) I Chamber Paced 0—None A—Atrium V—Ventricle D—Dual

II Chamber Sensed

III Response to Sensing

IV Rate Modulation and Programmability

V Antitachycardia Features

0—None A—Atrium V—Ventricle D—Dual

0—None I—Inhibited T—Triggered D—Dual

0—None I—Inhibited M—Multiple C—Communicating R—Rate modulation

0—None P—Antitachycardiac pacing S—Shock D—Dual

A1

A2

242

B1

B2

C1

C2

Figure 13–1  A, Various radiographs of an implanted pacemaker and automatic implantable cardioverter-defibrillator (AICD) show battery and lead wires. Posteroanterior (PA; 1) and lateral (2) chest radiographs demonstrate a biventricular pacing system. There are three leads—the first is positioned in the right atrium, the second is in the right ventricular apex, and the third courses posteriorly in the coronary sinus and into the posterolateral cardiac vein. B, PA chest radiographs of a dual-chamber pacemaker. (1) Ventricular lead is passing through an atrial septal defect into the left ventricle. (2) The lead is repositioned in the right ventricular apex. C, A dual-chamber implantable cardioverter-defibrillator (ICD) has been implanted using active fixation leads via a transvenous approach to place the atrial lead in the systemic venous atrium and the ventricular lead across the baffle into the morphologic left ventricle.

13  ● 

Assessment of implantable devices

D1

D2

243 Illustrated fracture

E Figure 13–1, cont’d D, PA chest radiograph of a patient with a dual-chamber pacemaker (1). The atrial lead, originally positioned in a right atrial appendage position, is clearly no longer positioned in the right atrial appendage. Lateral view (2) also shows definite dislodgment of the atrial lead. E, Close-up view of a portion of the PA chest radiograph of a patient with a single-chamber pacemaker. The lead has fractured (a subtle finding) where it passes below the clavicle (arrow). The patient presented with intermittent ventricular failure to capture and intermittent failure to output on the ventricular lead. Impedance was intermittently measured at more than 9999 ohms. Inset, Diagram of the fracture site.  (E inset, Courtesy of Telectronics Pacing Systems, Englewood, CO.)

present, the rate regulation feature is engaged and modulated through motion sensors installed within a pulse generator device, with a corresponding increase or decrease in pacing rate based upon the degree of motion sensed by the pacemaker device. Other physiologic sensors that may be installed as part of the pacemaker system include those designed to sense minute ventilation, Q-T interval, temperature, venous oxygen saturation, and right ventricular contractions. The latter sensors generally require that additional leads be placed.

AICD Characteristics The basic components of an AICD include sensing electrodes, defibrillation electrodes, and a pulse generator (Fig.

13–2), which can be seen on a chest x-ray. Transvenous electrodes have obviated the prior need for surgical placement. They are inserted into the pectoralis muscle. Many transvenous systems consist of a single lead containing a distal sensing electrode and one or more defibrillation electrodes in the right atrium and ventricle.17 Leads are inserted through the subclavian, axillary, or cephalic vein into the right ventricular apex. The left side is preferred because of a smoother venous route to the heart and a more favorable shocking vector.18 In an effort to improve defibrillation efficiency, an additional defibrillation coil may be used.18 Various placements of AICDs are demonstrated in Figure 13–3. The pulse generator is a sealed titanium casing that encloses a lithium–silver–vanadium oxide battery, voltage converters and resistor, capacitors to store charges, micropro-

CARDIAC PROCEDURES ● 

III 

244

cessors and integrated circuits to control the analysis of the rhythm and delivery of therapy, memory chips to store electrographic data and a telemetry module.19 Whereas a pacemaker can draw the required voltage for its function from its component battery, the energy requirements necessary for defibrillation require a battery that is prohibitively large.14 Subsequently, an AICD contains a capacitor that maximizes the required voltage by transferring energy from the battery prior to discharge. To achieve the required energy, AICDs use capacitors that are charged over 3 to 10 seconds by the battery and then release this energy rapidly for defibrillation.17 The maximal output is 30 J in most units and 45 J in higher-energy units.14 This energy is high enough that a discharge is very obvious and often distressing to the patient.

A

B Figure 13–2  A, AICD. B, Implantable pacemaker.  (A, SOLETRA device, Courtesy of Medtronics Inc., Minneapolis, MN.)

Most AICDs use a system in which the pulse generator is part of the shocking circuit, often described as a “can” technology, and most of them have a dual-coil lead with a proximal coil in the superior vena cava and a distal coil in the right ventricle.20 Current flows in a three-dimensional configuration from the distal coil to both the proximal coil and the generator.21 This dispersion of the electrical field increases the likelihood of depolarizing the entire myocardium at once, leading to successful defibrillation.21 AIDCs may have the same programming capabilities as pacemakers and can be single-chambered, dual-chambered or used with cardiac resynchronization therapy.22 Single-chamber devices have only a right ventricular lead. They often have had difficulty identifying atrial arrhythmias, resulting in the inappropriate defibrillation of atrial tachycardias. Dual-chamber AICDs have a right atrial and a right ventricular lead and have improved ability to discriminate rhythms. In most studies, the dual devices have offered improved discrimination between ventricular and supraventricular arrythmias, decreasing inappropriate shocks due to rapid supraventricular rhythms or physiologic sinus tachycardia.23 Approximately 50% of AICDs implanted in the United States are dual-chamber devices.24 Cardiac resynchronization devices add an additional left ventricular lead that is placed in the coronary sinus or epicardium. In patients requiring both AICD and pacemaker functions, these devices were both placed together. The advent of technology has allowed placement of a single device that can perform both pacemaker and defibrillator functions. AICDs use a combination of antitachycardia pacing, lowenergy cardioversion, defibrillation, and bradycardic pacing in a combination also known as tiered therapy. They are programmed with specific algorithms that identify and treat specific rhythms. The ventricular arrhythmias may initially be converted (or have attempts at conversion) with antitachycardic pacing as opposed to immediate defibrillation. This overdrive pacing may terminate the rhythm without the need for electrical defibrillation in up to 90% of events. This is most successful for terminating monomorphic ventricular tachycardia with a rate of less than 200.1 It is better tolerated by patients than cardioversion and reduces the risk of induced atrial fibrillation.25 These events may be silent and not felt by the patient and only discovered by interrogating the device. If unsuccessful, the next intervention may be low-energy cardioversion ( 5 mm Concordant S-T segment elevation > 1 mm S-T segment depression > 1 mm in precordial leads V1–V3

Sensitivity (%)

● 

Electrocardiographic Criterion

13 

TABLE 13–5  Criteria for Electrocardiographic Diagnosis of Acute Myocardial Infarction in the Setting of   Ventricular Paced Rhythm

condition may continue until the underlying atrial tachycardia is terminated by intervention. A third instance of pacemaker-mediated tachycardia occurs in patients with AICD units possessing a backup antibradycardia pacing capability. It appears that in such patients, if the pacemaker feature is switched on and an ectopic ventricular stimulus is delivered after a sudden pause in the intrinsic ventricular depolarization cycle, a ventricular tachyarrhythmia may be triggered.2,37

Diagnosis of Acute Myocardial Infarction in the Presence of a Paced Cardiac Rhythm Patients undergoing active ventricular pacing from an implanted pacemaker device will normally possess ECGs that resemble a left bundle branch block pattern. As a result, the electrocardiographic diagnosis of acute ischemic changes is equally challenging in both populations. Sgarbossa and coworkers38 published a series of criteria in 1996 that offers some utility in the interpretation of ECGs in patients with active ventricular pacing, in whom acute coronary syndromes are suspected. These criteria are depicted in Table 13–5.15

AICD-Unique Malfunctions Issues with sensing problems, lead migration, and battery failure are similar to pacemaker complications and most occur within 3 months after implantation.39 A potential malfunction unique to the AICD is the inappropriate or lack of defibrillation of the device. The AICD may not terminate ventricular arrhythmias, which may or may not be the result of device malfunction. AICD malfunction may be a result of battery depletion, component failure, undersigning, or lead malfunction. Failure-tocardiovert/defibrillate that occurs in the setting of a functioning AICD system may be caused by inappropriate cutoff rates, failure to satisfy multiple detection criteria, completed and exhaustion of therapies, and cross-inhibition by a separate pacemaker.18 The advent of AV, or dual chamber AICD devices, has improved the sensitivity of arrhythmia detection, preventing the delivery of inappropriate shocks. Inappropriate AICD-delivered shocks occur in 20% to 25% of patients40 and are the most common adverse events observed in AICD patients.34 The main causes are atrial arrhythmias, sinus tachycardia, nonsustained ventricular tachycardia, lead fracture or EMI, or electrical storm.40 By definition, this phenomenon occurs when three or more shocks

251

CARDIAC PROCEDURES ● 

III 

TABLE 13–6  Outcome of Automatic Implantable Cardioverter-Defibrillator Placement

TABLE 13–7  Method for Inactivation of Automatic Implantable Cardioverter-Defibrillator

Estimated Events over a 5-Yr Period after AICD Placement for Current Criteria*

1. Determine the orientation of the device in the abdominal pocket, radiographically or by palpation. 2. Place a ring magnet over the upper right hand corner of the device. 3. A beeping tone will sound, which corresponds with the sensing of QRS complexes. 4. Leave the magnet in place for at least 30 sec. 5. When the beeping changes to a continuous tone, the device is inactivated. 6. Remove the magnet.

30 patients will die anyway owing to underlying disease. 7–8 patients will be saved by AICD. 10–20 will have a shock delivered that is not needed. 5–15 will have an ICD complication. The rest will not experience the device firing. Some will ask to have it removed to allow natural death. *Left ventricular ejection fraction < 30%–35% and anticipated survival with good functional capacity beyond 1 yr. AICD, automatic implantable cardioverter-defibrillator; ICD, implantable cardioverter-defibrillator. Adapted from Stevenson LW, Desai AS: Selecting patients for discussion of the ICD as primary prevention for sudden death in heart failure. J Card Fail 12:407, 2006.

252

are delivered in a 24-hour period and this constitutes a medical emergency.41 AICD patients who experience this phenomenon may have end-stage cardiac failure and possess an increased risk for sudden cardiac death, particularly within the first 3 months after the event. Specific etiologies for electrical storm are unclear, but are associated with ventricular tachycardia in the setting of left ventricular ejection fractions below 30%, and occur more frequently in patients with demonstrated coronary artery disease who have not as yet undergone revascularization procedures. Suggested initial treatment includes administration of amiodarone and β2-adrenergic antagonists to pharmacologically suppress arrhythmias and to obtain urgent cardiology consultation.41 The AICD may discharge inappropriately in response to rapid supraventricular rhythms such as atrial fibrillation, supraventricular tachycardia, or even sinus tachycardia. Multiple shocks may be a manifestation of inefficient tachycardia termination such as an inappropriately low-energy delivery at the first shock, increased defibrillation thresholds, and migration or dislodgment of the defibrillation lead system or a defibrillator system failure. Shocks that occur every few minutes may suggest that recurring ventricular tachyarrythmias are being appropriately terminated (see Fig. 13–6). AICD discharges in the setting of chest pain may be a result of myocardial ischemia–induced tachyarrhythmias.30 As mentioned previously, electrocardiographic abnormalities noted immediately after shocks should be interpreted with caution because ST elevation or depression can occur immediately after a shock.31 If the patient receives shocks in association with chest pain, ischemia is suggested but other causes including hypokalemia, hypomagnesemia, drug-induced proarrythmias, or drugs that can prolong the Q-T interval (such as phenothiazines) should also be considered as underlying causes.18 In some settings, the AICD may fail to sense sustained ventricular tachycardia or fibrillation. These failures may occur as the result of an intrinsic arrhythmia rate below the programmed detection rate, usually resulting from concurrent pharmacologic therapy. If the patient is hemodynamically stable, it may be advantageous for the cardiologist to interrogate the pacer prior to initiating further antiarrhythmic therapy. If unsuccessful, or if the patient is experiencing a nonperfusing ventricular arrhythmia, other pharmacologic interventions include procainamide42 or amiodarone. Table

Adapted from Munter DW, DeLacey WA: Automatic implantable cardioverterdefibrillators. Emerg Med Clin North Am 12:579, 1994. Used with permission.

13–6 depicts the outcome of AICD placement in a general population. Use of a Magnet for AICD Inactivation The patient who is experiencing inappropriate AICD discharges in the ED can be treated by magnet inactivation of the device, similar to the approach described earlier for the pacemaker patient. If the patient is experiencing recurrent rhythms that require AICD activation, do not inactivate the device because it is functioning as required. Technique.  The method for inactivating an AICD device is outlined in Table 13–7. The orientation of the device in the abdominal pocket should be determined, with the lead connections normally cephalad. A ring magnet is then placed over the corner adjacent to the lead connections (usually the upper right hand corner of the device). A series of beeping tones, which correspond to the sensed QRS complexes, will sound. In the absence of organized QRS activity, random beeps will sound.43 When the magnet is left in place for 30 seconds, a continuous beep is heard. This indicates that the AICD is inactivated. The magnet should then be removed, and the AICD will remain inactivated. The AICD may be reactivated by applying the magnet for 30 seconds and re­moving it when the steady beep changes to intermittent beeping. Note that unlike a pacemaker where a magnet will turn a demand pacemaker to fixed rate pacemaker, a magnet will not affect the pacing function of an AICD.

Mental Health Issues Related to Implanted Pacemakers and AICDs Patients with these devices may manifest a number of anxietyrelated complications including adjustment disorder, panic attacks, depression, imaginary shock, and defibrillator dependence abuse or withdrawal.44–46 These patients may benefit from psychiatric referral either as an outpatient or as part of the admission evaluation if applicable. The conditions may be severe enough that the device is removed by patient request.

Implantable Pacemaker and AICD Recalls Since 1990, there have been approximately 29 FDA safety alerts and recalls affecting nearly 337,000 AICDs.47 These advisories occur from unanticipated device failures that are identified after product release and widespread clinical use.48

Out-of-Hospital AICD Discharge Patients are told to adhere to standard advice defining an appropriate response to out-of-hospital AICD discharges (Table 13–9). Many, however, come to the ED for evaluation after every shock. There is no standard ED intervention mandated by historical information, and clinical decisions are made on an individual basis based on the current scenario. Options include prolonged ED observation, consultation, cardiac monitoring, laboratory testing (such as electrolyte and cardiac enzymes) or interrogation.

Disposition Criteria In the majority of cases, patients presenting to EDs with pacemaker complications or malfunctions will be symptomatic. As such, and regardless of the clinical requirement for admission, they will likely require device interrogation and possible recalibration or replacement by a cardiologist. In most cases, this will be accomplished by hospital admission. With regard to AICD malfunctions and disposition, patients who present with a single shock and no other specific complaints or comorbidity can be discharged with follow-up in 24 to 48 hours. Patients with symptoms concerning for

Copy machines Electric blankets Household appliances (microwaves, washer/dryer) DVD/CD players TVs Personal computers Remote controls Heating pads Some Evidence of Interaction

Assessment of implantable devices

Given the plethora of new technologies, there is always concern about the interactions of EMI with pacemakers and AICDs. The sources of EMI comprise a significant spectrum and may involve radiated and conducted sources. The most common response of implanted devices to EMI is inappropriate inhibition or triggering of pacemaker stimuli, reversion to asynchronous pacing, and spurious AICD tachyarrhythmia detection. Reprogramming of operating parameters and permanent damage to the device circuitry or the electrode-tissue interference can also occur but are much less frequent.49 Additional adverse effects that may occur include inhibition of bradycardia pacing, inadvertent delivery of a shock, or antitachycardia pacing, The use of hermetic shielding in metal cases, filtering, interference rejection circuits, and bipolar sensing have helped to mitigate most of this interference.49 Nonetheless, the clinician should be familiar with common sources of EMI that may affect pacemakers and AICDs. Several caveats will help avoid the deleterious effects of EMI upon implantable devices. Cell phones should not be kept in a pocket over the device. When in use, they should be held at least 6 inches away from the device. In the case of hands-free headphones, when employed these devices should be placed in the ear opposite the implanted pacemaker or AICD. In addition, patients should be advised not to linger in theft detection areas or airport metal detectors. Table 13–849–51 identifies different devices and the corresponding potential EMI that can occur with pacemakers and AICDs.

Generally Felt to be Safe with EMI

● 

EMI and Implantable Devices

TABLE 13–8  Electromagnetic Interference Sources and Their Potential Effects on Implanted Pacemakers and Automatic Implantable Cardioverter-Defibrillators

13 

The decision to remove the devices is complex, and there has been difficulty reaching consensus on the optimal management of these recalled devices. The decision to replace these devices should be multifactorial, taking into consideration the estimated device malfunction rate, anticipated consequences of device failure, the individual center’s procedural risk of complications from generator change, and patient preferences and desired level of risk tolerance.48

Cell phones Induction ovens Power toothbrushes Battery-powered, cordless power tools Arc welding equipment Chain saws Drills Hedge clippers Lawn mowers Leaf blowers Snow blowers High-voltage lines Theft detection systems Airport scanners/metal detectors Avoid because of Interaction Electrolysis MRI Jackhammers Medical Devices requiring Caution Electrocautery (especially unipolar) equipment High-energy radiation sources TENS units MRI scanners Body fat measuring scales Diathermy equipment Electrolysis equipment Spinal cord stimulators Direct current external cardioversion/defibrillation equipment Radiofrequency catheter ablation equipment Lithotripsy equipment Safe Medical Devices CT scanners Dental drills Diagnostic x-ray machines Electrocardiography equipment Ultrasound equipment CT, computed tomography; EMI, electromagnetic interference; MRI, magnetic resonance imaging; TENS, transcutaneous electrical nerve stimulation. Adapted from Cardiosource, www.cardiosource.com, with permission.

ischemia, potentially lethal arrhythmias, or symptomatic illness should be admitted, and expeditious specialty consultation obtained. Patients who present with multiple shocks will need admission for observation and interrogation of their AICD device. Interrogation reveals significant information about the device, such as why it fired, the rhythm history, and

253

CARDIAC PROCEDURES ● 

III 

254

TABLE 13–9  Follow-up of the Patient with an AICD Follow up assessments of patients with an AICD are made on a routine basis, every 3–6 months, and when device discharge occurs. An analysis of any prior clinical event and testing of defibrillation function are readily accomplished. Internet-based remote follow-up systems may replace some office follow-up. Follow-up can occur remotely using vendor specific equipment, to interrogate and upload data. Remote follow-up, however, permits only device interrogation and retrieval of diagnostic data, but not threshold testing or reprogramming. Device interrogation includes: • Determination of pacing and sensing thresholds • Analysis of recorded episodes of arrhythmia detection and AICD activation, including pacing and delivered shocks. Data includes the date and time of each episode and a stored EKG from the event. • Battery status Defibrillator discharge—An appropriate shock is delivered in about 50 percent of patients by two years post-implantation. Patients may not sense anti-tachycardic (overdrive) pacing to terminate arrhythmias. Not all episodes of defibrillator discharge require immediate medical evaluation, although many patients immediately come to the ED. Patients with a first shock may be seen on an urgent or elective basis to ascertain the specifics of the event and to determine if the device is functioning properly. Discharges that are accompanied by changes in cognition (syncope, seizure, or loss of consciousness) require ED evaluation. Per guidelines, patients who have had a single AICD discharge with immediate return to baseline clinical status and no associated symptoms (e.g., chest pain, shortness of breath, or lightheadedness) may have the device interrogated within one to two days. Patients who receive frequent or clusters of shocks are either appropriate (due to recurrent VT) or inappropriate (due to atrial fibrillation or supraventricular tachycardia or to device malfunction). Such patients generally require emergent evaluation and hospital admission to determine the cause. Additional therapy (such as an antiarrhythmic drug or catheter ablation) may be required.

Figure 13–8  Sophisticated bedside interrogation of an AICD can be accomplished in a few hours by calling the manufacturer. Using a noninvasive AICD interrogating device, the clinician can determine the patient’s name, diagnosis, AICD settings, and surgeon, and can view recent and remote information about the heart and AICD activity. This takes away all of the guesswork when trying to determine what happened to the patient to prompt an AICD-related ED visit. Such information can also be obtained periodically over the phone in lieu of an ED or office visit.

an accurate assessment of the underlying problem (Fig. 13–8).

Acknowledgments The authors and editors wish to thank David Munter for his work on previous editions of this chapter.

REFERENCES  c a n

be found on

E x p e rt C o n s u lt

Richard A. Harrigan, Theodore C. Chan, and William J. Brady

The electrocardiogram (ECG) is a graphic recording of the electrical activity of the heart. The standard ECG is obtained by applying electrodes over the chest and limbs that record the electrical activity of the cardiac cycle. Developed nearly 100 years ago, the ECG remains the most important initial diagnostic tool for the assessment of myocardial disease, ischemia, and cardiac dysrhythmias. Electrocardiography is performed widely throughout the health care field, including ambulances, ambulatory clinics, emergency departments (EDs), and in-patient hospital units. Standard electrocardiography machines are small, selfcontained, and portable, allowing them to be employed in virtually any setting. As a result, clinicians, nurses, and many other health care providers should be familiar with the procedure of standard 12-lead electrocardiography. Emergency clinicians should also be familiar with the alternative leads and other accessory techniques available in electrocardiography, as well as the pitfalls of lead misplacement, misconnection, and tracing artifacts. BACKGROUND  c a n

be found on

E x p e rt C o n s u lt

INDICATIONS The most frequent indication for electrocardiography performance in the ED is the presence of chest pain. Other common indications include abnormal rhythm, palpitations, dyspnea, syncope, diagnosis-based (e.g., acute coronary syndrome [ACS], suspected pulmonary embolism), and system-related (e.g., “rule-out myocardial infarction [MI]” protocol, admission purposes, and operative clearance) indications.9 The ECG is used to help establish a diagnosis, select appropriate therapy, determine the response to treatment, assist in the correct disposition for the patient, and help predict risk of both cardiovascular complication and death. The initial 12-lead ECG obtained in the ED can be an important tool for determination of cardiovascular risk and, as such, the choice of in-hospital admission location. Brush and coworkers10 classified the initial ECG into high- and lowrisk groups. The low-risk electrocardiographic group had normal ECGs, nonspecific ST-T-wave changes, or no change when compared with a previous ECG. High-risk ECGs had significant abnormalities or confounding patterns—such as pathologic Q-waves, ischemic ST segment or T-wave changes, left ventricular hypertrophy, left bundle branch block, or ventricular paced rhythms. Patients with initial ECGs classified as low risk had a 14% incidence of acute myocardial infarction

BASIC EQUIPMENT The 12-Lead ECG Although there is variability depending upon the workplace, most ECGs in use today are three-channel recorders with computer memory. Such multichannel systems, recording electrical events in several leads concurrently, offer advantages over the antiquated single-channel recorder systems— capturing transient events on multiple leads simultaneously; banking the data in computer memory for storage, comparison, and transmission; and allowing for data presentation on a single sheet of paper.14 The electrocardiographic tracing is printed in a standardized manner on a standardized paper by the electrocardiograph, which has default settings regarding the speed with which the paper moves through the machine as well as the amplitude of the deflections to be made on the tracing. Electrocardiographic paper is divided into a grid, with a series of horizontal and vertical lines; the thin lines are 1 mm apart, and the thick lines are separated by 5 mm. At the standard paper speed of 25 mm/sec, each vertical thin line thus represents 0.04 sec (or 40 msec), and the thick vertical lines correspond to 0.20 sec (or 200 msec). Recordings from each of the 12 leads typically are displayed for 2.5 seconds by default setting; the leads appearing horizontally adjacent to each other are separated by a small vertical hash mark to represent lead change. The standard ECG includes 12 leads derived from 10 electrodes placed on the patient; each is color-coded and represented by a two-character abbreviation (Table 14–1). The placement of limb leads on the left and right arms (LA and RA, respectively) and the left and right legs (LL and RL, respectively) by color can be recalled with the help of several mnemonics. These include the following: “Christmas trees

Basic electrocardiographic techniques

Basic Electrocardiographic Techniques

(AMI), 0.6% incidence of life-threatening complications, and a 0% mortality rate. Patients with initial ECGs classified as high risk had a 42% incidence of AMI, 14% life-threatening complications, and 10% mortality rate.10 Another approach to risk prediction involves a simple calculation of the number of electrocardiographic leads with ST segment deviation (elevation or depression)—with an increasing number of leads being associated with higher risk. Along similar lines, the clinician is also able to predict risk with a summation of the total millivolts of ST segment deviation; once again, higher totals are associated with greater risk.10 The limitations of the ECG must be recognized, however. The ECG is widely reported to have a sensitivity for AMI of only approximately 55%; in one study of 1000 patients with ischemic symptoms, that sensitivity improved to 68% with serial ECGs and ST segment trend monitoring.11 In another series, the sensitivity of the ECG for AMI ranged from 43% to 65% over a 12-hour period after ischemic symptom onset, yet the negative predictive value of a normal ECG (defined as normal or with nonspecific changes or isolated fascicular blocks) for AMI did not improve above 93% during this period.12 In a large series of over 10,000 patients of whom 889 (8%) were ultimately diagnosed with AMI, 19 (2%) were inappropriately discharged from the ED. A nonischemic ECG emerged as one of five risk factors for that inappropriate disposition decision (along with female, 30%) risk of heart block. Transcutaneous pacing is generally a temporizing measure that may precede transvenous cardiac pacing. Although it is not an expectation that all emergency clinicians will be adept at placing emergency cardiac pacemakers, many have mastered the techniques and are the only clinicians available to perform this life-saving procedure.

EMERGENCY TRANSVENOUS CARDIAC PACING The transvenous method of endocardial pacing is commonly used and is both safe and effective. In skilled hands, the semifloating transvenous catheter is successfully placed under electrocardiographic (ECG) guidance in 80% of patients.1 The technique can be performed in less than 20 minutes in 72% of patients and in less than 5 minutes in 30% of patients. However, in some instances, anatomic, logistical, and hemodynamic impediments can prohibit successful pacing by even the most skilled clinician. As with other medical procedures, it should not be performed without a thorough understanding of its indications, contraindications, and complications.2 However, because this is essentially a blind procedure, a certain amount of luck and chance are consequential to a successful outcome. Therefore, sometimes this procedure simply will not be successful, either because the condition is not amenable to pacing (e.g., asystole, drug overdose) or because of technical difficulties inherent with the procedure. BACKGROUND  c a n

be found on

E x p e rt C o n s u lt

Indications The purpose of cardiac pacing is to stimulate effective cardiac depolarization. In most cases, the specific indications for cardiac pacing are clear; however, some controversial areas remain. The decision to pace on an emergent basis requires

Emergency cardiac pacing

Emergency Cardiac Pacing

knowledge of the presence or absence of hemodynamic compromise, the etiology of the rhythm disturbance, the status of the AV conduction system, and the type of dysrhythmia. The clinician caring for the patient is in the best position to decide on the value, or nonvalue, of pacing, based on nuances of the clinical scenario that are not possible to unravel by any theoretical discussion. Controversy exists throughout the literature, and this dis­ cussion is not meant to set a standard of care for individual circumstances. In general, the indications can be grouped into those that cause either tachycardias or bradycardias (Table 15–3). Transcutaneous cardiac pacing (TCP) has become the mainstay of emergent cardiac pacing and is often used pending place­­ ment of the transvenous catheter or to determine whether potentially terminal bradyasystolic rhythms will respond to pacing.

● 

15

15 

C H A P T E R

Bradycardias

Sinus Node Dysfunction.  Sinus node dysfunction may manifest as sinus arrest, tachybrady (sick sinus) syndrome, or sinus bradycardia. Whereas symptomatic sinus node dysfunction is a common indication for elective permanent pacing, it is seldom cause for emergency pacemaker insertion. In acute myocardial infarction (AMI), 17% of patients will experience sinus bradycardia.14 It occurs more frequently in inferior than in anterior infarction and has a relatively good prognosis when accompanied by a hemodynamically tolerable escape rhythm. However, sinus bradycardia is not a benign rhythm in this situation; it has a mortality rate of 2% with inferior infarction and 9% with anterior infarction.15 Sinus node dysfunction frequently responds to medical therapy but requires prompt pacing if this fails. Asystolic Arrest.  Transvenous pacing in the asystolic or bradyasystolic patient has little value. In 1 study of 13 patients who had suffered cardiac arrest, capture of the myocardium was noted in 4 patients, but there were no survivors.16 Transvenous pacing alone may also not be effective in postcountershock pulseless bradyarrhythmias.17 This failure of pacing has also been demonstrated with transcutaneous pacemakers, suggesting that failure of effective pacing is primarily related to the state of the myocardial tissue.16 Cardiac pacing may be used as a “last-ditch” effort in bradyasystolic patients but is rarely successful and is not considered standard practice. Early pacing is essential when done for this purpose if success is to be achieved18 (see later in this section). Most importantly, given the recognition of the importance of maximizing chest compressions during cardiopulmonary resuscitation (CPR), interrupting CPR in order to institute emergency pacing is not recommended.19 AV Block.  AV block is the classic indication for pacemaker therapy. In symptomatic patients without myocardial infarction (MI) and in the asymptomatic patient with a ventricular rate below 40, pacemaker therapy is indicated.20 In patients with AMI, 15% to 19% progress to heart block: approximately 8% develop first-degree block, 5% develop second-degree block, and 6% develop third-degree block.21 First-degree block progresses to second- or thirddegree block 33% of the time, and second-degree block progresses to third-degree block about one third of the time.22 AV block occurring during anterior infarction is believed to occur because of diffuse ischemia to the septum and infranodal conduction tissue. These patients tend to progress to high-degree block without warning, and a pacemaker is often placed prophylactically. Some patients are prophylacti-

269

CARDIAC PROCEDURES

First Letter

Second Letter

Third Letter

Chamber-Paced

Chamber-Sensed

Sensing Response

Programmability

A = atrium V = ventricle D = dual O = none

T = triggered I = inhibited D = dual (A-triggered and V-inhibited) O = none

P = simple M = multiprogrammable R = rate adaptive C = communicating O = none

TABLE 15–2  History of Transvenous Pacing Date

270

Fourth Letter

A = atrium V = ventricle D = dual O = none

III 

● 

TABLE 15–1  Four-Letter Pacemaker Code

Investigator

1700 1951

Early investigators Callaghan & Bigelow

1952

Zoll

1958

Falkmann & Walkins

1959

Furman & Robinson

1964

Vogel et al

1965 1966 1967

Kimball & Killip Goetz et al Zuckerman et al

1969 1973

Rosenberg et al Schnitzler et al

Event First restimulation studies First transvenous approach in dogs Transcutaneous cardiac stimulator Implanted pacing wires after surgery First transvenous pacer in humans Flexible electrocardiographic catheter without fluoroscopy First bedside transvenous pacing Demand pacemaker developed Use of demand pacemaker clinically Semifloating pacing catheter Balloon-tipped pacers

TABLE 15–3  Indications for Cardiac Pacing Bradycardias Without myocardial infarction Symptomatic sinus node dysfunction (sinus arrest, tachybrady [sick sinus] syndrome, sinus bradycardia) Second- and third-degree heart block Atrial fibrillation with slow ventricular response With myocardial infarction Symptomatic sinus node dysfunction Mobitz II second- and third-degree heart block New left bundle branch block, right bundle branch block with left axis deviation, bifascicular block, or alternating bundle branch block Prophylaxis: cardiac catheterization, after open heart surgery, threatened bradycardia during drug trials for tachydysrhythmias Malfunction of implanted pacemaker Tachycardias Supraventricular dysrhythmias Ventricular dysrhythmias Prophylaxis: cardiac catheterization, after open heart surgery

cally paced on a temporary basis, even in the absence of hemodynamic compromise. During inferior infarction, early septal ischemia is the exception and, typically, block develops sequentially from first-degree to Mobitz type I second-degree, then to third-

degree AV block. These conduction abnormalities frequently result in hemodynamically tolerable escape rhythms because of sparing of the bundle branches. The hemodynamically unstable patient who is unresponsive to medical therapy should be paced promptly. Whether and when the stable patient should be paced is unclear, but placing a transcutaneous pacer is one option that can be tried before placing a transvenous pacing catheter. Trauma.  Pacing is not a standard intervention in traumatic cardiac arrest. In selected cases, it may be considered. Several rhythm and conduction disturbances have been documented in the patient with nonpenetrating chest trauma. In these patients, traumatic injury to the specialized conduction system may predispose the patient to life-threatening dysrhythmias and blocks that can be treated by cardiac pacing.23 Hypovolemia and hypotension can cause ischemia of conduction tissue and cardiac dysfunction.24 Marked bradyarrhythmias that persist even after vigorous volume replacement may rarely respond to cardiac pacing in patients with such trauma.25 Bundle Branch Block and Ischemia Bundle branch block occurring in AMI is associated with a higher mortality rate and a greater incidence of third-degree heart block than uncomplicated infarction. Atkins and colleagues26 noted that 18% of patients with MI had bundle branch block. Of these patients, complete heart block developed in 43% who had right bundle branch block (RBBB) and left axis deviation, in 17% who had left bundle branch block (LBBB), in 19% who had left anterior hemiblock, and in 6% who had no conduction block. The investigators concluded that RBBB with left axis deviation should be paced prophylactically. A study by Hindman and associates27 confirmed the natural history of bundle branch block during MI. In their study, the presence or absence of first-degree AV block, the type of bundle branch block, and the age of the block (new vs. old) were used to determine the relative risk of progression to type II second-degree or third-degree block (Table 15–4). Because of the increased risk, consider pacing the following conduction blocks: new-onset LBBB, RBBB with left axis deviation or other bifascicular block, and alternating bundle branch block.27 Although controversial, one author recommends prophylactic pacing for all new bundle branch blocks when MI is evident.28 Whether to place a transvenous pacemaker prophylactically in patients with LBBB before insertion of a flow-directed pulmonary artery catheter (PAC) remains controversial. Some researchers strongly advocate this procedure because of the

25 12 20 25 19 13 14 27 29 44 13 25 26 23

ABBB, alternating bundle branch block; AVB, atrioventricular block; BBB, bundle branch block; LAFB, left anterior fascicular hemiblock; LBBB, left bundle branch block; LPFB, left posterior fascicular hemiblock; RBBB, right bundle branch block. Reprinted by permission of the American Heart Association from Hindman MC, Wagner GS, JaRo M, et al: The clinical significance of bundle branch block complicating acute myocardial infarction. 2. Indications of temporary and permanent pacemaker insertion. Circulation 58:690, 1978.

risk of transient RBBB and life-threatening complete heart block associated with PAC placement.29 One study notes that this risk is low in patients with prior LBBB but continues to recommend temporary catheter placement for all cases of new LBBB.30 One solution to this problem is to place a transcutaneous pacemaker before catheterization as an emergency measure should heart block develop. In these cases, a temporary transvenous pacemaker can be placed in a semielective manner when needed.31 In any event, the trend toward decreased PAC use, particularly outside of the critical care setting, makes it unlikely that this will be an issue in the ED.32 One final point to bear in mind regarding bradydysrythmias in the setting of AMI is that most of the investigations into the use of temporary pacing were done in the prethrombolytic era. Modern treatment of AMI is substantially different, but more recent studies, particularly of prophylactic pacing, are lacking. Tachycardias Hemodynamically compromising tachycardias are usually treated by medical means or electrical cardioversion. Since 1980, there has been an increasing interest in pacing therapy for symptomatic tachycardias. Supraventricular dysrhythmias, with the exception of atrial fibrillation, respond well to atrial pacing. By “overdrive” pacing the atria at rates 10 to 20 beats/ min faster than the underlying rhythm, the atria become entrained, and when the rate is slowed, the rhythm frequently returns to normal sinus. A similar procedure is done for ventricular dysrhythmias.33 Overdrive pacing is especially useful for recurrent prolonged Q-T interval arrhythmias such as

Cardiac Pacing in Drug-Induced Dysrhythmias Significant dysrhythmias can occur from excessive therapeutic medication (often in combination therapy) and from overdose of cardioactive medications. Because these drugs have direct effects on the myocardial pacemaker and conduction system cells, cardiac pacing is usually of little therapeutic value. Both bradycardias and tachycardias may result. Tachycardic rhythms from amphetamines, cocaine, anticholinergics, cyclic antidepressants, theophylline, and others do not benefit from cardiac pacing. Drug-induced torsades de pointes may theoretically be overdriven by pacing, but data on this technique are lacking. Any drug that affects the central nervous system (e.g., opiates, sedative-hypnotics, clonidine) may produce bradycardia. Rare causes of toxin-induced bradycardia include organophosphate poisoning, various cholinergic drugs, ciguatera poisoning, and rarely, plant toxins. Cardiac pacing is not used for bradycardias from these sources; rather, the underlying central nervous system depression is addressed. Severe bradycardia and heart block often accompany overdose of digitalis preparations, β-adrenergic blockers, and calcium channel blockers. Although intuitively attrac­­ tive, cardiac pacing is generally not effective in serious toxininduced bradycardias, even though there are case reports of successes.36–39 In β-blocker overdose, pacing may increase heart rate but rarely benefits blood pressure or cardiac output. Worsening of the blood pressure may be seen from loss of atrial contractions with ventricular pacing. Likewise, calcium channel blocker overdose and digitalis-induced bradycardia and heart block rarely benefit from cardiac pacing. Pharmacologic interventions, such as digoxin-specific Fab, glucagon, calcium, inotropic medications, and vasopressors, remain the mainstay in the treatment of drug-induced dysrhythmias. Given the lack of success of pacing, possible downsides, and the greater effectiveness of specific antidotes, it is not standard to routinely attempt transvenous cardiac pacing in the setting of drug overdose. However, as a last resort, cardiac pacing can be supported.40

Contraindications The presence of a prosthetic tricuspid valve is generally considered to be an absolute contraindication to transvenous cardiac pacing.41 Also, severe hypothermia will occasionally result in ventricular fibrillation when pacing is attempted. Because ventricular fibrillation under these conditions is difficult to convert, caution is advised when considering pacing the severely hypothermic and bradycardic patient. Rapid and careful rewarming is often recommended first, followed by pacing if the patient’s condition does not improve.

Equipment Several items are required to insert a transvenous pacemaker adequately. Like most special procedures, a prearranged tray is convenient. The usual components required to insert a transvenous cardiac pacemaker are listed in Table 15–5.

Emergency cardiac pacing

Infarct location   Anterior   Indeterminate   Inferior or posterior PR interval   >0.20 sec   ≤0.20 sec Type of BBB   LBBB   RBBB   RBBB + LAFB   RBBB + LPFB   ABBB Onset of BBB   Definitely old   Possibly new   Probably new   Definitely new

Progressing to HighDegree AVB (%)

● 

Patients

those seen with quinidine toxicity or torsades de pointes.34 Although this is an attractive thought, there is no reported experience with these techniques in the ED. Transvenous pacing also is useful in patients with digitalis-induced dysrhythmias in whom direct current (DC) cardioversion may be dangerous or in patients in whom there is further concern about myocardial depression with drugs.35

15 

TABLE 15–4  The Influence of Different Variables on the Risk of High-Degree Atrioventricular Block in Patients with Bundle Branch Block during Myocardial Infarction

271

CARDIAC PROCEDURES ● 

III 

TABLE 15–5  Suggested Transvenous Cardiac Pacemaker Equipment Pacemaker Tray* 10-mL syringe 1% lidocaine Alcohol wipes Povidone-iodine (Betadine) Several gauze pads 4 sterile drapes No. 11 scalpel blade 0.9 normal saline–2 ampules Sterile gloves Needle holder Two 22-gauge needles Scissors (suture) Two 4-0 silk sutures on needles Sterile basin Introducer set (sheath, guidewire, dilator, introducer needle)

Patient cable

Rate control Output control Connector terminals Sensitivity

Electrical Hardware Insulated connecting wire with alligator clamps at each end (or a male-to-male adapter) Fresh battery and a spare Pacing generator Pacing catheter* 12-lead electrocardiographic machine (well grounded) *Some or all of these components are available in prepackaged sets.

272

Pacing Generator Many different pacing generators are available, but in general, they all have the same basic features. The controls frequently will have a locking feature or cover to prevent the generator from being switched off or reprogrammed inadvertently. An amperage control allows the operator to vary the amount of electrical current delivered to the myocardium, usually 0.1 to 20 mA. Increasing the setting increases the output and improves the likelihood of capture. The pacing control mode is determined by adjusting the gain setting for the sensing function of the generator. By increasing the sensitivity, one can convert the unit from a fixed-rate (asynchronous mode) to a demand (synchronous mode) pacemaker. The typical pacing generator has a sensitivity setting that ranges from about 0.5 to 20 mV. The voltage setting represents the minimum strength of electrical signal that the pacer is able to detect. Decreasing the setting increases the sensitivity and improves the likelihood of sensing myocardial depolarization. In the fixed-rate mode, the unit fires despite the underlying intrinsic rhythm; that is, the unit does not sense any intrinsic electrical activity. In the fulldemand mode, however, the pacemaker senses the underlying ventricular depolarizations and the unit does not fire as long as the patient’s ventricular rate is equal to or faster than the set rate of the pacing generator. A sensing indicator meter and rate control knob are also present. Temporary pacing generators are battery operated; thus, it is always good practice to install a fresh battery whenever pacing is anticipated. An example of a pacing generator is shown in Figure 15–1. Pacing Catheters and Electrodes Several sizes and brands of pacing catheters are available. In general, most range from 3 to 5 French in size and are approximately 100 cm in length. Lines are marked along the catheter surface at approximately 10-cm intervals; these can be used to

Figure 15–1  Pacemaker energy source controls and connections.

– Connection adapter to pacer box +

Electrode connector

Pacing catheter Balloon Extensible sleeve Introducer sheath Figure 15–2  Balloon-tipped pacing catheter.

estimate catheter position during insertion. Pacing catheters differ with respect to their stiffness, electrode configurations, floating characteristics, and other qualities. For emergency pacing, the semifloating bipolar electrode catheter with a balloon tip is used most frequently (Fig. 15–2). The balloon holds approximately 1.5 mL of air, and the air injection port has a locking lever to secure balloon expansion. Before insertion, the balloon is checked for air leakage by inflating it and immersing it in sterile water. The presence of an air leak is noted by a stream of bubbles rising to the surface of the water. An inflated balloon helps the catheter “float” into the heart, even in low-flow states, but is obviously not advantageous in the cardiac arrest situation. For all practical purposes, temporary transvenous pacing is accomplished with a bipolar pacing catheter. The terms unipolar and bipolar refer to the number of electrodes in contact with that portion of the heart that is to be stimulated.

Emergency cardiac pacing

Introducer Sheath An introducer set or sheath is required for venous access. Some pacing catheters are prepackaged with the appropriate equipment, whereas others require a separate set. The introducer set is used to enhance passage of the pacing catheter through the skin, subcutaneous tissue, and vessel wall. The sheath must be larger than the pacing catheter in order to allow it to pass. The size of the pacing catheter refers to its outside diameter, and the size of the introducer refers to its inside diameter. Thus a 5-French pacing catheter will fit through a 5-French introducer. Introducer sheaths are available with a perforated elastic seal covering the opening

● 

ECG Machine An ECG machine can be used to record the heart’s inherent electrical activity during pacer insertion and to aid in localization of the catheter tip without fluoroscopy. The ECG machine must be well grounded to prevent leakage of alternating current, which can cause ventricular fibrillation. Such leakage should be suspected if interference of 50 to 60 cycles/ sec (Hertz) is noted on the ECG. The ECG machine should be placed in such a manner as to allow easy visibility of the rhythm during insertion. One method is to place the machine near the level of the patient’s midthorax facing the operator, on either side of the patient as logistics and operator preference allow (Fig. 15–3). Note that the operator stands at the head of the patient during internal jugular or subclavian vein passage of the catheter and at the midabdomen for femoral or brachiocephalic vein insertion. Newer patient monitors may be equipped with suitable ECG connections to allow their use in place of a stand-alone ECG machine. Because these patients will already be attached to a monitor, it may prove convenient to use the same piece of equipment to assist with pacemaker insertion.

15 

All pacemaker systems must have both a positive (anode) and a negative (cathode) electrode; hence, all stimulation is bipolar. In the typical bipolar catheter used for temporary transvenous pacing, the cathode (stimulating electrode) is at the tip of the pacing catheter. The anode is located 1 to 2 cm proximal to the tip, and a balloon or an insulated wire separates the two electrodes. The distinction between the unipolar and the bipolar pacing catheter is that a bipolar catheter has both electrodes in relatively close proximity on the catheter, and both may contact the endocardium. In the bipolar catheter, the electrodes are usually stainless steel or platinum rings that encircle the pacing catheter. When properly positioned, both electrodes will be within the right ventricle so that a field of electrical excitation is set up between the electrodes. With the bipolar catheter, the cathode does not need to be in direct contact with the endocardium for pacing to occur, although it is preferable to have direct contact. A unipolar system is also effective but is used infrequently for temporary transvenous pacing. In a unipolar system, the cathode is at the tip of the pacing catheter, and the anode is located in one of three places: (1) in the pacing generator itself, (2) more proximally on the catheter (outside the ventricle), or (3) on the patient’s chest. The bipolar system may be converted to a unipolar system by simply disconnecting the positive proximal connection of the bipolar catheter from the pacing generator and running a new wire from the positive (pacing generator) terminal to the patient’s chest wall. Such a conversion may be required in the unlikely event of failure of one lead of the bipolar system.

Figure 15–3  Position of an electrocardiographic (ECG) device during insertion of a pacemaker catheter through the left subclavian vein.

through which the pacing catheter is passed (pacer port). The seal allows the catheter to be manipulated while preventing blood from escaping or air from entering the vein. A side port allows the sheath to be used for central venous access. A makeshift sheath can be fashioned with an appropriate-sized intravenous (IV) catheter. For the 3-French balloon-tipped catheter, a 14-gauge, 1.5- to 2-in. IV catheter is suitable. The 4-French balloon-tipped catheter will also fit through a 14gauge catheter or needle. However, without a seal over the hub, blood will leak from the end of the IV catheter. Overall, the key to success with this procedure is preparation. In a typical ED, there is often a variety of vascular access kits and devices, not all of which will work well, if at all, for passing a pacing catheter. It is imperative that one examine all the components of the tray before starting the procedure and ensure that all wires, sheaths, dilators, and syringes fit as expected. Ideally, all of the equipment and accessories needed for emergency pacemaker insertion should be kept together in a designated location.

Procedure A checklist for the preparation and initial setup of the pacing generator is shown in Box 15–1. It may be useful to have a copy of this or a similar list stored with the pacemaker to have on hand in emergent situations. Patient Preparation Patient instruction is an extremely important aspect of any procedure. Frequently, there is not enough time to give patients a detailed explanation or to obtain written informed consent. Nonetheless, sufficient information should be provided so that the patient feels at ease. It is always prudent to obtain and document informed consent from the patient, if possible, prior to any invasive procedure or to document that the circumstances did not allow informed consent. Patients should be assured that they will feel no discomfort after the venipuncture site has been anesthetized and that they will feel

273

CARDIAC PROCEDURES ● 

III 

better when the catheter is in place and is functional. Continued reassurance is required during the procedure because patients are usually facing away from the operator and their faces are often covered; thus, they may be unsure of what is occurring. Sedation and analgesia should be considered when appropriate. All operators should wear surgical masks, caps, gloves, and gowns to decrease the risk of infection before catheter placement. Patients should be prepared and draped in the usual sterile fashion. This aseptic precaution should also be explained to the patient. Site Selection The four venous channels that provide an easy access to the right ventricle are the brachial, subclavian, femoral, and internal jugular veins (Table 15–6). The route selected is often one of personal or institutional preference. The right internal jugular and the left subclavian veins have the straightest anatomic pathway to the right ventricle and are generally preferred for temporary transvenous pacing. In some centers, a particular site

BOX 15–1

274

Checklist for Temporary Transvenous Pacing Generator*

• Insert new battery • Turn pacemaker ON • Set RATE (80 beats/min) OUTPUT (5 mA) and SENSITIVITY (3 mV)† • Connect patient cable to pacemaker • Open both connector terminals on patient cable • Insert PROXIMAL (+) pin of pacing catheter into POSITIVE (+) connector terminal on patient cable • Tighten connection firmly • Use alligator clips to connect DISTAL (–) pin of pacing catheter to lead V1 of ECG machine • When catheter is in position, remove DISTAL (–) pin from V1 and insert it into the NEGATIVE (–) connector terminal on patient cable • Tighten connection firmly *Adjust pacemaker settings as needed to achieve proper capture and sensitivity (see text). † Guidelines only. Follow recommendations of device manufacturer if different. ECG, electrcardiographic.

is preferred for permanent transvenous pacemaker placement, and if possible, this site should be avoided for temporary placement. The subclavian vein can be accessed by both an infraclavicular and a supraclavicular approach; the infraclavicular approach is most commonly reported for all temporary transvenous pacemaker insertions. This route is preferred because of its easy accessibility, close proximity to the heart, and ease in catheter maintenance and stability. The supraclavicular approach has been described in the literature for several years and has gained popularity among some clinicians.42,43 The left subclavian vein is preferred because of the less acute angle traversed when compared with the right-sided approach, but either side may be used.42,43 The internal jugular approach may also be used. In this case, the right internal jugular vein is preferred because of the direct line to the superior vena cava. Problems with this approach include dislodgment of the pacemaker with movement of the head, carotid artery puncture, and thrombophlebitis. During CPR, the use of the right internal jugular vein and the left subclavian veins for pacemaker insertion have been demonstrated to result in the highest rates of proper placement in the right ventricle.44 The right internal jugular vein is the more direct route of the two and may be the most appropriate site. Femoral veins, like neck veins, are compressible and easily catheterized. Problems include easy dislodgment, infection, and increased risk of thrombophlebitis.45,46 Brachial vein catheterization is easy to perform but results in a high incidence of infection and vessel thrombosis.47 In addition, the catheter is easily dislodged with arm motion. This approach is seldom used in the emergency setting. Although the left subclavian and right internal jugular veins are the preferred routes for access, in an emergency situation, the clinician should use the approach with which she or he is most experienced so as to minimize the time spent in cannulating the vein and reduce the potential for complications from the venipuncture. Skin Preparation and Venous Access The skin over the venipuncture site is cleaned twice with an antiseptic solution such as chlorhexidine or povidone-iodine.

TABLE 15–6  Advantages and Disadvantages of Pacemaker Placement Sites Venous Channels

Advantages

Brachial

Very safe route Vessel easily accessible, either by cutdown or percutaneous approach Compressible

Subclavian

Direct access to right heart (especially via left subclavian) Rapid insertion time Good patient mobility Direct access to right heart Rapid insertion time Compressible

Femoral

Internal jugular

Direct access to right heart (especially via right internal jugular) Rapid insertion time Compressible

Disadvantages Often requires cutdown Easily displaced and poor patient mobility Not reusable if cutdown technique is performed Catheter is more difficult to advance than with central or larger vessels Pneumothorax and other intrathoracic trauma are possible Noncompressible Increased incidence of thrombophlebitis Can be dislodged by leg movement and poor patient mobility Infection Possible carotid artery puncture Dislodgment with movement of the head Thrombophlebitis

15 

Cable to pacemaker generator

● 

Left innominate vein Pacing catheter

– Superior vena cava

Distal (negative) lead of pacing catheter

Introducer sheath

Figure 15–4  Insertion of the pacing catheter through the introducer sheath.

A wide area is prepared because of the tendency for guidewires and catheters to spring from the hands of the unsuspecting operator. Similarly, wide draping is carried out in the standard manner to maintain a sterile field and to allow clear visibility of the venipuncture site. The infraclavicular approach is used in this chapter to illustrate venous access, although the mechanics are generally the same for other vascular approaches. Occasionally, a patient who already has a central venous line in place requires the emergent placement of a pacing catheter. An existing central venous pressure (CVP) line can be used to place the pacing catheter if the catheter lumen is large enough to accept a guidewire. The CVP line should be withdrawn 3 to 5 cm to expose an area of sterile tubing. The tubing is transected through a sterile area while being held firmly at the skin level. A guidewire can then be passed through the tubing, and the tubing can be withdrawn, leaving only the wire in the vein. The guidewire and the tubing should never be released, because embolization may result. With the guidewire in place, a dilator and introducer sheath can then be passed together over the guidewire, as is done in the Seldinger technique. The dilator and guidewire are then removed and the pacing catheter can be passed through the introducer sheath (Fig. 15–4). One key additional step to help preserve sterility while manipulating the pacing catheter is to attach an extensible sleeve on the end of the introducer prior to inserting the pacing catheter (see Fig. 15–2). In this way, the pacing catheter can be advanced and withdrawn multiple times without fear of contamination. Bedside ultrasound can be useful as an aid to securing central venous access and its use in the setting of emergency transvenous pacing has been reported.48,49 Pacemaker Placement ECG Guidance.  The patient should be connected to the limb leads of an ECG machine, and the indicator should be turned to record the chest (V) lead. With newer ECG machines, the pacemaker may be attached to any of the V leads (usually V1 or V5) that are displayed during rhythm monitoring. The distal terminal of the pacing catheter (the cathode, or lead marked “negative”, “−,” or “distal”) must be connected to the V lead of the ECG machine by a male-to-male connector or an insulated wire with an alligator clip on each end (Fig. 15–5). The pacing catheter is thus an exploring electrode that creates a unipolar electrode for intracardiac ECG recording. The electrocardiogram recorded from the electrode tip localizes the position of the tip of the pacing electrode. As the tracing on the ECG machine may be slightly delayed, advancement of the catheter after initial insertion must be carefully evaluated. If a balloontipped catheter is used, the balloon is inflated with air after

V1 lead of ECG machine Wire with alligator clamps

Emergency cardiac pacing

+

Figure 15–5  Using alligator clips to connect the pacemaker to the V1 lead of an ECG machine.

the catheter enters the superior vena cava (~10–12 cm for a subclavian or internal jugular insertion). The inflation port should be locked and the syringe left attached. The pacing catheter should be advanced both quickly and smoothly. The V lead should be monitored, and the P-wave and QRS complex should be observed to ascertain the location of the pacing catheter tip. The use of an electrocardiogram to guide the placement of a pacing catheter is based on two concepts. First, the complex will vary in size depending on which chamber is entered. For example, when the tip of the pacing catheter is in the atrium, one will see large Pwaves, often larger than the corresponding QRS complex. Second, the sum of the electrical forces will be negative if the depolarization is moving away from the catheter tip and positive if the depolarization is moving toward the catheter tip. Therefore, if the catheter tip is above the atrium, both the Pwave and the QRS complex will be negative (i.e., the electrical forces of a normally beating heart will be moving away from the catheter tip). As the tip progresses inferiorly in the atrium, the P-wave will become isoelectric (biphasic) and will eventually become positive as the wave of atrial depolarization advances toward the catheter tip. The electrocardiogram resembles an aVR lead initially when in the left subclavian vein (Fig. 15–6A) or midsuperior vena cava (see Fig. 15–6B). At the high right atrium, both the P-wave and the QRS complex are negative; the P-wave is larger than the QRS complex and is deeply inverted (see Fig. 15–6C and D). As the center of the atrium is approached, the P-wave becomes large and biphasic (see Fig. 15–6E). As the catheter approaches the lower atrium (see Fig. 15–6F), the P-wave becomes smaller and upright. The QRS complex is fairly normal. When striking the right atrial wall, an injury pattern with a P-Ta segment is seen (see Fig. 15–6G). As the electrode passes through the tricuspid valve, the P-wave becomes smaller and the QRS complex becomes larger (see Fig. 15–6H). Placement in the inferior vena cava may be recognized by a change in the morphology of the P-wave and a decrease in the amplitude of both the P-wave and the QRS complex (see Fig. 15–6I). Once the pacing catheter is in the desired position, deflate the balloon by unlocking the port, observing that the syringe refills with air spontaneously, and then removing the syringe. One should avoid drawing back on the syringe because this may cause balloon rupture. If the syringe does not refill spontaneously, the operator should suspect that the balloon might be ruptured. The balloon should not be inflated and the

275

CARDIAC PROCEDURES ● 

III 

A

Left subclavian v.

B

Mid superior vena cava

High right atrium

C

High right atrium

D

QRS

P P-Ta

E

Mid right atrium

F

Low right atrium

G

Right mid-atrium (against wall)

H

Right ventricle (free)

276

I

Inferior vena cava

J

Right ventricle (against wall)

K

Pulmonary artery

Figure 15–6  A–K, Intracardiac electrocardiography: Electrical signals of atrial and ventricular depolarization and repolarization from different vascular and intracardiac locations (see text).  (A–F and H–K, From Bing OH, McDowell JW, Hantman J, et al: Pacemaker placement by electrocardiographic monitoring. N Engl J Med 287:651, 1972; G, from Goldberger E: Treatment of Cardiac Emergencies, 3rd ed. St. Louis, CV Mosby, 1982, p 252.)

pacing catheter should be withdrawn and the balloon checked for leaks. If a leak is found, a new pacing catheter should be used. After successful passage of the catheter into the right ventricle, the tip should be advanced until contact is made with the endocardial wall. When this occurs, the QRS segment will show ST segment elevation (see Fig. 15–6J). Ideally, the tip of the catheter should be lodged in the trabeculae at the apex of the right ventricle; however, pacing may be successful if the catheter is in various other positions within the ventricle or outflow tract. If the pacer enters the pulmonary artery outflow tract, the P-wave again becomes negative and the QRS amplitude diminishes (see Fig. 15–6K). If the catheter is in the pulmonary artery, the pacing catheter should be withdrawn into the right ventricle and readvanced. Sometimes, a clockwise or counterclockwise twist of the catheter will redirect its path in a more favorable direction. If catheter-induced ectopy develops, the catheter should be slightly withdrawn until the ectopy stops; then it should be readvanced. Occasionally, an antidys-

rhythmic drug such as lidocaine may need to be given to desensitize the myocardium. Once ventricular endocardial contact is made, the catheter is disconnected from the ECG machine and the distal lead is now connected to the negative terminal on the pacing generator. The pacing generator is then set to a rate of 80 beats/min or 10 beats/min faster than the underlying ventricular rhythm, whichever is higher. The full-demand mode is selected, with an output of about 5 mA. The pacing generator is then turned on. The patient should be assessed for electrical and mechanical capture. Electrical capture will be manifest on the ECG monitor as a pacer spike followed by a QRS complex. If the pacer spike is seen but no QRS follows, capture is not occurring. Mechanical capture means that a pacer spike with its corresponding QRS triggers a myocardial contraction. This can be assessed by checking that a palpable pulse is present that is equal to the rate set on the pacemaker. If complete capture does not occur or if it is intermittent, the pacer will need to be repositioned. When proper capture occurs, the pacer should be assessed for optimal positioning. This is done by

The complications of emergency transvenous cardiac pacing are numerous13,55 and represent a compendium of those related to central venous catheterization, those related to right-sided heart catheterization, and those unique to the pacing catheter itself (Table 15–7). Problems Related to Central Venous Catheterization Inadvertent arterial puncture is a well-known complication of the percutaneous approach to the venous system.56 This problem is usually recognized quickly because of the rapid return of arterial blood. Firm compression over the puncture site will almost always result in hemostasis in 5 minutes or less. Venous thrombosis and thrombophlebitis are also potential problems with central venous catheterization. Thrombophlebitis, which occurs early after insertion, is an uncommon complication. Thrombosis of the innominate vein is also a rare problem, with pulmonary embolism an even more

Emergency cardiac pacing

Testing Sensing The sensing function should be tested in patients who have underlying rhythms. Set the rate at about 10 beats/min greater than the endogenous rhythm, place the pacemaker in asynchronous mode (minimum sensitivity, which is the maximum setting on the sensitivity voltage control), and ensure that there is complete capture. Then adjust the sensitivity control to its midposition or approximately 3 mV, and gradually decrease the rate until pacing is suppressed by the patient’s intrinsic rhythm. The sensing indicator on the pacing generator should signal each time a native beat is sensed and should be in synchrony with each QRS on the ECG monitor. If the pacer fails to sense the intrinsic rhythm, increase the sensitivity (decrease the millivolts) until the pacer is suppressed. Conversely, if the sensing indicator is triggered by P- or T-waves or by artifact, decrease the sensitivity until only the QRS is sensed. Once the sensitivity threshold is determined, set the millivoltage to about half of that value.

Complications

● 

Testing Threshold The threshold is the minimum current necessary to obtain capture. Ideally, this is less than 1.0 mA and usually between 0.3 and 0.7 mA. If the threshold is in this ideal range, good contact with the endocardium can be presumed. To determine the threshold, the pacing generator should be set to maximum sensitivity (full-demand mode) at 5 mA output with a rate of approximately 80 beats/min (or at least 10 beats/min greater than the patient’s intrinsic rate). The current (output) should then be reduced slowly until capture is lost. This current is the threshold. This maneuver should be carried out 2 or 3 times to ensure that this value is consistent; the amperage should then be increased to 2.5 times the threshold to ensure consistency of capture (usually between 2 and 3 mA).

Securing and Final Assessment After the pacemaker’s position has been tested for electrical accuracy, it must be secured in place. If a sealed introducer sheath was used, the hub should be fixed firmly to the skin with suture (e.g., 4-0 nylon or silk). A fastening suture should be sewn to the skin and the hub tied securely in place. If a plain introducer was used, it should be withdrawn to prevent leakage (Fig. 15–8) and the catheter should be sutured in place. In either case, the excess pacing catheter should be coiled and secured in a sterile manner underneath a large sterile dressing. Pacemaker function should again be assessed, and a chest film should be taken to ensure proper positioning. Ideal positioning of the pacing catheter is at the apex of the right ventricle (Fig. 15–9). A 12-lead electrocardiogram should be obtained after transvenous pacemaker placement. If the catheter is within the right ventricle, a left bundle branch pattern with left axis deviation should be evident in paced beats (Fig. 15–10). If an RBBB pattern is noted, coronary sinus placement or left ventricular pacing due to septal penetration should be suspected. With a properly functioning ventricular pacemaker, large cannon waves may be noted on inspection of the venous pulsations at the neck. When the pacemaker achieves ventricular capture, there may be times when the atria contract against a closed tricuspid valve resulting in a cannon wave. On auscultation of the heart, a slight murmur secondary to tricuspid insufficiency from the catheter interfering with the tricuspid valve apparatus may be evident.52 A clicking sound heard best during expiration after each pacemaker impulse may also be noted here and is believed to represent either intercostal or diaphragmatic muscular contractions caused by the pacemaker.53 Note that this can also be a sign of cardiac perforation.54 On auscultation of the second heart sound, paradoxical splitting may be noted. This represents a delay in closure of the aortic valve because of delayed left ventricular depolarization. As in any procedure, the patient should then be assessed for improvement in his or her clinical status. An evaluation of vital signs, mentation, congestive symptoms, and urinary output must be noted. In addition, complications secondary to the procedure should be sought and treated as needed.

15 

testing the thresholds for pacing and sensing and by physical examination, electrocardiography, and chest radiographs. The sequence of events is demonstrated in detail in Figure 15–7. Catheter Placement in Low-Flow States.  If the cardiac output is too low to “float” a pacing catheter or if the patient is in extremis, there may not be enough time to advance a pacing catheter using the previously described techniques. Such a situation would be asystole or complete heart block with malignant ventricular escape rhythms (although one can make a case for transcutaneous pacing in such conditions). In such emergent situations, the pacing catheter is connected to the energy source, the output is turned to the maximum amperage, and the asynchronous mode is selected. The catheter is then blindly advanced in the hope that it will enter the right ventricle and that pacing will be accomplished. The pacing catheter is rotated, advanced, withdrawn, or otherwise manipulated according to the clinical response. The right internal jugular approach is the most practical access route in this situation. Ultrasound Guidance.  As bedside ultrasound has become more widely available in the ED, new uses have been discovered. One promising technique involves using ultrasound to assist with the placement of emergency transvenous pacing catheters.50,51 Cardiac image ultrasound may also help demonstrate whether or not mechanical capture has been achieved. The advantages of ultrasound over fluoroscopy are its safety and ready availability. Further experience will be necessary to confirm its utility.

277

CARDIAC PROCEDURES ● 

III 

A

B

C

D

E

F

278

Figure 15–7  How to pass a temporary transvenous pacemaker.  Note: this is a 2-person procedure. The operator attends to pacer placement. The assistant observes the patient, the monitor, coordinates equipment, and orders appropriate drug therapy.A, Intravenous access obtained with an introducer in the right internal jugular vein provides direct access to the right ventricle via the superior vena cava and right atrium. The left subclavian vein is the next best choice. B, Attach the still-compressed sterile sheath to the introducer hub, making sure that the connector of the sheath is firmly attached to the hub of the introducer. Open the hub of the introducer by turning it counterclockwise to allow passage of the pacing wire. Inflate then deflate the balloon on the pacing wire before it is introduced to test it for integrity. There is a valve that keeps the balloon inflated; it must be turned to inflate/deflate the balloon. Use 1.2–1.5 mL of air for the balloon. C, An assistant attaches the proximal pacing wire to the nonsterile energy source. Use the demand mode and turn on the pacer output to the highest level, rate about 80/min. With the balloon deflated, insert the pacing wire into the still collapsed sheath and into the hub of the introducer. D, Slowly advance the wire through the introducer. Inflate the balloon when the tip of the pacing wire is in the superior vena cava and continue to advance. Close the valve to keep the balloon inflated. E, Watch the electrocardiogram and look for capture, demonstrated by a wide QRS pattern after each pacer spike. The right ventricle should be encountered at 15–20 cm as noted by markings on the pacing wire. Misplacement and coiling of the wire may preclude placement into the right ventricle. If no capture is seen by 25 cm, withdraw the wire and try again—this is a blind procedure and luck plays a role. When consistent capture is seen, deflate the balloon and advance the wire 1–2 cm more to seat the wire in the endocardium. F, Tighten the valve on the sheath introducer to stop subsequent movement of the wire, and extend the sheath its full length. If required, suture the wire in place. Turn off the energy, then slowly turn it up to determine pacing threshold (first sign of capture). Set the output at two to three times the stimulation threshold and set the desired rate. Leave the pacer in the demand mode until stability is assured. Obtain a chest x-ray and 12-lead electrocardiogram.  (A–F, From Thomsen T, Setnik G [eds]: Procedures Consult—Emergency Medicine Module. Copyright 2008 Elsevier Inc. All rights reserved.)

Site of incision

15  ● 

Subclavian vein

Superior vena cava

Wire

RA

RV LV Apex of the heart

A Terminals to pacemaker +

Emergency cardiac pacing

Pacemaker with lithium battery and timer

Cannula −

Connector

Figure 15–8  Transvenous pacer in right ventricle. LV, left ventricle; RA, right atrium; RV, right ventricle.

uncommon event.57 Femoral vein thrombosis, however, appears to be a much more common event associated with femoral vein catheterization.45,58 Studies using noninvasive techniques have shown a 37% incidence of femoral vein thrombosis, with 55% of these having ventilation-perfusion scan evidence of pulmonary embolism.58 Pneumothorax is consistently a problem with the various approaches to the veins at the base of the neck. The decision to place a chest tube in patients with this complication depends on the extent of the air leak and the clinical status of the patient. In addition, laceration of the subclavian vein with hemothorax,59 thoracic duct laceration with chylothorax, air embolism, wound infections, pneumomediastinum, hydromediastinum, hemomediastinum,60 phrenic nerve injury,61 and fracture of the guide wire with embolization62,63 are all potential complications.34,59 Complications of Right-Sided Heart Catheterization A common complication of the pacing catheter is dysrhythmia, with premature ventricular contractions being a common occurrence. One study noted a 1.5% incidence of serious dysrhythmias with a balloon-tipped catheter using ECG guidance, compared with a 32% incidence with the semirigid catheter using fluoroscopic guidance, suggesting that the balloon catheter was the preferred type of catheter.12 Another study noted a 6% incidence of ventricular tachycardia during insertion.45 The ischemic heart is more prone to dysrhythmias than the nonischemic heart.64 The therapy for catheterinduced ectopy during insertion involves repositioning the catheter in the ventricle. This usually stops the ectopy; however, if after repeated attempts, it is found that the catheter cannot be passed without ectopy, myocardial suppressant therapy may be used to desensitize the myocardium. Misplacement of the pacing catheter has been well studied. Passage of the catheter into the pulmonary artery can be diagnosed electrocardiographically by observing the return of an inverted P-wave and the decrease in the voltage of the QRS complex. Misplacement in the coronary sinus may occur and should be suspected in the patient in whom a paced RBBB pattern on the electrocardiogram is seen with right ventricular pacing (Fig. 15–11). Rarely, an RBBB pattern can be seen with a normal right ventricular position; therefore, all RBBB patterns do not represent coronary sinus pacing.65 Further

RA RV

B

C Figure 15–9  A, Schematic of proper position of pacer lead in apex of right ventricle. Normal pacemaker position in apex of right ventricle on posteroanterior (B) and lateral (C) chest films.  (A, From Chabner DA: The Language of Medicine, 6th ed. Philadelphia, Saunders, 2001.)

279

CARDIAC PROCEDURES

AVR

V1

V4

II

AVL

V2

V5

III

AVF

V3

V6

III 

● 

I

Figure 15–10  ECG pattern of right ventricular pacemaker.

280

TABLE 15–7  Complications of Transvenous Cardiac Pacing Year

Reference

Patients (N) 1

111

Catheter

Route

Flexon steelwire vs. unipolar semifloating (ECG) 3-Fr bipolar semifloating balloon (ECG) 6-Fr bipolar (fluoroscopy)

96 Subclavian 5 Basilic 1 External jugular Antecubital vein

12 inconsistent pacing, 3 local infection, 2 pneumothorax, 1 subclavian artery puncture; 16% complication rate 2 PVCs, stable pacing, no thrombophlebitis

Femoral

2 ventricular tachycardia, 2 perforations, 2 required repositionings, 1 questionable thrombophlebitis and pulmonary embolism, 1 local infection 12 ventricular tachycardia and fibrillation in 9 patients, 3 perforations in 2 patients; local hematoma, abscess, and bleeding in 30%; 16.9% complication rate 25% deep vein thrombosis

1969

Rosenberg et al

1973

Schnitzler et al11

17

1973

Weinstein et al46

100

1973

Lumia & Rios*

142 insertions in 113 patients

Bipolar (fluoroscopy)

61 Brachial 81 Femoral

1980

Pandian et al†

20

Femoral

1980

Nolewajka et al58

29

5-Fr bipolar (fluoroscopy) 6-Fr Cordis (fluoroscopy)

1981

Lang et al12

111

Balloon, semifloating vs. semirigid

Subclavian

1982

Austin et al45

113 insertions in 100 patients

4–7-Fr bipolar (fluoroscopy)

Brachial Femoral

Femoral

Result

34% venous thrombosis by venogram with 60% of these with pulmonary embolism by  scan V Q Serious dysrhythmia: 1.5% balloon-tipped, 20.4% semirigid Catheter displacement: 13.6% ± 4.4 days balloon-tipped; 32% ± 1.9 days semirigid Failure to sense or pace in 37%; repositioning in 37% of brachial insertions; repositioning in 9% of femoral insertions; fever, sepsis, local infection only in femoral insertions; 20% complication rate

 ventilation-perfusion. ECG, electrocardiogram; PVC, premature ventricular contraction; V Q, *Lumia FJ, Rios JC: Temporary transvenous pacemaker therapy: An analysis of complications. Chest 64:604, 1973. † Pandian NG, Kosowsky BD, Gurewich V: Transfemoral temporary pacing and deep venous thrombosis. Am Heart J 100:847, 1980.

V4

II

AVL

V2

V5

III

AVF

V3

V6

evidence for coronary sinus location can be obtained by viewing the lateral chest film. Normally, the catheter tip should point anteriorly toward the apex of the heart; however, with coronary sinus placement, the catheter tip is displaced posteriorly and several centimeters away from the sternum (Fig. 15–12). Other potential forms of misplacement include left ventricular pacing through an atrial septal defect or a ventricular septal defect, septal puncture, extraluminal insertion, and arterial insertions.66 Perforation of the ventricle is a well-described complication that can result in loss of capture,67 hemopericardium, and tamponade.68,69 Reported symptoms and signs of this problem include chest pain, pericardial friction rub, and diaphragmatic or chest wall muscular pacing.70 At least one case of a postpericardiotomy-like syndrome and two cases of endocardial friction rub have been reported without perforation.71,72 Pericardial perforation is suggested radiographically when the pacing catheter is outside or abuts the cardiac silhouette and is not in proper position within the right ventricular cavity (Fig. 15–13).73 ECG clues include a change in the QRS and T-wave axis or a failure to properly sense. In suspected cases, a two-dimensional echocardiogram usually demonstrates the catheter’s extracardiac position. Simply pulling back the catheter and repositioning it in the right ventricle can usually treat uncomplicated perforation. During the insertion of a temporary pacing catheter when a nonfunctioning permanent catheter is in place, there is a small risk of entanglement or knotting.74 This potential also exists with other central lines and PACs. Even without the presence of other lines, the pacing catheter can become knotted.75 Frequently, these lines can be untangled under fluoroscopy using specialized catheters. Local and systemic infections,47 balloon rupture, pulmonary infarction,76 phrenic nerve pacing,77 and rupture of the chordae tendineae are also potential complications.76

Figure 15–11  Coronary sinus pacing. Note the paced right bundle branch block pattern.

Emergency cardiac pacing

V1

● 

AVR

15 

I

281

A

B Figure 15–12  Coronary sinus position. A, Posteroanterior view. B, Lateral view.  (A and B, From Goldberger E: Treatment of Cardiac Emergencies, 3rd ed. St. Louis, CV Mosby, 1982.)

Complications of the Pacing Electrode The complications related to the pacing electrode can be separated into three groups: mechanical, organic, and electrical. Mechanical failures include displacement, fracture of the catheter, and loose leads. Displacement can result in intermit-

CARDIAC PROCEDURES III 

● 

Indications and Contraindications

Figure 15–13  A pacing catheter that is outside or abuts the cardiac silhouette and is not properly positioned within the right ventricular cavity suggests myocardial perforation.  (From Tarver RD, Gillespie KR: The misplaced tube. Emerg Med Clin North Am 20:97, 1988.)

282

tent or complete loss of capture or improper sensing, malignant dysrhythmias, diaphragmatic pacing, or perforation. Displacement should be suspected with changes in amplitude, with vector changes greater than 90°, or with a change in threshold.78 Frequently, catheter fractures may be detected by a careful review of the chest film or may be suspected because of a change in the sensing threshold. As with displacement, catheter fractures may result in intermittent or complete loss of capture. Organic causes of pacemaker failure result in changes in the threshold or sensing function.79 Progressive inflammation, fibrosis, and thrombosis may result in more than a doubling of the original threshold.80 However, this process takes several weeks and is of no concern in the setting of ED pacemaker placement. Electrical problems with pacing in the past have included pacemaker generator failure, dysrhythmias, and outside interference. Modern devices are extremely reliable and resistant to outside interference. Although ventricular tachycardia and ventricular fibrillation have been reported to result from pacemakers, these are rare. Because of this, patients who present with such dysrhythmias should be evaluated for a nonpacemaker-induced etiology.81 Defibrillation and cardioversion are safe in patients who have temporary pacemakers.

EMERGENCY TCP TCP is a rapid, minimally invasive method of emergency cardiac pacing that may temporarily substitute for transvenous pacing. Electrodes are applied to the skin of the anterior and posterior chest walls, and pacing is initiated with a portable pulse generator. In an emergency setting, this pacing technique is faster and easier to initiate than transvenous pacing. Pulse generators are sufficiently portable to be used in EDs, hospital wards, intensive care units, and mobile paramedic vehicles. BACKGROUND  c a n

be found on

E x p e rt C o n s u lt

General indications for cardiac pacing are discussed earlier. TCP is the fastest and easiest method of emergency pacing. This technique is useful for initial stabilization of the patient in the ED who requires emergency pacing while arrangements or decisions for transvenous pacemaker insertion are being made. The equipment is readily mastered, and the procedure is fast and minimally invasive.91,94 Refinements in equipment have made TCP the emergency pacing procedure of choice. TCP is also widely used in the prehospital environment as well as in hospital in the cardiac catheterization laboratory, operating room, intensive care units, and on general medical floors.95–97 The technique may be preferable to transvenous pacing in patients who have received thrombolytic agents or other anticoagulants. No central venous puncture, with the attendant risk of hemorrhage, is required. Limited experience suggests that TCP also may be useful in the treatment of refractory tachydysrhythmias by overdrive pacing.98–102 Although small pediatric electrodes for TCP have been developed, experience with pediatric TCP has been limited.103,104 TCP is indicated for the treatment of hemodynamically significant bradydysrhythmias that have not responded to medical therapy. Hemodynamically significant implies hypotension, anginal chest pain, pulmonary edema, or evidence of decreased cerebral perfusion. This technique is temporary and is indicated for short intervals as a bridge until transvenous pacing can be initiated or until the underlying cause of the bradydysrhythmia (e.g., hyperkalemia,94 drug overdose105) can be reversed. Although generally unsuccessful, TCP may be attempted in the treatment of asystolic cardiac arrest. In this setting, the technique is efficacious only if used early after arrest onset (generally within 10 min).106,107 TCP is not indicated for treatment of prolonged arrest victims with a final morbid rhythm of asystole.104,108–110 Delay from the onset of arrest to the initiation of pacing is a major problem that limits the usefulness of TCP in prehospital care. Hedges and associates107 reported that everyday availability of pacing increased the number of patients who received pacing within 10 minutes of hemodynamic decompensation and increased long-term patient survival as well. Prehospital pacing may be most useful in the treatment of the patient with a hemodynamically significant bradycardia who has not yet progressed to cardiac arrest (e.g., heart block in the setting of AMI) or in the patient who arrests after the arrival of prehospital providers.106,107 In conscious patients with hemodynamically stable bradycardias, TCP may not be necessary. It is reasonable to attach electrodes to such patients and to leave the pacemaker in standby mode against the possibility of hemodynamic deterioration while further efforts at treatment of the patient’s underlying disorder are being made. This approach has been used successfully in patients with new heart block in the setting of cardiac ischemia.111 Generally, when a transvenous pacemaker becomes available, transvenous pacing is preferred because of better patient tolerance.

Equipment Since their reintroduction, transcutaneous pacemakers have undergone rapid evolution and are now standard equipment in most EDs as well as other hospital settings and the prehospital environment. The pacemakers introduced in the early

C

D

Emergency cardiac pacing

B

EKG leads

● 

A

deliver electrical impulses for pacing. Newer combined defibrillator-pacemakers can use a single set of electrodes for ECG monitoring, pacing, and defibrillation. This approach makes use of the device simpler, although the ECG waveform and analysis may be suboptimal. Provisions generally are

15 

1980s tended to be asynchronous devices with a limited selection of rate and output parameters. Units introduced more recently have demand mode pacing and more output options and are often combined with a defibrillator in a single unit. Combined defibrillator-pacers offer advantages in cost and ease and rapidity of use when compared with stand-alone devices. An example of a combined unit is shown in Figure 15–14. All transcutaneous pacemakers have similar basic features. Most allow operation in either a fixed rate (asynchronous) or a demand mode (VVI). Most allow rate selection in a range from 30 to 200 beats/min. Current output is usually adjustable from 0 to 200 mA. If an ECG monitor is not an integral part of the unit, an output adapter to a separate monitor is required to “blank” the large electrical spike from the pacemaker impulse and allow interpretation of the much smaller ECG complex. Without blanking pro­ tection, the standard ECG machine is swamped by the pacemaker spike and is uninterpretable. This could be disastrous because the large pacing artifacts can mask treatable ventricular fibrillation (Fig. 15–15). Pulse durations on available units vary from 20 to 40 msec and are not adjustable by the operator. Two sets of patient electrodes are usually required for operation of the device. One set of standard ECG electrodes is used for monitoring. The much larger pacing electrodes

Transcutaneous pads

Figure 15–14  Combined pacemaker-defibrillator-ECG monitor. ECG leads are on the left. Transcutaneous pacing electrodes are connected on the right. Note that the connector for the pacing electrodes is outside of the package, allowing for rapidity and ease of use while preserving shelf life of the electrodes. Figure 15–15  A–C, The top three rhythm strips are taken from a standard wall-mounted ECG monitor. They all demonstrate large pacer spikes without capture. The underlying rhythm cannot be determined and could be treatable ventricular fibrillation. D, The bottom rhythm strip demonstrates a tracing on the same patient with the external pacer monitor (special dampening). Note that the pacing spikes are much smaller, and it is easily seen that the underlying rhythm is asystole, without pacer capture. The presence of a T-wave after the QRS complex is a good indicator of ventricular capture.

283

CARDIAC PROCEDURES ● 

III 

made for separate ECG monitoring electrodes for use as desired by the operator. Along with the widespread use of TCP come problems arising out of lack of equipment standardization. Pacing electrodes placed on a patient prehospital may be incompatible with the transcutaneous pacemaker used in the ED, and likewise, the equipment in the ED may differ from the in-patient units. Efforts should be made to establish a single standard for pacing electrode connectors within an institution and out to the prehospital environment if possible. In order to facilitate setup for capture, the pacing electrodes should be connected to the pacemaker at all times. With conventional packaging, the leads are inside the packet with the pads, which means that the packet must be opened to allow connection to the pacing unit. However, exposure to the air causes the electrodes to dry out and lose their conductivity thus requiring continual replacement of the unused electrodes. Newer packaging, as shown in Figure 15–14, leaves the connectors outside of the packet, thus allowing connection to the pacemaker while preserving the shelf life of the pacing electrodes.

Technique

284

Pad Placement The pacing electrodes are self-adhesive and positioned as shown in Figure 15–16. Care should be taken to avoid placing the electrodes over an implanted pacemaker or defibrillator, and any transdermal drug delivery patches should be removed if they are in the way. Excessive hair may be removed if time permits. The anterior electrode (cathode or negative electrode) is placed as close as possible to the point of maximal impulse on the left anterior chest wall, and the second electrode is placed directly posterior to the anterior electrode (see Fig. 15–16A). The posterior electrode serves as the ground. An alternative arrangement for the pacing electrodes is shown in Figure 15–16B. On females, the electrode must be placed beneath the breast. Data regarding optimum electrode placement are scarce, so selection can be made based on the clinician’s preference and the patient’s habitus.112,113 Nonetheless, suboptimal capture owing to electrode placement may be rectified with a small change in electrode position. Although the polarity of the electrodes does not appear to be important for defibrillation, at least one study indicated that it might be for pacing.114 The electrodes are labeled by the manufacturer to indicate which should be placed over the precordium; it is

A

B

prudent to observe this recommendation. ECG electrodes (if used) are placed on the chest wall and/or limbs as required and connected to the instrument cable. Some clinicians prophylactically apply pacing electrodes to all critically ill patients with bradycardia to facilitate immediate TCP should decompensation occur. There is little risk of electrical injury to health care providers during TCP. Power delivered during each impulse is less than 1/1000 of that delivered during defibrillation.115 Chest compressions (CPR) can be administered directly over the insulated electrodes while pacing.116 Inadvertent contact with the active pacing surface results in only a mild shock. Pacing Bradycardic Rhythms To initiate TCP, the pacing electrodes are applied and the device is activated. In the setting of bradyasystolic arrest, it is reasonable to turn the stimulating current to maximal output and then decrease the output as appropriate after capture is achieved. Clinicians should slowly increase the output from minimal settings until capture is achieved in patients with a hemodynamically compromising bradycardia who are not in cardiac arrest. Rate and current (output) selections are adjustable (Fig. 15–17). Generally, a heart rate of 60 to 70 beats/min will maintain an adequate blood pressure (by blood pressure cuff or arterial catheter) and cerebral perfusion. Assessment of electrical capture can be made by monitoring the electrocardiogram on the filtered monitor of the pacing unit (Fig. 15–18). Mechanical capture is assessed by palpating the pulse as in transvenous pacing. Owing to muscular contractions triggered by the pacer, carotid pulses may be difficult to assess; palpating the femoral pulse may be easier. In addition, bedside ultrasound may prove useful in determining ventricular capture.117,118 Ideally, pacing should be continued at an output level just above the threshold of initial electrical capture so as to minimize discomfort. One study in 16 normal male volunteers who were paced without sedation noted cardiac capture at a mean current of 54 mA (range, 42–60 mA).119 Most subjects could tolerate pacing at their capture threshold; only 1 subject required discontinuation of pacing at 60 mA because of intolerable pain. Heller and coworkers120 compared subjective pain perception and capture thresholds in 10 volunteers paced with five different transcutaneous pacers. Capture rates (40%–80%), thresholds (66.5–104 mA), and subjective discomfort varied from pacemaker to pacemaker.

C

Figure 15–16  Correct placement of transcutaneous pacemaker electrodes. A and B, Anterior-posterior positions. C, Anterior-lateral positions (see text).

● 

Emergency cardiac pacing

Overdrive Pacing Overdrive pacing98–102 of ventricular tachycardia or paroxysmal supraventricular tachycardia is performed in patients who are stable enough to tolerate the brief delay associated with the necessary preparation for this technique. Little data exist on the efficacy or use of this procedure in the ED. The patient is sedated as explained earlier, pacing and monitoring electrode pads are placed in the standard position as detailed earlier, and brief trains (6–10 beats) of asynchronous pacing are initiated. The pacer rate must be set approximately 20 to 60 pulses/min greater than the dysrhythmia rate.123 Generally, an impulse rate of 200 pulses/min is used for ventricular tachycardias (rate generally 150–180 beats/min), and a rate of 240 to 280 pulses/min is used for paroxysmal supraventricular tachycardias (rate commonly 200–250 beats/min). Because rhythm acceleration is possible during overdrive pacing, it is essential that full resuscitation equipment, including a defibrillator, be available at the bedside.

15 

contraction.111,119,120 Analgesia with incremental doses of an opioid agent (fentanyl seems ideal), sedation with a benzodiazepine compound, or both, will make this discomfort tolerable until transvenous pacing can be instituted.

Failure to capture with TCP may be related to electrode placement or patient size. Patients with barrel-shaped chests and large amounts of intrathoracic air conduct electricity poorly and may prove refractory to capture. In one study, the scarring associated with thoracotomy was found to nearly double the pacing threshold.121 A large pericardial effusion or tamponade also will increase the output required for capture.122 Failure to electrically capture with a transcutaneous device in these settings is an indication to consider immediate transvenous pacer placement. Patients who are conscious or who regain consciousness during TCP will experience discomfort because of muscle

Complications The major potential complication of TCP is failure to recognize the presence of underlying treatable ventricular fibrillation. This complication is primarily due to the size of the pacing artifact in the ECG screen, a technical problem inherent in systems without appropriate dampening circuitry.

Figure 15–17  Rate and current (output) controls on a transcutaneous pacemaker.

285 25 FEB 88

LEAD I

Figure 15–18  Assessing ECG capture with transcutaneous pacing. Note that the monitor has been adapted to accommodate the large pacing artifact so as not to obscure the underlying ventricular activity.

SIZE 1.0

HR=41

SIZE 1.0

HR=43

35 MA

HR=71

60 MA

Bradycardia: no pacing 25 FEB 88

LEAD I

Pacing below threshold (35 mA): no capture 25 FEB 88

LEAD I

SIZE 1.0

Pacing above threshold (60 mA): with capture (pacing-pulse marker

)

CARDIAC PROCEDURES ● 

III 

286

A rare complication of TCP is the induction of ventricular fibrillation. Studies of fibrillation thresholds using large precordial electrodes have shown that the longer impulse durations used in modern devices seem to decrease the chance of inducing ventricular fibrillation with TCP. Nonetheless, asynchronous TCP for tachydysrhythmias has been associated with rhythm acceleration and development of ventricular fibrillation.101 Studies looking at prolonged TCP in humans have not been extensive. Zoll and colleagues87 reported 25 humans paced for up to 108 hours with impulses of 20-msec duration. Pacer-induced dysrhythmias did not occur. Leatham and colleagues88 paced one patient for 68 hours with impulses of 20-msec duration. The patient died 2 days after pacing was discontinued. Pathologic examination revealed no evidence of pacer-induced myocardial damage. Madsen and colleagues124

paced 10 healthy volunteers at threshold for 30 minutes and found no enzyme or echocardiographic abnormalities. TCP appears unlikely to produce cardiac injury with short-term use in the ED. Soft tissue discomfort with the potential for injury may still occur with current transcutaneous pacemakers. Most patients are able to tolerate the discomfort, especially after sedation and analgesia. Nonetheless, prolonged use may still induce local cutaneous injury particularly in pediatric patients owing to the use of smaller electrodes.125,126 Patients who cannot tolerate TCP or who will need long-term pacing are candidates for transvenous pacing. REFERENCES  c a n

be found on

E x p e rt C o n s u lt

Richard J. Harper

Pericardiotomy was first performed under direct vision in 1815. Twenty-five years later, the procedure was performed blindly with a trocar on a patient with pericardial tamponade from malignancy.1 By the end of the 19th century, the trocarand-cannula method of pericardiocentesis was commonly used, and in 1911, the subxiphoid approach was described in writing. Pericardiocentesis, with or without electrocardiographic (ECG) assistance, has a significant morbidity rate, reportedly 15% to 20%.2,3 Such a procedure is, however, often required in the emergency department (ED) in life-threatening situations. The use of ultrasound to diagnose pericardial effusion and guide needle placement has become the standard for elective pericardiocentesis. It has been shown to reduce the incidence of complications (0.5%–3.7%)4–6 and shorten the time to diagnose clinically significant effusions.7–11 Although large pericardial effusions are a rare cause of hemodynamic instability in the ED, a lower threshold for performing bedside ultrasonography may increase the detection of effusions before they become hemodynamically significant.8 Even if tamponade physiology is present, ultrasound (echocardiographic) diagnosis and guidance are essential. The ECG-assisted blind pericardiocentesis technique remains an option only for truly emergent pericardiocentesis when a lengthy delay associated with obtaining and organizing ultrasound or fluoroscopic assistance could result in a poor clinical outcome, rapid deterioration, or death.

PERICARDIOCENTESIS IN PULSELESS ELECTRICAL ACTIVITY Pulseless electrical activity (PEA) is the most common clinical scenario in which truly emergent pericardiocentesis is required. Always consider cardiac tamponade in a patient with PEA. In one series of 20 patients with PEA, 3 had tamponade requiring emergent pericardiocentesis, and another 5 had some degree of pericardial effusion.12 Whereas blind pericardiocentesis is acceptable in PEA, ultrasonography can make the diagnosis at the bedside and increase safety by guiding the procedure.

CAUSES OF PERICARDIAL EFFUSION AND TAMPONADE The medical literature categorizes a pericardial fluid collection as either acute hemopericardium, largely secondary to trauma, or pericardial effusion from other causes. This categorization is based on the fact that these two clinical entities differ in their time course, etiology, and treatment.

The most common cause of a pericardial effusion is cardiac surgery and other cardiac procedures.13 Although the overall incidence of this complication is low,14,15 the frequent performance of these procedures results in a significant number of cases. The major risk factor associated with tamponade after cardiac surgery is anticoagulation. Other causes of acute hemopericardium include coagulopathies, cardiovascular catastrophes, and acute injury resulting from either blunt or penetrating trauma. All of these latter causes result in rapid accumulation of whole blood in the pericardial sac. The blood accumulates too fast for the relatively inelastic pericardial sac to stretch and accommodate the fluid. This results in cardiac tamponade from the collection of a small volume of fluid within an essentially normal pericardial size.

Pericardiocentesis

Pericardiocentesis

Acute Hemopericardium

● 

16

16 

C H A P T E R

Penetrating Trauma Traumatic tamponade caused by penetrating trauma may be the result of an obvious external injury such as a knife or gunshot wound, or it may be insidious, as seen with iatrogenic cardiac perforation during a cardiac or vascular procedure. In external penetrating trauma, pericardial tamponade is most commonly the result of a stab wound.16 Approximately 80% to 90% of stab wounds to the heart demonstrate tamponade,16,17 compared with 20% of gunshot wounds. Stab wounds cause tamponade more often because the pericardial rent is small enough to seal and blood is trapped in the pericardial space.16,18 Larger pericardial wounds from gunshots generally drain into the pleural space and produce a hemothorax.19 Cardiac tamponade is often suspected with anterior chest wounds, but it is imperative to remember that any penetrating wound of the chest, back, or upper abdomen may involve the heart. Iatrogenic causes of cardiac tamponade are relatively uncommon but well-known complications of invasive or diagnostic procedures. Pacemaker insertion (either transthoracic or transvenous) and cardiac catheterization, including valvuloplasty and angioplasty, are two of the main causes for the inadvertent penetration of cardiac chambers or coronary vessels.20–22 Penetration of vascular structures is common during transthoracic pacemaker placement.23 Tamponade is a complication of cardiac surgery, but it is usually anticipated. Mediastinal or pericardial drainage helps to control and prevent it.20,24 Pericardiocentesis itself can cause tamponade by lacerating the myocardium or coronary vessels.25,26 Cardiac tamponade may result from perforation of the right atrium or, less commonly, of the right ventricle or superior vena cava by a central venous pressure (CVP) catheter or subclavian hemodialysis catheter.27 The diagnosis is often delayed and, therefore, often fatal.28 Perforation may occur during placement or, more commonly, 1 to 2 days later when the catheter erodes through tissue, particularly if a catheter made of stiff material is used or if the left internal jugular vein approach is used.29 Tamponade from CVP line placement is seldom seen in the ED but must be considered when a patient with a CVP line suddenly decompensates. Consider tamponade when a patient deteriorates hemodynamically after an invasive diagnostic or therapeutic procedure involving the heart. To prevent this complication, be sure to place CVP catheters in the superior vena cava rather than the right atrium or ventricle.

287

CARDIAC PROCEDURES ● 

III 

Blunt Trauma Blunt trauma may cause hemopericardium, often as the result of major chest injury with associated rib and sternal fractures. Cases have been reported, however, in which tamponade occurred in blunt trauma with no obvious signs of injury to the thorax.30 Such incidents may be more common than are clinically recognized, judging by the reports of constrictive pericarditis and pericardial defects found months to years later in trauma patients who were not originally noted to have effusion. Pericardial effusion due to blunt trauma may also be a late finding, becoming symptomatic 12 to 15 days after trauma.31 Severe deceleration injury may cause tamponade as a result of aortic or caval injury.32 This appears to be an uncommon development, with two case series reporting tamponade in 3.6% (1 of 28 patients) and 2.3% (1 of 43 patients) of victims of aortic injury.33 Theoretically, cardiopulmonary resuscitation (CPR) can cause pericardial effusion secondary to the blunt trauma of chest compressions, broken ribs, or intracardiac injections. Early studies reported pericardial effusion in 1% to 3% of CPR survivors.34 Echocardiographic studies showed small cardiac effusions (but not tamponade) in 12% of survivors, only 4% of whom had received intracardiac injections.35 Thus, although case reports of tamponade exist,36,37 CPR and intracardiac drug injections are unlikely to cause significant effusion, much less tamponade.

in a moderately hypotensive patient.44,45 In many cases of small nonhemorrhagic effusion, tamponade does not occur, and the effusion may resolve with treatment of the underly­­ ing disease or may be managed successfully by elective pericardiocentesis. Many disease processes, ranging from the common to the rare (Table 16–1), can cause pericardial effusion. The cause of nonhemorrhagic tamponade may not be obvious on

TABLE 16–1  Causes of Pericardial Effusion Neoplasm

Pericarditis

Connective tissue disease

Nontraumatic Hemopericardium 288

Nontraumatic but acute hemopericardium caused by a bleeding diathesis, aortic dissection, and ventricular rupture behaves much like traumatic tamponade because of its acute nature. This type of hemopericardium is less obvious in etiology than hemopericardium caused by external trauma. Tamponade from aortic dissection or ventricular rupture is usually fatal. Bleeding diathesis may cause spontaneous bleeding into the pericardial sac. The incidence of spontaneous pericardial tamponade in anticoagulated patients has been reported to range from 2.5% to 11%.20,38 Thrombolytic therapy has also been implicated in tamponade secondary to bleeding diathesis. In one series of 392 patients, all with large anterior myocardial infarctions, only 4 (1%) developed tamponade secondary to hemopericardium without ventricular rupture.39 A dissection of the ascending aorta may extend around the base of the vessel into the pericardial sac, causing rapid, and usually fatal, tamponade. This pathologic abnormality may be due to conditions such as syphilis, Marfan syndrome, or atherosclerosis. Infection may create pseudoaneurysms of the aorta, which can also present as tamponade.40 Ventricular rupture after myocardial infarction is a common source of acute hemopericardium. Although the prognosis is grim, survival is possible with prompt recognition and definitive treatment.41,42

Nonhemorrhagic Effusions Nonhemorrhagic effusions usually accumulate slowly, allowing the pericardium to stretch and accommodate up to 2000 mL of fluid.43 Because the effusion accumulates slowly, often over weeks to months, the circulatory system adapts, thus allowing more time for evaluation and treatment, even

Metabolic disorders

Cardiac disease

Drugs

Trauma

Miscellaneous

Mesothelioma Lung Breast Melanoma Lymphoma Radiation (especially after Hodgkin’s disease) Viral Bacterial Staphylococcus Pneumococcus Haemophilus Fungal Tuberculosis Amebiasis Toxoplasmosis Idiopathic Systemic lupus erythematosus Scleroderma Rheumatoid arthritis Acute rheumatic fever Myxedema Uremia Cholesterol pericarditis Bleeding diatheses Acute myocardial infarction Dissecting aortic aneurysm Congestive heart failure Coronary aneurysm Hydralazine Phenytoin Anticoagulants Procainamide Minoxidil Blunt Major trauma Closed-chest cardiopulmonary resuscitation Penetrating Major penetrating trauma Intracardiac injections Transthoracic and transvenous pacing wires Pericardiocentesis Cardiac catheterization Central venous pressure catheter Serum sickness Chylous effusion Löffler syndrome Reiter syndrome Behçet syndrome Pancreatitis Postpericardiotomy Amyloidosis Ascites

Data from Guberman BA, Fowler NO, Engel PJ, et al: Cardiac tamponade in medical patients. Circulation 64:633, 1981; and Pories WJ, Caudiani VA: Cardiac tamponade. Surg Clin North Am 55:573, 1975.

— 16 7.5 18 9 2.5

32 — 4 — — —

12 5 2.5 1.5 12.5 13.5 — —

2 9 12.5 — — 14 — 7.5

— — —

4 4 11

—

2

*Note: Various complications related to human immunodeficiency virus (HIV) infections are now probably the most common causes of large nonhemorrhagic pericardial effusions. Effusions related to bacterial, viral, and mycobacterial infections and Kaposi sarcoma and lymphoma are common.

examination in the ED, and tamponade is frequently misdiagnosed as congestive heart failure or respiratory disease. Although neoplasm has generally been the most common underlying cause of nonhemorrhagic effusion,38,46 human immunodeficiency virus (HIV) has been implicated as a common etiology of large nonhemorrhagic pericardial effusion and tamponade47,48 (Table 16–2). HIV-related effusions have been ascribed to many opportunistic bacterial and viral infections, with mycobacterial infections being the most common.47–49 Kaposi sarcoma and lymphoma50,51 have caused noninfectious pericardial effusions in HIV patients. Cancer is a prominent cause of nonhemorrhagic effusions; the pericardium is involved in 20% of patients with disseminated tumors52 and 8% of all patients with cancer.53 There is primary pericardial involvement in 69% of acute leukemias, in 64% of malignant melanomas, and in 24% of lymphomas; however, the incidence of actual tamponade in these malignancies is not known. Of metastases to the pericardium, 35% originate in the lung, 35% in the breast, 15% in lymphomas, and less than 3% in each of the other cancers.53 Thus, any patient who is known to have one of these malignancies should be considered at risk for tamponade. Metastasis to the heart is usually a late finding in cancer, and foci located elsewhere are usually evident.54 Classic findings of tamponade, such as pulsus paradoxus, are frequently absent in cancer patients with tamponade, and their symptoms are usually attributed to their malignancy.53 Radiation pericarditis, particularly after treatment for Hodgkin’s disease, is a common cause of effusion.43 Effusion occurs in approximately 5% of those patients who receive 4000 rad to the heart.

Pericardiocentesis

Guberman   et al46 (56 Patients) (%)

● 

Neoplastic disease Pericardial invasion Radiation pericarditis Etiology uncertain Traumatic hemopericardium Hemopericardium, nontraumatic Rheumatic disease Uremia/dialysis Bacterial infection Congestive heart failure Uncertain etiology Idiopathic pericarditis Cardiac infarction Iatrogenic diagnostic procedures Myxedema Aneurysm Anticoagulation and cardiac disease Postpericardiotomy

Krikorian & Hancock38 (120 Patients) (%)

Approximately 15% to 20% of patients on dialysis for renal failure develop pericarditis, and 35% of those with pericarditis develop tamponade.55,56 Up to 7% of patients on chronic dialysis may have effusions, sometimes of 1 L or more.54 Pericardial effusion in renal failure may be managed with dialysis alone in many cases. Thirty percent of myxedema patients have pericardial effusions, but few have tamponade.46 Most of the other etiologies listed in Table 16–1 are isolated case reports, and their exact incidences have not been determined.

16 

TABLE 16–2  Etiology of Pericardial Effusion in Two Studies*

Other Causes of Pericardial Tamponade An interesting but rare cause of cardiac tamponade is pneumopericardium. Pneumopericardium is most commonly seen with pneumothorax and pneumomediastinum as a complication of respiratory therapy in infants, but it may also occur from similar barotrauma in adults.57 Pneumopericardium also occurs spontaneously in asthma,58 after blunt chest injury,59,60 and even after high-speed motorcycle rides.61 Pneumopericardium is usually benign, but tension pneumopericardium has been reported as a cause of life-threatening tamponade after blunt chest trauma.60,62 The appearance of life-threatening pneumopericardium and tamponade has also been described immediately63 and 6 days after penetrating chest trauma.64

PATHOPHYSIOLOGY OF TAMPONADE The pericardium is a tough, leathery sac that normally contains about 25 to 35 mL of serous fluid.65 It is not rapidly elastic, although it does demonstrate stress relaxation within minutes of increased intrapericardial pressure, providing a slight ability to accommodate sudden increases in fluid.66 As fluid accumulates, the first 80 to 120 mL is easily accommodated without significantly affecting pericardial pressure (Fig. 16–1).67 However, if an additional 20 to 40 mL is rapidly accumulated, the intrapericardial pressure almost doubles, thus frequently leading to sudden decompensation. With effusions that develop over weeks to months, the pericardium lengthens circumferentially to a huge size and can accommodate liters of fluid. Pericardial compliance, which helps determine the pressure-volume response curve (Fig. 16–2),65 varies considerably in different individuals and in various disease states. The pressure-volume relationship demonstrates hysteresis; the withdrawal of a quantity of fluid drops the pressure more than the addition of the same amount of fluid raised the pressure. As pericardial fluid accumulates, the increased intrapericardial pressure is transmitted across the myocardial wall and causes compression of the atria, vena cava, and pulmonary veins. This reduces right ventricular filling in diastole, producing decreased stroke volume and cardiac output.68 Pulse pressure narrows as reflex sympathetic stimulation increases. Severe tamponade is produced with intrapericardial pressures of 15 to 20 mm Hg.69 As stroke volume decreases, heart rate increases to maintain cardiac output. Sympathetic discharge causes both arterial and venous vasoconstriction.69,70 Vasoconstriction increases venous pressure, which helps to restore the normal venous-atrial and atrioventricular filling gradients. These compensatory mechanisms are often effective and may permit establishment of a new homeostasis with normal cardiac output.

289

Pericardial sac Rt. atrium BP

18 16

(A) Compensated zone

14 12

150

10

125

8

100

6

75

4

50

2

25

0

0

120

(B) Tamponade zone

Aortic systolic pressure

30

RV systolic pressure mm Hg

Rt. atrial and intrapericardial pressure (mmHg)

CARDIAC PROCEDURES ● 

Time: Hours, days or weeks depending on rapidity of accumulation and multiple patient variables

Mean systematic B.P. (mmHg)

III 

20.9 Kg dog 20

–2 –4 20

60

100

140

180

220

sure re s pres ressu Venou ht atrial p ig e r sur Mean pres tolic s a i RV d

260

mL saline injected into pericardial sac Figure 16–1  Production of cardiac tamponade by injections of saline into the pericardial sac.  Although pericardial space can acutely accommodate 80 to 120 mL of fluid without a significant increase in pericardial pressure, note the steep increases in pressure and the drop in blood pressure at about 200 mL of saline. Once critical volumes are reached, very small increases cause significant hemodynamic compromise.  (From Fowler NO: Physiology of cardiac tamponade and pulsus paradoxus. II: Physiological, circulatory, and pharmacological responses in cardiac tamponade. Mod Concepts Cardiovasc Dis 47:116, 1978. Reproduced by permission of the American Heart Association, Inc.)

290

10

Pericardial pressure (cm H2O)

8 6 4

In Out

2 0 –2 –4 40

80

120

160

200

240

Pericardial volume in mL Figure 16–2  Relationship of intrapericardial pressure to volume of pericardial fluid. Note that pressure drops rapidly when a small amount of fluid is removed.  (From Pories W, Gaudiani V: Cardiac tamponade. Surg Clin North Am 55:573, 1975. Reproduced by permission.)

With chronic effusion and in early tamponade, cardiac contractility is not affected and myocardial perfusion is normal.68,71,72 As pressure continues to increase, coronary perfusion pressure drops; thus, in its later stages, tamponade causes myocardial ischemia. Before hypotension occurs, left ventricular blood flow has already decreased to 37%.73 For comparable degrees of hypotension, experimental animals in

0

Increasing pericardial fluid Venoatrial gradient Atrioventricular gradient

Figure 16–3  Summary of physiologic changes in tamponade. RV, right ventricle.  (From Shoemaker WC, Carey JS, Yao ST, et al: Hemodynamic monitoring for physiological evaluation, diagnosis, and therapy of acute hemopericardial tamponade from penetrating wounds. J Trauma 13:36, 1973; and Spodick D: Acute cardiac tamponade: Pathologic physiology, diagnosis, and management. Prog Cardiovasc Dis 10:65, 1967. Reproduced by permission.)

hemorrhagic shock have five times greater coronary blood flow than animals in cardiac tamponade.73 Severe experimental tamponade is followed by large increases in creatine kinase MB and microscopic evidence of cardiac injury resulting from ischemia.74 As intrapericardial pressure continues to rise, the heart’s compensatory mechanisms fail. Myocardial ischemia and perhaps lactic acidosis from poor tissue perfusion may be the triggering events that disrupt the delicate equilibrium.75 Atrial pressure rises rapidly (Fig. 16–3). The atria and pulmonary circulation, being at much lower pressure than the systemic arterial pressure, are more vulnerable to the rising intrapericardial pressure. A “pressure plateau” occurs in which right atrial pressure, right ventricular diastolic pressure, pulmonary artery diastolic pressure, and pulmonary capillary wedge pressure are virtually identical. This equalization of pressures leads to the echocardiographic hallmark of tamponade: right ventricular collapse. At this point, hypotension is severe, bradycardia is common, and PEA may occur. Unless intrapericardial pressure is immediately decreased, pulmonary blood flow ceases and cardiac arrest follows.75 Total blood volume affects cardiac compensation, and it is possible to encounter a “low-pressure” cardiac tamponade.76 The hypovolemic patient with tamponade has a decreased venous pressure, which not only decreases cardiac output but also may obscure the diagnosis because distended neck veins or an elevated CVP is not present. In a patient with a chronic pericardial effusion, the onset of hypovolemia can lower filling pressure enough to precipitate tamponade, and

The classic physical findings of tamponade were first characterized by Beck in 1935.82 He described two triads, one for acute and one for chronic compression. The chronic compression triad consists of increased CVP, ascites, and a small, quiet heart. The acute compression triad consists of increased CVP, decreased arterial pressure, and muffled heart sounds. Unfortunately, although almost 90% of patients have one or more signs,16 only about one third demonstrate the complete triad.75,83 The simultaneous occurrence of all three physical signs is a very late manifestation of tamponade and is usually seen most consistently shortly before cardiac arrest. Careful hemodynamic monitoring reveals earlier changes that indicate the progression of tamponade (Table 16–3).84 In grade I tamponade, cardiac output and arterial pressure are normal, but CVP and heart rate are increased. In grade II tamponade, blood pressure is normal or slightly decreased, and CVP and heart rate are increased. In grade III tamponade, the classic findings of Beck’s acute triad occur. Although this sequence represents the natural history of acute tamponade, the time course varies. Some patients are stable at a given stage for hours; others proceed to cardiac arrest within minutes.75,84 Unfortunately, not all patients with early tamponade respond with a predictable pattern of change in vital signs. Brown and coworkers85 found that 6 of 18 patients with tamponade, defined through right heart catheterization, responded to tamponade with elevated systolic blood pressure. After pericardiocentesis, these patients had a marked reduction in systolic blood pressure accompanied by increased cardiac output. All of these patients had previously been hypertensive.

Pulsus Paradoxus (see also Chapter 1, Vital Signs Measurement) Pulsus paradoxus is defined as an exaggeration of the normal inspiratory fall in blood pressure70,83 (Fig. 16–4A). A paradoxical pulse (pressure) is one of the classic physical signs of tamponade, but it is not pathognomonic. It is also seen in pulmonary emphysema, asthma, labored respirations, obesity, cardiac failure, constrictive pericarditis, pulmonary embolism, and cardiogenic shock.16,44,75 Although occasionally useful and

TABLE 16–3  Shoemaker System of Grading Cardiac Tamponade Grade

Pericardial Volume (mL)

Cardiac Index

Stroke Index

I

200

↓↓

↓↓

Mean Arterial Pressure

Central Venous Pressure

Heart Rate

Beck’s Triad

Normal

↑

↑

Normal or ↓

↑ (≥12 cm H2O)

↑

↓↓

↑↑ (≤30–40 cm H2O)

↓

Venous distention hypotension, muffled heart sounds usually not present May or may not be present Usually present

From Shoemaker WC, Carey SJ, Yao ST, et al: Hemodynamic monitoring for physiologic evaluation, diagnosis, and therapy of acute hemopericardial tamponade from penetrating wounds. J Trauma 13:36, 1973.

Pericardiocentesis

Pericardial effusion is rarely diagnosed based on physical findings. In contrast, pericardial tamponade can be diagnosed based on clinical criteria, but specific clinical signs are often absent. Particularly in the setting of acute hemorrhagic tamponade, the time from the first signs of tamponade to full arrest may be brief.79 Classic clinical findings have been described for tamponade. However, these findings are often obvious only when the patient is unstable owing to tamponade. Ideally, tamponade is diagnosed early, when the patient suffers no more than dyspnea, weakness, or perhaps right heart failure. It is common to attribute respiratory symptoms (e.g., dyspnea on exertion) to a more common condition such as heart failure or pulmonary pathology and to overlook pericardial effusion until the classic late signs (e.g., hypotension) appear.80 Acute pericardial tamponade may be fatal before it is clinically recognized, and deterioration can be exceedingly rapid. Often, therapeutic pathways must be undertaken without enough data to definitively ensure the specific diagnosis. The condition may closely resemble tension pneumothorax, acute hemothorax, hypovolemia, pulmonary edema, aortic dissection, or pulmonary embolism. Severe right ventricular contusion can mimic the findings of tamponade.81 The patient is often agitated or panic-stricken, confused, uncooperative, restless, cyanotic, diaphoretic, and acutely short of breath. In

Physical Signs

● 

DIAGNOSIS OF CARDIAC TAMPONADE Patient Profile and Symptoms

the late stages, the patient is moribund. Hypotension in the presence of severe cyanosis and distended neck veins is a helpful but late finding.

16 

conversely, providing additional volume may temporarily offset increased pericardial pressure. Ventilation and blood CO2 levels have significant effects on cardiac tamponade. This is of particular significance because trauma patients with tamponade may also have respiratory impairment. Pericardial pressure decreases 3 to 6 mm Hg with a hypocarbia of 24 torr and increases 2 to 4 mm Hg when the Pco2 reaches 57 torr.77 This degree of hypercarbia-induced pericardial pressure rise can decrease cardiac output by 25%. Similarly, fluctuations in intrapleural pressure induced by intermittent positive-pressure ventilation are transmitted to the pericardial space and can reduce cardiac output another 25%.78 The clinical implications of these findings are that patients suspected of having tamponade should normally be allowed to breathe spontaneously under careful monitoring and should not be ventilated with positive pressure unless it is absolutely necessary, because their hemodynamic status may deteriorate precipitously.

291

Pressure mm Hg

15 5

PC

PV

“EFG”

0

LA Intrapericardial pressure

–5

LV

Intrathoracic pressure

–10 Normal

III 

● 

CARDIAC PROCEDURES

Pulmonay wedge pressure

[Insp]

Pulmonay wedge pressure

[Insp]

PC PV

Pressure mm Hg

15

LA

“EFG”

10

Intrapericardial pressure

5

LV

0 Intrathoracic pressure

–5 –10

A

Tamponade

200 45°

180 120

Inspiration

80 40

292

B

Expiration

0

Figure 16–4  A, top, The normal situation in which changes in intrathoracic pressure are transmitted to both the pericardial sac and the pulmonary veins. The effective filling gradient (EFG) changes only slightly during respiration. Bottom, Cardiac tamponade in which changes in intrathoracic pressure are transmitted to the pulmonary veins but not to the pericardial sac. The EFG falls during inspiration. Insp, inspiration; LA, left atrium; LV, left ventricle; PC, pulmonary capillaries; PV, pulmonary veins. B, Normally, systolic blood pressure drops slightly during inspiration. To measure pulsus paradoxus, the patient breathes normally while lying at a 45° angle. The blood pressure cuff is inflated well above systolic pressure and slowly deflated. When the pulse is first heard only during expiration, this is the upper value. The cuff is deflated until the pulse is heard during both inspiration and expiration, and this is the lower value. The difference in the two values is the amount of pulsus paradoxus. A difference of more than 12 mm Hg is abnormal.  (A, From Bunnell IL, Holand JF, Griffith GT, Greene DG: Am J Med 25:640, 1960.)

academically attractive, measuring the paradoxical pulse is difficult and time-consuming, and any frightened, hypotensive patient with labored breathing can demonstrate this finding (see Fig. 16–4B). Eschewing this test in the patient in entremis is quite acceptable. If the difference between inspiratory and expiratory systolic blood pressures is greater than 12 mm Hg, the paradoxical pulse is abnormally high.86 Most patients with proven tamponade will demonstrate a difference of 20 to 30 mm Hg or more during the respiratory cycle.16,44,75 This may not be true of patients with very narrow pulse pressures (typical of grade III tamponade); they will have a “deceptively small” paradoxical pulse of 5 to 15 mm Hg. The decreased pulsus paradoxus with hypotension occurs because the paradoxical pulse is a function of actual pulse pressure, and the inspiratory

systolic pressure may be below the level at which diastolic sounds disappear.70 For this reason, the ratio of the paradoxical pulse to the pulse pressure is a more reliable measure. A paradoxical pulse greater than 50% of the pulse pressure is abnormal.70 Pulsus paradoxus may also be suspected from the use of pulse oximetry.87,88 If the highest value of the upper plethysmographic peak of the pulse oximetry waveform is decreased during inspiration (ratio ≥ 1.5), pulsus paradoxus may be suspected. Pulsus paradoxus in tamponade has been correlated with the degree of impairment of cardiac output. In atraumatic patients, a 15% pulsus paradoxus in the face of relative hypotension was found in 97% of patients with moderate or severe tamponade and only 6% of patients with absent or mild tamponade.86 A similar study of right ventricular diastolic collapse by echocardiography found that an abnormal pulsus paradoxus had a sensitivity of 79%, a specificity of 40%, a positive predictive value of 81%, and a negative predictive value of 40%.89 The absence of a paradoxical pulse does not rule out tamponade. Although the mean paradoxical pulse was 49 mm Hg in one series of nonhemorrhagic tamponade,46 23% of the patients had a paradoxical pulse of less than 20 mm Hg, and 1 patient had no measurable paradoxical pulse. An abnormal pulsus paradoxus has been reported to be absent in tamponade when there is an atrial septal defect, aortic insufficiency, localized collections of pericardial blood, or extreme tamponade with hypotension.76 It may also be absent when left ventricular diastolic pressure is intrinsically elevated because of poor left ventricular compliance. In traumatic tamponade, pulsus paradoxus is unreliable.76,90–92 In one study of 197 cases of tamponade caused by trauma, only 8.6% of the diagnoses were made by finding an abnormal pulsus paradoxus.93 Although the absence of pulsus paradoxus rules against severe tamponade, it does not completely rule it out. Whether time is taken to determine pulsus paradoxus depends on the patient’s status. If the patient is moribund or rapidly deteriorating, taking time to check this parameter is obviously poor clinical judgment.

Venous Distention Venous distention, reflecting increased CVP, is also a late sign in cardiac tamponade (see Fig. 16–5C). Neck vein distention may be masked by vasoconstriction as a result of vasopressors (e.g., dopamine), intrinsic sympathetic discharge, or hypovolemia.44,75,84,91 Neck vein distention may be obvious clinically, but the measured CVP is more reliable than the presence of venous distention. The CVP reading should take into account positive-pressure ventilation and the effects of a Valsalva maneuver. Most patients with significant tamponade will have a CVP of 12 to 14 cm H2O or greater.91 Hypovolemia changes the intrapericardial pressure-volume curve in tamponade and will lower the CVP reading at any given stage in the tamponade process. Animal studies have documented that right atrial pressure can be normal in tamponade when hypovolemia is present. One case of low-pressure cardiac tamponade was reported in a patient with no jugular venous distention, no paradoxical pulse, and a right atrial pressure of 8 mm Hg.76 Thus, although the initial CVP reading is useful and diagnostic if grossly elevated (e.g., 20–30 cm H2O),91,94 a series of CVP readings looking for an upward trend is the most sensitive diagnostic

16  ● 

Pericardiocentesis

PE

RV

RA C

A

B

C

Figure 16–5  A, Chest radiograph shows an enlarged, globular cardiac silhouette (“water-bottle heart”) in a patient with tamponade due to a malignant effusion. The chest x-ray has minimal value in diagnosing tamponade but is usually abnormal when significant chronic effusions are present. B, Ultrasound: Apical view of a large pericardial effusion in early ventricular diastole; marked right atrial collapse is seen. C, collapsed segment of the right atrial wall; PE, pericardial effusion; RA, right atrium; RV, right ventricle. C, Markedly distended neck veins may be seen with cardiac tamponade, but they are not universal, especially in hypovolemic trauma patients.

tool.91 A rising CVP, especially when there is persistent hypotension, is extremely suggestive of tamponade in the trauma patient. In the rare case of the hypovolemic patient in whom tamponade is suspected but who demonstrates a low CVP, a fluid challenge will help clarify the situation and will also improve cardiac output, at least temporarily.76

Ancillary Testing Use routine chest radiographs and electrocardiograms to increase the level of suspicion for pericardial effusion and tamponade, but make the diagnosis by noninvasive means with computed tomography (CT) or, preferably, cardiac ultrasound (Fig. 16–5A and B). If available, use bedside ultrasound because it is the fastest and most reliable for the emergency clinician to demonstrate a significant pericardial effusion, although it may not be diagnostic of tamponade.

Chest Radiographs Chest radiographs are not useful in the diagnosis of acute traumatic tamponade because the cardiac size and shape do not change acutely. However, the radiographs may reveal

other important findings such as hemothorax, bullet location, or even pneumopericardium. In the patient without trauma and with chronic effusion, a chest film often reveals an enlarged, saclike “water-bottle” cardiac shadow (see Fig. 16–5A). Unfortunately, it is difficult to differentiate pericardial from myocardial enlargement, and radiographs cannot be used to distinguish between simple pericardial effusion and tamponade.

Electrocardiograms Electrocardiograms may suggest, but should not be used to diagnose, pericardial effusion or cardiac tamponade (Fig. 16–6). Most classic ECG changes, such as PR segment depression, low-voltage QRS complexes, and electrical alternans, have acceptable specificity but poor sensitivity for pericardial effusion or tamponade.44,95,96 Low voltage is defined as QRS amplitude of 5 mV or less in all limb leads (or a sum of the limb lead QRS amplitude ≤ 30 mV), and PR depression is defined as depression of 1 mV or greater in at least 1 lead other than aVR. In a study correlating the electrocardiogram with echocardiographic evaluation, ECG signs had an overall sensitivity of only 1% to 17% and a specificity of 89% to

293

I

aVR

V1

V4

II

aVL

V2

V5

III

aVF

V3

V6

III 

● 

CARDIAC PROCEDURES

ELECTRICAL ALTERNANS IN PERICARDIAL TAMPONADE

II

Figure 16–6  Electrical alternans may develop in patients with pericardial effusion and cardiac tamponade. Notice the beat-to-beat alternation in the P-QRS-T axis; this is caused by the periodic swinging motion of the heart in a large pericardial effusion. Relatively low QRS voltage and sinus tachycardia are also present. Overall, the electrocardiogram (ECG) has a low sensitivity for pericardial effusion or tamponade. Note that electrical alternans may be more evident in the V leads.

294

100% for pericardial effusion.44 Others have demonstrated significantly higher sensitivity (i.e., in the range of 32%–68%) for voltage criteria.97 PR segment depression is the most common ECG finding in pericardial tamponade, and low voltage is most commonly associated with a moderate to large effusion. It is important to note that none of the ECG findings differentiate tamponade from effusion. Electrical alternans is caused by pendulum motion of the heart within the pericardial sac.98 Alternans of the QRS complex has been seen in about 22% of medical tamponade cases80 but in only 5% of cancer patients with tamponade.53 Electrical alternans of both the P-wave and the QRS complex (total electrical alternans) is a rare finding, but when seen, is thought to be pathognomonic of tamponade (see Fig. 16–6).44,99 As with electrical alternans, low voltage may be a finding associated with tamponade but not simple effusion.100

Echocardiography Echocardiography is the best available tool for diagnosing pericardial effusion and has the further advantage of being noninvasive101 (see Fig. 16–5B). Echocardiography is very sensitive in the diagnosis of pericardial effusion and tamponade.102,103 Patients presenting with acute and subacute cardiac tamponade usually do not display the classic triad of hypotension, neck vein distention, and muffled heart tones. Waiting for their appearance to order a confirmatory test will delay diagnosis and jeopardize the safety of the patient. Image the heart whenever a clinically significant pericardial effusion is suspected. Scenarios warranting consideration of a pericardial effusion include a patient with hypotension of unclear etiology, particularly those patients with known malignancies, recent myocardial infarction, and end-stage renal disease, as well as victims of trauma, both blunt and penetrating.8,9,11,104,105

Use early demonstration of a large pericardial effusion to guide further work-up, support early consultation from either cardiology or cardiothoracic surgery, and facilitate therapeutic drainage for those patients who remain hypotensive in spite of fluid resuscitation. For patients in extremis, perform pericardiocentesis immediately. Use the clinician most experienced in both sonography and aspiration of the pericardial sac to perform the procedure. For patients with large effusions who are relatively stable, management options are greater and may include a pericardial window. Consultation with either cardiology or cardiothoracic surgery is advised prior to performing aspiration on stable patients; but this is always a clinical judgment issue best made by the clinician at the bedside. The disadvantages of echocardiography are that it requires ultrasound equipment and is dependent on a skilled operator who is specifically trained in echocardiography. Even when immediately available, echocardiography may take at least 5 minutes, which may be too much time for a patient who is deteriorating rapidly. If the patient is not in full arrest and ultrasound equipment is available, ultrasonography should always be used to diagnose effusion and tamponade and to guide the procedure. Pericardial fluid is relatively easy to demonstrate with bedside ultrasonography, but because many ill patients will demonstrate some pericardial fluid, bedside ultrasonography may not differentiate incidental fluid from tamponade. Ultrasonography may also be misleading in showing loculated effusion with tamponade associated with the early period after cardiac surgery. In one series, transthoracic echocardiography failed to visualize 60% of effusions.106 These effusions were visualized through the use of transesophageal echocardiography. It is possible to mistake an epicardial fat pad for a pericardial effusion. This may be avoided by careful attention to detail. First, the epicardial fat pad is an anterior structure. Clot in the anterior pericardial space suggests that

INDICATIONS FOR PERICARDIOCENTESIS There are two indications for pericardiocentesis: (1) to diagnose the cause or presence of a pericardial effusion and (2) to relieve tamponade. The former is an elective procedure and ideally should be accomplished under ultrasound guidance. The latter may be semielective and performed with ultra-

The use of pericardiocentesis for diagnosis of the etiology of nonhemorrhagic effusions is widespread, although opinions of its utility vary.43,109,110 Neoplastic cells, blood, bacteria, viruses, and chyle can be sought. Measurement of pericardial fluid pH can be helpful, because inflammatory fluid is significantly more acidotic than noninflammatory fluid.111 When a specific etiology is suspected, additional diagnostic testing may be useful (e.g., adenosine deaminase in tuberculosis and carcinoembryonic antigen in suspected malignancy).112 The diagnostic accuracy of pericardiocentesis varies greatly from series to series, depending on the vigor with which a definitive etiology was sought and the prevalence of certain etiologies in the patient population under consideration. In one large series, fluid was obtained in 90% of the aspirations, but a specific etiologic diagnosis was obtained in only 24% of the fluid specimens.38 Certain diagnoses are unlikely to be made from pericardial fluid. Pericardial fluid has been shown to give false-negative cytologic results in certain cases of lymphoma and mesothelioma.38 In HIV patients, effusions caused by Kaposi sarcoma and cytomegalovirus have been diagnosed by pericardial biopsy after fluid studies were nondiagnostic.113,114 An alternative diagnostic tool is subxiphoid pericardiotomy. This technique, performed in the operating suite, obtains both fluid and a pericardial biopsy specimen. It is more likely to provide a definitive diagnosis and has been performed safely without general anesthesia.115,116 In a prospective series of 57 patients, 36% obtained a definitive diagnosis; 40%, a probable diagnosis; 16%, a possible diagnosis; and 7% remained undiagnosed with subxiphoid pericardiotomy.117 Although it is uncertain whether this technique is safer than ultrasound-guided pericardiocentesis, published reports show a low rate of complications in experienced hands.117 Regardless of technique, the need to sample small effusions or obtain pericardial tissue has been questioned. A prospective series found a diagnostic rate of 6% with pericardial fluid and 5% with pericardial tissue when a small persistent effusion was sampled for the specific purpose of diagnosis.110 In contrast, when patients from the same population had therapeutic intervention for tamponade, the yields from fluid and tissue were 54% and 22%, respectively.110 The use of pericardiocentesis as a diagnostic tool in traumatic tamponade is limited. When used diagnostically to determine the presence of pericardial bleeding in trauma, the procedure has a false-negative rate of between 20% and 40%.91,118–120 The reason for the high false-negative rate (defined as no blood aspirated) is well demonstrated by typical stab wounds of the heart.17,121 Ninety-six percent of the patients had blood in the pericardium, but it was clotted in 41% of the patients and partially clotted in another 24%. In only 19% was the blood completely fluid and thus capable of giving a true-positive result on pericardiocentesis.

Therapeutic Pericardiocentesis Tamponade of Uncertain Etiology The primary reason for performing pericardiocentesis in the ED is as part of the treatment for cardiac arrest or in periar-

Pericardiocentesis

At some institutions, CT is much more readily available than echocardiography. However, it requires that the patient be transported to the site of the CT equipment and patient stability must be considered. If clinically indicated, CT is effective in defining the presence and extent of pericardial effusion in the stable patient.107 In certain circumstances, CT can provide a more definitive diagnosis than echocardiography. In one series, eight equivocal echocardiograms were evaluated by CT.108 Two patients thought to have pericardial effusion by ultrasonography were found by CT to have pleural effusions. Another patient with pericardial effusion by ultrasonography was found by CT to have an epicardial lipoma. CT defined three loculated pleural effusions not seen by ultrasonography. A final two patients had hemopericardium visualized by CT but not ultrasonography. In circumstances in which the patient is stable and ultrasonography produces equivocal results or is not available, CT may provide a definitive diagnosis of pericardial effusion.

Diagnostic Pericardiocentesis

● 

CT

sound guidance or emergent and performed blindly or with ECG assistance.

16 

the effusion is circumferential and, therefore, should also be seen in the dependent portion of the pericardial space. This is best demonstrated using the parasternal long-axis view. Alternatively, the probe can be aligned longitudinally so that the inferior vena cava is visualized as it enters the right atrium. The right side of the heart can be seen adjacent to the diaphragm, and blood or fluid within the pericardial sac can be readily identified as long as it is not loculated.7 Second, the inferior vena cava should collapse when the patient sniffs; a collapse of less than 50% indicates increased intrathoracic pressure and possibly tamponade. Third, blood clotting is a dynamic process, with clots continuously forming and being broken down. If blood is present within the pericardial sac, careful examination should reveal fronds of clot waving within an anechoic (black) pericardial space. Finally, and most importantly, an anterior fat pad should not cause collapse of the right ventricular free wall. If, after careful examination, doubt still exists as to the presence of an effusion, hemodynamically stable patients should have a formal echocardiogram or CT performed. Also, remember that fluid within the pericardial space is not always pathologic. A small (5

Dilator Size (Fr) 28 32 36 38 40

From Emslander HC, Bonadio W, Klatzo M: Efficacy of esophageal bougienage by emergency physicians in pediatric coin ingestion. Ann Emerg Med 27:726, 1996.

SPECIAL SITUATIONS Childhood Coin Ingestions Coins are among the most commonly ingested objects in preschool-aged children. In most cases, the ingestion is quickly realized by a caretaker, and in the majority of cases, the coins pass uneventfully.112 Rarely, an esophageal coin can

Esophageal foreign bodies

Equipment

● 

Strict patient selection is paramount for successful and uncomplicated bougienage. The criteria have changed little since initially proposed and define a group in whom a round, smooth object can be forcibly passed into the stomach with little risk.108,111 Although many swallowed objects meet this description, only coins hold clear supportive evidence in the literature. Selection criteria are the following: a single, smooth FB, lodged less than 24 hours, in a patient with no respiratory distress or history of esophageal disease including prior FB or surgery. The procedure is contraindicated in patients who do not satisfy all criteria. It is important to ascertain time period of esophageal impaction to avoid the procedure when there may be underlying esophageal injury. For this reason, some advocate requirements for clearly witnessed ingestions less than 24 hours before presentation.109 Plain radiographs are indicated to verify coin location and the absence of multiple esophageal bodies. Preprocedure esophagograms are not required.

plications, before initiation. Small children may require physical restraint with bed sheets or papoose. All patients require guidance and reassurance throughout the procedure. Topical anesthesia may be achieved with gargled 2% to 4% viscous lidocaine (Xylocaine), atomized 2% lidocaine, or topical benzocaine (20% Hurricane spray, Cetacaine). Sedation is generally not needed. Anxiolysis may be of use in some cases but must be weighed against the potential for aspiration from uncontrolled secretions or induced vomiting. Blind esophageal bougienage is relatively straightforward and resembles placement of an orogastric tube. Patients may be positioned prone or seated upright for the procedure. The distance from nose to midepigastrium approximates the length to reach the stomach and should be noted before passage. Ask the patient to flex the head slightly forward and protrude the tongue. A tongue blade may be used to displace the tongue in uncooperative children. Pass the well-lubricated, appropriately sized bougie posteriorly along the roof of the mouth following the natural curve of the soft palate caudally to the hypopharynx. The patient will momentarily gag as the bougie meets resistance at the level of the cricopharyngeus muscle. Encourage the patient to swallow and gently pass the dilator through the cricopharyngeus muscle. Asking the patient to phonate helps exclude accidental laryngeal intubation; marked hoarseness or inability to phonate indicates airway obstruction and incorrect placement. Once past the cricopharyngeus muscle, extend the head to enable the bougie to pass distally to the stomach with little resistance. Withdraw the bougie after a single pass. Terminate the procedure immediately for pain or resistance to advancement. A postprocedure radiograph of the chest and upper abdomen documents coin location and should be scrutinized for evidence of complications. Routine postprocedure esophagograms are not indicated unless a complication is clinically suspected. Barring a suspicion of complications, release asymptomatic patients to home with appropriate precautions including the need to return for signs of respiratory compromise, chest or abdominal pain, dysphagia, hematemesis, persistent vomiting, or other concerns. Follow-up abdominal radiographs may be performed to document passage of the coin if it is not identified in the feces within 1 week. For adult patients, follow-up is mandated owing to the 65% to 80% chance of underlying esophageal disorders.9

39 

Indications and Contraindications

727

GASTROINTESTINAL PROCEDURES ● 

VII 

be clandestine for many weeks or months, producing a variety of vague respiratory or GI symptoms.113 In addition, many coin ingestions are not witnessed,39 so maintain a high index of suspicion for children presenting with dysphagia, drooling, or crying who may have esophageal FBs, most likely coins. Most coins pass from the esophagus to the stomach with only transient symptoms. The child may be in pain for a few minutes as the coin migrates, but on arrival in the ED, the child is often asymptomatic. Coins that remain in the esophagus are likely to, but do not always, produce continued symptoms (e.g., drooling, pain, dysphagia, refusal to eat or drink). Rarely, esophageal coins can produce airway distress by external compression of the trachea, simulating an asthmatic attack. Coins below the diaphragm are asymptomatic, and the presence of pain or symptoms requires further evaluation. Coins in the trachea produce immediate and obvious respiratory distress. The first clinical decision is whether to obtain a radiograph. Although some authors recommend that asymptomatic children not be radiographed,34 it is important to remember that up to 44% of children with esophageal coins may be asymptomatic. Therefore, it is prudent to perform plain radiographs on all children with a suggestive history.36 In most cases, a single film that includes the pharynx, esophagus, and stomach will suffice to prove or exclude an ingested coin. Another advantage of obtaining radiographs is to rule out multiple FB ingestions, which are not uncommon in chil-

dren.114 Only a single PA chest film is needed to prove the presence of a coin, but a lateral projection is also suggested. If the flat surface of the coin is seen (see Figs. 39–2 to 39–4), this orientation ensures an esophageal position. If the edge of the coin is seen, this orientation suggests that it traversed the vocal cords, but a coin in the airway is not subtle and produces obvious distress. It is advisable to also routinely obtain a lateral radiograph to determine whether multiple coins are stacked on top of each other (Fig. 39–10). Once a coin’s presence has been documented, a decision concerning removal must be made. The approach varies, and there is no agreed upon standard. Overall, about 25% of coins will pass spontaneously, even if the coin is proximal. Observation for 8 to 16 hours is a reasonable approach for asymptomatic children if ingestion has been within 24 hours. Coins in the upper and middle third of the esophagus are less likely to pass spontaneously, and some prefer to remove them as soon as the diagnosis is made.18 Coins in the distal esophagus will pass spontaneously in one third to one half of patients within 24 hours.18,35 The decision on managing these patients depends on various factors: clinician comfort and experience with removal techniques, local protocols and procedures developed by the medical staff of each institution, and comfort level of the caretakers with various therapeutic options. Regardless of the approach, a radiograph should be taken just before surgical removal to ensure that spontaneous passage has not occurred.

728

A

B

Figure 39–10  A, Lateral radiograph of the child shows four stacked coins at the same location. PA view suggested a single coin. Multiple swallowed coins are common in children. It is important to obtain both PA and lateral films to ascertain the exact number and location of swallowed coins. B, A single coin was seen on the PA chest film, but this lateral film suggests three coins. However, they do not seem to be stacked directly on top of each other. This digital radiograph was accidentally exposed three times; actually, only one coin was swallowed and xrayed three times during minimal movement.

Esophageal bougienage

Symptomatic (stridor, drooling, severe pain, coughing, respiratory distress)

Foley catheter removal

Outpatient 12–24 hour observation if middle or lower esophagus, reliable and able to eat and drink

Coin below diaphragm: discharge and follow-up in 5–7 days if not passed, sooner if symptoms develop.

Immediate removal (endoscopy, foley catheter)

Inpatient 12–24 hour observation if upper esophagus

Endoscopic removal with general anesthesia

If successful, discharge with follow-up. If unsuccessful, endoscopic removal. Figure 39–11  Flow diagram outlines an approach to the management of swallowed coins.

One suggested protocol (Fig. 39–11) involves radiologically localizing the coin and, if the child is symptomatic, immediately removing the coin. If the patient is asymptomatic, the coin may be removed immediately or the patient may be observed either as an inpatient or at home. If the child is asymptomatic, one common practice is to allow the child to drink a carbonated beverage and eat a small amount of soft food in the ED, wait about 1 to 2 hours, and perform another radiograph. If sent home with an asymptomatic retained FB, the patient is allowed to eat or drink but should be rechecked in 12 to 24 hours with the knowledge that up to 50% of asymptomatic coin FBs will pass into the stomach spontaneously.18,35,115 The techniques of esophageal bougienage and Foley catheter removal have been described earlier. Both are options for single coins in the esophagus present for less than 24 to 48 hours. Another option for coins at the LES is pharmacologic relaxation of the sphincter to aid passage into the stomach. The most common method to remove esophageal coins in use today is esophagoscopy. About half of ingested coins are in the stomach at the time of first investigation, and such patients can be safely released home to allow for almost certain spontaneous passage with a normal diet. Spontaneous passage of a coin from the stomach to the anus usually requires 5 to 7 days. There is no need for routine cathartic. Parents should be advised to check the stool for the coin and return for repeat radiographs if the

Esophageal foreign bodies

Asymptomatic (or only mild pain/anxiety)

No coin: discharge or consider non radiopaque object

● 

Coin in esophagus

coin is not found in 1 to 2 weeks. Most coins are passed unknowingly by the patient. Any abdominal discomfort or distention warrants reevaluation in the ED. If a follow-up radiograph demonstrates a persistent coin in the intestines for more than 3 to 4 weeks, an obstructive lesion may be present, and further evaluation is warranted. Finally, there are theoretical concerns about U.S. pennies, which contain 97.5% zinc. Theoretically, zinc can lead to mucosal ulceration from the caustic nature of zinc116; however, evidence to date suggests no increased risk from ingested pennies.117

39 

Obtain routine radiograph to include neck, chest, and abdomen

Fish or Chicken Bones in the Throat Patients who complain of a “bone” in their throat usually present to the ED within several hours of the onset of symptoms and usually have tried a home remedy, such as swallowing a piece of bread. These patients are typically able to pinpoint the location of their discomfort and present with an FB sensation, exacerbated with swallowing. Patients who are markedly symptomatic, vomiting, or unable to swallow require definitive therapy. Those with minor complaints may be safely evaluated over a few days, often as outpatients. In cooperative patients, a careful examination of the oropharynx, with either direct or indirect laryngoscopy or both, should be made. If the bone is seen, it should be removed with forceps (Fig. 39–12). If the patient feels pain in the upper throat, special attention is directed to the tonsils because bones often lodge in this area. Strands of saliva may mimic a bone, and small bones may be difficult to see. More commonly, the area of complaint is below the oropharynx. In these patients, indirect laryngoscopy or nasopharyngoscopy should be the first step, once again removing the bone if one is seen. Most patients presenting with an oropharyngeal FB will not have an easily identified or visualized object on examination. These patients present a diagnostic dilemma for several reasons. Only 17% to 25% of patients complaining of an FB sensation after eating chicken or fish have an endoscopically proven bone present, and only 29% to 50% of endoscopically proven bones are seen on plain films.64,118,119 The symptoms in those patients with an FB sensation, but no FB on endoscopy, are believed to be due to esophageal abrasions. For these reasons, a two-tiered, but individualized, approach to managing these patients is proposed.6,9,33,70 The patient receives a physical examination and the bone is removed if seen. Carefully examine the tonsils, posterior pharynx, and base of the tongue, which are common places for bones to lodge. If the bone is removed, and symptoms disappear, no further intervention is required and follow-up is as needed. The removal of a bone usually provides immediate and complete relief of symptoms. Persistent symptoms are cause for further evaluation based on individual circumstances. If no bone is seen on physical examination, the bone may have passed after causing local irritation that persists, or the bone is present and not visualized due to location or consistency. Minor symptoms in the upper throat likely represent persistent local irritation. Minimally symptomatic patients can be discharged and followed up in 24 hours. Those with complaints of an FB below the visulized pharynx, or very bothersome persistent symptoms, should be evaluated with a CT scan of the neck, or possibly the chest, if symptoms are distal. Positive scans are an indication for endoscopic removal of the bone. If the CT scan reveals no FB or postbone com-

729

GASTROINTESTINAL PROCEDURES ● 

VII 

A

B

Figure 39–12  Many fish bones become impaled in the soft tissues of the upper digestive tract. This woman felt a bone catch in her throat while eating fish. As is often the case, she was able to consistently localize the FB to the right submandibular area (A), suggesting that it could be seen with direct visualization. With only a tongue blade, local anesthetic spray, and good lighting, a fish bone was found embedded in the tonsil and was easily removed with forceps (B). Removal provided immediate and total relief, as is usually the case. Strands of saliva can mimic a fish bone, so be careful when probing and grasping.

730

plication, and the patient is stable, she or he is discharged home with follow-up within 24 hours. Patients with oropharyngeal abrasions will usually be asymptomatic at that time. If still symptomatic on follow-up, endoscopy is advocated. A small bone lodged in the esophagus for a few days is annoying and painful, but it is generally not an emergency. However, impacted bones can cause serious sequelae, often weeks later, and continued complaints cannot be ignored. Importantly, a lodged bone will not dissolve and rarely passes spontaneously once lodged in the mucosa. Referral, and possible endoscopy, is necessary if complaints persist for more than 2 to 3 days, even if the examination and CT scan are negative.

Sharp Objects in the Esophagus Sharp objects cause the majority of complications seen in patients with esophageal FBs. These objects include tacks, pins, open paperclips, bobby pins, toothpicks, and razor blades (Fig. 39–13). They will usually not pass spontaneously and should be removed. The only appropriate removal technique is under direct visualization with endoscopy. Attempts at radiographic localization are appropriate for metallic or radiopaque FBs. Most objects in the stomach, even those considered problematic, will transit the remainder of the GI tract if less than 6 cm in length or 2 cm in diameter. If larger than this, consult a gastroenterologist. If radiographs show the FB in the esophagus, endoscopic removal is indicated, and attempts to remove such objects in the ED by other methods are not indicated. Complication rates for endoscopic removal of sharp FBs range from 0% to 3%.8,12,40

Nonradiopaque Objects in the Esophagus Objects such as toothpicks, aluminum tabs from beverage cans, plastic, and food boluses cannot be visualized on plain radiographs and will normally not pass spontaneously. Toothpicks cause a higher percentage of complications than any other type of esophageal FB. Localization of the FB may be accomplished by esophagogram with contrast material, although the yield is low with toothpicks. As with fish bones, toothpicks often lodge in the tonsils or posterior pharynx and can be seen on direct vision.

Impacted Food Bolus A large bolus of food may become impacted in the esophagus, usually at the LES. This occurs most frequently in the elderly, those intoxicated while eating, or those with dentures. Often, there is underlying esophageal pathology, such as a stricture or web, even in the young (see Fig. 39–6). The diagnosis is usually straightforward, and patients may be in significant distress, gagging, and unable to swallow. A barium swallow may be used to confirm the diagnosis, but this is rarely necessary. Proceeding directly to endoscopy appears most reasonable. Food boluses may be amenable to pharmacologic relaxation of the LES, but the definitive intervention is endoscopy to both remove the bolus and to evaluate the esophagus for pathology. The specific approach, however, is varied and not standard. The most logical ED approach is initial aggressive symptmatic relief (judicious narcotics, sedatives, antiemetics), followed by attempts at pharmacologic manipulation of the LES. If the bolus passes, esophageal evaluation can be performed

39  ● 

Esophageal foreign bodies

A

B

Figure 39–13  A, PA radiograph of an open safety pin lodged in the upper esophagus. Sharp foreign bodies in the esophagus are best removed with endoscopic visualization. B, This 10-year-old child came to the emergency department (ED) with severe chest pain. No history of an FB was given. Even when the radiograph demonstrated this metallic object in the esophagus, how it got there remained a mystery. Objects such as this are removed under anesthesia with an endoscope, and no ED intervention, except for pain relief, is indicated.

at follow-up. Papain is contraindicated. An esophagogram can be performed but seems unnecessary if the diagnosis is obvious (it usually is) and endoscopy is available or planned. A barium swallow should not be used to delay definitive treatment. Removal of impacted food is an urgent issue but need not be done immediately upon presentation, or in the middle of the night with inadequate resources. Often, pain relief and a few hours of relaxation will allow the bolus to slowly break. Vomiting occasionally dislodges the impaction.

Button Battery Ingestion Button batteries lodged in the esophagus should be considered an emergency because of the potential for serious morbidity and mortality.22,24,120 These batteries range in size from 7 to 25 mm and are radiopaque (Fig. 39–14). Batteries appear as round densities, similar to an impacted coin, but some demonstrate a “double-contour” configuration. It is important to distinguish between a coin and a button battery, because button batteries require immediate removal. Batteries consist of two metal plates joined by a plastic seal. Internally, they contain an electrolyte solution (usually concentrated sodium or potassium hydroxide) and a heavy metal, such as mercuric oxide, silver oxide, zinc, or lithium. If ingested, these batteries often lodge in the esophagus. The mechanisms of injury include electrolyte leakage, injury from electrical current, heavy metal toxicity, and pressure

necrosis. Of particular concern is the development of a corrosive esophagitis or perforation as a result of caustic injury and prolonged mucosal pressure. Although essentially harmless in the stomach and intestines, batteries lodged in the esophagus should be considered an emergency situation because even new batteries demonstrate corrosion, leakage, and mucosal necrosis within a few hours of contact with the esophagus (see Fig. 39–14).23,24 Esophageal impaction mandates immediate removal. Options include Foley catheter removal, esophageal bougienage, or esophagoscopy. Esophagoscopy allows for direct esophageal evaluation and a more controlled extraction. In addition, the “invasive” nature of batteries may lead to rapid edema, making the catheter technique more difficult. Once in the stomach, button batteries do not require removal. They may be followed radiographically to demonstrate passage, with little risk of GI injury or heavy metal poisoning, even if the battery opens.121–124

The Patient in Distress Pharyngeal or upper esophageal FBs can cause respiratory embarrassment or respiratory arrest, usually in infants and the elderly. The Heimlich maneuver was developed for just such circumstances and can be attempted in the ED when the situation is appropriate. There are no data on the best intervention for the unknown choking patient who arrives in extremis

731

GASTROINTESTINAL PROCEDURES ● 

VII 

C

A

732

B

D

Figure 39–14  A and B, Button batteries have a wide range of sizes and can mimic coins on radiographs. Note that the battery (upper x-ray) has a double-density circular appearance at the border, whereas the coin has a homogeneous density with smooth borders. C and D, Endoscopic mucosal injury and size comparison of battery with coins.  (A and B from Kost KM, Shapior RS: Button battery ingestion. A case report and review of the literature. J Otolaryngol 16:4, 1987.) Figure 39–15  Diagnostic algorithm for the symptomatic assessment of the patient with dysphagia. (Adapted from Saud BM, Szyjkowski RD: A diagnostic approach to dysphagia. Clin Fam Pract 6:525, 2004.)

Dysphagia

Difficulty initiating swallows (includes coughing, choking, and nasal regurgitation)

Food stops or “sticks” after swallowed Esophageal dysphagia

Oropharyngeal dysphagia

Solid or liquid food

Solid food only Mechanical obstruction

Intermittent

Neuromuscular disorder

Progressive

Intermittent

Progressive

Bread/ steak

Chronic heartburn No weight loss

Age >50 Weight loss

Chest pain

Chronic heartburn

Bland regurgitation Weight loss

Lower esophageal ring

Peptic stricture

Carcinoma

Diffuse esophageal spasm

Scleroderma

Achalasia

REFERENCES  c a n

be found on

Esophageal foreign bodies

Foreign body insult to the esophagus are usually straightforward, but patients may present to the ED with a complaint of a lump in the throat or difficulty swallowing, with no apparent reason. One common cause is acute uvulitis (see Chapter 64, Otolaryngologic Procedures). Such complaints require an examination and an investigation based on the clinical encounter and individual circumstances. The complete evaluation of these complaints is beyond the scope of this chapter, but initial modalities available to the clinician to evaluate these complaints are barium swallow, CT scan, and pharyngoscopy/

● 

ED Evaluation of Dysphagia/Lump in the Throat

laryngoscopy. The need for consultation is based on the clinical scenario. Figure 39–15 is a suggested approach to the patient with dysphagia. If no cause is suspected by history or examination, globus pharyngeus may be the cause. This may be associated with anxiety or a panic attack. The sensation of a painless lump in the throat is called globus pharyngeus or globus hystericus. It has many causes other than foreign bodies. Palpate, visualize, or review the anatomical structures in the area: the chin, laryngeal cartilage, cricothyroid cartilage, tracheal rings, sternum and cricopharyngeal muscle. Foreign body sensation may be caused by infection, acid reflux, esophageal spasm, esophageal strictures, pill esophagitis, benign and malignant tumors, hiatal hernia, scleroderma, and many other causes. Globus sensation may also persist after a foreign body has been completely removed due to mucosal injury. Neurologic causes include botulism, myasthenia gravis, cerebral vascular accident, and amyotrophic lateral sclerosis. If the patient is otherwise well appearing and able to drink liquids and keep hydrated, referral to a gastroenterologist as an outpatient is standard.

39 

to the ED, often with no history. The first intervention is to ensure an adequate airway, which can be obvious by the situation or may require laryngoscopy or other means of direct visualization. For the infant who is rushed to the ED in extremis with a known FB, it seems reasonable to hold the infant by the legs, head down, and attempt a Heimlich maneuver and/or sweep the pharynx with a finger to remove an obstruction. However, blind intervention into the pharynx has the potential to change a partial obstruction into a complete obstruction. Forceps may be required to remove obstructions under direct vision. Because FBs may mimic multiple other pathologies, the approach to the acutely choking patient is indeed challenging and every situation individual.

E x p e rt C o n s u lt

733

40

Nasogastric and Feeding Tube Placement Leonard E. Samuels

VII 

● 

GASTROINTESTINAL PROCEDURES

C H A P T E R

Nasogastric (NG) intubation is commonly used to evaluate or treat bowel obstruction, ileus, or gastric hemorrhage; pre- or postoperatively, or to administer food or medication into the gastrointestinal tract. Patients with long-term feeding tube complications and those requiring replacement or other manipulation of tubes frequently present to the emergency department (ED).

PROPERTIES OF NG AND FEEDING TUBES

734

Polypropylene is the most common material used for Levin and Salem sump NG tubes (Fig. 40–1), but it is too rigid for long-term use as a feeding tube. Polypropylene tubes are less likely to kink than others, but are more capable of creating a false passage during placement. Latex (rubber) tubes are moderately firm, require greater lubrication for passage, are relatively thick-walled, and induce a greater foreign body reaction than tubes of other common materials. Latex, especially in latex balloons, deteriorates more rapidly than other materials.1 Foley catheters are primarily latex, although silicone Foley catheters are available for those patients with latex allergies. Silicone tubes are thin-walled, pliable, and nonreactive; however, the walls of silicone tubes are weaker and may rupture if fluid is introduced into a kinked tube.2 Polyurethane tubes are nonreactive and relatively durable. Rigidity varies from manufacturer to manufacturer, depending on tube thickness. A stylet may aid in the passage of polyurethane and silicone tubes, but it increases rigidity and the potential for tissue dissection, especially in tubes that have a small distal end bulb.3 Some feeding tubes have weights, usually made of tungsten, which are nontoxic if released into the gastrointestinal tract.

NG TUBE PLACEMENT Indications and Contraindications The simplest NG tube is the Levin tube, which has a single lumen and multiple distal “eyes.” The advantage of the Levin tube is its relatively large internal diameter (ID) in proportion to its external diameter. The theoretical disadvantage is that a Levin tube should not be left hooked up to suction after the initial contents of the stomach have been drained because the suction will cause the stomach to invaginate into the eyes of the tube, blocking future tube function and potentially causing injury to the stomach lining. Levin tubes are, therefore, rarely used in the ED. The Salem sump tube is preferred over the Levin tube for chronic use as a drainage device because it has a separate (blue-colored) channel that vents the distal main

lumen to the atmosphere (Fig. 40–2). This vent helps to prevent excessive vacuum at the tube tip. Note that both intermittent suction and wall-unit vacuum can exceed the venting capacity of the second lumen, so the vacuum setting should be less than 120 mm Hg.4 The major indication for NG tube placement is to aspirate the stomach contents, particularly to differentiate upper from lower gastrointestinal bleeding. Except when frankly bloody fluid is obtained, the sensitivity and specificity of aspiration to detect upper intestinal bleeding are not good.5–7 Use of Hemoccult or guaiac cards to detect bleeding in gastric aspirates is unreliable because false-positive tests are frequently obtained.5 Although variceal rupture has occurred during insertion of instruments into the esophagus, several studies suggest that NG tube passage is generally safe, even in the presence of esophageal varices.8,9 NG suction is indicated in patients in whom vomiting is likely to be recurrent or dangerous, such as with a paralytic ileus, bowel obstruction, or acute gastric dilation. The trauma patient may need an NG tube as part of the evaluation for gastrointestinal injury or to decompress the stomach before surgery or peritoneal lavage. A radiopaque NG tube may help delineate transdiaphragmatic hernia of the stomach after trauma. A deviated NG tube is a nonspecific sign of traumatic aortic rupture. NG tubes are contraindicated in patients with special predispositions to injury from tube placement. Patients with facial fractures who have a cribriform plate injury may suffer intracranial penetration with a blindly placed nasal tube.10 A severe coagulopathy is a relative contraindication for passage of an NG tube. For patients with a coagulopathy or significant facial or head trauma, an NG tube passed through the mouth may be a better alternative (Fig. 40–3). Patients who have esophageal strictures or a history of alkali ingestion may suffer esophageal perforation. Gagging will decrease venous return and increase cervical and intracranial venous pressure. Comatose patients may vomit during or after NG tube placement. Indwelling NG tubes predispose patients to pulmonary aspiration because of tube-induced hypersalivation, depressed cough reflex, or mechanical or physiologic impairment of the glottis.11 Aspiration is also quite common with nasoenteral feedings in debilitated patients, hence the use of a gastrostomy feeding tube for this condition. An NG tube should be avoided when possible in patients with gastric bypass surgery or lap banding procedures. Despite their traditional use, NG tubes are not routinely required in patients with mild to moderate pancreatitis,11,12 and NG tubes may actually prolong hyperamylasemia and pain. Extended irrigation of the stomach with water in a patient with upper gastrointestinal hemorrhage can lower serum potassium levels,14 and animal studies suggest that cold water lavage can cause, rather than control, bleeding.15,16 No study has shown irrigation to be effective in the control of bleeding,6,17 and vigorous lavage with cold water may lower the body temperature. An NG tube may be used to instill air into the stomach for documentation of a suspected gastric perforation by enhancing visualization of free air under the diaphragm on an upright chest film.

Equipment Passage of standard NG tubes or feeding tubes can be messy and may be accompanied by coughing, retching, sneezing, bleeding, and spilled water or stomach fluid. For this reason, both patient and clinician should be gowned; cleanup may be

40 

3

● 

1

Figure 40–1  Salem sump tube.  This tube contains a second lumen that allows venting during continuous suction. 1, Gastric end with suction eyes. 2, Pigtail extension (blue) of the air vent lumen. 3, Connector for attachment of suction lumen to vacuum line.

Air vent pigtail (may be used as cap for suction lumen when tube is not in use)

CROSS SECTION

Depth marking

A

Vent lumen

B

Vent lumen Suction drainage lumen

Air

5-in-1 adapter

Figure 40–3  A nasogastric (NG) tube may enter the cranium or facial soft tissues in patients with severe head or facial trauma. Those with a coagulopathy may experience significant bleeding from nasal or pharyngeal trauma during passage of an NG tube. In such cases, a standard NG tube inserted through the mouth may be a better alternative.

Nasogastric and feeding tube placement

2

Sentinel eye bisects sentinel line

Drainage eyes

Suction drainage lumen

Gastric contents

Figure 40–2  Diagram of the Salem sump tube.  A, General design. B, Diagram of double-lumen principle for suction.  (A and B, Courtesy of the Argyle Division of Sherwood Medical, St. Louis.)

reduced if the bib area is covered with a towel and a supply of tissues or washcloths is available. For standard NG tube placement, a piston or bulb syringe (with a catheter slip-tip) should be available. NG feeding tubes should have a compatible 50- or 60-mL syringe (some are Luer compatible and others are slip-tip compatible). Tape torn in 4-inch strips or a commercial NG tube holder (e.g., Suction Tube Attachment Device, Hollister, Libertyville, IL) should be handy for securing the tube after placement. Cotton-tipped applicators and tincture of benzoin may be helpful to secure the tube to the nose if the skin is greasy. Make sure the feeding tube is designed for duodenal passage if that is desired—such tubes are usually longer than regular feeding tubes.

735

Figure 40–4  Estimation of tube insertion depth.  Before inserting the NG tube, the clinician should estimate the length of tubing that will be required to ensure intragastric positioning without excess coiling. Holding the tube against the patient’s body, measure the distance from the tip of the xiphoid to the earlobe. Add the distance from the earlobe to the tip of the nose. Then add another 15 cm. Note the total distance using markers on the tube or attach a piece of tape to the tube.  (From Thomsen T, Setnik G [eds]: Procedures Consult—Emergency Medicine Module. Copyright 2008 Elsevier Inc. All rights reserved.)

Procedure Explain the procedure to the patient. Written informed consent is not standard. If the patient is alert, raise the head of the bed so that the patient is upright. Place a towel over the patient’s chest to protect the gown, and place an emesis basin on the patient’s lap.18 Position the tube (typically a 16or 18-Fr sump) so that the insertion distance can be estimated, and mark the distance with tape or by noting the markers printed on the proximal tube. A simple method for measurement is to measure the tube from the xiphoid to the earlobe and then to the tip of the nose. Then add 15 cm (6 inches) to this number (Fig. 40–4).19 It is a common error to fail to estimate the proper length of the tube before passage, resulting in the tip of the tube in the esophagus or the tube coiled excessively in the stomach. Check the nares for obstruction. Assess patency by direct visualization, by gentle digital nasal examination, or by having the patient sniff while first one and then the other

GASTROINTESTINAL PROCEDURES ● 

VII 

nostril is occluded. Pass the tube down the more patent naris.

After topical vasoconstrictor and anesthetic are administered, lubricate the tube with viscous lidocaine or lidocaine jelly.23 Lubrication and anesthesia of the nares can be facilitated by using a syringe (without needle) filled with 5 mL of anesthetic lubricant, such as 2% lidocaine gel (see Fig. 40–5C). Simply putting anesthetic jelly on the tube before insertion will not provide any anesthesia. Topical anesthetics are generally quite safe, but pay attention to the total dose of administered anesthetic to avoid toxicity.24 Note that each milliliter of a 10% lidocaine solution contains 100 mg of lidocaine and can be absorbed systemically. Topical benzocaine has been known to rarely cause methemoglobinemia in the relatively small amounts used for endoscopy.25 Ondansetron (Zofran 4 mg) or metoclopramide (Reglan 10 mg) given intravenously 15 minutes before NG tube passage can reduce nausea and gagging, and secondarily, it improves the pain and procedural prolongation that gagging engenders. Metoclopromide may have additional benefits on the discomfort of NG tube insertion, unrelated to its antinausea effects. Ondansetron is preferred for nausea because metoclopromide can cause agitation or facial and tongue spasm, but these can be rapidly reversed with 25 mg diphenhydramine intravenously. Under direct vision, not blind forcing, insert the tube gently into the naris along the floor of the nose, under the inferior turbinate, and not upward toward the nasal bridge (Fig. 40–6A). If mild resistance is felt in the posterior nasopharynx,

Relief of Discomfort Ameliorate the pain and gagging associated with tube placement by using vasoconstrictors, topical anesthetics, and antiemetics. Because patients rate NG tube placement as very painful, one of the most painful procedures performed in the ED, use these adjuncts whenever the time and clinical situation permit (Fig. 40–5A). Spray topical vasoconstrictors, such as phenylephrine (Neo-Synephrine 0.5%) or oxymetazoline (Afrin 0.05%) into both nares at first in case one side proves to be problematic. The nares, nasopharynx, and oropharynx should all be anesthetized at least 5 minutes before the procedure. Gagging is reduced if the pharynx is anesthetized as well as the nose. Combinations of tetracaine, butyl aminobenzoate, and benzocaine (Cetacaine), nebulized or atomized (spray cans/bottles) lidocaine (4% or 10%), and lidocaine gels (2%) are most commonly used. Lidocaine preparations of 10% are most useful. Lidocaine may be nebulized and delivered by face mask with the equipment used to administer bronchodilators to asthmatics. This method has been found to be superior to lidocaine spray to reduce gagging and vomiting and to increase the chance of successful passage.20,21 Cullen and coworkers22 concluded that nebulized nasal and pharyngeal lidocaine (4 mL of 10%) reduced NG tube passage discomfort better than placebo, without lidocaine toxicity (see Fig. 40–5B).

736

B

A

C

Figure 40–5  A, NG tube insertion has been termed one of the most painful and unpleasant procedures performed in the emergency department (ED), and it should not be used unless specifically indicated. Whenever possible, some form of topical anesthesia to both the nose and the pharynx should be used at least 5 min before passing an NG tube. B, Nebulizing 3–4 mL of 10% lidocaine (note concentration) reduces both nasal and pharyngeal discomfort. Have the patient alternately breathe through the nose and mouth. C, The method is effective primarily for the nasal opening, so encourage the patient to swallow. Fill a syringe barrel with 5 mL of 2% viscous lidocaine. Without using a needle, squirt the solution along the floor of the nose and allow it to drip into the nasopharynx and be swallowed. This method works best with the patient supine.

Confirmation of Tube Placement

After

B

B

C C

A

A = inferior turbinate B = nasal septum C = pathway of the NGT

B Figure 40–6  A, The NG tube is passed under the inferior turbinate, made more patent after vasoconstrictors applied in the nose. The operator should actually look into the nose during this insertion to properly guide the tube, not force it blindly. B, Drinking water by straw during passage seems to help the tube go down. Once the esophagus is entered, rapid advancement is more tolerable and successful than slow placement.  (A and B, From Thomsen T, Setnik G [eds]: Procedures Consult—Emergency Medicine Module. Copyright 2008 Elsevier Inc. All rights reserved.)

apply gentle pressure to overcome this resistance. If significant resistance is encountered, it is better to try the other nostril because bleeding or dissection into retropharyngeal tissue may occur if force is used. Once the tube passes into the oropharynx, pause to help the patient regain composure and enhance the chances for cooperation with the rest of the procedure. If the patient is alert and cooperative, ask him or her to sip water from a straw and swallow while you advance the tube into and down the esophagus (see Fig. 40–6B). This often helps ease passage of the tube. Once the tube is in the nasopharynx, flex the patient’s neck to direct the tube into the esophagus rather than the trachea. Withdraw the tube promptly into the oropharynx if the patient has excessive choking, gagging, coughing, a change in voice, or the appearance of condensation on the inner aspect of the tube. This indicates the possibility of passage of the tube into the trachea. Inspect the tube by way of the mouth to detect coiling or respiratory passage. If the tube is lateral to the midline, this suggests correct position in the esophagus.26 Once the tube is in the esophagus, advance it rapidly to the previously determined depth. Passing the tube slowly prolongs discomfort and may precipitate more gagging.18

Securing the Tube The NG tube is generally secured to the patient with tape attached to both tube and nose (Fig. 40–7A). A butterfly bandage (or tape on each side of the nose) that then coils around the NG tube is a typical approach. The nose and the tube should both be clean and possibly prepared with tincture of benzoin. If a tape should let go or require repositioning, both the tape and the tincture of benzoin must be replaced. It is wise to also secure the tube to the patient’s gown, so that a tug on the tube will encounter this resistance before pulling on the material securing the tube to the patient’s nose. A rubber band tied around the tube with a slipknot (see Fig. 40–7B) and pinned to the gown near the patient’s shoulder is effective. It is critical to ensure that the tube is secured in such a way that it does not press on the medial or lateral nostril. Necrosis or bleeding can result if a tube is not secured correctly. When a Salem sump is used, the blue pigtail must be kept above the level of the fluid in the patient’s stomach or stomach contents may leak back through the vent lumen. If a patient

Nasogastric and feeding tube placement

A

● 

A

Before the NG tube is secured, confirm successful placement by nonradiographic means or by auscultation. Use more than one method when in doubt because all confirmation methods have some possibility of error. Radiographic evaluation is the most definitive way to confirm the position of an NG tube, but it is not standard to routinely obtain x-ray confirmation. A quick and simple method is to insufflate air into the NG tube and auscultate for a rush of air over the stomach. If increased pressure is required to instill the air or if no sounds are heard, the tube may be malpositioned or kinked. Suspect an esophageal location if the patient immediately burps upon insufflation. If the patient is awake, it will be immediately discernible if the tube is in the trachea or lungs. Unfortunately, if the patient is comatose, struggling, or demented, the tube may pass into the lungs unrecognized. Insufflation is often insufficient to detect this type of malpositioning.3 The insufflation test is also unreliable in detecting a tube that has advanced past the stomach and into the small bowel.27 In such patients, radiographic verification may be prudent. Aspiration of stomach contents, especially if pH tested, is more reliable, and can be performed if positioning is in question. If the pH is less than 4, there is an approximately 95% chance that the tube is in the stomach and nonrespiratory placement is almost guaranteed.28 Whereas aspirated fluid can occasionally be obtained from the lung or pleural space, the pH should be 6.0 or higher.28,29 Approximately 2% of patients have an alkaline stomach pH30 with causes including duodenal reflux, antacids, H2 blockers, or recent instillation of formula or medications.30,31 If awake and cooperative, ask the patient to talk. If the patient cannot speak, suspect respiratory placement. Note that with small-bore tubes, patients may still be able to speak despite tracheal placement.31 Once correct tube position is tentatively confirmed, secure the tube. If the patient requires abdominal or chest radiographs for other diagnostic purposes, place the NG tube before obtaining the films. An NG tube deviated to the right may occasionally be seen in patients with traumatic rupture of the aorta, but this is not a reliable indicator.

40 

Before

737

GASTROINTESTINAL PROCEDURES

NG tube

VII 

● 

Slit ET tube

A Safety pin and rubber band

Separate ET tube

Tape

738

Figure 40–7  A, The tape on the bridge of the nose keeps the tube in the middle of the nasal opening, away from the skin, preventing irritation. B, Attach the NG tube to the patient’s gown using a rubber band and a safety pin so that the first tug on the tube pulls the gown and not the tape holding the tube in the patient’s nose.  (A, From Thomsen T, Setnik G [eds]: Procedures Consult—Emergency Medicine Module. Copyright 2008 Elsevier Inc. All rights reserved.)

needs to ambulate with a sump tube in place, the blue pigtail can be fitted into the plastic connector at the end of the suction lumen, creating a closed loop that should not leak.

Placement Issues If the patient is intubated, deflate the balloon of the endotracheal (ET) tube briefly to allow passage of the NG tube. In the unconscious patient, the NG tube is easily misplaced into the pulmonary tree, but misplacement is unlikely if the patient is intubated. This complication may be missed during the procedure because gag and cough reflexes may be suppressed and the patient cannot talk. In addition, the absence of swallowing may prevent successful passage of the tube. Several techniques may be used to successfully pass an NG tube in a difficult unconscious patient. If initial attempts fail, place the NG tube through a naris into the oropharynx. Visualize the tip of the tube with a laryngoscope, grasp it with Magill forceps, and pull it out of the mouth. Select an ET tube with an ID that is slightly larger than the external diameter of the NG tube. Slit it along its lesser curvature from the proximal end to a point 3 cm from its distal end. Pass the slit ET tube (generally 8 mm ID) through the mouth into the esophagus.32 Alternatively, pass a 7-mm ID slit ET tube directly through the nose into the esophagus.33 Passage into the esophagus is facilitated by the stiffness of the larger ET tube and does not require active swallowing. Thread the tip of the NG tube into the ET tube and advance it into the stomach (Fig. 40–8). Remove the slit ET tube from the esophagus. When the distal

Figure 40–8  Diagrammatic representation of the separation of the NG tube from the guiding endotracheal (ET) tube through the slit in the guiding ET tube. The NG tube has first been passed through the nose and is pulled out through the mouth. The tip of the tube is then threaded into the guiding ET tube to ensure passage down the esophagus. The guiding ET tube is removed from the esophagus before being separated from the NG tube. Note the prior placement of another ET tube in the trachea (partially shown) to avert passage of the guiding ET tube into the trachea.  (Modified from Sprague DH, Carter SR: An alternate method for nasogastric tube insertion. Anesthesiology 53:436, 1980.)

part of the ET tube is visible, slit the unslit 3-cm distal part with scissors. Remove the ET tube, and the NG tube will remain in place.34 Advance any slack tubing with forceps or pull it back nasally, depending on the final depth required for the NG tube. The technique can also be performed by passing the slit ET tube nasally, which also saves the trouble of orally advancing or nasally retracting any slack tubing.33,34 In a particularly passive, sedated, unconscious, or toothless patient, guiding the NG tube with the fingers in the pharynx is occasionally successful (Fig. 40–9).35 Displacing the larynx forward by manually gripping and lifting the thyroid cartilage can aid tube insertion,36 as can simple jaw elevation. A soft nasopharyngeal airway, well lubricated, is at times easier to pass nasally than the NG tube, and then the lubricated NG tube can be passed through it. In addition, it affords some protection to the nasal mucosa if multiple attempts to pass the NG tube are necessary, or if it is particularly important to minimize bleeding or trauma. Cooling an NG tube increases its rigidity, and coiling it can increase the tube curvature, both of which may help pass the tube. Ultimately, if all other methods fail, place a flexible fiberoptic bronchoscope or esophagoscope into and through the esophagus under direct visualization.37 Thread a guidewire into the stomach. Place the NG tube over the guidewire into the stomach and then remove the guidewire.38

Complications Complications of standard NG tube placement are similar to problems noted with NG feeding tube placement. The complications related to tube misplacement are discussed in that section. In addition, the clinician placing the NG tube in the patient with neck injuries should be cautious of potentiating

40  ● 

Nasogastric and feeding tube placement

Nasogastric tube Endotracheal tube

Index and third finger depress tongue and guide NG tube

Pull jaw forward

NG tube

A

A

739

B B Figure 40–9  A, The passage of an NG tube through the nose of an intubated patient. An ET tube is in the trachea via the mouth. Place the second and third fingers in the posterior pharynx. Depress the tongue with the fingers. Guide the NG tube down the esophagus by passing it through the second and third fingers that are in the posterior pharynx. Importantly, place the thumb under the jaw and pull the jaw forward. B, In this intubated patient, the Video Laryngoscope (Glidescope, Verathon, Bothell, WA) can be used to manipulate the larynx to allow for visualized NE tube passage.

cervical spine injuries with excessive motion during passage (especially in association with coughing and gagging in the awake patient). Furthermore, passage of an NG tube in the awake patient with a penetrating neck wound may exacerbate hemorrhage should coughing or gagging result. Particularly serious forms of tube misplacement are pulmonary placement (Fig. 40–10) and intracranial placement (Fig. 40–11). A tension gastrothorax can develop in patients with an intrathoracic stomach. The tension gastrothorax can occupy much of the left hemithorax, displacing the heart and lungs and causing a clinical syndrome identical to tension pneumothorax. Whereas successful passage of an NG tube will relieve a tension gastrothorax, the high pressures of a tension gastro-

Figure 40–10  A, Levin tube inadvertently placed in the right main stem bronchus; an alveolar infiltrate consistent with early pneumonia is also shown. B, Proper position of the NG tube is best verified by an x-ray.  (A, From Johnson JC: Letter to the editor: Back to basics for morbidity-free nasogastric intubation. JACEP 8:289, 1979; B, from Thomsen T, Setnik G [eds]: Procedures Consult—Emergency Medicine Module. Copyright 2008 Elsevier Inc. All rights reserved.)

thorax often develop because torsion of the stomach in the chest prevents egress of air; that torsion may prevent ingress of the therapeutic NG tube. The condition is rare enough that further emergent therapy is based on case reports rather than substantial series. Relief of the tension gastrothorax has been accomplished with transthoracic puncture of the stomach with a 16-gauge catheter over needle. The catheter over needle was inserted in the second intercostal space in the midclavicular line, and then the needle was removed. The catheter was left in place attached to intravenous tubing with the distal end under a water seal.39 A single 16-gauge puncture of the stomach is unlikely to leak and cause pleuritis; such punctures have long been used in percutaneous endoscopic gastrostomy (PEG) tube placement. Inserting a chest tube into the stomach is not advisable because gastric fluid may leak into the pleural space. Once the tension on the stomach is relieved, it may be possible to pass the NG tube to prevent reoccurrence of the problem. The stomach, no longer tense

GASTROINTESTINAL PROCEDURES ● 

VII 

Figure 40–11  Anteroposterior and lateral skull radiographs demonstrate intracranial insertion of an NG tube in a patient with multiple skull fractures.  (From Johnson JC: Letter to the editor: Back to basics for morbidity-free nasogastric intubation. JACEP 8:289, 1979.)

740

and wedged in the chest, can twist to allow the tube to pass. Surgical correction of the condition permitting intrathoracic herniation of the stomach is the definitive treatment to prevent recurrence of tension gastrothorax. NG tubes, when in place for prolonged periods, are a common cause of innocuous gastric bleeding and gastric erosions.

REPLACEMENT OF NASOENTERIC FEEDING TUBES Indications and Contraindications The most common indication for feeding tube replacement in the ED is unintentional removal of a preexisting feeding tube. In one prospective study, 38% of tubes were removed unintentionally. Although some of these tubes had fallen out or had been coughed out, more than half were pulled out by the patient.40 Tube rupture, deterioration, or clogging may also necessitate replacement. Management of a clogged or nonirrigating feeding tube is discussed in the section on clogged feeding tubes.

Feeding Tube Site Three major classes of enteral feeding tubes are in common use; classification is according to the site of insertion. Tubes can enter through the nares, a cervical ostomy, or an abdominal ostomy. Enteral tubes are often categorized by the location of the tip of the tube. Tubes may terminate primarily in the stomach, such as a gastrostomy (G) or PEG tube. They may terminate in the small intestine, a jejunosotomy (J) tube, or in both the stomach and the small intestine, a PEG-J tube. To confuse the issue, some tubes enter the stomach and terminate in the stomach (G tube) or in the proximal small bowel (J tube), whereas some tubes enter the gastrointestinal tract directly through the small bowel wall (J tube). Practically speaking, almost all gastric tubes are PEG tubes. They are placed endoscopically with local anesthesia and without a surgical incision. J tubes,

conversely, are placed surgically under general anesthesia, require a surgical incision, and result in a surgical scar at the insertion site. Gastric feeding results in better digestion than intestinal feeding whereas J tubes are less likely to result in reflux and aspiration. Normally about 20% of gastric antral contents pass into the duodenum, with 80% refluxing back into the body of the stomach for further mixing.41 If the feeding tube is placed in the antrum of the stomach or in the small bowel, enteral feeding solution passing into the small bowel may not be tolerated, resulting in diarrhea and paradoxical decreased nutrition.41,42 The most common rationale for small intestinal feeding is to reduce regurgitation and aspiration.43–46

Procedure Nasoenteric feeding tube replacement requires greater time and effort if the patient is uncooperative or has a physically obstructing lesion. Nasoenteric feeding tube migration into the duodenal bulb generally requires patient positioning in the right decubitus position for about an hour after successful intragastric passage.47 The clinician should explain the procedure to the patient before tube passage. It is generally advisable to restrain the hands of demented, impaired, or otherwise uncooperative patients. Prepare the nares before passage of the tube similar to the procedure for primary NG tube placement. If a feeding tube stylet is used, lubricate and insert it into the feeding tube before introducing it into the nares. Tube stylets can be lubricated with water-soluble jelly. If using Dobbhoff, EntriFlex (Biosearch) or another tube with preapplied lubricant, you may need to activate the lubricant with a 5-mL flush of water. Never allow the stylet to protrude beyond the end of the feeding tube because these stiff, small-diameter wires have the capacity to scratch the esophagus and allow for the creation of a false passage. The stylet may lock into position on the tube at the proximal end and should be properly secured.

Feeding tube

Anchor tube

A

B

Pulmonary intubation is an uncommon but well-known and potentially fatal complication of nasal feeding tube insertion (see Fig. 40–10). Coughing and respiratory distress are the most common symptoms of respiratory passage of an NG tube, but there may be relatively few apparent symptoms in a demented or comatose patient.54 Decreased mentation and an absent cough reflex are predisposing factors for unrecognized nasopulmonary intubation with NG tubes.3 A small end bulb (e.g., 2.7 mm diameter) can slip past a tracheal high-volume, low-pressure cuff and pass easily to the lung periphery.3,54,55 A pneumothorax may result when an NG tube dissects into or is withdrawn from the pulmonary parenchyma.56 Bloody aspirate from a tube should heighten awareness of possible tissue damage. A clogged or nonfunctional NG tube may be difficult to remove. Fluoroscopy may allow careful insertion of a guidewire or stylet into an in situ tube to facilitate removal. Fluoroscopy may also identify the mechanical problem interfering with the tube’s removal. Bent-double segments are probably the most common cause. Knots, although uncommon, do occur. Do not use excessive force to remove an NG tube because serious injury to the patient may result. Premature removal of the NG tube is the most frequent complication of feeding tube use. To help prevent removal by an uncooperative patient, secure the NG tube to a loop anchor passed in the same naris. The anchor works by aversive stimulation of the soft palate and nose with distraction of the NG tube, rather than by mechanical stabilization of the tube. Sax and Bower57 recommend a technique for creating a separate NG tube anchor. Cut a soft weighted nasoenteric tube approximately 12 inches from the top. Pass a heavy (2-0) silk suture through the tube to exit the side hole. Insert the guidewire with care, because it must not protrude from the inserted end. Sedate the patient if uncooperative. Insert the tube through the anesthetized naris into the nasopharynx, grasp it with Magill forceps and pull to remove it through the mouth (Fig. 40–12A). Trim excess tubing without cutting the silk suture. Make a closed loop by tying the silk suture in front of the nose. Leave the loop long enough that it does not apply Figure 40–12  Placement of an NG tube anchor to secure a companion NG or feeding tube in an uncooperative patient who repeatedly pulls out the feeding tube. A, Using forceps, grasp the tube in the pharynx and pull it out through the mouth. This will serve as an anchor tube. B, Tie the ends of the short anchor tube together to form a loop, and tie the companion NG or feeding tube to the anchor loop.

Nasogastric and feeding tube placement

Auscultatory confirmation of tube placement can be misleading, so confirm proper placement of the tube with a radiograph before feeding.3 However, radiographic confirmation of tube placement may also be misleading. In viewing the radiograph, it is particularly important to study the area around the carina. An esophageal tube shows at most a mild change in course, whereas a tracheally placed tube usually deviates significantly as it travels into the right or left main stem bronchus. The end of an NG tube may appear to be in the stomach yet actually may be in the left lung behind and below the top of the diaphragm.48 When a stylet has been used for passage, leave it in the feeding tube for the radiograph, because the tube’s course is not always visible without it. The stylets of most tubes are designed to allow insufflation and aspiration while in place. Even when stomach entry is certain, the intestinal location may be misleading on radiograph. A nasoenteric tube may lie completely to the left of midline and yet have its tip in the duodenum, or it may have a position overlying the right abdomen yet not have entered the duodenum. A contrast study is necessary to ascertain duodenal position when pulmonary placement has been ruled out.47,49 Examine the radiograph also for the presence of mediastinal air or a pneumothorax, which may suggest pulmonary or esophageal puncture. An esophageal puncture should be evaluated with endoscopy and may require surgery, depending on the size of the rent. The end bulb of most nasoduodenal tubes will pass into the duodenum after patient positioning in the right decubitus position for an hour. Some researchers recommend pretreatment with metoclopramide to enhance gastric emptying.50–52 One investigator47 found that metoclopramide enhances duodenal passage of nasogastrically placed feeding tubes in diabetic, but not in nondiabetic, patients. Gastric antral motility in diabetics is often impaired; metoclopramide helps restore normal synchronized activity in these patients but has little effect on emptying in subjects who have normal antral function. The usual dose of metoclopramide is 10 mg adminis-

Complications

● 

Placement Confirmation

tered intravenously. Also, 3 mg/kg of erythromycin lactobionate given intravenously over 1 hour works similarly and may be effective even if metoclopramide fails.53 Endoscopy or fluoroscopy may be necessary if positioning and metoclopramide are not successful.

40 

When the patient is uncooperative or cannot drink, introduce 5 to 15 mL of water into the mouth or into the proximal end of the feeding tube with a syringe; this may induce swallowing and facilitate tube passage. Although the patient may not swallow for several minutes, wait for her or him to swallow because this may mean the difference between a coiled or pulmonary tube placement and successful passage.

741

GASTROINTESTINAL PROCEDURES ● 

VII 

continuous pressure to the nose or palate while at rest. Pass the nasal feeding tube through the same nostril and secure it to the loop (see Fig. 40–12B). This anchor is simpler to construct and more comfortable than anchors that pass through the opposite nostril.57 Complications of properly placed nasoenteric tubes include nasopharyngeal erosions, esophageal reflux, tracheoesophageal fistulas, gagging, rupture of esophageal varices, and otitis media.58 One survey of nasogastrically fed patients found that the most distressing features of having an NG tube for feeding were deprivation of tasting, drinking, and chewing of food; soreness of the nose; rhinitis; esophagitis; mouth breathing; and the sight of other patients who were eating.59 Checking feeding tolerance is difficult with small-gauge feeding tubes. Aspiration of tubes to check for residuals is not recommended with tubes of 9 French size or smaller. Aspiration is likely to clog the tubes because they collapse under pressure and because relatively small particles can occlude the tube. For the same reasons, the residual is likely to be inaccurate.43

Parotid gland Stylohyoid m. Hyoid bone Internal laryngeal n. Carotid artery Sternocleidomastoid m.

Digastric m. Hyoid bone Sternohyoid m. Omohyoid m.

A

Patient Instructions

742

To maintain catheter patency, small tubes should be flushed with 20 to 30 mL of tap water at least two to three times daily and after administration of medication.43,49 Water is a more effective irrigant than cranberry juice.60 Medications should be in liquid form or be completely dissolved, or they may clog the tube. Methods of dealing with a clogged tube are discussed subsequently. The tube should be anchored to the nose and face in such a way that it is not in contact with the skin at the nasal opening. This reduces tube discomfort and prevents necrosis of the alae, nares, and distal septum. Patients who exhibit a tendency to pull on their tubes need adequate restraints. Patients receiving tube feedings should have their heads elevated to at least 30° above the horizontal.43

Cut edge of esophagus

PHARYNGOSTOMY AND ESOPHAGOSTOMY FEEDING TUBES Cervical pharyngostomy and cervical esophagostomy have both been developed relatively recently. Cervical esophagostomies are generally performed at the time of cervical or maxillofacial operations. Malignant growths of the proximal esophagus, head, or neck are the primary indications for esophagostomy. Cervical esophagostomies may eventually evolve a permanent sinus, allowing the feeding tube to be removed between meals. Such tubes will unlikely be replaced in the ED, but the concept of these feeding tubes is illustrated in Figure 40–13. Complications of pharyngostomy and esophagostomy include local soft tissue irritation, accidental extubation because of excess length of the external tube, pulmonary aspiration from vomiting, arterial erosion with exsanguination, and esophagitis or stricture of the esophagus from reflux.

G, GASTROENTEROSTOMY, DUODENOSTOMY, AND J TUBES Since the turn of the century, more than 30 different operative techniques have been described for tube gastrostomy.61 The “pull” technique, an endoscopic percutaneous procedure,

B Figure 40–13  A, Pharyngostomy feeding tube. B, Pathway for an esophagostomy or pharyngostomy feeding tube.

performed under conscious sedation, is now the most common method of gastrostomy placement. This has been termed a PEG tube. In the “pull” technique, an endoscope is passed into the patient’s stomach and the contents are aspirated. The procedure is described in Figure 40–14. Feeding tubes may also be placed directly into the jejunum or advanced into the duodenum or jejunum via the stomach. Rarely, an operative procedure is performed to suture a jejunal tube into the lumen of the small bowel (Fig. 40–15). Tube duodenostomies are also created for duodenal decompression after partial gastrectomy with Billroth II anastomoses.62 Permanent jejunostomies are rarely used. Tube jejunostomy is indicated when the proximal bowel has a fistula or is obstructed, when recovery of small bowel motility is anticipated long before recovery of gastric motility, and after a gastrectomy.61,62

A

Nasogastric and feeding tube placement

Light at end of scope

● 

Endoscope

40 

Finger indents wall

Figure 40–14  Placement of a percutaneous endoscopic gastrostomy (PEG) tube.  A, Under conscious sedation, pass a lighted endoscope into the stomach. Indent the skin with a finger to determine the optimal puncture site where the stomach and abdominal wall are closest, with no bowel between. B, Fill a syringe with saline and advance it percutaneously at the selected entry point, and the tip of the needle is seen entering the gastric lumen through the endoscope. If air is aspirated and no needle tip is seen, the needle is in the bowel, not the stomach. C, Push and pull the scope/snare/feeding tube combination into position. D, Pull the head of the feeding tube into contact with the gastric mucosa. E, Use an external bolster or crossbar to keep the tube snug against the skin and gastric wall, but not so tight that it causes ischemia of the intervening tissue.

B Feeding tube

3mm

Head of feeding tube

D

C

Subcutaneous tissue

Anchoring suture Dacron cuff

Jejunum Tacking suture

Figure 40–15  Formation of the Witzel tunnel and final permanent jejunal catheter placement.  (From Wiedeman JE, Smith VC: Use of the Hickman catheter for jejunal feedings in children. Surg Gynecol Obstet 162:69, 1986.)

Contraindications for gastrostomy feeding include severe gastroesophageal reflux, upper gastrointestinal fistulas, repeated aspiration of gastric contents, and intestinal or gastric outlet obstruction.61 Jejunal feeding is contraindicated if the highly osmolar feeding solutions required for jejunal feeding are poorly tolerated and cause copious diarrhea.

E

Indications and Contraindications for Tube Replacement The nursing home patient with a nonfunctioning or displaced feeding tube represents a common ED presentation. The clinician cannot always determine the location of the original feeding tube by simply looking at the patient who arrives in the ED for tube replacement. Nevertheless, the emergency clinician should attempt to ensure that the terminal end of a replaced tube is in the same viscus as the original. External inspection may or may not reveal where a feeding tube should terminate (Fig. 40–16). Contrast studies and fluoroscopy usually provide such information (Fig. 40–17). A de Pezzer (mushroom) or Foley G tube is designed only for intragastric termination. Some tubes have two lumina, one terminating in the stomach for decompression and the other in the small bowel for feeding. These can be confused with tubes that have two entrances to one lumen (one for continuous feeding and the other for medications) and tubes that have a second lumen leading to an inflatable balloon. Foley catheters are not ideal as long-term feeding tubes. They clog easily, and the balloon disintegrates in stomach acid. They may be used temporarily but should be replaced with specialized feeding tubes when feasible. A call from the nursing home indicating that a tube has been pulled out should be answered by the advice that a Foley catheter be immediately used to keep the stoma open. Always inflate the balloon with saline, and use a bolster to prevent tube migra-

743

2

3

4

A

VII 

● 

GASTROINTESTINAL PROCEDURES

1

Safety plug and strap

Feeding port

Access port

Safety Luer plug fitment

C Balloon volume

Balloon inflation valve (Color coded for size) Tube fr. size

Corflo-triple gastrostomy tube

Centimeter markings Movable retention bolster

744

Balloon

9” (23 cm) overall length

Exit ports and radiopaque tip

B

tion (Fig. 40–18). An original PEG tube is pictured in Figure 40–19. This long tube has a mushroom end that is removed by traction. The clinician has a few options when faced with the task of replacing a feeding tube. Unfortunately, old records or nursing home personnel rarely give specific information that is helpful to the emergency clinician. If only a stoma exists, one may request that the nursing home describe or send the prior tube to the ED. If no surgical scar is seen at the stoma site, the tube is almost certainly a G tube or a G tube that terminated in the jejunum. When in doubt, passing a Foley catheter without balloon inflation, taping it to the skin, and referring the patient to a consultant or the original referring clinician is appropriate. Some type of tube must be placed to stent the stoma, otherwise the stoma will quickly close (in a matter of hours), and the patient might require a more complicated procedure to regain access. The only real concern of placing a gastric tube into the jejunum is that the balloon will produce intestinal obstruction if it is fully inflated. If the tube is nonfunctioning yet still in place, the clinician must make a judgment as to the risk versus benefit of removal and replacement versus an attempt at unclogging

Figure 40–16  A, Various types of gastrostomy tubes. 1, Silicone catheter (American Endoscopy [Bard Interventional Products, Billerica, MA]). 2, Polyurethane catheter with collapsible foam flange (to collapse, tube should be cut) (VIASYS MedSystems, Wheeling, IL). 3, Latex catheter with a movable external bolster and an internal mushroom or de Pezzer–type flange on the end (American Endoscopy [Bard Interventional Products, Billerica, MA]). 4, Balloon (Foley) catheter (Wilson-Cook Co., Winston-Salem, NC). B, A user-friendly gastrostomy tube is supplied by VIASYS MedSystems (Wheeling, IL), the CORFLODUAL GT gastrostomy tube, packaged with lubricant, a prefilled syringe for inflating the balloon, and an extension set. The color-coded inflation valve indicates tube size (12–24 Fr). The silicone tube uses a retention balloon and a movable bolster. Note that the retention bolster is designed to prevent inward migration of the tube and not to be an anchoring device sutured to the skin. C, A gastric balloon jejunal feeding tube enters the stomach and delivers feedings into the jejunum.

the tube (see subsequent discussion on unclogging). The major concern is that a new tube may be misplaced (i.e., into the peritoneal cavity). If it appears that a skin incision was used to place the tube, it is unlikely that the patient has an easily removable tube. If the patient has signs of a complication (e.g., infection, ileus, intestinal obstruction), surgical consultation is warranted. Note that a migrated tube, with the balloon or tube obstructing the gastric outlet, is a common cause of gastric distention, persistent vomiting, or signs of intestinal obstruction. This is easily remedied by simply withdrawing the tube and ensuring that a bolster is functioning (Fig. 40–20). Most PEG tubes do not have sutures joining the stomach with the abdominal wall, so there is potential for a replaced tube to end up in the peritoneal cavity. Adhesions, however, usually keep the stomach appropriately positioned, but only after the tract has matured. Nonoperative tube replacement techniques are safe only through an established tract between the skin and the bowel. Catheter replacement should not be attempted in the immediate postoperative period. A simple gastrostomy takes about a week to form a tract.63 A stable tract may take 2 or 3 weeks longer to form if healing is compro-

40  ● 

Nasogastric and feeding tube placement

Figure 40–17  A, Without old records, the exact type and positioning of this nonfunctioning feeding tube are unknown. The operative scar on the abdominal wall suggests an implanted tube, not a simple gastric tube. B, Injection of contrast before tube removal demonstrates the tube tip ending in the small bowel, not the stomach. C, If the tube has been removed, and questions remain about type of tube and circumstances of placement, a new tube is best placed under fluoroscopy with guidewire assistance.

A

745

B

C

mised. Poor nutrition is the most common element compromising wound healing in patients requiring a feeding tube. A Witzel tunnel may take up to 3 weeks after the operation to mature sufficiently for safe nonoperative tube replacement.

Equipment for Replacing a Dislodged Tube Equipment for feeding tube replacement into a matured site includes gloves, stethoscope, feeding tube, external bolster, lubricant, basin, and a syringe that fits the tube. Tincture of benzoin, tape, and absorbent dressing material may be used to dress the wound, although many are better left undressed.64 Some feeding tubes require special plugs or connectors. Others need to be pinched with a clamp when not in use to prevent leakage. Some tubes are placed with the aid of accompanying guidewires or stents.

The easiest tube to replace is one that has been removed in the ED or dislodged for only a few hours. The stoma closes quite quickly, so replacement is best done as soon as possible. The stoma site can be gently probed by a cotton applicator to determine patency and direction of the tract (Fig. 40–21). In selected cases, a hemostat can gently dilate the opening to accept a replacement tube. All such attempts should be done carefully to avoid creating a false tract. Local anesthetic around the stoma may be used if exploration causes pain, and bleeding is common. After the tube is passed, restrain the hands to avoid tube removal by uncooperative patients.

Transabdominal Feeding Tube Removal A feeding tube may have to be removed because it is irreversibly clogged, leaking, or broken; persistently develops kinks;

VII 

● 

GASTROINTESTINAL PROCEDURES

Figure 40–18  A, Foley catheter is not an ideal feeding tube but can be used temporarily to maintain stoma patency, lasting only about a few months owing to disintegration of the latex. Always inflate the balloon with saline and use a bolster. B, To make an external bolster for a feeding tube to prevent tube migration: Step 1—Cut a 3-cm segment of tubing from the proximal segment of another Foley catheter. Step 2—Bend the tubing in half and cut to create a hole on each side of the segment. Step 3—Insert a hemostat through the holes in the completed bolster and grasp the feeding tube. Step 4—Advance the tube through the bolster to 1 cm above the skin of the external abdomen.

A

746

Step 1

Step 2

Feeding tube is pulled through hole in bolster

Step 3

B

Step 4

40  ● 

B

C

D

Nasogastric and feeding tube placement

A

747

E

F

Figure 40–19  A, This type of tube serves as the original PEG device. It has a mushroom head, not a balloon. B, When replaced, a balloon-type tube is used. C, This original tube is leaking because the mushroom tip has been pulled out of the stomach lumen and is lodged in the soft tissue of the abdominal wall. D, If the tube tract has matured (at least 2 wk after placement), it may be removed by traction/countertraction. Significant force may be required, and be prepared for a pop and splattering of gastric contents. E, It is easy to determine if there is a balloon at the end of a PEG tube that cannot be removed. Simply cut the tube and if there is no additional port or channel to inflate the balloon (F), it must be the type of tube that can be removed by traction.

A

B

Figure 40–20  A, Whenever possible, use a formal feeding tube instead of a Foley catheter. B, A common dilemma: This patient had persistent vomiting after tube feedings, and gastric distention. The tube had simply migrated distally (note comparison of new tube and positioning of indwelling one) because the bolster was too far proximal. Withdrawing the tube and repositioning the bolster alleviated the problem.

GASTROINTESTINAL PROCEDURES VII 

● 

B

A

D

748

C F

E

Figure 40–21  A, The feeding tube was dislodged at 11 pm, and by 9 am, the stoma was too tight for easy tube replacement. It was accomplished under fluoroscopic guidance, always the best option in questionable cases. B, The stoma opening and direction of the tract can be investigated by gently probing the site and tract with a Q-tip; in this case, it easily entered the stomach. C, This tight stoma was carefully dilated under local anesthesia with a hemostat. A false passage can easily be created. This area usually readily bleeds. D, To give a Foley catheter rigidity to aid in passage, the end of a Q-tip was inserted in the side port of the distal catheter, and traction was applied to the catheter. E, If a de Pezzer catheter is used, an ET tube stylet distends the flange for passage, and the tip re-forms once in the stomach. F, This patient removed her recently replaced feeding tube, with balloon inflated, while still in the ED awaiting transfer. This could have been avoided if her hands were restrained.

A

B

C

D

Figure 40–23  A, If a Foley balloon will not deflate, use traction to bring the balloon against the abdominal wall. Gently pass a small needle along the course of the catheter, puncturing as many times as necessary. Wait a few minutes for the fluid to egress. B, Once the balloon is deflated, it can be withdrawn. Note the encrusted condition of this longstanding Foley catheter used as a PEG tube. C, Occasionally the wire from a central line kit can clear the lumen and allow deflation. D, If the valve mechanism malfunctions, cut the catheter and attempt to drain the balloon by placing a needle in the inflation channel, by flushing and withdrawing fluid.

Nasogastric and feeding tube placement

Figure 40–22  Gentle, firm traction using the flat part of the opposite hand for countertraction will remove most PEG tubes, even those with internal mushroom bumpers. Modest force may be required, and be prepared for a sudden pop and splattering of gastric contents.

● 

puncture a Foley balloon, apply traction to the catheter to draw the balloon up against the ostomy (Fig. 40–23). Using the taut feeding tube as a guide, pass a small-gauge needle along the tube to puncture the balloon. It may be necessary to try again on the other side of the catheter, because the balloon may be asymmetrically inflated, and contact with the needle may be established on one side and not the other. Be careful not to track away from the ostomy into the patient’s abdominal wall or to cause separate punctures of the stomach. Allow a minute for the balloon to deflate before another attempt is made at traction removal. Large, nondeflating balloons should probably be punctured, whereas small balloons may be removed with traction.

40 

too large or too small; causes a hypersensitivity reaction; is associated with an abscess; or is not the appropriate length for feeding into the desired viscus. Before a new transabdominal feeding tube is inserted, the old tube must be removed. Most, but not all, tubes can be removed without endoscopy. It is imperative to know whether the tube in place is safe to remove before attempting to remove it. Standard de Pezzer or mushroom catheters that have been modified with bolsters or rings at the time of endoscopic or surgical insertion may no longer be safe to remove with traction. Tubes are occasionally secured with sutures or rigid internal bumpers or stays. It is rare, however, to encounter a tube that cannot be removed with traction/countertraction. Modest force may be required; use a hand for countertraction, and be prepared for a pop and splattering of gastric contents (Fig. 40–22). This causes the tube and end mushroom to narrow, and the tube should come out easily. The inner crossbar, if present, may remain in the stomach when the rest of the feeding tube complex is removed by traction. Obstruction from the crossbar, which will pass in the stool, has yet to be reported for adults. In small children, obstruction is a possibility, and the crossbar should be removed by endoscopy.65,66 Recently placed feeding tubes may need to be left in until a tract has formed (1–2 wk, depending on the procedure) even if the tube is nonfunctional. A simple Foley catheter G tube is easiest to remove. Deflate the Foley balloon and the tube should slide right out. If the Foley balloon cannot be deflated, cutting the tube may allow the balloon to deflate. Do not cut the catheter so close to the abdomen that it will be impossible to maintain a grip on it for a traction removal if the balloon still does not deflate. The balloon may also be punctured to cause it to deflate. To

749

GASTROINTESTINAL PROCEDURES ● 

VII 

750

If it is not possible to pull the inner bolster or mushroom out through the ostomy, cut the tube at the skin, push the remaining short stump into the stomach, and rely on later rectal passage. Although obstruction or impaction is infrequent, it can occur, and this alternative has the potential to be problematic with children or patients who have had previous impaction, potential for bowel obstruction, or stoolpassing problems. Rigid or large internal mushrooms and bolsters, the very kind that cause the most difficulty with percutaneous removal, also are more likely to cause difficulty with rectal passage. In no case should a device be released into the gut with a long length of tubing attached. Remember that double-part tubes may have an additional length of tubing for duodenal or jejunal feeding that extends far past the inner bolster. Korula and Harma67 reported the successful intestinal elimination of 63 of 64 gastrostomy tubes that were cut at the skin entrance and advanced into the stomach through the stoma. These cases included tubes with internal bumpers, and success occurred regardless of the nature of the patient’s underlying medical disorder, age, or method of original tube placement. However, no patient had suspected obstruction or potential for obstruction (e.g., no prior radiotherapy, inflammatory bowel disease). The 1 lodged tube required endoscopic removal from the pylorus. In most cases, tube passage was documented by sequential radiographs, with a mean interval to passage of 24 days (range, 4–181 days). Some clinicians and surgeons may strongly condemn cutting off the tube at the skin, even when the risks posed by the procedure are very low. It is always advisable to contact the patient’s private clinician before cutting the tube. In some cases, endoscopic retrieval of the tube remnant will be preferred to allowing rectal passage, and the tube should not be cut until just before or during endoscopy to ensure that migration does not occur before endoscopy.

Securing a Transabdominal Feeding Tube If a bolster is used, no additional means of securing the tube is necessary if the patient is not prone to pulling it out. Some clinicians tape tubes to the skin rather than using a bolster, or use special adhesive devices designed to control the tube and prevent ingress such as Drain/Tube Attachment Device (Hollister, Inc., Libertyville, IL) or Flexi-Trak (ConvaTec, Skillman, NJ). Tape is particularly vulnerable to problems because the stress the tape places on the skin as the tube pulls on the tape can lead to skin damage. Strong adhesive tapes can also damage the skin on removal. Tape that is not sufficiently adhesive can let go, particularly if it gets wet. Because tape is less durable, various home or nursing home caregivers who are less skilled may retape the tube (Fig. 40–24). If the

Tape

tape is replaced at home, it may be placed under too much tension, causing abdominal wall thinning at the stoma. If too much slack is allowed, the tube may get pulled in. Strong tape can also damage the tube during tape changes if it comes to adhere too strongly to the tube. Special adhesive devices or bolsters are preferred.

Verification of Placement There is no universally agreed-upon standard with regard to performing a confirmatory contrast study for all easily replaced feeding tubes. Some clinicians verify position routinely with a contrast radiograph, whereas others use the clinical criteria outlined earlier. The editors advocate radiographic verification in the majority of cases, a procedure easily performed and interpreted in the ED. The routine use of postplacement contrast radiography to confirm proper placement should be mandatory when the tube tract is immature (i.e., 1 cm) are more likely to retrieve particulate matter successfully, but the tube size is such that whole pills are unlikely to pass (Fig. 42–3). Smaller, more flexible tubes may kink and are significantly more difficult to pass. An NG tube may be passed through the mouth or nose, but orogastric hoses should not be passed through the nose. Because most pills disintegrate in the stomach in a few minutes, significant amounts of particulate matter may be retrieved with a large-bore NG tube such as an 18-French Salem sump tube. NG tubes are considerably easier to pass and are less traumatic for the

Figure 42–4  Gastric lavage in a child is always problematic. Obviously, an adult-sized large-bore oral gastric tube cannot be used, but a nasogastric (NG) tube may suffice. Some pediatric textbooks recommend a 24-Fr oral gastric tube for toddlers, and a 36-Fr tube for adolescents. In this case, a child was found with an open bottle of digoxin, and it could not be determined whether ingestion had occurred. She would not drink charcoal. The 18-Fr NG tube was used to attempt to aspirate digoxin from the stomach (none was recovered) and to instill charcoal. Some would suggest the oral route for this tube, but it was passed rather easily through the nose. An NG tube is not ideal for some ingestants (iron, sustained-release products), but most pills quickly dissolve in the stomach and the small particles can easily be removed with an NG tube. Although lavage may have been reasonable in this scenario, a potent and safe antidote for digoxin does exist. The common routine practice of passing an NG tube in a child who is unwilling to drink charcoal is controversial and while intuitively reasonable, it is of unproven value and likely done far too often for benign ingestions.  (Reprinted with permission from Seckl MJ, Rustin GJ, Newlands ES, et al. Pulmonary embolism, pulmonary hypertension, and choriocarcinoma. Lancet 338:1313, 1991.)

patient. NG tubes are preferred for liquid ingestions and in children (Fig. 42–4). In most cases, a 36- to 40-French or 30 English-gauge tube (external diameter, 12–13.3 mm) should be used in adults and a 24- to 28-French-gauge (diameter, 7.8–9.3 mm) tube in

42  ● 

Decontamination of the poisoned patient Figure 42–6  Once the pharynx has been entered, put the patient’s chin on the chest to facilitate passage of the tube into the esophagus. If the patient begins to vomit, withdraw the tube immediately to allow the patient to close the epiglottis and lessen the chance of aspiration.

Figure 42–5  Measure and mark the appropriate depth of gastric lavage tube prior to passage. This ensures that the tip is in the stomach and that there is no excess tubing to hinder fluid egress or to allow kinking/knotting of the tube.

children.22 Before passage, the length of the tube required to enter the stomach should be estimated by approximating the distance from the corner of the mouth to the midepigastrium; premeasurement avoids the curling and kinking of excess hose in the stomach (Fig. 42–5). Passage of an excessive length of hose may cause gastric distention, bruising, and perforation, whereas passage of an insufficient length of hose may result in lavage of the esophagus and the increased risk for emesis and aspiration. Commercial lavage systems are available and often use either a gravity fill-and-empty system with a Y connector or a closed irrigation syringe system. Alternatively, an irrigation syringe can be used for intermittent lavage fluid input and withdrawal. Technique Lubricate the gastric tube and pass it gently to avoid damage to the posterior pharynx. Use a bite block or an oral airway to avoid the patient’s chewing the orogastric tube and biting the fingers of the inserter. If the patient is obtunded or paralyzed, extend the jaw to facilitate passage. Never use force to pass the tube. Once the pharynx has been entered, put the patient’s chin on the chest to facilitate passage of the tube into the esophagus (Fig. 42–6). Cough, stridor, or cyanosis indicates that the tube has entered the trachea; withdraw the tube immediately and reattempt passage. Once the tube is passed, confirm that it is in the stomach. Intragastric placement is usually evident on clinical grounds and confirmed by auscultation of the stomach during injection of air with a 50-mL syringe and aspiration of gastric contents. In the intubated or obtunded patient or the young child, confirm the tube position radiographically before lavaging, although this is not routinely performed. A misplaced tube may irrigate the esophagus with a tube that has doubled back on itself during passage. The most serious complication is inadvertent passage

of the tube into the lungs. Tracheal passage of a lavage tube should be readily obvious in the awake patient before lavage, and obtunded patients are intubated, obviating this problem. If an awake patient begins to vomit during the lavage, immediately remove the tube to allow the patient to protect the airway. Before gastric irrigation, remove the gastric contents by careful gastric aspiration with repeated repositioning of the tube tip. With the Y connector closed system, perform lavage by clamping the drainage arm of the Y adapter and infusing aliquots of fluid into the stomach from a reservoir (Fig. 42–7). Clamp the reservoir arm of the Y, and then open the drainage arm to permit gravity drainage of the stomach contents. Repeat this procedure. Some resistance is produced by the Y connector and tubing. Apply suction intermittently to the drainage tubing to enhance stomach emptying. Lavage can be performed adequately with tap water in adults. Because electrolyte disturbance has occurred in children who were lavaged with tap water, prewarmed (45°C) normal saline is generally recommended for children.25–27 Warmed lavage fluid increases the solubility of most substances, delays gastric emptying, and theoretically, should increase the effectiveness of the procedure.28,29 Repeatedly introduce small aliquots of lavage solution (200–300 mL in adults and 10 mL/kg body weight in children up to a maximum of 300 mL) into the stomach and then remove them. Larger amounts of fluids create the potential for an increased risk of washing gastric contents into the duodenum or the lungs. Much smaller amounts are not clinically practical because of the dead space in the tubing (~50 mL in the 36-Fr hose) and the increase in time that is required. The amount of fluid that is returned should approximate the amount introduced. Manual agitation of the patient’s stomach by gently “kneading” the stomach with a hand placed on the abdomen may increase recovery.28 Continue lavage until the fluid becomes clear. After gastric aspiration and lavage have been completed, administer a slurry of activated charcoal through the gastric tube. When no longer needed, clamp off the gastric tube

763

GASTROINTESTINAL PROCEDURES ● 

VII 

A

Figure 42–7  A, Example of a Y connector closed system with the patient in a left lateral decubitus position. B, Patients on a ventilator or intubated with airway protection may be lavaged in the supine position, but an awake nonintubated patient is never lavaged in the supine position.

764

during its removal to avoid “dribbling” fluid into the airway. With the increasing use of repetitive doses of activated charcoal, the gastric tube is often left in place after the lavage procedure is completed. Because this large tube is irritating and may predispose the patient to gagging, drooling, or aspiration, it should be removed. The alert patient should take subsequent doses orally as necessary. The patient who remains obtunded may receive additional doses via a standard NG tube. Complications A correctly performed procedure in the appropriate environment is generally safe but numerous complications have been associated with gastric lavage.30 The complications can be divided into those caused by mechanical trauma and those resulting from the lavage fluid. Depending on the route selected for tube insertion, damage to the nasal mucosa, turbinates, pharynx, esophagus, and stomach has been reported.31–34 After tube insertion, it is imperative to confirm correct placement. Scalzo and associates35 found radiographically that 7 of 14 children had improper tube placement (too high or too low) despite positive gastric auscultation in all cases. Radiographic confirmation of tube placement should be considered in young children and intubated patients. Instillation of lavage fluid and charcoal into the lungs through tubes inadvertently misplaced within the airways has been reported.36 During lavage, changes in cardiorespiratory function have been noted. Thompson and coworkers37 reported that during lavage, 36% of patients had atrial or ventricular ectopy, 4.8% had transient ST elevation, and 29% had a fall in oxygen tension to 60 torr or less. Patients at greatest risk for these findings included the elderly, smokers, those with lung disease, or those with cyclic antidepressant overdose. Laryngospasm may also occur during gastric lavage.22

B

The lavage fluid itself is a potential source of complications. The large amount of fluid administered during lavage has been reported to cause patient fluid and electrolyte disturbances. These disturbances have been seen with the use of both hypertonic and hypotonic lavage fluids in the pediatric population.25–27 Hypothermia is a possible complication if the lavage fluid is not prewarmed. Pulmonary aspiration of gastric contents or lavage fluid is the primary potential risk during gastric lavage, especially in patients with compromised airway protective reflexes.38–40 Merigian and colleagues20 reported a 10% incidence of aspiration pneumonia in patients who received gastric lavage. This risk is reduced by using small aliquots of lavage fluid, adequately positioning the patient, and intubating patients with compromised airway protective reflexes. If the lavage tube cannot be easily removed, do not force it. Kinking or knotting of the tube can occur, but occasionally, a tube may become stuck because of lower esophageal spasm. If fluoroscopy or a radiograph demonstrates no deformation to the lavage tube, 1 to 2 mg of intravenous glucagon can be infused in an attempt to relieve lower esophageal spasm.41 Surgical removal may be necessary if the gastric tube is deformed by kinking or knotting.

Activated Charcoal Background Activated charcoal is a carbon product that is subjected to heat and oxidized to increase the surface area. It has the capacity to adsorb substances onto the porous surface of the charcoal. Activated charcoal acts both by adsorbing a wide range of toxins present in the gastrointestinal tract and by enhancing toxin elimination if systemic absorption has already occurred. It enhances elimination by creating a concentration gradient between the contents of the bowel and the circulation, but it

●

Up to 1 year: 1 g/kg of body weight. 1 year to 12 years: 25 to 50 g. ● Older than 12 years: 25 to 100 g. ●

POSITION STATEMENT: SINGLE-DOSE ACTIVATED CHARCOAL American Academy of Clinical Toxicology; European Association of Poisons Centres and Clinical Toxicologists Single-dose activated charcoal should not be administered routinely in the management of poisoned patients. Based on volunteer studies, the effectiveness of activated charcoal decreases with time; the greatest benefit is within 1 hour of ingestion. The administration of activated charcoal may be considered if a patient has ingested a potentially toxic amount of a poison (which is known to be adsorbed to charcoal) up to 1 hour previously; there are insufficient data to support or exclude its use after 1 hour of ingestion. There is no evidence that the administration of activated charcoal improves clinical outcome. Unless a patient has an intact or protected airway, the administration of charcoal is contraindicated. Figure 42–8  Position statement: single-dose activated charcoal.  (From the American Academy of Clinical Toxicology; European Association of Poisons Centres and Clinical Toxicologists. Published in Clin Toxicol 35:721, 1997.)

If the ingestion were, for example, clonidine (0.1-mg tablets) or digoxin (0.25-mg tablets), this regimen would be more than adequate for even a massive overdose to achieve the desired 10 : 1 ratio. If the ingestion consisted of a large number of 325-mg aspirin tablets, or 240-mg verapamil tablets, the dosing regimen could be insufficient. If toxic medications with a high milligram dosage are ingested, it would be prudent to administer more charcoal than indicated by these guidelines. There is no known benefit of mixing charcoal with a cathartic (i.e., sorbitol), and the combination is not suggested. Sorbitol increases the incidence of vomiting. Because in many formulations the contents settle with time, shake the preparation vigorously before administering it to the patient. Follow this by rinsing the container with a small amount of tap water before administering it to the patient to allow ingestion of the full dose.45 Aqueous activated charcoal has a gritty texture that most patients find unpleas-

Decontamination of the poisoned patient

Contraindications The administration of charcoal is contraindicated in any person who demonstrates compromised airway protective reflexes, unless he or she is intubated.43 It is absolutely con-

Technique There is no universally accurate dose for charcoal. A 10 : 1 ratio (charcoal-to-toxin) is recommended if the amount of ingestion is known. Charcoal dosing should be considered in light of the specific ingestion, but the recommended empirical doses of single-dose activated charcoal (standard aqueous products, such as Liqui-Char) are as follows:43

● 

Indications For years, the administration of a single dose of oral activated charcoal for essentially all overdoses has been routine. Clearly, charcoal binds many toxins in the gut, thereby decreasing some systemic absorption. Despite a lack of scientific data demonstrating a decrease in morbidity and mortality, and without firm evidence to support its widespread use, charcoal is a reasonable intervention for most poisoned patients presenting to the ED if it can be easily and safely administered (Fig. 42–8). The exact indications are not established, and no universally accepted standard of care has been promulgated.43 A single dose of activated charcoal is indicated if the clinician estimates that a clinically significant fraction of the ingested substance remains in the gastrointestinal tract, the toxin is adsorbed by charcoal, and further absorption may result in clinical deterioration. This will usually be a clinical decision, because adequate historical data may often be lacking. It may also be administered by multiple dosing if the clinician anti­ cipates that the charcoal will result in increased clearance of an already absorbed drug. It is most effective within the first 60 minutes after oral overdose and decreases in effectiveness over time. Charcoal is generally considered to provide superior gut decontamination over gastric lavage. There is no definitive evidence that administration of activated charcoal improves outcome.

traindicated in persons who have ingested corrosive substances (acids or alkalis). Not only does charcoal provide no benefit in a corrosive ingestion, but its administration could precipitate vomiting, obscure endoscopic visualization, and lead to complications if a perforation developed and charcoal entered the mediastinum, peritoneum, or pleural space. Charcoal should be avoided in cases of a pure aliphatic petroleum distillate ingestion. Hydrocarbons are not well adsorbed by activated charcoal, and its administration could lead to further aspiration risk. Many hydrocarbons are potential systemic toxins (e.g., carbon tetrachloride and benzene) or are mixed with other potentially significant toxins such as pesticides. In these cases, data are lacking, but charcoal administration can be considered. Caution should be exercised in using charcoal in patients with medical conditions that could be further compromised by charcoal ingestion, such as gastrointestinal perforation or bleeding. Charcoal is not indicated for isolated ingestions of ethanol, iron, or lithium because these substances are not adsorbed. If the airway is not secure, charcoal should be given with caution to minimally symptomatic patients who have ingested a toxin that may suddenly induce seizures. Because it is often impossible to determine the exact nature of an ingestion, a liberal use policy is advocated for potentially mixed overdoses. Charcoal administration by paramedics and other emergency response personnel should be performed with caution.44 The same indications and contraindications apply as for those patients who are in the hospital. The motion of the ambulance during transport may make the patient more prone to emesis. Either the spilling of charcoal or the vomiting of charcoal may result in significant contamination of the transport vehicle and subsequently place that vehicle out of commission until it can be cleaned.

42 

also has the potential of interrupting enterohepatic circulation if the particular toxin is secreted in the bile and enters the gastrointestinal tract before reabsorption.42 Oral activated charcoal is given as a single dose or in multiple doses. The adsorptive capacity of charcoal depends on the inherent properties of the toxin and the local milieu, such as pH. Adsorption begins within minutes of contact with a toxin, but may not reach equilibrium for 20 to 30 minutes. Desorption of toxins from charcoal occurs over time, although this has little clinical significance for most patients and can be overcome by administering additional charcoal.

765

GASTROINTESTINAL PROCEDURES ● 

VII 

A

Figure 42–9  Radiographic confirmation of NG tube placement may be performed before lavage or instillation of charcoal if there is uncertainty about the position of the tube. Tracheal placement of a lavage tube is usually readily evident. Vomiting during lavage suggests that the tube has curved back into the esophagus. A confirmatory radiograph is suggested in the obtunded patient if correct gastric placement is questioned. Tracheal intubation precludes passage of a tube into the lungs, but it does not ensure proper gastric placement.

766

ant; attempts have been made to improve the taste and texture. Mixing activated charcoal with chocolate milk, chocolate- or cherry-flavored syrup, or ice cream may increase palatability, but mixing with these additives has been suggested, though not proved, to cause a decrease in the adsorptive capacity of activated charcoal.46 Rangan and colleagues47 reported no decrease in adsorption after mixing superactivated charcoal with a noncaffeinated cola. Scharman and associates48 demonstrated that a regular, sugared cola was favored by children over a diet cola, but only 20% of the time were they able to cajole even nonpoisoned children younger than 3 years to drink a therapeutic amount of flavored charcoal. Give activated charcoal orally if the patient is awake and cooperative and by NG tube if the patient is unconscious. If an NG tube is inserted, it is imperative to verify correct placement (Fig. 42–9). Confirm correct tube placement radiographically before administering charcoal, especially in obtunded or intubated patients. Instillation of charcoal into the lungs has been reported after inadvertent misplacement within the airways36 and massive aspiration can be fatal49 (Fig. 42–10). Intubation is protective but it is not uncommon to see some charcoal in the airway even if the patient has been intubated. The common tactic of passing an NG tube in the awake but uncooperative patient merely to administer charcoal is controversial. Such a scenario is more likely to result in trauma from the tube placement, a misplaced tube, or subsequent emesis from the rapid administration of charcoal. Given the unproven efficacy of charcoal, the authors advise against the routine insertion of an NG tube simply to administer charcoal in the awake and minimally symptomatic patient. Such a decision is, however, a clinical one that must be made by the clinician and based on the entire clinical milieu (Fig. 42–11).

B Figure 42–10  A, Vomiting can be expected following charcoal administration, especially if sorbitol is added (not recommended). This patient rapidly became drowsy after charcoal administration, vomited, but fortunately did not aspirate. B, In another patient who aspirated, the charcoal can be seen at the carina with a fiberoptic scope. Massive aspiration can be fatal. Intubation does not totally protect against minimal charcoal aspiration.

Complications The administration of activated charcoal is not without risks and complications. Published reports have demonstrated adverse effects associated with activated charcoal therapy, including childhood deaths.50 The most common complications of charcoal administration include constipation, diarrhea, and vomiting.51,52 Bowel perforation has been described in a patient with diverticular disease.53 Pulmonary aspiration of activated charcoal is a dreaded complication that can result in pneumonitis, obstruction of the respiratory tree, bronchiolitis obliterans,33,54–56 acute lung injury, and barotrauma.50 Risk factors for serious aspiration are large amounts of charcoal instilled over a short period of time, multiple-dose charcoal in the setting of an ileus, charcoal administration in a patient who becomes obtunded, charcoal that is inappropriately diluted, or the forced administration of charcoal via an NG tube, especially in a restrained supine patient. These complications can be prevented by prudent dosing of charcoal and associated cathartic therapy, as well as monitoring of fluid and electrolyte status, abdominal examinations, and clinical condition. Trivial aspirations of charcoal are common and

POSITION STATEMENT: POSITION STATEMENT AND PRACTICE GUIDELINES ON THE USE OF MULTI-DOSE ACTIVATED CHARCOAL IN THE TREATMENT OF ACUTE POISONING

Decontamination of the poisoned patient

Indications The use of multiple-dose activated charcoal (MDAC) may be indicated in select cases59 (Fig. 42–12). Its use has been advo-

● 

Multiple Doses of Activated Charcoal

42 

usually innocuous even if the patient is intubated. Studies show a 4% to 39% incidence of aspiration pneumonia in intubated patients who received activated charcoal while intubated.57 It has been shown that even in the face of a protected airway with a cuffed endotracheal tube, vomiting can lead to pulmonary aspiration of the charcoal.50 This can lead to a significant increase in lung microvascular permeability, causing lung edema and pulmonary compromise.58

American Academy of Clinical Toxicology; European Association of Poisons Centres and Clinical Toxicologists

A

B Figure 42–11  A, If the overdose patient will voluntarily drink charcoal, there are few reasons to withhold it, even though a definite clinical benefit in the routine case cannot be proved. If a patient will not drink charcoal, patient management becomes controversial. Passing an NG tube in a struggling patient or in a recalcitrant child merely to instill the unproven, but theoretically useful, antidote is not supported by scientific data. Nonetheless, it remains a common procedure. Although not always easy or pleasant, such an intervention is usually safe. Pulmonary aspiration, even in the awake patient, is the major downside. Restrained supine patients are at greatest risk for aspiration, and that position should be avoided, even in the initially awake patient. B, Charcoal that is voluntarily swallowed or instilled via an oral-gastric lavage tube or NG tube can induce emesis. This occurs in both the obtunded and the awake patient. In this instance, the patient was unconscious from the overdose and the airway was protected with prior tracheal intubation. Although the intubation procedure does not totally exclude pulmonary aspiration and it carries some morbidity in its own right, it is recommended prior to charcoal use in the patient who is not able to fully protect the airway. Patients who initially are asymptomatic or minimally affected but have ingested drugs that have the potential to produce rapid deterioration, seizures, or loss of airway protection make decisions on the use of charcoal difficult for the clinician. In borderline cases, some experienced clinicians avoid the use of charcoal altogether.

Although many studies in animals and volunteers have demonstrated that multiple-dose activated charcoal increases drug elimination significantly, this therapy has not yet been shown in a controlled study in poisoned patients to reduce morbidity and mortality. Further studies are required to establish its role and the optimal dosage regimen of charcoal to be administered. Based on experimental and clinical studies, multiple-dose activated charcoal should be considered only if a patient has ingested a life-threatening amount of carbamazepine, dapsone, Phenobarbital, quinine, or theophylline. With all of these drugs there are data to confirm enhanced elimination, though no controlled studies have demonstrated clinical benefit. Although volunteer studies have demonstrated that multiple-dose activated charcoal increases the elimination of amitriptyline, dextropropoxyphene, digitoxin, digoxin, disopyramide, nadolol, phenylbutazone, phenytoin, piroxicam, and sotalol, there are insufficient clinical data to support or exclude the use of this therapy. The use of multiple-dose charcoal in salicylate poisoning is controversial. One animal study and 2 of 4 volunteer studies did not demonstrate increased salicylate clearance with multiple-dose charcoal therapy. Data in poisoned patients are insufficient presently to recommend the use of multiple-dose charcoal therapy for salicylate poisoning. Multiple-dose activated charcoal did not increase the elimination of astemizole, chlorpropamide, doxepin, imipramine, meprobamate, methotrexate, phenytoin, sodium valproate, tobramycin, and vancomycin in experimental and/or clinical studies. Unless a patient has an intact or protected airway, the administration of multiple-dose activated charcoal is contraindicated. It should not be used in the presence of an intestinal obstruction. The need for concurrent administration of cathartics remains unproven and is not recommended. In particular, cathartics should not be administered to young children because of the propensity of laxatives to cause fluid and electrolyte imbalance. In conclusion, based on experimental and clinical studies, multiple-dose activated charcoal should be considered only if a patient has ingested a life-threatening amount of carbamazepine, dapsone, Phenobarbital, quinine, or theophylline. Figure 42–12  Position statement and practice guidelines on the use of multidose activated charcoal in the treatment of acute poisoning.  (From the American Academy of Clinical Toxicology; European Association of Poisons Centres and Clinical Toxicologists. Published in Clin Toxicol 37:731, 1999.)

767

Aspirin Caffeine Carbamazepine Cyclosporine Dapsone Digoxin Disopyramide Nadolol

Phenobarbital Phenytoin Quinine Sotalol Sustained-release thallium Theophylline Valproate Vancomycin

VII 

● 

GASTROINTESTINAL PROCEDURES

TABLE 42–2  Drugs Whose Serum Clearance May Be Enhanced by Multiple Doses of Activated Charcoal

768

cated for two purposes: first, to prevent continued absorption of a drug that may still be present within the gastrointestinal tract; second, to increase the serum clearance of a drug that has already been absorbed (Table 42–2). MDAC prevents continued absorption by either binding a drug that may be present throughout the gastrointestinal tract or binding a drug that exists as extended-release or enteric-coated preparations. MDAC enhances elimination of a drug by interrupting enterobiliary recirculation or augmenting enterocapillary exsorption.52 By interrupting enterobiliary recirculation, charcoal binds to an active drug that is secreted by the biliary system, subsequently preventing reabsorption. By augmentation of enterocapillary exsorption, charcoal produces sink conditions that drive diffusion of drug from the capillaries into the entraluminal space, where it is subsequently eliminated. This process is called intestinal dialysis.60 Drug characteristics that are associated with enhanced systemic clearance with MDAC include a low intrinsic clearance, a prolonged distributive phase, low protein binding, and a small volume of distribution.61 MDAC has been shown to increase total body clearance of multiple drugs, including carbamazepine,62–64 dapsone,65,66 phenobarbital,64,67–73 quinine,74,75 and theophylline.42,59,76–87 Despite the reported increase in drug clearance associated with the use of MDAC, improved clinical outcomes have not been definitively demonstrated. For example, Pond and coworkers72 described 10 comatose patients following phenobarbital overdose who were randomized to receive either single-dose activated charcoal or MDAC. Despite the fact that the MDAC group had a significantly shorter phenobarbital serum halflife, no difference was found between the groups in regard to the duration of intubation or hospitalization. Contraindications MDAC is contraindicated if there is evidence of bowel obstruction. An ileus is a relative contraindication. Many ill patients who develop an ileus may be selected candidates for MDAC if the airway is protected. The administration of MDAC is contraindicated in any patient who does not have an intact or protected airway. MDAC should be avoided in patients who have repetitive emesis, especially when associated with decreased mental status or a decreased gag reflex. The concurrent use of cathartics with MDAC remains unproved and is not recommended.88 MDAC with cathartics should not be administered to young children because of the propensity for laxatives to cause fluid and electrolyte imbalance. For example, MDAC with sorbitol has been associated with hypernatremia and dehydration,89,90 and MDAC with magnesium cathartics has been associated with hypermagnesemia, neuromuscular weakness, and coma.91,92

Technique Give 1 g/kg (≤100 g) for the first dose of charcoal. If a cathartic is used, administer it only with the first dose of charcoal to decrease the risk of cathartic-induced electrolyte abnormalities that can develop, especially in children.89–92 Follow the initial dose of charcoal by 0.5 g/kg (≤50 g) every 4 hours. Stop giving MDAC if repeat examination reveals an absence of bowel sounds or a distended abdomen. In this case, consider placing an NG tube and put it on low intermittent suction. Patients receiving MDAC may be at increased risk for emesis because of the larger total dose of activated charcoal received. The use of antiemetics may help decrease the incidence of vomiting associated with MDAC.76,93,94 Charcoal therapy should be continued until there is clinical improvement and plasma drug levels have fallen to acceptable levels. Complications The complications encountered in single-dose activated charcoal are also encountered in MDAC. In addition, there have been reports of gastrointestinal obstruction and perforation from MDAC therapy, especially in conjunction with the ingestion of drugs with anticholinergic properties.95–99

Cathartics Background The use of cathartics is intended to decrease the absorption of substances by accelerating the expulsion of the poison from the gastrointestinal tract. Cathartics are often used in conjunction with activated charcoal owing to charcoal’s side effect of constipation. The mechanism of action of cathartics is such that, theoretically, it would minimize the possibility of desorption of drug bound to activated charcoal. There is little evidence that a single dose of aqueous activated charcoal is significantly constipating; however, cathartics are often given for this potential problem. The majority of data suggest negligible clinical benefit from cathartic use.100,101 Indications The routine administration of a cathartic in combination with activated charcoal is not endorsed by the American Academy of Clinical Toxicology or the European Association of Poison Centres and Clinical Toxicologists.102 The administration of a cathartic alone has no role in the management of the poisoned patient. Contraindications Cathartics are contraindicated if there is volume depletion, hypotension, significant electrolyte imbalance, corrosive ingestion, ileus, recent bowel surgery, intestinal obstruction, or perforation. The administration of cathartics is also contraindicated with patients who do not have an intact or protected airway. They should be avoided in patients who have repetitive emesis, especially when associated with decreased mental status or a decreased gag reflex. Cathartics should be used cautiously in young children and the elderly because of the propensity for laxatives to cause fluid and electrolyte imbalance. Technique There are two types of osmotic cathartics: saccharide cathartics (sorbitol) and saline cathartics (magnesium citrate, magnesium sulfate, and sodium sulfate). The optimal dose of

Whole bowel irrigation (WBI) should not be used routinely in the management of the poisoned patient. Although some volunteer studies have shown substantial decreases in the bioavailability of ingested drugs, no controlled clinical trials have been performed and there is no conclusive evidence that WBI improves the outcome of the poisoned patient. Based on volunteer studies, WBI may be considered for potentially toxic ingestions of sustained-release or enteric-coated drugs. There are insufficient data to support or exclude the use of WBI for potentially toxic ingestions of iron, lead, zinc, or packets of illicit drugs; WBI remains a theoretical option for these ingestions. WBI is contraindicated in patients with bowel obstruction, perforation, ileus, and in patients with hemodynamic instability or compromised unprotected airways. WBI should be used cautiously in debilitated patients, or in patients with medical conditions that may be further compromised by its use. A single dose of activated charcoal administered prior to WBI does not appear to decrease the binding capacity of charcoal or to alter the osmotic properties of WBI solution. Administration of charcoal during WBI appears to decrease the binding capacity of charcoal.

Complications The administration of sorbitol has been associated with vomiting, abdominal cramps, nausea, diaphoresis, and transient hypotension.103–105 Because the sorbitol content varies between different charcoal/sorbitol combination products, pay attention to the sorbitol content in each brand to avoid excessive sorbitol administration. Be aware that multiple doses of sorbitol have been associated with volume depletion.89 Multiple doses of magnesium-containing cathartics have been associated with severe hypermagnesemia.91,92 Children are particularly susceptible to the adverse effects of cathartics, and therefore, use caution or totally avoid using cathartics in children.

WBI Background WBI involves the enteral administration of an osmotically balanced polyethylene glycol electrolyte solution (PEG-ES) in a sufficient amount and rate to physically flush ingested substances through the gastrointestinal tract, purging the toxin before absorption can occur.12 PEG-ES (CoLyte, GoLYTELY) is isosmotic, is not systemically absorbed, and will not cause electrolyte or fluid shifts. Available data suggest that the large volumes of this solution needed to mechanically propel pills, drug packets, or other substances through the gastrointestinal tract are safe, including in pregnant women and in young children.106,107 The clinical data on the efficacy of WBI remains limited. Ly and colleagues108 found that the effect of WBI on the reduction of acetominophen concentration versus time was not statistically significant. However, WBI did have a mechanical effect on radiopaque markers in the gastrointestinal tract with 8 of 10 subjects’ markers congregating in the right hemicolon after WBI. WBI was shown to mobilize lead BB pellets in a child to the large bowel, where less absorption occurs and the foreign bodies could be removed by colonoscopic intervention.109 In addition, PEG-ES may play a role in the pharmacologic conversion of some toxins. For example, it has been shown that the relatively high pH of PEG-ES increases the rate of spontaneous conversion of cocaine to its inactive metabolite benzoylecgonine.110 Indications WBI may be considered for ingestions of exceedingly large quantities of potentially toxic substances, ingestions of toxins that are poorly adsorbed to activated charcoal (e.g., iron, lithium), ingestions of delayed-release formulations, late presentation after ingestion of a toxin, pharmacobezoars, and in body stuffers or packers12,110–113 (Fig. 42–13). WBI remains a theoretical option for these ingestions and is often performed on body packers who have ingested many times the lethal amount of heroin or cocaine (Fig. 42–14). No definitive evidence exists that WBI improves the outcome of the poisoned

Decontamination of the poisoned patient

American Academy of Clinical Toxicology; European Association of Poisons Centres and Clinical Toxicologists

● 

Position Statement: Whole Bowel Irrigation

42 

sorbitol or magnesium citrate remains to be determined. The recommended dose of sorbitol is approximately 1 to 2 g/kg of body weight or 1 to 2 mL/kg of 70% sorbitol in adults and 4.3 mL/kg of 35% sorbitol in children (single administration only).102 Many charcoal formulations come premixed with sorbitol, but the sorbitol content varies considerably. The recommended dose of magnesium citrate is 250 mL of 10% solution in an adult and 4 mL/kg body weight of 10% solution in a child. Multiple doses of cathartics should be avoided.

Figure 42–13  Position statement: whole bowel irrigation (WBI).  (From the American Academy of Clinical Toxicology; European Association of Poisons Centres and Clinical Toxicologists. Published in Clin Toxicol 35:753, 1997.)

patient.114 Although not a proven procedure, WBI is often suggested by toxicologists, and its use in select cases is intuitively reasonable and supported by the authors. The most common indication for WBI in the ED is for the treatment of toxic sustained-release medications (such as iron, calcium channel blockers, β-blockers, theophylline, and lithium) and iron tablets (Fig. 42–15).115 Contraindications WBI is contraindicated in patients with gastrointestinal obstruction, perforation, ileus, and corrosive ingestion. It should also be avoided in patients with hemodynamic instability or an unprotected airway.115 WBI should also be avoided with patients who have repetitive emesis, especially when associated with decreased mental status or a decreased gag reflex. WBI should be used cautiously in debilitated patients. Technique PEG-ES is marketed in a powder form. Add tap water to make a total volume of 4 L. The recommended rate of administration is12,115 ●

9 months to 6 years: 500 mL/hr. 6 years to 12 years: 1000 mL/hr. ● Older than 12 years: 1500 to 2000 mL/hr. ●

Cooperative patients with intact airway protective reflexes may drink the solution. The large volume and taste often limit even the most motivated patient’s ability to comply. If the patient is unable or unwilling to drink this solution, administer it through a small-bore NG tube after placement is con-

769

GASTROINTESTINAL PROCEDURES ● 

VII 

Figure 42–15  WBI is commonly recommended for the treatment of iron ingestion. These radiographs depict the effect of 5 hr of WBI. Note the marked decrease of radiopaque pills (arrows) in the gastrointestinal tract. Intact pills were recovered in the rectal effluent.

A

770

B Figure 42–14  A, This body packer attempted to smuggle more than 50 packets of heroin. All packets were passed intact after 12 hr of whole bowel irrigation. B, Note the integrity of the carefully wrapped packets that were passed.

firmed. Even cooperative patients have difficulty drinking adequate fluid for effective whole bowel irrigation. Because it is common for WBI to be delayed while the patient and medical personnel attempt to administer the large volumes of oral WBI solution required to be effective, it is suggested that NG instillation be instituted early in the ED course (Fig. 42–16). Unconscious patients with protected airways may receive WBI via an NG tube. In one study, patients vomited shortly after beginning WBI infusion at a rate of 1.5 to 2L/hr. Antiemetics, such as metoclopramide, as well as gradually advancing the infusion rate over 60 minutes can help ease this side effect.108 Prewarming the irrigant to a temperature of approximately 37°C avoids the potential complication of hypothermia. To collect the waste products, ask an awake patient to

sit on a commode. In an obtunded patient, insert a rectal tube to collect the waste. Many toxicologists recommend adding two to three bottles of activated charcoal to each liter of WBI solution. The benefit is unproved, but there is little theoretical downside to this technique, and it is supported by the editors. The binding capacity of charcoal is decreased when combined with PEG-ES, but the clinical consequences of this observation are unknown. Empirically, metoclopramide may be coadministered to decrease nausea and facilitate gastrointestinal passage. The end point of WBI is the arrival of clear rectal effluent and/or resolution of toxic effect.115 There are rare case reports of late purging of drug packets, plant parts, and tablets after the arrival of clear effluent.111,116 Radiographic studies may also be beneficial to determine the end point in body packers or in patients who have ingested radiopaque medications. Complications Few complications from WBI therapy, especially pertaining to acute poisonings, have been reported. Nausea, vomiting, abdominal cramps, and bloating have been described.117 Nausea and vomiting may make administration of WBI difficult. Antiemetics and a 15- to 30-minute break followed by a slower rate may allow readministration. As discussed with the other methods of decontamination, attention should be directed to the airway and the potential for aspiration. Administration of a large amount of chilled or room-temperature WBI fluid to pediatric patients could potentially cause hypothermia. Consider warmed fluids in these patients. If activated charcoal is administered concurrently with WBI, there might be a desorption of toxin from charcoal.118–120

DERMAL DECONTAMINATION Background Numerous hazardous material (HAZMAT) incidents occur each year in the United States. In 2004, 7744 HAZMAT events in 15 states involving thousands of substances were reported to the Hazardous Substances Emergency Events Surveillance System (HSEESS). HAZMAT events frequently result in injuries, and the ED treatment of contaminated HAZMAT patients is not a rare event. Many of these patients, including those involved in past terrorist events, transport themselves to the ED. For example, in the Tokyo sarin gas attack, 93% of 498 patients reporting to St. Luke’s Hospital

Figure 42–16  It is very difficult for even the most motivated patient to drink an effective volume of WBI solution. To enhance compliance and to decrease vomiting, polyethylene glycol electrolyte solution (PEG-ES) may be slowly and continuously administered via an NG tube. An empty bag of saline is hung on an intravenous pole, the corner of the bag is removed, and the PEG-ES is poured into the bag. Standard intravenous tubing is connected to the proximal end of an NG tube and the solution is infused continuously. In this picture, charcoal has been added to the WBI solution. Metoclopramide was coadministered to reduce nausea.

Decontamination of the poisoned patient

B

● 

A

42 

arrived by means other than ambulance.121 The risk of injury to medical personnel incurred while treating contaminated patients is significant. Of the patients reported to HSEESS, emergency responders accounted for 10% of injuries and hospital personnel for 4.1% of injuries.122 After the Tokyo attack, 13 of 15 clinicians (87%) reported symptoms while treating patients in the ED and 23% of involved hospital staff complained of acute poisoning symptoms.123 Burgess and associates124 reported that 13% of Washington state emergency care facilities had evacuated their ED or another part of the hospital for contamination during a 5-year period. Ghilarducci and coworkers125 surveyed level 1 trauma centers in the United States and reported that only 6% had the necessary equipment required for safe decontamination. Less than 36% of emergency medicine staff had received appropriate training in handling the contaminated patient, and 5.6% had experienced injuries to their staff due to contact with contaminated patients during a 1-year period. It is imperative that EDs have plans in place to handle patients who are exposed to potential toxins, provide adequate decontamination facilities, and ensure the safety of the treating medical staff.126

Technique There are a number of key components in the management of hazardous materials incidents and the care of the contaminated patients who present to the ED.127 These components should include early recognition of a HAZMAT event, rapid activation of a plan to manage contaminated patients, initiation of primary triage, appropriate patient registration, patient decontamination, secondary triage, and final treatment. First, the ED must be able to recognize that an event has occurred before contaminated patients gain entrance into the health care facility. Communication with local fire, police, and paramedics provides early detection of such events and allows preparation before patients arrive. Security should be arranged to prevent contaminated patients from entering the hospital, and a “lockdown” of the facility should be considered. Second, the ED should have the authority to activate a plan expeditiously to prepare the decontamination facility and allow appropriate preselected personnel to don personal protective equipment (PPE). If necessary, the hospital disaster plan should be activated quickly at the discretion of the ED clinician who is in contact with scene operations and incoming patients. Specific data to determine the appropriate level of PPE to maintain hospital worker protection remain limited. Minimum PPE for hospital-based decontamination (level C) consists of a splash-proof, chemical-resistant suit with tape, double-layer protective gloves, and a powered air-purifying respirator per the National Institute of Occupational Safety and Health. Higher levels of protection, such as a level A Self-Contained Breathing Apparatus (SCBA) fully encapsulated chemical-resistant suit or level B SCBA chemical-resistant suit, is recommended with unknown chemical and biologic exposures and for entering hot zones, but these are not readily available in EDs.128,129 Fortunately, most chemical exposures are known. For those that occur in the workplace, Material Safety Data Sheets can be obtained and either the local poison center or the Agency for Toxic Substances and Disease Registry (ATSDR) can be contacted for advice on what level of protection is appropriate. Third, appropriate primary triage should occur. Contaminated patients should not enter the ED until proper

771

GASTROINTESTINAL PROCEDURES ● 

VII 

772

decontamination has occurred to ensure that the hospital staff will not get secondary contamination. Appropriate triage should then occur, with experienced personnel performing an initial brief assessment of each patient. The triage and decontamination areas should be organized into several “zones” to prevent further contamination. The “hot” zone is the location with the highest level of contaminant, or where the incident occurred. In most cases of hospital-based decontamination, there is no hot zone because the patients have been removed from the initial chemical insult. On average, patients arrive at the ED 20 minutes after the event and have had significant off-gassing by this time; however, the majority of patients have self-transported and have received no prehospital decontamination by Emergency Medical Services/HAZMAT. Basic life-saving treatments, airway/hemorrhage control, antidote administration (e.g., for cyanide or nerve agents), and decontamination occur in the “warm” zone. The “cold” zone is safe from contaminant.128 Fourth, a brief sign-in process in the warm zone should capture the patient’s name and date of birth, with full registration to occur after decontamination. Contaminated clothing and valuables should be placed in an impervious bag to avoid potential off-gassing.130,131 Fifth, decontamination should be performed. The hospital ED should have preexisting HAZMAT incident protocols that designate the decontamination area and the triage and decontamination team. Ideally, a hospital should have a permanent decontamination facility capable of handling a small number of chemically exposed patients and, in addition, a large portable unit for mass casualties. The decontamination area should meet several qualifications: (1) it should be secured to prevent spread to other areas of the hospital, (2) the ventilation system should be separate from the rest of the hospital or it should be shut off to prevent airborne spread of contaminants, and (3) provisions must be made to collect the rinsate from contaminated patients to prevent contamination of the facility and water supply. At most facilities, the best place to begin initial treatment and evaluation is outdoors (Fig. 42–17). Portable decontamination facilities are available, but their cost may be prohibitive for many institutions. A practical alternative is to have a warm shower nozzle, soap, and wading pool available outside the entrance to the ED. A tent or screen can provide privacy. The first priority in decontaminating a patient is to remove her or his clothing while both maintaining privacy and preventing hypothermia. This step is the most important in the decontamination process and can reduce the contaminant by 75% to 90%. Cut the clothes off rather than pulling them off, if possible. Place all clothing and valuables in labeled bags, as mentioned earlier. Brush off solids with a soft brush or towel. Irrigate the skin with copious amounts of warm water and cleanse the skin with soap. Although some agents (e.g., metallic sodium, potassium, cesium, and rubidium) may react with water, irrigation is still more beneficial to immediately decontaminate the skin than to delay treatment time. Starting from head to toe, irrigate the exposed skin and hair for 10 to 15 minutes. Scrub with a soft surgical sponge, being careful not to abrade the skin. Irrigate wounds for an additional 5 to 10 minutes with water or saline. Remove contact lenses and irrigate the eyes for 10 to 15 minutes with saline. Direct irrigation away from the medial canthus to avoid

Figure 42–17  During a hazardous materials (HAZMAT) incident, a decontamination tent with personnel in protective gear is assembled outside the emergency department entrance.

forcing contaminants into the lacrimal duct. With strongly alkaline substances, irrigate for longer times. Irrigate the nares and the ear canals with frequent suctioning if contamination is suspected. Clean underneath the fingernails with a brush. Avoid using stiff brushes and abrasives because they may enhance dermal absorption of the toxin and can produce skin lesions that may be mistaken for chemical injuries. Sponges and disposable towels are effective alternatives. Secondary triage should occur after decontamination. Transfer patients with major or moderate casualties to areas designated for such cases. Send patients with minor or no injuries to appropriate holding areas for further evaluation. Medical care at this stage depends on the toxin to which the patient has been exposed and the potential toxicity of that agent. Wounds, after copious irrigation, may need thorough exploration and possibly surgical removal of the contaminant. In order for the ED to care for the contaminated patient, protocols should be in place and regularly rehearsed by the facility. Train staff in the procedures and protocols, establish communication between community agencies and the hospitals, regularly inspect equipment, and rehearse setups. Obtain template protocols both from peer-reviewed medical literature and in the government literature if needed.132,133 For example, guidelines for managing HAZMAT incidents are available from the Emergency Response and Consultation Branch (E57), Division of Health Assessment and Consultation, Agency for Toxic Substances and Disease Registry, 1600 Clifton Road NE, Atlanta, GA 30333. In addition, prompted by the 2001 terrorist attacks, the U.S. Department of Veterans Affairs and several policy experts developed a “comprehensive hospital-wide emergency mass casualty decontamination program.” This program has been applied at most Veterans Affairs medical centers and demonstrates a costeffective protocol suitable for implementation at other U.S. hospitals.134

REFERENCES  c a n

be found on

E x p e rt C o n s u lt

Michael S. Runyon and John A. Marx

Paracentesis and diagnostic peritoneal lavage (DPL) constitute the two primary intraperitoneal procedures. They are fundamentally similar in purpose and design; however, the former is generally reserved for medical concerns and the latter for traumatic pathology. There is no mandate for emergency clinicians to perform these procedures if local custom is to refer to specialists, but many emergency clinicians routinely perform these procedures.

DPL Root and colleagues1 introduced DPL in 1964. It has withstood the passage of more than 4 decades and remains a useful tool in the management of penetrating torso trauma. Following a blunt mechanism of injury, its greatest utility is as a triage tool in the assessment of the hemodynamically unstable multiply injured patient. The intent is to rapidly discover or exclude the presence of intraperitoneal hemorrhage (IPH). This purpose is identical with that of ultrasound (US) in the diagnostic armamentarium of the emergency clinician evaluating the blunt trauma patient.2 Although commonly referred to as diagnostic peritoneal lavage, this procedure has two distinct components: peritoneal aspiration and peritoneal lavage. Peritoneal aspiration, in which an attempt is made to retrieve free intraperitoneal blood, precedes lavage. A finding of intraperitoneal blood presages intraperitoneal organ injury and precludes the need for subsequent lavage. In the lavage portion, normal saline is introduced by catheter into the peritoneal cavity, recovered by gravity, and analyzed. Peritoneal lavage can be used as a therapeutic tool in hypothermia and as a means of removing toxins.3 It has also been used as a diagnostic instrument for suspected intraabdominal infection and nontraumatic sources of hemorrhage.4,5 However, its primary use is as a determinant for the need for laparotomy after trauma, and this chapter focuses on that use.

Indications Blunt Trauma Prior to the advent of computed tomography (CT) and US, DPL was the sole diagnostic option to supplement physical examination for predicting the need for operative intervention (Table 43–1). It was integral both to the reduction of unnecessary laparotomies and to the discovery of unsuspected and life-threatening intra-abdominal hemorrhage in patients with significant closed head injury.6,7 In a number of respected centers in the United States, DPL continues to be a focal diagnostic instrument.8 It serves two primary functions.9 First, it can rapidly determine or

Peritoneal procedures

Peritoneal Procedures

● 

43

43 

C H A P T E R

exclude the presence of IPH (Table 43–2). Thus, the patient with a critical closed head injury, the unstable motor vehicle crash victim with multiple potential sources of blood loss, or the patient with pelvic fracture and retroperitoneal hemorrhage can be appropriately routed to life-saving laparotomy.10,11 Furthermore, given its exquisite sensitivity, a negative peritoneal aspiration allows the clinician to proceed to alternative management steps and the patient to forego unnecessary laparotomy. Second, DPL has been used in less exigent circumstances as a means of predicting solid or hollow visceral injury requiring laparotomy.12,13 However, in this venue, its sensitivity to the presence of hemorrhage may prompt unnecessary laparotomy in patients with self-limited lacerations of the liver, spleen,14–17 or mesentery.17 CT scan specifically evaluates all intraperitoneal structures as well as the retroperitoneum, a region inaccessible to DPL. Because the resolution and the speed with which it can be undertaken have vastly improved, CT has become an invaluable adjunct in the management of blunt trauma18,19 and has largely replaced DPL in the stable patient. It is most useful in the identification of injury to solid organs with accompanying IPH and greatly assists nonoperative management of those injuries. The ability of CT to discern hollow viscus and pancreatic pathology has improved but remains inconsistent.20,21 With regard to hollow viscus injury, it is when serial clinical evaluations cannot be performed that gut perforation leads to preventable mortality. This is especially true in the patient with severe closed head injury or high spinal cord injury in whom physical assessment of the abdomen is quite compromised. It is for these express scenarios that some authorities recommend the performance of DPL. The clinician’s concern should be heightened if the US or CT demonstrates minimal amounts of free intraperitoneal fluid without evidence of solid organ damage.22,23 Experience with US in North America is meager in comparison with that in Western Europe (notably Germany) and Asia (notably Japan). In the past, US in the United States had been used exclusively for the detection and serial examination of traumatic pancreatic pseudocysts. Two paradigms have brought US to the forefront. First, this modality has been adopted as the primary triage instrument, in lieu of DPL, for the detection of IPH on the basis of identifying which pouches and gutters are fluid-filled.24–27 Clinical success in this role has been mixed with reported sensitivities for IPH of 65% to 95%.28–34 In addition, to be useful in this role, a competent technician, interpreter, and equipment must be present in real time. It has been demonstrated that emergency clinicians and surgeons can be trained in this technique to a level of competence sufficient for this need.35 In centers that rely upon US, DPL should serve as a reliable study when US equipment is unavailable, the US is technically difficult, or when the results of the US are indeterminate, especially when the patient demonstrates hemo­ dynamic compromise. Second, US can determine injury to solid viscera such as the liver, spleen, kidneys, and pancreas. This requires con­ siderably greater expertise, and in most centers, US has not supplanted CT for this purpose.36 DPL is a readily available procedure that can be conducted rapidly in the safe confines of the emergency department (ED). The ability to undertake CT, in particular, or to a lesser extent, US in a similar manner requires careful consideration of clinical circumstances, equipment location, and the capabilities of available personnel (Table 43–3 and Fig. 43–1).37,38

773

TABLE 43–3  Diagnostic Studies in Blunt Abdominal Trauma

Manifestation

Scenario

Unstable vital signs with strongly suspected abdominal injury Unequivocal peritoneal irritation Pneumoperitoneum

VII 

● 

GASTROINTESTINAL PROCEDURES

TABLE 43–1  Clinical Indications for Laparotomy after Blunt Trauma

Evidence of diaphragmatic injury Significant gastrointestinal bleeding

Pitfall Alternate sources shock

Unreliable Insensitive; may be due to cardiopulmonary source or invasive procedures (diagnostic peritoneal lavage, laparoscopy) Nonspecific Uncommon, unknown accuracy

TABLE 43–2  Principal Indications for Diagnostic Peritoneal Lavage in Abdominal Trauma

774

Determine diaphragmatic violation Rapidly determine presence of IPH Determine presence of HVI

Primary Study

Alternate/  Compensatory

Hemodynamically Unstable

From Marx J, Isenhour J: Abdominal trauma. In Marx JA, Hockberger RS, Walls RM, et al [eds]: Rosen’s Emergency Medicine: Concepts and Clinical Practice, 6th ed. St. Louis, CV Mosby, 2006, p 509.

Indication

Study Purpose

Clinical Scenario Penetrating injury to low chest, upper abdomen Ultrasound unavailable, indeterminate, or negative for freefluid in an unstable patient CT nondiagnostic and SCE unreliable: • Head injury with altered mental status • Alcohol intoxication • Drug intoxication • Spinal cord injury

CT, computed tomography; HVI, hollow viscus injury; IPH, intraperitoneal hemorrhage; SCE, serial clinical evaluations.

Penetrating Trauma The advent of DPL was seminal in the promotion of selective management for penetrating abdominal injury. Here its role is more dominant than for blunt trauma owing to the far greater likelihood of occult injury to hollow viscera and the diaphragm after a penetrating mechanism.39,40 Instruments and missiles may penetrate the abdominal cavity via the anterior abdominal wall, flank, back, or low chest.41 The intraperitoneal space is vulnerable if penetration occurs as high as the fourth intercostal space anteriorly and the sixth or seventh laterally and posteriorly, because the diaphragm may rise to these levels in the expiratory phase of respiration.42 Coincident thoracic penetration has occurred in up to 46% of abdominal injuries.43–45 The likelihood of retroperitoneal injury increases when the entry site is over the flank or back, but the prospect of intraperitoneal pathology remains considerable with cited incidences of up to 43% for the flank and 14% for the back (Table 43–4).46–48 Stab Wounds.  Because only one fourth to one third of patients who sustain stab wounds to the anterior abdomen

General Pelvic fracture

IPH IPH

US US

DPA DPA*

OI†‡ OI

CT CT||

SPEs, DPL SPEs, DPL¶

OI, HVI IPH

CT,|| DPL¶ US, DPL

SPEs** CT††

Hemodynamically Stable General Nonoperative management§ CHI BAI

*Positive peritoneal aspirate mandates laparotomy, positive red blood cell count only, warrants attention to pelvic fracture. † To discover fluid/blood suggesting injury. ‡ US for OI is much less reliable than for IPH. § Institutional capability should be carefully considered. || CT is less reliable for HVI than for solid visceral injury. ¶ Complementary to CT if HVI suspected. ** SPEs are unreliable in the patient with CHI. †† May be more appropriate if helical CT is primary study for BAI or can be rapidly acquired. BAI, blunt aortic injury; CHI, closed-head injury; CT, computed tomography; DPA, diagnostic peritoneal aspiration; DPL, diagnostic peritoneal lavage; HVI, hollow viscus injury; IPH, intraperitoneal hemorrhage; OI, organ injury; SPEs, serial physical examinations; US, ultrasonography. Adapted From Marx J, Isenhour J: Abdominal trauma. In Marx JA, Hockberger RS, Walls RM, et al (eds): Rosen’s Emergency Medicine: Concepts and Clinical Practice, 6th ed. St. Louis, CV Mosby, 2006, p507.

TABLE 43–4  Injury Likelihood by Entry Site Intraperitoneal Anterior abdomen Flank Back Low chest

Retroperitoneal

Diaphragm

++

+

+

+ + +

++ ++ +

+ + ++

From Marx JA: Diagnostic peritoneal lavage. In Ivatury RR, Cayten CG (eds): The Textbook of Penetrating Trauma. Baltimore, Williams & Wilkins, 1996, p 336.

require laparotomy, diagnostic algorithms are used to decrease the rate of unnecessary operation.39,44,49,50 An optimal approach would not sacrifice sensitivity for morbid intraperitoneal injury. A pathway using a combination of clinical mandates, local wound exploration, and DPL is well established (Fig. 43–2).51 These clinical mandates are reasonably accurate predictors of significant intraperitoneal injury (Table 43–5). Thus, the presence of one or more mandates suggests the need for urgent laparotomy and precludes the undertaking of other diagnostic studies. DPL fills three roles in the evaluation of patients with abdominal stab wounds (see Table 43–2): (1) rapid determination of the presence of hemoperitoneum, (2) discovery of intraperitoneal injury requiring operation in stable patients, and (3) the establishment of diaphragmatic violation. As is the

ANTERIOR ABDOMEN STAB WOUND ALGORITHM

43 

BLUNT ABDOMINAL TRAUMA ALGORITHM

● 

Peritoneal procedures

BAT mechanism

Clinical mandate for LAP?

Clinical mandate for LAP?

Yes

Yes

No

No Peritoneal entry? Hemodynamically unstable?

(LWE)* Yes

No ?

Yes

No

IPH? (US, DPA)*

Injury? (CT, DPL, SPEs, LPY)†

Unreliable examination?† Yes‡

Yes

No

Yes

IP injury?

Yes

No Abdominal tenderness?

(CT, DPL, US, SPEs)‡ Yes

No

No

Injury requires LAP?§ Yes LAPAROTOMY

LAPAROTOMY

No OBSERVE

DISCHARGE

Figure 43–2  Anterior abdomen stab wound algorithm. *Plain films, ultrasonography, LPY, and CT can also assess peritoneal entry. † CT, DPL, SPEs, or LPY can be used in singular or complementary fashion depending on the clinical scenario. ‡ Expectant management of injuries is infrequently attempted. CT, computed tomography; D/C, discharge; DPL, diagnostic peritoneal lavage; LAP, laparotomy; LPY, laparoscopy; LWE, local wound exploration; SPE, serial physical examination.  (From Marx J, Isenhour J: Abdominal trauma. In Marx JA, Hockberger RS, Walls RM, et al [eds]: Rosen’s Emergency Medicine: Concepts and Clinical Practice, 6th ed. St. Louis, CV Mosby, 2006, p 503.)

No OBSERVE¶

DISCHARGE

Figure 43–1  Blunt abdominal trauma (BAT) algorithm. *Determined by unequivocal free IP fluid on US or positive peritoneal aspiration on DPA. † Can be unreliable because of closed-head injury, intoxicants, distracting injury, or spinal cord injury. ‡ One or more studies may be indicated. § Need for LAP is based on clinical scenario, diagnostic studies, and institutional resources. ¶ Duration of observation should be 6–24 hr depending on whether diagnostic tests have been performed, the results of the tests, and clinical circumstances including the absence of factors rendering the examination unreliable. CT, computed tomography; D/C, discharge; DPA, diagnostic peritoneal aspiration; DPL, diagnostic peritoneal lavage; IP, intraperitoneal; IPH, intraperitoneal hemorrhage; LAP, laparotomy; SPE, serial physical examination; US, ultrasound.  (From Marx J, Isenhour J: Abdominal trauma. In Marx JA, Hockberger RS, Walls RM, et al [eds]: Rosen’s Emergency Medicine: Concepts and Clinical Practice, 6th ed. St. Louis, CV Mosby, 2006, p 508.)

case in blunt trauma patients, DPL can be invaluable as a rapid triage tool when the source of hemodynamic instability is not known. Pericardial tamponade, intrathoracic hemorrhage, and IPH may be contributory to hemodynamic instability or wholly causal. Again, as for blunt trauma evaluation, US is the only diagnostic modality for IPH that is competitive for this role and carries the added advantage of scanning for intrapericardial and intrathoracic hemorrhage.45 In the determination of injury after stab wounds, DPL carries 90% accuracy.52–54 Serial examinations,55–57 CT, and laparoscopy58–61 are alternative modalities in specific circumstances and centers.62 Diaphragmatic rents created by stab wounds are generally small; thus, at the outset, they do not create apparent clinical or radiologic abnormalities.63,64 However, morbidity due to delayed herniation of bowel is common and substantive.65 Physical examination is notoriously insensitive and DPL is currently the most sensitive means of discerning this injury in

775

GASTROINTESTINAL PROCEDURES

Manifestation Hemodynamic instability Peritoneal signs Evisceration Diaphragmatic injury Gastrointestinal and vaginal hemorrhage Impalement in situ Intraperitoneal air

Premise

Pitfall

Major solid visceral or vascular injury Intraperitoneal injury Additional bowel, other injury Diaphragmatic herniation Proximal gut or uterine injury Vascular impalement Hollow viscus perforation

Thorax, mediastinum Unreliable, especially immediately postinjury No injury in one fourth to one third of stab wound cases Rare clinical, radiographic findings Uncommon, unknown accuracy High operative risk, pregnancy Insensitive; may be caused by intraperitoneal entry only or be due to cardiopulmonary source

Modified from Marx JA: Diagnostic peritoneal lavage. In Ivatury RR, Cayten CG (eds): The Textbook of Penetrating Trauma. Baltimore, Williams & Wilkins, 1996.

VII 

● 

TABLE 43–5  Clinical Indications for Laparotomy after Penetrating Trauma

776

the immediate post-trauma phase.52 Higher-generation CTs such as 64-slice scanners have extraordinary resolution and may be able to detect even small diaphragmatic tears, but data are not yet available. For these small wounds, magnetic resonance imaging (MRI) may be diagnostic, but owing to safety and accessibility concerns, it should be reserved for the nonacute phase of management. Laparoscopy has demonstrated promise in experienced hands.58,59 Gunshot Wounds.  Multiple organ injury is the rule after gunshot wounds, and mortality is significantly greater when compared with that for stab wounds.62 The diagnostic approach is more conservative for gunshot wounds because, in some studies, the likelihood of intraperitoneal injury requiring operative intervention exceeds 90% when the projectile has entered the intraperitoneal cavity (Fig. 43–3).67 If clinical mandates are met (see Table 43–5) or if peritoneal violation has occurred, most centers proceed to laparotomy.51 One series, however, cited intra-abdominal injury in 70% to 80% of cases, supporting the contention that nonoperative management could be applied to a substantial percentage of patients.68 In a separate cohort of 152 patients sustaining solid organ injury from penetrating abdominal trauma (70% gunshot wounds and 30% stab wounds), 27% were successfully managed without laparotomy after selection by a protocol combining clinical examination and CT scanning.69 DPL is reserved for two circumstances: (1) the wound tract is neither obviously superficial nor intraperitoneal, and (2) penetration occurred in the low chest, where diaphragmatic injury is more likely yet the possibility of intraperitoneal injury exists.

Contraindications DPL can be undertaken in virtually any patient irrespective of age or comorbid illness. Adjustment of the technique and site of performance allows relative contraindications to be overcome. Relative contraindications include prior abdominal surgery or infections, obesity, coagulopathy, and second- or third-trimester pregnancy. The sole absolute contraindication is when clinical mandates for urgent laparotomy already exist.

Technique Preliminary Steps Decompress the stomach and bladder to prevent inadvertent injury. Place the patient in the supine position and administer sedation and analgesia as appropriate. Perform DPL according to compliance with standards for body fluid precautions.

ABDOMINAL GUNSHOT WOUND ALGORITHM

Clinical mandate for LAP?

Yes

No

Peritoneal entry?*

Yes†

No‡ ?

Injury? (DPL, CT, LPY, SPEs)§ Yes¶ LAPAROTOMY

No OBSERVE

DISCHARGE

Figure 43–3  Abdominal gunshot wound algorithm. *Can be assessed by missile path, plain films, local wound exploration, ultrasonography, and LAP. † Most centers proceed to LAP if peritoneal entry is suspected. ‡ Patients with documented superficial and low-velocity injuries can be discharged; unknown-depth or high-velocity injuries require further tests or observation. § DPL, CT, LPY, or SPEs can be used in singular or complementary fashion depending on the clinical scenario. ¶ Expectant management of injuries caused by gunshot wounds is rarely attempted. CT, computed tomography; D/C, discharge; DPL, diagnostic peritoneal lavage; LAP, laparotomy; LPY, laparoscopy; SPE, serial physical examination.  (From Marx J, Isenhour J: Abdominal Trauma. In Marx JA, Hockberger RS, Walls RM, et al [eds]: Rosen’s Emergency Medicine: Concepts and Clinical Practice, 6th ed. St. Louis, CV Mosby, 2006, p 506.)

Observe sterile technique throughout the procedure. Before making the skin incisions described later, prepare the site with standard skin antiseptics and drape appropriately. Prophylactic antibiotics are not indicated for routine DPL because local and systemic infections are rare.70

Figure 43–5  Standard equipment for open lavage technique.

Peritoneal procedures

Figure 43–4  Local anesthesia is introduced at the incision or puncture site. The patient is supine with the head of the bed elevated slightly.

● 

DPL Catheter Placement There are two basic methods for DPL: open and closed. The two open techniques are semiopen and fully open, and they typically require an assistant. DPL is clearly within the diagnostic armamentarium of the emergency clinician and surgeon. It may be undertaken by either or both in keeping with clinical policies established at the particular trauma center.

Semiopen Technique.  In the semiopen method, make a skin incision, 4 to 6 cm in length with a No. 11 scalpel. Using Army-Navy retractors, proceed with blunt dissection to expose the rectus fascia (Fig. 43–6A and B). With the infraumbilical incision in the midline, continue blunt dissection until the linea alba is seen. Its crossing bands of crural fibers may be apparent.71 Make a small 2- to 3-mm opening in the linea alba with a No. 15 scalpel blade (see Fig. 43–6C). You may notice a tough, gritty sensation when cutting the linea alba with the scalpel. Place towel clips through this opening to grasp each side of the rectus fascia (see Fig. 43–6D). Ask an assistant to lift the two towel clips and carefully advance the catheter and trocar in a 45° to 60° caudad orientation. Proceed through the peritoneum into the peritoneal cavity (see Fig. 43–6E and F ).72 One method to decrease the likelihood of penetrating underlying viscera is to hold the fingers low on the cathetertrocar instrument such that on entering the abdominal peritoneum, the fingers will prevent deep penetration. Excessive pressure during trocar penetration is a common error. Apply steady one-finger pressure to the handle sufficient to “pop” through the peritoneum. After controlled peritoneal penetration of 0.5 to 1.0 cm in the midline, retract the trocar 1.0 to 2.0 cm within the catheter, and advance the catheter carefully toward the pelvis. Some operators advance the catheter toward the right or left side of the pelvis. Use a slight twisting motion during advancement to minimize visceral or omental injury. The fully open technique extends the semiopen technique by one step. Lengthen the opening in the linea alba, to open the peritoneum, and use direct visualization to advance the catheter placement into the peritoneal cavity. The trocar is unnecessary with the open technique. The two open techniques can be accomplished with a single technician, but it is useful to have an assistant help with retraction and handling of instruments. The fully open method is the more technically demanding and timeconsuming. Reserve this method for clinical circumstances in which neither the closed nor the semiopen technique is deemed safe or has been attempted and failed. Examples of these circumstances include pelvic fracture, pregnancy, prior abdominal surgery, infections, and obesity. Closed Technique.  For the closed techniques, introduce the catheter into the peritoneal space in a blind percutaneous fashion.73 Utilize the simple Seldinger (guidewire) method, in which a small-gauge guide needle is inserted into the peritoneal cavity in the midline just inferior to the umbilicus (Fig. 43–7A). Pass a flexible wire through the needle (see Fig. 43-7B), and remove the needle but not the wire. Advance a soft catheter over the wire and into the peritoneal cavity. Make a small stab with a No. 11 scalpel at the entry site of the wire to allow easier passage of the catheter through the abdominal wall (see Fig. 43–7C and D). Rotate the catheter while pushing it over the guidewire to facilitate entry into the peritoneal cavity. Place the catheter into the right or left pelvic gutter. Always control the guidewire to avert intra-abdominal migration of the wire. Withdraw the wire and aspirate for blood using a 10-cc syringe. Follow this with peritoneal lavage when necessary. Proponents of the guidewire technique promote its ease and rapidity.74–78 Those who prefer the semiopen method argue that the time to peritoneal aspiration, the more critical interval, is minimally different and that this method may have fewer complications and thus be more accurate than the guidewire technique.79–82 Note that for both

43 

Infiltrate the area for incision and dissection with a local anesthetic such as 1% lidocaine with epinephrine (Fig. 43–4). Delay the incision for more than 30 seconds after local anesthetic infiltration to permit local vasospasm, which minimizes wound bleeding during the procedure. Standard equipment for an open peritoneal lavage catheter placement is shown in Figure 43–5.

777

GASTROINTESTINAL PROCEDURES

Sterile drapes

Subcutaneous fat

Army-Navy retractor

VII 

● 

4-6 cm incision

Rectus fascia (note crurap fibers)

A

Towel clamp lifts peritoneum

B

Traction

Incision

778

2-3 mm incision in fascia

C

D

Traction

E

F

Figure 43–6  A, After bladder decompression (generally by Foley catheter placement), make a 4- to 6-cm long vertical infraumbilical incision with a No. 11 scalpel. B, Carry blunt dissection using Army-Navy retractors down to the rectus fascia. Crossing bands of crural fibers may be seen. C, Make a 2- to 3-mm incision through the rectus fascia in the midline (linea alba) with a No. 15 scalpel. D, Grasp each side of the rectus fascia with towel clips and lift it prior to insertion of the trocar and DPL catheter. E, Pass the trocar with DPL catheter at a 45° caudad angle into the fascial opening and through the peritoneum. In the fully open method, extend the incision in the rectus fascia until the peritoneum is directly visualized. Incise the peritoneum and place the catheter alone into the peritoneal cavity. F, As soon as the peritoneum has been entered, gently advance the catheter into the peritoneal cavity while withdrawing the trocar. It is often helpful to advance the catheter with a slight twisting motion and to direct it toward either the right or the left pelvic gutter.

43 

Guide wire

● 

Peritoneal procedures

A

B Guide wire

Small puncture

779

C

D

Figure 43–7  A, For the closed DPL method using a guidewire (Seldinger technique), insert the needle into the peritoneal cavity in the midline just below the umbilicus and aim it slightly caudad. B, Pass the flexible guidewire through the needle and into the peritoneal cavity. Ideally, direct the wire toward the right or left pelvic gutter. Withdraw the needle while stabilizing the wire with the free hand at all times. C, Make a stab incision with a No. 11 scalpel immediately below the wire to permit easier passage of the DPL catheter. D, Direct the DPL catheter over the wire and into the peritoneal cavity using a slight twisting motion. Stabilize the wire at all times and remove it after catheter placement, directing it toward the right or left pelvic gutter when advancing.

semiopen or closed approaches, the time to aspiration should be no more than 2 to 5 minutes. Site The optimum location for DPL is at the infraumbilical ring, at the inferior border of the umbilicus (Table 43–6). Here, between the rectus abdominis muscles, there is adherence of the peritoneum and relative lack of vascularity and preperitoneal fat.71 Closed DPL should always be conducted here. In the event of second- or third-trimester pregnancy, a suprauterine approach is used. If there is midline scarring, a fully open technique at the lateral border of the rectus abdominis in the left lower quadrant may be necessary. The left side is preferred to avoid later confusion about whether an appendectomy has been performed. It is interesting to note that Moore and associates83 found no increase in complications or misclassified lavages when the closed technique was used in a small series of patients with prior abdominal surgery. In the

TABLE 43–6  Preferred Site of Diagnostic Peritoneal Lavage Clinical Circumstance Standard adult Standard pediatric Second- and third-trimester pregnancy Midline scarring Pelvic fracture Penetrating trauma

Site

Method

Infraumbilical midline Infraumbilical midline Suprauterine

C or SO C or SO FO

Left lower quadrant Supraumbilical Infraumbilical midline*

FO FO C or SO

*The stab wound or gunshot wound site should be avoided. C, Closed; FO, fully open; SO, semiopen.

GASTROINTESTINAL PROCEDURES ● 

VII 

presence of a pelvic fracture, use a fully open supraumbilical approach. This greatly decreases the likelihood of passing the catheter through retroperitoneal hematoma that has dissected from the fracture anteriorly and across the abdominal wall.84 In penetrating trauma, do not perform the DPL through the stab or missile entry site. This approach can contaminate the intraperitoneal cavity and potentially exacerbate abdominal wall bleeding, leading to a false-positive result. Aspiration and Lavage Once the catheter has been placed successfully into the peritoneal cavity, attach the right-angle adapter, extension tubing, and a non–Luer-Lok syringe and attempt aspiration (Fig. 43–8). If 10 mL of blood are aspirated, the test is positive, and the procedure is terminated. In penetrating trauma, the acquisition of lesser amounts may be meaningful because of the tendency for the diaphragm and bowel to hemorrhage minimally when injured. However, there are no established rules in this regard. If little to no blood is aspirated, lavage the peritoneal cavity with either normal saline or lactated Ringer’s solution (Fig. 43–9). Apply a blood pressure cuff or blood infusion pump to the plastic intravenous (IV) bag to speed the influx

(i.e., decrease lavage time) if necessary. Large-bore infusion tubing (e.g., urologic irrigation tubing sets, such as the Abbott No. 6543 cystoscopy/irrigation set) also shortens fluid influx time. Infuse 1 L of fluid in adults or 15 mL/kg in children. When possible, roll or shift the patient from side to side after the infusion to increase mixing. Place the IV bag or bottle on the floor (or below abdominal level), and allow the fluid to return by gravity. If the fluid does not return, there may be several reasons. Some IV tubing contains a one-way valve. If tubing with a valve was used in error, replace with valveless tubing and reattach it to the IV bag. Another reason for poor return is inadequate suction. To correct this problem, insert a needle into the second opening at the bottom of the IV bag or into the head of the IV bottle for aspiration of 10 mL of air. Alternatively, the catheter may be adherent to the peritoneum. If so, try to relieve some of the pressure in the IV bottle by gently wiggling and twisting the catheter or applying abdominal pressure to aid the return of flow. The return of 700 mL or more in the adult is generally accepted as adequate for interpretation of findings. However, as little as 10% to 20% of the infusate may give a representative sample for both gross and microscopic determinations. Send 10 mL of fluid from the return to the laboratory for cell count analysis and send another 10 mL for enzyme analysis (see “Interpretation,” later in this chapter). Some operators prefer to leave the catheter in place until the returned fluid is analyzed so they may relavage when the initial results are borderline or an occult bowel perforation is suspected.

Complications

780

Local and Systemic Local wound complications, including infection, hematoma, and dehiscence, have occurred in only 0.3% of patients in two large series.52,85 Dehiscence with evisceration is an even more rare condition.86 Systemic infection has been described rarely (Table 43–7). Figure 43–8  After attachment of the right-angle connector and extension tubing, attempt aspiration of the peritoneum.

Figure 43–9  If the aspiration is negative, instill normal saline or Ringer’s lactate solution through the catheter. The intravenous tubing should have no valves in place. After the infusion, place the fluid bag on the floor to allow it to fill with peritoneal effluent via gravity.

Intraperitoneal Iatrogenic intraperitoneal injury can be inflicted by the trocar, wire, and rarely, the catheter. Virtually any structure in the peritoneal cavity can be breached, including the small and large bowel, the bladder, and major vessels. Typically, if the needle is the culprit, and even if the trocar is responsible, injury to these structures is minimal and self-limiting, and observation of the patient is sufficient. Technical Failure Inability to recover peritoneal aspirate or lavage fluid can result in a false-negative interpretation. This can occur in several circumstances. It follows unwitting placement of the catheter into the preperitoneal space, which is less likely to occur with either open technique. Compartmentalization of fluid by adhesions or obstructing omentum can impede egress of fluid. When a fully open supraumbilical or suprauterine technique is used, the catheter may be too short to access the depths of the intraperitoneal cavity. Finally, large diaphragmatic tears typical of blunt pathophysiology allow flow of lavage fluid from the intraperitoneal to the thoracic cavity. Saunders and coworkers87 compared percutaneous DPL versus the open technique in a prospective, randomized trial. Fluid obtained by the two techniques had similar test performance for intra-abdominal pathology. The open technique

Local and Systemic Hematoma incision site Dehiscence incision site Local wound infection Systemic infection

Local wound care Local wound care As indicated As indicated

Intraperitoneal Injury Bowel Bladder Vascular

Observe, usually self-limited Observe, usually self-limited Observe, usually self-limited

Technical Failure

Positive Blunt trauma Stab wound   Anterior abdomen   Flank   Back   Low chest Gunshot wound

Indeterminate

100,000*

20,000–100,000

100,000 100,000 100,000 5000–10,000 5000–10,000

20,000–100,000 20,000–100,000 20,000–100,000 1000–5000 1000–5000

Peritoneal procedures

Comments

● 

Category

TABLE 43–8  Diagnostic Peritoneal Lavage Red Blood Cell Criteria (per mm3)

43 

TABLE 43–7  Diagnostic Peritoneal Lavage Complications

*In a hemodynamically stable patient with pelvic fracture and positive or equivocal red blood cell count, computed tomography should be obtained to corroborate or refute intraperitoneal injury. From Marx J, Isenhour J: Abdominal Trauma. In Marx JA, Hockberger RS, Walls RM, et al (eds): Rosen’s Emergency Medicine: Concepts and Clinical Practice, 6th ed. St. Louis, CV Mosby, 2006, p 500.

Inability to Recover Fluid* Preperitoneal catheter placement Compartmentalization of fluid Obstructed catheter Diaphragm injury “Short” catheter (supraumbilical or suprauterine approach)

Repeat DPL US, CT Gentle catheter manipulation Reverse Trendelenburg; consider US, CT Trendelenburg

Intraperitoneal Hemorrhage† Iatrogenic injury Stab wound abdominal wall bleed Pelvic fracture (RBC count)

As indicated by clinical markers As indicated by clinical markers Complementary CT

*May lead to false-negative DPL † May lead to false-positive DPL CT, computed tomography; DPL, diagnostic peritoneal lavage; RBC, red blood cell; US, ultrasound.

took, on average, more than 4 minutes longer, but the percutaneous approach had an 11.2% (vs. 3.8% with the open approach) technical failure rate. False-positive findings can occur in two ways. First, iatrogenic misadventure can be responsible. Second, in penetrating trauma, particularly stab wounds, bleeding from the abdominal wall injury site into the peritoneal cavity can lead to positive findings when no injury to intraperitoneal structures has occurred.53

Interpretation Gross Blood The recovery of 10 mL or more blood via aspiration is considered a positive finding. Lesser volume aspirates are generally discarded and are not factored into lavage analysis. Grossly bloody aspirates are typically indicative of solid visceral or vascular injury, with a positive predictive value greater than 90%.88,89 Aspiration of blood is responsible for approximately 80% of true-positive DPL findings in blunt trauma and for 50% of those after stab wounds.52 A positive aspiration in the blunt trauma patient who is hemodynamically stable or has been resuscitated to apparent stability need not mandate urgent operation. Unnecessary laparotomy will occur if there has been minimal and selflimited damage to the liver, spleen, bowel serosa, or mesen-

tery.90 In this situation, CT and clinical indicators should be used in concert with the DPL findings. Red Blood Cell Count The recommended red blood cell (RBC) threshold varies according to mechanism and, in the case of stab wounds, the external site of injury (Table 43–8). The optimum criterion will deliver excellent sensitivity, a high positive-predictive value, and a minimum incidence of unnecessary laparotomy. Negative laparotomy incurs a prolongation of hospitalization and increases the cost of care, in addition to creating the potential for procedural complications.91,92 RBC counts greater than 105/mm3 (i.e., >105/µL) are generally considered positive with a blunt mechanism or after stab wounds to the anterior abdomen, flank, or back. Counts of 20,000 to 100,000/mm3 should be considered indeterminate.52,54,93,94 For stab wounds to the low chest, where the diaphragm is at increased risk of injury, the RBC criterion should be lowered to 5000/mm3 to maximize sensitivity for isolated injury to this structure.44,52,95,96 With gunshot wounds to the abdomen or low chest, the same RBC criterion of 5000/mm3 is applied. This is intended to increase the sensitivity of the test, because intraperitoneal entry by a missile carries a likelihood of intraperitoneal injury of 90% or greater.44,68,97,98 An uncomplicated DPL should not create more than several hundred to several thousand RBCs in the peritoneal lavage fluid. The incidence of false-positive RBC interpretation in the setting of pelvic fracture is considerable. However, aspiration of free blood in the critical pelvic fracture patient predicts active IPH in greater than 80% of cases.99 A positive RBC count should generally prompt corroboration or refutation of intraperitoneal injury by CT. In this fashion, needed pelvic angiography and embolization will not be delayed unnecessarily should active intraperitoneal bleeding not be found (Fig. 43–10). White Blood Cell Count An inflammatory peritoneal response occurs to a multitude of stimuli, including stool, blood, and enzymes.100 The white blood cell (WBC) count in lavage effluent was formerly touted to predict small bowel injury but has since been proved unreliable.101 It is insensitive in the immediate postinjury period, because 3 to 5 hours is necessary before the test becomes positive (Table 43–9).102,103 Moreover, a positive finding is

781

GASTROINTESTINAL PROCEDURES

PELVIC FRACTURE AND BLUNT ABDOMINAL TRAUMA ALGORITHM

TABLE 43–9  Diagnostic Peritoneal Lavage Non–Red Blood Cell Criteria

Pelvic Fx* LAM (IU/L) LAP (IU/L) WBCs (per mm3)

Positive

Indeterminate

≥20 ≥3 >500

10–19 NA 250–500

LAM, lavage amylase; LAP, lavage alkaline phosphatase; NA, not applicable; WBCs, white blood cells. From Marx JA: Diagnostic peritoneal lavage. In Ivatury RR, Cayten CG (eds): The Textbook of Penetrating Trauma. Baltimore, Williams & Wilkins, 1996, p 337.

VII 

● 

Hemodynamically unstable?

Yes

No

mal intestine. Amylase is contained in the latter only. Perforation of small bowel allows access of these two markers into the peritoneal cavity, where they can be recovered by peritoneal lavage.105–107 Although levels of the two markers usually rise in tandem, lavage amylase has been shown to be a more accurate marker than lavage alkaline phosphatase (see Table 43–9). In contradistinction to the WBC count, these tests will be positive in the immediate postinjury period. However, they may not be economical if used on a mandatory rather than a selective basis. Neither is helpful in discerning the presence of pancreatic pathology.

IPH? (US, DPA)† Yes

No Angiography and pelvic Fx stabilization

IP injury?

Miscellaneous Routine bile staining, Gram stain, and microscopy to identify vegetable fibers are rarely productive and are of untested accuracy. Deck and Porter108 reported that finding urine in the lavage fluid, as evidenced by a straw color, and creatinine in the peritoneal fluid should suggest an intraperitoneal bladder or collecting system injury.

(CT, DPL)‡ Yes

782

No

Injury requires LAP?§

Yes LAPAROTOMY

Conclusion No OBSERVE

DISCHARGE¶

Figure 43–10  Pelvic Fx and blunt abdominal trauma algorithm. *Certain pelvic Fx are more likely to cause pelvic vascular disruption and subsequent retroperitoneal hemorrhage. † Determined by unequivocal free intraperitoneal fluid on US or positive peritoneal aspiration on DPA. ‡ One or more studies may be indicated. SPEs are generally considered unreliable owing to the presence of pelvic Fx. § Need for LAP is based on clinical scenario, diagnostic studies, and institutional resources. ¶D/C from the perspective of need for further consideration for LAP. CT, computed tomography; D/C, discharge; DPA, diagnostic peritoneal aspiration; DPL, diagnostic peritoneal lavage; Fx, fracture; IP, intraperitoneal; IPH, intraperitoneal hemorrhage; LAP, laparotomy; SPE, serial physical examination; US, ultrasonogram.  (From Marx J, Isenhour J: Abdominal Trauma. In Marx JA, Hockberger RS, Walls RM, et al. [eds]: Rosen’s Emergency Medicine: Concepts and Clinical Practice, 6th ed. St. Louis, CV Mosby, 2006, p 510.)

likely to be falsely so.102,104 Therefore, the WBC level in and of itself should not determine the need for laparotomy. Enzymes Alkaline phosphatase is contained in intramural small bowel as well as in hepatobiliary secretions released into the proxi-

In the current era of readily available advanced imaging techniques such as CT and US, the DPL maintains a diminished, but important, role in the evaluation of the injured patient. It can identify life-threatening IPH in the unstable patient when US is unavailable, indeterminate, or negative for free-fluid. In addition, DPL is a marker for hollow viscus injury when the CT is nondiagnostic in patients in whom serial clinical evaluations are impractical or unreliable, as in the presence of severe traumatic brain and spinal cord injuries. Finally, DPL remains more accurate than CT in identifying occult diaphragmatic injuries and is useful in high-risk patients without other indications for surgical abdominal exploration. It should be used in common sense fashion. The noted laboratory parameters are guidelines and should not be embraced to the exclusion of pertinent clinical features. Optimal strategies depend largely on the capability of an institution’s resources and personnel in each clinical scenario.

PARACENTESIS Ascites connotes an abnormal accumulation of fluid within the peritoneal cavity. The word derives from the Greek askos, which means “bag” or “sack.” It is a symptom with important diagnostic, therapeutic, and prognostic implications. Therapeutic abdominal paracentesis is one of the oldest medical procedures, dating to approximately 20 bc. Paracentesis was first described in modern medical literature by Saloman109 at the beginning of the 20th century, and it became

Differential Diagnosis The etiologies of ascites can be categorized in several ways. On a structural basis, these are divided into diseases of the peritoneum and diseases not involving the peritoneum. The former group includes infections, neoplasms, collagen vascular diseases, and idiopathic causes. The latter includes cirrhosis, congestive heart failure, nephrotic syndrome, protein-losing enteropathy, malnutrition, myxedema, pancreatic disease, ovarian disease, chylous effusion, Budd-Chiari syndrome, and hepatic venous occlusive disease.119,126 Pathophysiologic categories are found in Table 43–10. In the United States, parenchymal liver pathology is overwhelmingly the most likely cause. Within this group, alcoholic liver disease is responsible for approximately 80% of cases (Table 43–11).127,128 Finally, ascites can be classified on the basis of a serum-ascites albumin gradient, that is, the difference between albumin values obtained simultaneously from serum and ascites samples (Table 43–12).129

Indications and Contraindications Therapeutic paracentesis is often undertaken in the ED setting to relieve the cardiorespiratory and gastrointestinal manifestations of tense ascites.130–132 Diagnostic paracentesis is indicated in any patient whose ascites is of new onset or to disclose the presence of infection in patients with known or suspected ascites, particularly in the context of alcohol-related

I. Elevated hydrostatic pressure A. Cirrhosis B. Congestive heart failure C. Constrictive pericarditis D. Inferior vena cava obstruction E. Hepatic vein obstruction (Budd-Chiari syndrome) II. Decreased osmotic pressure A. Nephrotic syndrome B. Protein-losing enteropathy C. Malnutrition D. Cirrhosis or hepatic insufficiency III. Fluid production exceeding resorptive capacity A. Infections 1.  Bacterial 2.  Tuberculosis 3.  Parasitic B. Neoplasms

Peritoneal procedures

Determination of Ascites Small amounts of ascites may be asymptomatic. Larger collections typically cause a sense of abdominal fullness, anorexia, early satiety, and perhaps nausea and abdominal pain. Considerable accumulations create symptoms of respiratory distress by virtue of restricting lung capacity.120 The most predictive history and physical findings for excluding the diagnosis of ascites are the absence of ankle swelling and increased abdominal girth and the inability to demonstrate bulging flanks, flank dullness, or shifting dullness.121,122 Positive predictors for the diagnosis are a positive fluid wave, shifting dullness, or peripheral edema.122,123 Patients who lack obvious clinical markers may benefit from the performance of US, which can discern the presence of as little as 100 mL fluid.124 Endoscopic-guided US may discover as little as 10 mL. It is more sensitive than CT in this respect and can assist in the identification of malignancy.125 In addition, it is a useful adjunct to determine the location of fluid that may be compartmentalized by preexistent infection or surgical adhesions.

● 

Clinical Features

TABLE 43–10  Pathophysiologic Classification of Ascites

43 

a valued decompressive therapy. With the advent of diuretics in the early 1950s, paracentesis fell out of favor as a treatment option. Controlled clinical trials in the late 1980s up to the present have restored its reputation by demonstrating the safety and efficacy of large-volume paracentesis in adults and children.110–116 Because this mode is invasive and consumes clinician hours, it is generally reserved for the treatment of patients with chronic ascites who have tense ascites or whose condition is refractory to diuretic therapy.113,117 However, paracentesis remains an important diagnostic agent for patients with new-onset ascites or to determine the presence of worrisome conditions, notably infection, in those with preexistent ascites.118,119

From Runyon BA: Diseases of the peritoneum. In Wyngaarden JB, Smith LH (eds): Cecil Textbook of Medicine, 18th ed. Philadelphia, WB Saunders, 1988, pp 790–793.

TABLE 43–11  Causes of Ascites* Cause Parenchymal liver disease “Mixed” Malignancy Heart failure Tuberculosis Pancreatic Nephrogenous (“dialysis ascites”) Chlamydia Nephrotic Surgical peritonitis in the absence of liver disease

Patients (%) 80 5 10 5 2 1 1000 (70%); usually >70% lymphocytes

Pyogenic peritonitis

Turbid or purulent

If purulent, >1.016

If purulent, >2.5

Unusual

Congestive heart failure

Straw-colored

Variable, 40 yr, first episode of dislocation, traumatic mechanism of injury defined as fall greater than one flight of stairs, a fight or assault, or a motor vehicle crash) predicted clinically important fractures with a sensitivity of 97.7%.18 This study has not yet been prospectively validated. Anterior dislocations are not subtle on the routine anteroposterior (AP) radiograph, and this view detects the most important fracture to identify, that of the humeral neck. An adequate AP view, when combined with the typical clinical examination, allows for successful management of most ante-

E

875

MUSCULOSKELETAL PROCEDURES ● 

VIII 

Sp

Sp

C

C

G

HH

HH S

S

Hu

A

Pre-reduction

Hu

B

A

Post-reduction

Scapular spine Humeral head

B

C 876

Figure 49–8  A, Scapular Y view demonstrates an anterior dislocation (note that the humeral head is displaced inferior and medial). B, Scapular Y view after dislocation is reduced. Note that the humeral head bisects the two intersecting limbs of the scapular Y. C, Scapular Y view of a posterior dislocation with the humeral head laterally displaced. A posterior dislocation may be difficult to appreciate on an AP view because it is not inferiorly displaced and may appear to be in the glenoid fossa. C, coracoid; G, glenoid; HH, humeral head; SP, scapular spine. (A–C, From Thomsen T, Setnik G [eds]: Procedures Consult—Emergency Medicine Module. Copyright 2008 Elsevier Inc. All rights reserved.)

countertraction. This technique is problematic and is not recommended by some authors.3,11 Likewise, the Kocher method, which involves forceful leverage of the humerus, has an increased rate of complications and is generally discouraged in favor of other techniques.10,11 This section discusses several methods of reduction that are well studied, proven to be safe, and easy to master. Regardless of the reduction technique used, gradual, gentle application of the technique is essential. Although all of the techniques discussed are generally acceptable and many authors state that their techniques are quite painless,1–5 few studies have quantified the actual pain reported by patients.20 As noted previously, intra-articular lidocaine also may be used to reduce the pain of reduction (Fig. 49–9). In studies by Matthews and Roberts21 and Kosnick and colleagues,22 the use of intraarticular lidocaine was found to offer significant pain relief during reduction of anterior shoulder dislocations, making it a useful alternative to procedural sedation and analgesia. When using intra-articular lidocaine, any blood should be aspirated from the glenohumeral joint before injecting anesthetic. Note that 10 to 20 mL of 1% lidocaine has been used with the intra-articular technique, and it may take as long as

Figure 49–9  Intra-articular injection for the reduction of an acute anterior shoulder dislocation can be very effective.  A, After aspirating blood from the joint, 10–20 mL of 1% plain lidocaine is slowly injected through the lateral sulcus, aiming slightly caudad. B, Anterior view. Allow 15–20 min for the lidocaine to take effect. (A and B, From Matthews DE, Roberts T: Intra-articular lidocaine versus intravenous analgesic for reduction of acute anterior shoulder dislocations. Am J Sports Med 23:54, 1995. Reproduced by permission.)

15 to 20 minutes for adequate analgesia. Recently, Blaivas and Lyon23 reported the ED use of ultrasound-guided interscalene blocks for analgesia before reduction of shoulder dislocations. It is important to note that neither local or regional anesthesia produce muscle relaxation, but these may obviate the need for IV access and prolonged observation. Operator judgment is an important part of the decision as to whether reduction should be attempted without premedication. The advantages of such an approach include the avoidance of potential complications from drug therapy, reduced staff requirements, and theoretically, a more rapid patient disposition. Certainly, the patient who is markedly intoxicated may require little, if any, supplemental sedative therapy. However, all patients who are reluctant or too anxious to cooperate with an attempt at reduction without medication and those with a high degree of muscle spasm should receive premedication. Generally, only one attempt is made; if unsuccessful, further reduction attempts are made after the administration of IV sedation. When in doubt, it is best to use pharmacologic adjuncts (see Chapter 33, Systemic Analgesia and Sedation for Procedures). Several factors will help decide which technique is best in each situation. One factor is whether the patient will tolerate a reduction attempt without sedation, because attempts without sedation should not use forceful techniques such as traction-countertraction. The clinician’s comfort level with a given technique is always a factor, because the greatest success rates will likely result from techniques with which the clinician is most familiar. The time and resources available to the

2

A

B Figure 49–10  A, Stimson technique. This technique is often tried first, because it is the least traumatic if the patient can relax the shoulder muscles. 1, The patient is lying prone on the edge of the table. One must be careful that the sedated or intoxicated patient does not fall off the table. Belts or sheets can be used to secure the patient to the stretcher. 2, 5-kg weights are attached to the arm, and the patient maintains this position for 20–30 min, if necessary. 3, Occasionally, gentle external and internal rotation of the shoulder with manual traction aids reduction. B, Physician applying manual traction and rotation to aid reduction. (A, From DePalma AF: Management of Fractures and Dislocations: An Atlas. Philadelphia, WB Saunders, 1970, p 618. Reproduced by permission. B, from Thomsen T, Setnik G [eds]: Procedures Consult—Emergency Medicine Module. Copyright 2008 Elsevier Inc. All rights reserved.)

Management of common dislocations

3

● 

1

but it may take 20 to 30 minutes. Reduction may be facilitated by gentle external rotation of the extended arm. Variations of this method include the recommendation for flexion of the elbow to further relax the biceps tendon and the application of manual traction instead of weights.25,26 Rollinson27 allowed the arm to hang under its own weight after a supraclavicular block and reported a 91% success rate with usually no more than a gentle pull on the arm after 20 minutes in this position. Each variation of the Stimson method can be used in combination with the scapular manipulation technique described later. Indeed, a success rate of 96% has been reported using the combined prone position, hanging weights, IV drug therapy, and scapular manipulation.24 Disadvantages of the Stimson method include the time required and the danger of patients slipping off the elevated bed. A “seatbelt” strap or bedsheet may be placed around the patient and stretcher to avoid patient movement off the stretcher. In addition, a bed that elevates to a suitable height for the patient’s arm length, a convenient method to hang the weights, the weights themselves, and adequate staff to monitor the patient are often difficult to locate and organize in a busy ED. Scapular Manipulation Technique.  This method is popular owing to its ease of performance, reported safety, and acceptability to patients. To date, no complications from this technique have been reported in the literature.20,24,28 Shoulder reduction using this method focuses on repositioning the glenoid fossa rather than the humeral head, and it requires less force than many other methods.21 The success rate is high, generally greater than 90% in experienced hands.24,28 The initial maneuver for scapular manipulation is traction on the arm as it is held in 90° of forward flexion. This may be performed with the patient prone and the arm hanging down, as described in the Stimson method, with or without flexion of the elbow to 90° (Fig. 49–11A). Alternatively, this traction may be applied by the operator placing an outstretched arm over the seated patient’s midclavicle while pulling the injured extremity with the other arm (see Fig. 49–11B and C). Regardless of the means of arm traction, slight external rotation of the humerus may facilitate reduction by releasing the superior glenohumeral ligament and presenting a favorable profile of the humeral head to the glenoid fossa.29 The prone patient position is recommended for those not familiar with the technique, because it facilitates identification of the scapula for manipulation (medial rotation of the tip). Nonetheless, the technique can be performed with the patient supine, given that the patient’s shoulder is flexed to 90° and the scapula is exposed during gentle upward traction on the humerus.30 Although seated scapular manipulation offers the advantage of not requiring the patient to go through the awkward and potentially uncomfortable assumption of the prone position, it is a technically more difficult variation of scapular manipulation, especially if sedation is going to be necessary. When placing the patient in the prone position, it is important to place the injured shoulder over the edge of the bed to allow the arm to hang perpendicularly for the application of traction.28 After application of traction, the scapula is then manipulated to complete the reduction. Anderson and associates28 recommended manipulation of the scapula after the patient’s arm is relaxed; however, success is possible with no delay in the performance of this second step.20 Manipulation of the scapula is carried out by stabilizing the superior aspect of the

49 

clinician must be considered, because methods such as the Stimson maneuver require greater time and the availability of weights and straps. In addition, certain reduction techniques can be performed without assistance, whereas others require an additional person to apply countertraction or to help with manipulation of the scapula or humeral head. Ideally, the emergency clinician should become familiar with a number of different techniques for reducing anterior dislocations of the shoulder, because no single method has a 100% success rate nor is any technique ideal in every situation. Stimson Maneuver.  The Stimson maneuver (Fig. 49–10) is a classic technique that offers the advantage of not requiring an assistant. The patient is placed prone on an elevated stretcher and about 2.5 to 5.0 kg (5–10 lb) of weight is suspended from the wrist.10,11 The weights can be strapped to the wrist or a commercially available Velcro wrist splint can be placed and the weights hung from this with a hook.24 The slow, steady traction of this method often permits reduction,

877

MUSCULOSKELETAL PROCEDURES ● 

VIII 

B A

2

878

C

1

Figure 49–11  Scapular manipulation technique.  A, The inferior tip of the scapula is pushed medially and dorsally with the thumbs while the superior aspect of the scapula is stabilized with the fingers of the superior hand. Weights may be attached to the hand to apply hanging traction. B and C, With the patient seated, the operator applies traction with one hand and countertraction with the other, while an assistant rotates the scapula in the same manner. (B and C, from Thomsen T, Setnik G [eds]: Procedures Consult—Emergency Medicine Module. Copyright 2008 Elsevier Inc. All rights reserved.)

scapula with one hand and pushing the inferior tip of the scapula medially toward the spine (see Fig. 49–11A and B). The thumb of the hand stabilizing the superior aspect of the scapula can be placed along the lateral border of the scapula and used to assist the pressure applied by the thumb of the other hand. A small degree of dorsal displacement of the scapular tip is recommended as it is being pushed as far as possible in the medial direction.28 When the patient is properly positioned, with the affected arm hanging perpendicularly, the lateral border of the scapula may be difficult to find in larger subjects. This border is generally located quite laterally with the patient in this position, and it must be properly located before any reduction attempt. The reduction itself is occasionally so subtle that it may be missed by both the patient and the operator. A minor shift of the arm may be the only clue to the successful reduction. Careful palpation of the subclavicular area in order to locate the position of the humeral head before repositioning the patient may be used to determine the success of the reduction. A recently described variation of the seated scapular manipulation technique is the “Best of Both” (BOB) maneuver.31 In the BOB maneuver, the patient is positioned seated sideways on the stretcher with the unaffected shoulder and hip against the fully elevated head of the stretcher. The operator stands on the foot end of the gurney at the patient’s affected side and uses one hand to apply downward force on the patient’s proximal forearm. The operator’s other hand is used to grasp the patient’s hand in order to gently internally

or externally rotate the arm as needed. Once downward force is being applied, an assistant performs the scapular manipulation maneuver as described earlier.31 External Rotation Method.  This method offers the advantage of requiring only one person and no special equipment. The technique requires no strength or endurance on the part of the operator and is well tolerated by patients.3 The actual pain experienced by patients with this technique has not been quantified, but Plummer and Clinton3 stated that it can be performed with “little, if any sedation.” In this technique, the basic maneuver is slow, gentle external rotation of the fully adducted arm. In 1957, Parvin32 described a selfreduction external rotation technique in which the patient sits on a swivel-top chair and grasps a fixed post positioned waist high and slowly turns the body to enact external rotation. Parvin32 reported that the reduction usually takes place at 70° to 110° of external rotation. Since Parvin’s initial study, this method has been described with the patient supine and the affected arm adducted tightly to the side of the patient.1,33 The elbow is flexed to 90° and held in the adducted position with the operator’s hand closest to the patient. The other hand holds the patient’s wrist and guides the arm into slow and gentle external rotation (Fig. 49–12). The procedure may require several minutes, because each time the patient experiences pain, the procedure is momentarily halted. Although the report of Mirick and coworkers1 mentioned using the forearm as “a lever,” a later description clearly recommends allowing the forearm to “fall” under its own weight.3 No additional

49  ● 

Management of common dislocations

Figure 49–12  External rotation method.  No traction is applied and a slow, gentle approach is essential. First, the arm is adducted to the patient’s side. In one hand, the elbow is held flexed at 90° while the other hand grasps the wrist. Slowly and gently, the forearm is used as a lever to rotate the arm externally. Usually by the time the forearm has reached the coronal plane, the shoulder will have been reduced. (From Mirick MJ, Clinton JE, Ruiz E: External rotation method of shoulder dislocation reduction. JACEP 8:528, 1979, and Thomsen T, Setnik G [eds]: Procedures Consult—Emergency Medicine Module. Copyright 2008 Elsevier Inc. All rights reserved.)

Figure 49–13  Milch technique. Slow, steady abduction with overhead traction, external rotation (not shown), and direct pressure over the humeral head are the steps of the Milch technique. The procedure may take 3–4 min to complete, and the operator should avoid sudden, jerky manipulations. It may help to ask the patient to make a motion as if he or she is reaching up and picking an apple from a tree.

force should be applied to the forearm and no traction is exerted on the arm. The end point of the reduction may be difficult to identify, because reduction is frequently very subtle. It is therefore recommended to continue the external rotation until the forearm is near the coronal plane (lying on the bed, perpendicular to the body), a process that usually takes 5 to 10 minutes.3 If the patient notes persistent dislocation with full external rotation, steady gentle traction at the elbow may be added at this time. Reduction may occasionally be noted when the arm is rotated back internally.33 The success rate of this technique in three series performed by emergency clinicians was around 80%.1,33,34 Milch Technique.  Proponents of this method praise its gentle nature, high success rate, lack of complications, and tolerance by patients.2,5 It can be described as “reaching up to pull an apple from a tree.” The basic steps of this technique are abduction, external rotation, and gentle traction of the affected arm. Finally, if needed, the humeral head is pushed into the glenoid fossa with the thumb or fingers (Fig. 49–13). Milch,35 in describing this technique, wrote that the fully abducted arm was in a natural position in which there was little tension on the muscles of the shoulder girdle. He postulated that this was related to our ancestral “arboreal brachiation” (swinging from trees). The primary step in this technique is to have the affected arm abducted to an overhead position. Russell and colleagues29 had their patients raise the arm and put the hand behind the head as a first step. Although this seems odd, patients can usually do this quite readily with little assistance and be quite comfortable in this position. Alternatively, the operator may abduct the arm by grasping the patient’s arm at the elbow or the wrist. Lacey and Crawford36 found that the prone position, with the patient’s shoulder close to the end of the bed, facilitated this step. Once the arm is fully abducted, gentle longitudinal traction is applied with slight external rotation. If reduction does not occur quickly, the humeral head can be pushed upward into the glenoid fossa using the thumb or fingers of the other

hand. Beattie and associates2 reported a success rate of 70% with the Milch technique, but others report success rates of 90% or greater.5,29 Traction-Countertraction.  This method is commonly used in the ED, largely out of tradition, because it has a high rate of success and many emergency clinicians are most comfortable with it. Familiarity is an advantage of this technique, but it requires more than one operator, some degree of force, and occasionally, endurance. This technique is usually quite uncomfortable for the patient, and premedication is recommended before any attempt. With the patient supine, a sheet or strap is wrapped around the upper chest and under the axilla of the affected shoulder (Fig. 49–14). An assistant holds this sheet, preferably by wrapping the sheet around the waist to take advantage of body weight rather than arm strength, to apply the countertraction. The operator’s foot should not be used in the axilla to provide countertraction. Traction may then be applied to the extended arm, but this generally results in operator fatigue, especially if the operator relies on biceps strength to provide continuous traction. Preferably, the elbow of the affected side is flexed to 90° and a sheet or strap is wrapped around the proximal forearm and then around the operator’s back. The bed should be elevated to a point at which the sheet can sit at the level of the operator’s ischial tuberosities. This allows the operator to comfortably lean back and use the body weight to supply the force of traction, eliminating the possibility of operator fatigue. The portion of the sheet that is positioned on the patient’s forearm has a tendency to ride up; flexion of the elbow beyond 90° will minimize this problem. Alternatively, the operator merely leans backward with the arms fully extended, again using the continuous weight of the body rather than the strength of the biceps to provide constant traction. Once traction is applied, the operator must be patient, because the procedure may take a number of minutes to be successful. Inadequate premedication is noted by the patient who resists the procedure or is notably uncomfortable during

879

MUSCULOSKELETAL PROCEDURES ● 

VIII 

A

B 880 Gentle lateral traction

First traction

Then adduct arm

C Figure 49–14  A and B, Traction-countertraction method. This simple technique for reducing the dislocated shoulder applies gradual and steady traction along the axis of the dislocated limb. A bedsheet, wrapped around the supine patient’s upper chest wall and over the unaffected shoulder, is either tied or held by an assistant and acts as a fixed counterforce. A second bedsheet is placed around the patient’s flexed forearm, just distal to the flexed elbow, and securely tied behind the operator’s back. Note that a significant skin avulsion or friction burn may occur if there is excessive motion of the sheets, especially in the elderly patient with thin, delicate skin. With the patient’s forearm held in a neutral rotation and the hand in a vertical position, the operator applies traction by leaning back, rather than using the biceps to apply traction, which will soon fatigue most clinicians. C, Gentle lateral traction on the humerus, coupled with adduction of the arm, often helps reduction. (B and C, From Respet PB: A practical technique for reducing shoulder dislocations. J Musculoskel Med 5:29, 1988.)

the reduction attempt. The operator should not hesitate to order supplementary medications. Gentle, limited external rotation is sometimes useful to speed reduction.10 Applying traction to an arm that is slightly abducted from the patient’s body is often successful, but some operators prefer to slowly bring the arm medial to the patient’s midline while maintaining traction or to have an assistant apply a gentle lateral force to the midhumerus to direct the humeral head laterally. Successful reduction is usually presaged by slight lengthening of the arm as relaxation occurs, and a noticeable “clunk” may occur at the point of reduction. A brief fasciculation wave in the deltoid may also be seen at the time of reduction. Spaso Technique.  This technique was first reported by Spaso Miljesic as a simple, single-operator technique requiring minimal force.37 One published series reported an 87.5% success rate among premedicated patients when performed by junior house officers.38 The patient is placed in a supine position and the operator grasps the affected arm around the wrist or distal forearm. The affected arm is gently lifted vertically toward the ceiling, applying gentle vertical traction. While traction is continuously maintained, the arm is externally rotated (Fig. 49–15A). Reduction may be subtle, but is generally signaled by hearing or feeling a “clunk.” Completion of this technique may require several minutes of gentle traction, allowing the muscles of the patient’s shoulder to relax.38 Other Methods.  Poulsen39 reported a method termed the Eskimo technique, which may be performed in field settings. In this technique, the patient lies on the unaffected side and is lifted a short distance off the ground by grasping the abducted arm of the injured side (see Fig. 49–15B). The patient’s body weight acts to effect the reduction. Poulsen’s39 success rate was 74% in a series of 23 patients, all of whom were premedicated. Poulsen39 also postulated that this technique could place undue stress on the brachial plexus or axillary vessels. Use of this technique, when other options are available, should probably be reserved until a larger experience is reported. Noordeen and coworkers40 reported a simple method in which the patient sits sideways in a chair, with the affected arm draped over the backrest. The operator holds the arm with the wrist supinated, and the patient is instructed to stand up. The success rate was 72% in 32 patients treated in this manner. A variation of the chair technique, which was successful in 97% of 188 anterior shoulder dislocations, involves operator-applied traction to the patient’s flexed elbow by means of a cloth loop or stockinette.41 Standing beside the patient, the operator holds the involved elbow in 90° of flexion while stepping down on the cloth loop. The patient sits in the chair, and an assistant may help support the patient by applying countertraction under the involved arm. Postreduction Care After an attempt at reduction, the neurovascular status of the affected extremity should be rechecked and the results documented on the patient record. Indirect evidence that the reduction has been successful includes an immediate reduction in pain, restoration of the round shoulder contour, and increased passive mobility of the shoulder. No harm is done by putting the joint through a limited range of motion. If the patient can tolerate placement of the palm from the injured arm on the opposite shoulder, it is quite likely that the shoulder reduction was successful (see Fig. 49–15C). Postreduction radiographs are often recommended, with a careful search for new fractures. Although most greater

49  ● 

Management of common dislocations

A

C

B 881

D

E

Figure 49–15  A, Spaso technique. While an assistant maintains gentle vertical traction-countertraction, the affected arm is externally rotated by grasping the wrist or forearm. Reduction may be subtle. B, Eskimo technique. The patient’s body weight serves as countertraction. If the operator stands on a stool, biceps power is not needed. The operator’s arms are left fully extended while grasping the patient’s wrist, and the back is straightened to avoid operator arm fatigue. C, Regardless of the technique used, if a patient with a shoulder injury can place the palm of the injured arm on top of the contralateral shoulder, it is unlikely that a shoulder dislocation is present. Alternatively, completion of this maneuver after a reduction attempt provides strong evidence that the reduction was successful, even if the patient is still sedated. D and E, The best way to immobilize any reduced shoulder dislocation is uncertain and unlikely of consequence for a few days (see text). D, A typical shoulder immobilizer or a simple sling is appropriate pending referral and follow-up. (B, D, and E, From Thomsen T, Setnik G [eds]: Procedures Consult—Emergency Medicine Module. Copyright 2008 Elsevier Inc. All rights reserved.)

tuberosity fractures do not alter patient management, patients with greater tuberosity fractures displaced greater than 1 cm after closed reduction are almost always associated with a rotator cuff tear42 and should receive prompt orthopaedic consultation, because they may require operative repair. Traditional postreduction treatment has focused on the importance of preventing the shoulder from dislocating after discharge. This is best accomplished by immobilizing the joint using a commercially available shoulder immobilizer or

a sling and swath, which limits external rotation and abduction (see Chapter 50, Splinting Techniques). Orthopaedic follow-up is recommended for all anterior shoulder dislocations because the incidence of rotator cuff injury is as high as 38% and might complicate restoration of normal function.43 Younger patients are usually immobilized for approximately 3 weeks and can be instructed to follow up within 1 or 2 weeks of the event. As a general rule, the older the patient, the shorter the time of immobilization.10 Those older than 60

MUSCULOSKELETAL PROCEDURES ● 

VIII 

882

years should have early follow-up (e.g., 5–7 days) to allow for early mobilization and avoidance of persistent or permanent shoulder joint stiffness. Since the early 2000s, the wisdom of immobilization in internal rotation has been questioned. Several studies have shown that placing the arm in internal rotation actually increases labral detachment from the glenoid rim, whereas some degree of external rotation maximizes contact between the detached labrum and the glenoid rim.44–46 In one study, cadavers were used to measure the force of contact between the labrum and the glenoid rim in different arm positions. The authors of this study found that maximal contact force was actually generated in 45° of external rotation, whereas no contact force was generated with the arm in internal rotation.46 One prospective study showed that none of 20 patients immobilized in external rotation had recurrent dislocation after more than 1 year, compared with 6 of 20 patients immobilized in internal rotation.45 Despite this growing body of evidence, very little scientific data remain to guide the clinician on the most appropriate position for postreduction immobilization of anterior shoulder dislocations. In fact, a recent literature review designed to assess (1) whether traumatic anterior shoulder dislocations should be immobilized, (2) how long they should be immobilized, and (3) whether the position of immobilization affects outcomes was unable to provide any definitive answers.47 According to the author of this study, “much of this uncertainty is due to the limited size of the evidence base, which exhibited numerous methodological weaknesses (e.g., small sample sizes, no control groups, not evaluating findings against statistical tests).”47 As a result, it is not unreasonable to immobilize the extremity in a manner consistent with the orthopaedic surgeons at one’s institution until further evidence is presented. When in doubt, a simple sling or the traditional shoulder immobilizer will certainly suffice pending 5- to 7-day followup (see Fig. 49–15D and E). It is appropriate to prescribe oral analgesics (either nonsteroidal anti-inflammatory drugs or narcotics) appropriate for the amount of patient discomfort at the time of disposition and to instruct the patient to return for any worsening of the clinical condition. Periodically, one may encounter a return visit from a successfully treated patient who is in severe pain from a hemarthrosis. Trimmings48 reported excellent relief of pain by aspiration of the hemarthrosis 24 to 48 hours after shoulder reduction in a series of patients older than 60 years. This can be accomplished using the technique of arthrocentesis described in Chapter 53, Arthrocentesis. In addition, intra-articular instillation of 10 to 20 mL of 1% lidocaine (or longer-acting local anesthetic) as has been recommended for shoulder reduction may be helpful for further pain relief.

well past the original event.50 Patients with seizures may not experience obvious problems in the immediate postictal period owing to their altered mental status. Clinical Assessment Although clinically less obvious than anterior dislocations, posterior shoulder dislocations do present in a typical, recognizable manner. Mistakes may be made if the clinician is overly reliant on the AP radiographs, which are potentially misleading,50 and may result in misdiagnosing the injury as a soft tissue contusion or acromioclavicular (AC) strain. The principal sign of posterior dislocation is an arm that is somewhat fixed in adduction and internal rotation (Fig. 49–16). Abduction and external rotation are limited, and attempts to perform these movements generally elicit pain.10,12 Inspection and palpation reveal a loss of the normal anterior contour of the shoulder and a prominent coracoid and acromion. The shoulder is flattened anteriorly and rounded posteriorly, whereas the humeral head may be palpable.10,12 Comparison with the opposite shoulder should be undertaken with the understanding that this injury may occasionally occur bilaterally. Neurovascular assessment is performed in the standard manner, although such complications are unusual with posterior dislocations.

Normal shoulder

Full external rotation possible

Posterior dislocation Arm adducted elbow flexed (sitting poition) Unable to externally rotate shoulder

A

Posterior dislocation Elbow extended

Posterior Shoulder Dislocations Posterior shoulder dislocations account for less than 4% of all shoulder dislocations.12 Because they are so uncommon, posterior dislocations are easily overlooked and the emergency clinician must be knowledgeable about these injuries to avoid a misdiagnosis. Delays in diagnosis for weeks to months have been reported with posterior dislocations.49,50 This may lead to increased rates of dislocation arthropathy and chronic pain.13 The mechanism of injury is almost always indirect, with a combination of internal rotation, adduction, and flexion.10 Classic precipitating events include seizure, electrical shock, and falls. The patient may also present at a point

Appears to externally rotate shoulder B Figure 49–16  A posterior dislocation may be difficult to appreciate on x-rays. A, A clue to a posterior shoulder dislocation is the arm locked in adduction and internal rotation, with patient’s inability to rotate the shoulder externally with the elbow flexed at 90°. B, Note that extension of the elbow with supination of the forearm may obscure loss of the external rotation.

Management of common dislocations

Reduction Technique An acute posterior dislocation may be reduced by traction on the internally rotated and adducted arm combined with anteriorly directed pressure on the posterior aspect of the humeral head (Fig. 49–20).10,50 Generous premedication is generally indicated, and countertraction may be applied with a sheet looped in the affected axilla much as described for anterior dislocations. Kwon and Zuckerman10 recommend applying lateral traction on the upper humerus if the humeral head is locked on the posterior glenoid. Hawkins and colleagues50 suggested that posterior dislocations with an impression defect of the humeral head greater than 20% of the articular surface require open reduction. Posterior dislocations that have been diagnosed late are difficult to reduce in a closed manner, but an attempt with adequate premedication is generally indicated.50

● 

There may also be a compression fracture of the medial aspect of the humeral head, indicated by a dense line. This is known as the trough sign51 (see Fig. 49–17C). A fracture of the lesser tuberosity should always prompt a search for the presence of a posterior shoulder dislocation.49

49 

Radiologic Examination The key point regarding radiographs for posterior shoulder dislocations is the subtle nature of this dislocation on a single AP radiograph (Fig. 49–17A) and the diagnostic value of the scapular Y view (see Fig. 49–8C) or the axillary view (see Fig. 49–17B). The diagnosis of posterior shoulder dislocation using the axillary view is quite easy, whereas the routine AP and lateral views are difficult to interpret in around half of cases.50 The axillary view is generally available in the radiology department and can be obtained with as little as 20° to 30° of abduction, with the plate placed on the shoulder.50 In addition to easy visualization of the posteriorly situated humeral head, the axillary view often reveals an impression fracture of the humeral head (see Fig. 49–17B). Whereas the axillary view is diagnostic, clues to posterior dislocation do exist on the AP film. The internally rotated humeral head appears symmetrical on the AP film in the shape of a lightbulb as opposed to the normal club-shaped appearance created by the greater tuberosity51 (Figs. 49–18 and 49–19). With posterior dislocation, the space between the articular surface of the humeral head and the anterior glenoid rim is widened, and there is a decrease in the half-moon– shaped overlap of the head and the fossa (see Fig. 49–18).49,51

883

B

A

D

C

E

Figure 49–17  Posterior shoulder dislocation.  A, This patient has a posterior dislocation of the humerus. Because the dislocation is directly posterior, there is no superior or inferior displacement of the humeral head. On superficial observation, the head of the humerus appears to maintain a normal relationship with the glenoid fossa and the acromion process. However, definite abnormalities exist in this film. The space between the humeral head and the glenoid fossa is abnormal, and because of the extreme internal rotation of the humerus, the head and neck are seen end on. In this projection, the humeral head resembles a lightbulb. Impaction of the humeral head on the posterior rim of the glenoid (B) leads to the “trough sign” (arrow) seen on an AP radiograph (C). D, Trans-scapular radiograph from a third patient shows posterior dislocation of the humeral head (large arrow) relative to the glenoid (small arrow). E, Computed tomography (CT) scan from the patient in D shows impaction of the anterior humeral head on the posterior glenoid. (A, From Harris JH Jr, Harris WH: Radiology of Emergency Medicine, 2nd ed. Baltimore, Williams & Wilkins, 1981, p 629. Reproduced by permission.)

MUSCULOSKELETAL PROCEDURES VIII 

● 

A

Figure 49–20  Reduction of posterior shoulder dislocation. With countertraction applied (arrow), traction on the internally rotated and adducted arm is combined with anterior pressure on the humeral head to effect reduction. B Figure 49–18  A, Note the normal elliptical pattern of overlap produced by the head of the humerus and the glenoid fossa. B, In the patient with a posterior dislocation, this pattern is lost, and internal rotation of the greater tuberosity is also noted. (A and B, From Simon R, Koenigsknecht S: Orthopedics in Emergency Medicine. New York, Appleton-Century-Crofts, 1982, p 344. Reproduced by permission.)

884

Figure 49–19  “Lightbulb” appearance of the humeral head in a posterior dislocation. Posterior dislocation should be considered after significant trauma, electrical injuries, or seizures, or when the patient is in severe pain.

Postreduction Care As with anterior dislocations, a repeat neurovascular examination and radiographs are obtained after reduction attempts. As before, the patient’s ability to place the palm of the injured arm on the opposite shoulder is suggestive of a successful reduction. Given the rarity of these injuries, orthopaedic consultation is often sought early in the care of these patients. Certainly in a training environment, involvement of an orthopaedic resident benefits her or his education and should be considered early on.

Unusual Shoulder Dislocations Luxatio Erecta Inferior dislocations of the shoulder, known as luxatio erecta, are quite rare, but also quite obvious. The patient presents

with the arm locked in marked abduction with the flexed forearm lying on or behind the head52 (Fig. 49–21). Occasionally, the humerus may have less abduction, thus potentially obscuring the diagnosis.53 The humeral head can be palpated along the lateral chest wall. In this injury, the inferior capsule is almost always torn. Associated injuries include fractures of the greater tuberosity, acromion, clavicle, coracoid process, and glenoid rim. Neurovascular compression may be present, but this is usually reversed once reduction is accomplished.10 Long-term complications include adhesive capsulitis or recurrent dislocations. Overhead traction (generally with the arm in full abduction) is applied in the longitudinal direction of the arm, and cephalad pressure can be exerted over the humeral head much as in the Milch technique.10,53 Countertraction toward the patient’s feet can be applied using a sheet placed over the injured shoulder. After reduction, the abducted arm is brought into adduction against the body and the forearm is supinated.54 Alternatively, the “two-step” maneuver described by Nho and associates55 can be used to reduce inferior dislocations. The luxatio erecta is essentially converted to an anterior dislocation. To perform this maneuver, the operator places one hand on the medial condyle of the elbow and the other hand around the shaft of the humerus. Pushing anteriorly on the shaft of the humerus while stabilizing the medial condyle of the elbow, allows the operator to rotate the humeral head from an inferior to an anterior position. The authors then describe using the external rotation method to reduce what is now a typical anterior dislocation.55 Scapular dislocation or “locked scapula” is a rare condition that presents with an obvious protrusion of the lateral border of the scapula and significant swelling of the medial border due to tearing of the musculature.56 Reduction is accomplished by traction on the abducted arm and medial pressure on the scapula.56

AC JOINT SUBLUXATION AND DISLOCATIONS The AC joint is a true diarthrodial joint with a synovial cavity surrounded by a relatively lax capsule and the weak AC ligament. This structure allows for the gliding motion necessary for shoulder movement. The major stability of the AC joint comes from the coracoclavicular ligament, which has poste-

49  ● 

Management of common dislocations

A

B

2 3

Figure 49–21  A, Luxatio erecta is a rare shoulder dislocation that presents with the arm in marked abduction, usually held over the head. B, Radiographic appearance of luxatio erecta. C, Reduction is accomplished with longitudinal traction with the arm full abducted and countertraction applied downward. (A– C, From Thomsen T, Setnik G [eds]: Procedures Consult—Emergency Medicine Module. Copyright 2008 Elsevier Inc. All rights reserved.)

1

C

Incomplete tear

A

Figure 49–22  A–C, Grades I–III of acromioclavicular separation. (See text for description.) (From Heppenstall RB: Fractures and dislocations of the distal clavicle. Orthop Clin North Am 6:480, 1975.)

Incomplete tear

B

rior (conoid) and anterior (trapezoid) components. The mechanism of injury is generally from a direct force such as a fall on the point of the shoulder with the arm adducted.57 The six grades of injury to the AC joint are classified by degree or type (I–VI) (Fig. 49–22). First Degree (Type I).  This injury consists of a minor tear in the AC ligament. The coracoclavicular ligament is intact. The clinical findings are limited to tenderness in the area of the AC joint. Radiographs show little, if any, change in the position of the clavicle in relation to the acromion.42 The management of this condition consists of a sling for comfort, ice, and mild analgesics. Generally, symptoms subside with 7 to 10 days of rest.10 Orthopaedic referral is

C

generally not necessary unless return to normal function is delayed beyond 2 weeks. Second Degree (Type II).  In addition to a complete tear of the AC ligament, the coracoclavicular ligament is stretched or incompletely torn.42 The patient generally supports the injured arm and has slight swelling and definite tenderness over the AC joint. Radiographs demonstrate a definite change in the relationship of the distal clavicle to the acromion. However, in type II injuries, the inferior edge of the clavicle should not be separated from the acromion by more than half its diameter,42 and on radiographic examination, the coracoclavicular distance is the same as that on the uninjured side.10 This injury can be treated in a closed fashion with a sling.10

885

MUSCULOSKELETAL PROCEDURES ● 

VIII 

886

Orthopaedic referral is recommended, and some will use a sling-strap device that elevates the arm and depresses the clavicle for these injuries.42 Third Degree (Type III).  In this injury, the distal end of the clavicle is essentially free-floating, as both the AC and the coracoclavicular ligament are completely disrupted.42 The arm is supported by an uncomfortable patient and the distal clavicle is usually seen to be riding high above the acromion. The diagnosis is generally obvious, and radiographs are used mainly to rule out an associated fracture. Radiographic criteria for this degree of injury include an inferior border of the clavicle above the acromion or a discrepancy in the coracoclavicular distance compared with that of the normal side.10 These injuries require orthopaedic referral, and a fair bit of controversy exists regarding their subsequent management.10 Larsen and coworkers58 conducted a prospective, randomized trial of conservative versus operative management for significant AC separations and concluded that conservative management was generally better, with possible exceptions made for patients with significant cosmetic deformity and for those who frequently keep the arm at 90° of abduction. Whereas optimal therapy is still unclear, a logical approach would include ED treatment with a sling and early orthopaedic referral. Fourth, Fifth, and Sixth Degrees (Types IV–VI).  In type IV injury, the distal clavicle is free-floating and posteriorly displaced into the mass of the trapezius muscle. Type V injury is characterized by inferior displacement of the scapula with a marked increase (two to three times normal) in the coracoclavicular interspace.10 Types IV and V dislocations generally require surgery, and orthopaedic referral is necessary. Type VI injury involves dislocation of the distal clavicle inferiorly. Because this is usually the result of major trauma, multiple other fractures are often seen.10

Radiographic Examination The diagnosis is usually made clinically, with pain and local tenderness at the AC joint in the absence of other findings. Radiographs are generally indicated to rule out associated fractures and to aid in assessing the degree of injury. A single radiograph of the injured shoulder often suffices, but some clinicians prefer to obtain comparison views of the opposite shoulder. Although their efficacy has never been proved, it has been traditionally recommended that “weighted” films be obtained in suspected type I or II injuries. Weighted films are generally performed after routine “unweighted” radiographs and are obtained by strapping about 4.5 to 7.0 kg (10–15 lb) of weight to the patient’s wrists and repeating the radiographs. Weighted films are of questionable value in mild injuries and superfluous in obvious type III to VI injuries. Their use has been essentially abandoned in emergency medicine practice.59 In a prospective study of 70 type I or II injuries, the use of weights was associated with less evident separation in 7 cases, essentially producing a false-negative study compared with plain unweighted films. Only 3 injuries were recategorized as type III after the performance of weighted films.59

STERNOCLAVICULAR DISLOCATIONS Despite the fact that the sternoclavicular joint is the least stable joint in the body, sternoclavicular dislocations are rare.60 The primary supports of this joint are the sternocla-

vicular and costoclavicular ligaments. Anterior dislocations are much more common and usually the result of an indirect mechanism involving a blow thrusting the shoulder forward,42 or they may be atraumatic due to ligamentous laxity in teens and young adults.60 Posterior dislocations also usually result from a blow to the shoulder but can also be the result of a direct superior sternal or medial clavicular blow.60 Posterior sternoclavicular dislocation (also known as retrosternal dislocation as the medial end of the clavicle dislocates both posteriorly and medially) is potentially life threatening, because injury to the great vessels or compression of the airway might occur.60 Any suggestion of these complications should prompt immediate surgical consultation. The presentation of these injuries is usually straightforward, with pain, swelling, tenderness, and deformity of the joint. Plain radiographs of this joint are difficult to interpret and generally include an apical lordotic-type view with the radiographic tube angled at 45° cephalad. Confirmation of the diagnosis is best made using a thoracic computed tomography (CT) scan, which may also identify high rib fractures, pulmonary contusion, or pneumothorax.60,61 The use of CT angiography will also show vascular injury associated with posterior dislocations. Children may have epiphyseal disruption with retrosternal displacement of the medial clavicle.62 Closed reduction of both types of sternoclavicular dislocation involves placing a rolled blanket or a sandbag between the scapulae and applying traction to the 90°-abducted, 10°extended arm in line with the clavicle. The clavicle can then be pushed (anterior) or lifted (posterior) back into position.60 Posterior dislocations may be difficult to reduce and to maintain reduced in a closed manner. Therefore, some authors recommend reduction in an operating suite unless complications necessitate immediate reduction.60 Given the rarity of this injury and the potential for major underlying complications, early consultation is recommended in suspected posterior sternoclavicular dislocations. Once reduced, a clavicle strap may be used to immobilize both anterior and posterior dislocations.

ELBOW DISLOCATIONS The elbow is second only to the shoulder as a site for major joint dislocations in adults; it is the most commonly dislocated joint in children. Anatomically, the principal articulation of the humerus and ulna is a stable hinge joint with the intercondylar groove of the distal humerus nestled in the olecranon fossa. Owing to the stability of the elbow, any dislocation is expected to be accompanied by significant soft tissue damage, and associated fractures are common. Elbow dislocations are often simply divided into posterior and anterior dislocations (Fig. 49–23). However, there are actually several different types of elbow dislocations in addition to posterior and anterior. These include lateral, divergent, and isolated dislocations of the radius.63 In the rare divergent dislocations, the radius and ulna are dislocated in opposite directions, either anterior and posterior or medial and lateral.63 The most serious complication of elbow dislocation is a brachial artery injury. This injury is possible with any type of elbow dislocation and is a frequent occurrence in open dislocations.63 Vascular compromise can be delayed in onset, resulting from either unsuspected arterial injury or progressive soft tissue swelling. The circulatory status of the arm must be carefully monitored even after successful reduction.63 Although not absolute, patients with these injuries who mani-

49  ● 

Management of common dislocations

Posterior dislocation

Anterior dislocation

A

Posterior dislocation Figure 49–23  Classification of elbow dislocations. (From Simon R, Koenigsknecht S: Orthopedics in Emergency Medicine. New York, Appleton-Century-Crofts, 1982, p 333. Reproduced by permission.)

fest significant or immediate soft tissue swelling or hematoma formation or who have questionable vascular integrity or neurologic findings are often admitted to the hospital or ED observation unit for observation. In most cases, orthopaedic consultation should be sought prior to disposition. Injury to the median and ulnar nerves may be the result of stretch, severance, or entrapment. It is difficult to clinically distinguish these etiologies, and management of nerve injuries is frequently expectant.63 It is imperative to conduct a careful neurologic examination before and after reduction, because any increase in findings may indicate entrapment and the need for surgical intervention.63 Myositis ossificans is also a potential complication of this injury, which underscores the advisability of orthopaedic consultation early in the course of care.

887

B Humerus

Posterior Dislocations Posterior dislocations make up the vast majority of elbow dislocations.42 The usual mechanism is a fall on the outstretched hand with the arm in extension. The clinical examination is usually diagnostic unless severe soft tissue swelling is present. The patient presents with a shortened forearm that is held in flexion and the olecranon is prominent posteriorly. The normally tight triangular relationship of the olecranon and the epicondyles of the distal humerus is disturbed in a posterior dislocation. A defect may also be palpated above the prominence of the olecranon. Radiologic Examination Two radiographic views, an AP and a true lateral, should be obtained (Fig. 49–24A and B). The diagnosis is obvious with proper radiographs. A careful search for fractures of the distal humerus, radial head, and coronoid process must be undertaken, because they commonly occur in this injury.63 In children younger than 14 years, the fracture is usually a medial epicondyle separation, because the epiphyseal plate gives way before the medial collateral ligament of the elbow.

Capitellum Radial head

C

Radius

Olecranon process Ulna

Figure 49–24  Lateral (A) and AP (B) radiographs of a posterior elbow dislocation, the most common type. C, Normal postreduction radiograph. (A and B, From Thomsen T, Setnik G [eds]: Procedures Consult—Emergency Medicine Module. Copyright 2008 Elsevier Inc. All rights reserved.)

MUSCULOSKELETAL PROCEDURES ● 

VIII 

888

Postreduction radiographs (see Fig. 49–24C) are also necessary to confirm reduction and disclose any associated fractures.64 Reduction Techniques and Postreduction Care Although, as with shoulder reduction, some authors claim that their method of reduction is virtually painless,32,65 this has not been objectively documented. In general, patients with posterior elbow dislocations are quite uncomfortable, and it is beneficial to administer IV analgesics early in the course of care, preferably before positioning for radiographs. In addition to, or in lieu of, parenteral sedation and analgesia, some clinicians inject the elbow joint with a local anesthetic (e.g., 3–5 mL of 2% plain lidocaine) before attempting reduction. Prior to injection, the joint should be aspirated to remove blood. Traditional Traction Method.  The traditional method of reducing a posterior dislocation is to place the patient in the supine position and have an assistant stabilize the humerus with both hands66 (Fig. 49–25). The operator then grasps the wrist and applies slow and steady in-line traction. The elbow is slightly flexed to keep the triceps mechanism loose, and the wrist is held supinated as traction is applied. Reduction is usually signified by a “clunk” that is heard or felt. If this method is not successful after a reasonable period of traction (10 min), the forearm may be gently flexed to try to effect reduction. Alternatively, downward pressure on the proximal volar surface of the forearm may help free up the coronoid process. Alternatives.  Several authors have described variations of a prone method of reduction that are reportedly well tolerated by patients.32,65,66 In the method described by Minford and Beattie,65 the patient is positioned with the arm hanging over the padded back of a chair or over the edge of the bed. The clinician then simply applies pressure to the prominent posterior aspect of the olecranon to achieve reduction. Alternatively, traction may be applied with the elbow flexed over the edge of a chair by pulling down on the hand while using the thumb to guide the olecranon into place66 (Fig. 49–26A and B). Parvin32 positioned the patient as for the Stimson

method of shoulder relocation, prone on a stretcher with the arm hanging down, and applied gentle downward traction to the wrist (see Fig. 49–26C). Recommended Initial Approach.  A prone technique is advantageous because patients tolerate this position quite well. The elbow is allowed to hang flexed over the edge of the bed, and an assistant is positioned with his or her back toward the patient such that the humerus can be encircled with both hands and pressure applied with the thumbs to the posterior aspect of the olecranon. This pressure on the olecranon is intended to lift it up and away from the humerus. The operator applies longitudinal traction to the arm with the elbow in slight flexion. If reduction is not succeeding, an attempt may be made to flex the elbow or the assistant can be instructed to lift the humerus (see Fig. 49–26C). Reduction is generally noted by a definite “clunk.” Postreduction Care.  Once reduction is achieved, the elbow should immediately be put through a gentle range of motion to ensure that the reduction is stable and that there is no mechanical block to movement.63 An inability to move the elbow through a smooth range of motion after reduction is often caused by an entrapped medial epicondyle fracture fragment, which requires operative intervention.63 The elbow is generally immobilized in at least 90° of flexion with a longarm posterior splint. A complete recheck of the neurovascular status is performed along with postreduction radiographs. After reduction, any signs of delayed vascular compromise are first addressed by loosening the splint and decreasing the degree of flexion. This may restore the pulse.63 If not, immediate surgical consultation is necessary for emergent arteriogram or exploration of the brachial artery, or both.64 The risk of vascular compromise is a reason to consider in-hospital observation. Alternatively, some clinicians observe the patient in the ED or ED observation unit for 2 to 3 hours postreduction, eva­luating for delayed neurovascular compromise before discharge.

Anterior Dislocations Anterior dislocations of the elbow are quite rare; they usually result from a direct posterior blow to the olecranon with the

4 2

3 5

1 4

Figure 49–25  Manipulative reduction of a posterior elbow dislocation.  While an assistant holds the arm and makes steady countertraction (1), grasp the wrist with one hand and apply steady traction on the forearm in the position in which it lies (2). While traction is maintained, correct any lateral displacement with the other hand (3). While traction is maintained (4), gently flex the forearm (5). Note that with reduction, a clunk is usually felt and heard as the olecranon engages the articular surface of the humerus. (From DePalma AF: Management of Fractures and Dislocations: An Atlas. Philadelphia, WB Saunders, 1970, pp 793 and 794. Reproduced by permission.)

49  ● 

Management of common dislocations

2 1

C

A

Counteraction

Olecranon pressure

B

Figure 49–26  Additional methods to reduce posterior elbow dislocation.  A, Hanging method of elbow dislocation reduction (chair). Downward traction is applied to the hand while the olecranon is guided into place. B, The patient is placed in the prone position with the elbow hanging flexed over the edge of the bed. An assistant encircles the distal humerus and applies pressure to the olecranon with both thumbs in order to lift the olecranon up and away from the humerus. The operator applies longitudinal traction to the arm with the elbow in slight flexion. If reduction does not occur, the operator can flex the elbow or the assistant can lift the humerus. C, The patient is placed prone on the stretcher with the arm hanging down. Having the upper arm supported by the stretcher may help relax the elbow joint. Gentle downward traction may be applied (1) while the humerus is distracted to help move the olecranon into place (2). (A, From Lavine LS: A simple method of reducing dislocations of the elbow joint. J Bone Joint Surg Am 35:785, 1953.)

elbow flexed.63 On physical examination, the arm is extended and there is anterior tenting of the proximal forearm with prominence of the distal humerus posteriorly.63 These injuries are the result of a great deal of force; they are frequently open and accompanied by significant neurovascular injury. An avulsion of the triceps mechanism may also occur.63 Reduction of an anterior dislocation of the elbow involves in-line traction and backward pressure on the proximal forearm (Fig. 49–27). An assistant provides countertraction by grasping the humerus with both hands. Given the infrequent nature of anterior dislocations and the high probability of a severe associated injury, the emergency clinician should consider early orthopaedic consultation in such dislocations.

RADIAL HEAD SUBLUXATION (NURSEMAID’S ELBOW) Radial head subluxation is a common pediatric presentation generally occurring between the ages of 1 and 3 years. The mean age of presentation is just older than 2 years, but this entity has been reported in infants younger than 6 months67,68 and in older children up to the preteen years.69 There is a slight predilection for this injury to occur in girls and in the left arm.68,70 The classic mechanism of injury is longitudinal traction on the arm with the wrist in pronation, as occurs when the child is lifted up by the wrist.68 There is no support for the common assumption that a relatively small head of the radius in relation to the neck of the radius predisposes the young to this injury.71 The pathologic lesion is generally a tear in the attachment of the annular ligament to the periosteum of the radial neck, with the detached portion becoming trapped between the head of the radius and the capitellum.71

889 2

3 1

4

Figure 49–27  Manipulative reduction of anterior elbow dislocation.  Reduction is performed with the patient under deep sedation or local or general anesthesia. 1, An assistant grasps the arm and provides countertraction. 2, The operator grasps the wrist with one hand and applies traction in the line of the arm and, with the other hand, applies firm, steady pressure downward and backward on the upper end of the forearm (3). A clunk usually indicates that reduction is achieved. 4, The arm is flexed to 45° beyond a right angle. (From DePalma AF: Management of Fractures and Dislocations: An Atlas. Philadelphia, WB Saunders, 1970, p 796. Reproduced by permission.)

Clinical Assessment The history offered by the caretaker may not be that of the classic pulling-type mechanism. Schunk,68 in a series of 83 patients, reported that only 51% described such a mechanism, whereas 22% reported a fall. In patients younger than 6 months, the mechanism in the majority is simply rolling over

MUSCULOSKELETAL PROCEDURES ● 

VIII 

A

B

Figure 49–28  Typical presentation of a child with a subluxation of the radial head (nursemaid’s elbow). It may be difficult to determine exactly where the pathology exists, and often the wrist is thought to be the area of injury. This child will not use the injured arm but has minimal discomfort as long as the elbow is not manipulated. A, The affected arm hangs down at the side, slightly flexed and pronated. B, Once the subluxation is reduced, full activity is generally regained in a matter of minutes.

890

in bed.67 It is important to remember this and not to proceed with a child abuse investigation unless other suggestive features are present. The typical patient with a nursemaid’s elbow presents in no distress with the arm held slightly flexed and pronated at the side (Fig. 49–28). This has been termed the nursemaid’s position.72 The exact area of pain is often difficult to locate. The child will refuse to use the arm, and this may be the chief complaint.69 The older child will usually point to the dorsal aspect of the distal forearm when asked where it hurts. This may mislead one to suspect a buckle fracture of the distal radius. Although tenderness about the elbow has been reported occasionally, there is often little tenderness or swelling of the elbow region.68,69 In the cooperative child, the arm and shoulder are carefully palpated to discern any tenderness. Areas of focus on palpation should include the clavicle and the distal radius, because these are common sites of pediatric fractures. When patient anxiety interferes with a reliable assessment of tenderness in a child whose arm is in the classic nursemaid’s position, the examiner can stand at a distance and have the parent or caretaker palpate the extremity to ascertain tenderness. This may also be done in the cooperative patient to reassure the doubtful parent regarding the absence of a fracture. If no tenderness is noted by palpation, it is appropriate to attempt a reduction without prior radiographs.73 Although resistance to or pain with supination is a frequent finding in such patients,68 one need not test for this finding until the time of reduction. Radiographic Examination Radiographs are generally not needed in a child presenting with an arm in the nursemaid’s position that is nontender (or minimally tender in the radial head area) on palpation, regardless of the history.73 In these cases, radiographs are generally normal, and if obtained, the positioning of the child’s arm by the x-ray technician often effects reduction.68 However,

Frumkin71 described three cases of nursemaid’s elbow in which a line drawn through the longitudinal axis of the radius did not normally bisect the capitellum on prereduction radiographs, but did so after reduction. Radiographs are sometimes recommended if the child is not moving the arm normally 15 minutes after reduction.72 However, this timeframe may be too short because reuse can be delayed for more than 30 minutes, particularly in children who present some time after the injury. Quan and Marcuse69 recommended an approach in which no radiographs are obtained on the first visit, including in those children released from the ED prior to regaining full use of the arm. At the time of a 24-hour follow-up visit, radiographs are obtained only if repeat attempts at manipulation are not successful. Whereas this condition does not generally require x-rays, they can be valuable if external signs of trauma are present (e.g., swelling, abrasions, ecchymoses), or if the child does not use the arm normally within 24 hours after the subluxation is considered reduced. Other less common conditions that can present with similar findings are fractures, joint infections, tumors, or osteomyelitis. Reduction Techniques

Supination Method.  Reduction of a nursemaid’s elbow (Fig. 49–29) is generally performed without premedication. If the subluxation has been present for hours, oral or nasal midazolam can be a useful adjunct to overcome the child’s anxiety related to manipulation. It is important to explain to the caretaker that the reduction will likely cause the child discomfort, but that this is transient and a clue to the diagnosis. The child is positioned seated on the lap of an assistant (often the parent) who stabilizes the arm by holding the humerus adducted to the side. The operator then grasps the elbow with one hand placing the thumb over the region of the radial head. Although it has been stated that the thumb can apply pressure to the radial head, this positioning is mainly useful for palpation of the reduction “click.” The other hand grasps the wrist and is then used to supinate the extended forearm in a steady, deliberate manner. Slight traction before supination is generally recommended, but it is unclear whether this increases the likelihood of successful reduction. Once supinated, the arm can be flexed or extended; however, flexion is the most common maneuver and may actually be somewhat more successful than extension.68 An audible or palpable click signifies successful reduction, but is not always noted. Once the reduction has been performed, the child usually cries for a few minutes. Generally, the operator should leave the room and then return in 10 to 15 minutes to do a repeat examination. Full use of the arm should be evident (see Fig. 49–28B). Pronation Method.  This technique is performed with the child positioned as in the supination method. However, the forearm is not supinated. Instead, the forearm is rapidly hyperpronated and flexed. A recent study by McDonald and colleagues70 reported equal success rates using this technique and the supination technique. After Attempted Reduction.  If a click is detected, the child will generally regain use of the arm quickly (almost always by 30 min).69 Therefore, if a definite click is detected, it is reasonable to observe the child for up to 30 minutes prior to further intervention. If there is still no use at 30 minutes, the operator may try to determine whether supination is still painful, which would suggest the need for a repeat attempt. In those in whom a click is not detected, the majority will not

49 

Collateral ligament

REVISED

● 

Volar plate

A Accessory ligament

A

C

B Collateral ligament

Management of common dislocations

Axial traction with elbow extended

Volar plate

Figure 49–30  A and B, The collateral ligament–volar plate relationship. The metacarpophalangeal (MCP) and interphalangeal joints derive their strength from a combination of the two collateral ligaments and the volar plate. Dislocations of these joints require tearing of at least two parts of this three-part structure. C, Lateral view demonstrates the collateral ligament–volar plate relationship. (A and B, From Carter P [ed]: Common Hand Injuries and Infections. Philadelphia, WB Saunders, 1983, p 114. Reproduced by permission; C, redrawn from Eaton RG: Joint Injuries of the Hand. Springfield, IL, Charles C Thomas, 1972.)

B

D C Figure 49–29  Radial head subluxation.  A, Anatomically, this injury represents interposition of the torn annular ligament between the radial head and the capitellum. B, The supination method of reduction is performed by grasping the arm about the wrist and placing the other hand about the elbow with the thumb over the radial head. The forearm is then supinated (C) and then the arm is flexed (D) in one continuous motion. (A–D, From Fleisher GR, Ludwig S: Textbook of Pediatric Emergency Medicine. Baltimore, Williams & Wilkins, 1988, p 1322. Reproduced by permission.)

use the arm by 30 minutes.69 In these children, a repeat attempt at reduction is recommended after 10 to 15 minutes of nonuse. Two or more attempts are required to produce the click in up to 30% of patients.69 If the child has not regained the use of the arm after a few attempts and a reasonable period of time, some authors recommend that radiographs be performed.72 X-ray films also may help relieve parental anxiety. Alternatively, instructions should be given for 24-hour follow-up if normal function is not restored, with consideration for radiographs at the time of follow-up.69 In two series of patients with nursemaid’s elbow, of 10 patients released without normal arm use, 6 had spontaneous restoration of function, and the other 4 required remanipulation, which successfully restored function.68,69 The use of a posterior splint to protect the elbow of the child who refuses to use the arm after a presumed reduction is of uncertain value. However, some form of immobilization (e.g., splint, sling, or both) may be valuable in the child with significant residual discomfort after a prolonged period of subluxation or in whom recurrent subluxations have occurred. On occasion, a successful reduction painfully resubluxates with movement; in this case, immobilization and referral may

be necessary.71 If reduction has been achieved clinically and maintained in the ED, analgesics or a follow-up visit is unnecessary. Because other pathology can rarely mimic this condition (e.g., occult fractures, osteomyelitis, joint infection, tumors), full, unrestricted, and painless use of the arm must be evident by 24 hours. If not, further assessment is indicated.

HAND INJURIES The hand is an extremely common site of injury owing to the demands placed on it and the exposed nature of its location and use. Proper motion and function of the hand are intimately related to normal anatomic alignment.74 The emergency clinician must therefore be skilled in the diagnosis and management of dislocations about the hand. An improperly managed hand injury can result in significant disability that the patient is reminded of on a daily basis. Anatomically, the joints of the digits are quite similar and consist of a hinge joint with a tongue-in-groove–type arrangement.74 The soft tissue support includes two collateral ligaments attached to a volar plate (Fig. 49–30). The volar plate is dense fibrous connective tissue that is thickened at its distal attachment and thinner at its proximal attachment, to allow for folding with joint flexion.74,75 Dorsal dislocation of a digit requires failure of the volar plate, whereas lateral dislocation disrupts a collateral ligament and induces at least a partial tear in the volar plate (see Fig. 49–30). Radiographic examination of all hand injuries is relatively straightforward, including at least two views (AP and lateral) of the injured area. The most important radiographic error in evaluating joint injuries of the hand is failing to get a true lateral view of the injured joint.75 This may lead to missing a fracture or a loose body in the joint. Anesthesia is generally required for the proper management of dislocations about the hand. This is most often

891

MUSCULOSKELETAL PROCEDURES ● 

VIII 

892

accomplished by finger or wrist block, although a more proximal regional or Bier block may be used on occasion. Getting a secure grip on the digits may be difficult and may complicate the reduction. Wearing rubber gloves or wrapping gauze around the fingers may be useful.

Dislocations The opposable thumb is an essential structure for countless activities. Despite its strong ligamentous and capsular support, the exposed positioning of the thumb makes it a frequent site of dislocations and subluxations. The metacarpophalangeal (MCP) joint is similar to those of the fingers but has a stronger volar plate and collateral ligaments.75

A

Interphalangeal Joint Dislocation of the Thumb The single interphalangeal (IP) joint of the thumb has strong cutaneous-periosteal attachments, and dislocations of this type are therefore frequently open.75 Dislocations are generally dorsal and can be reduced in a manner similar to that for IP dislocations of the finger. The mechanism of injury is recreated by longitudinal traction and hyperextension to distract the phalanges. Reduction is accomplished by flexing the IP joint with continued traction and by applying direct pressure to the base of the distal phalanx.75 After reduction, the range of motion is tested and the stability of the reduction is ascertained. An adequate reduction that is documented on postreduction films is then splinted in slight flexion for 3 weeks.75 Orthopaedic referral is advisable.

B

MCP Joint Injury of the Thumb Dorsal Dislocation.  The MCP joint of the thumb can be dorsally dislocated by a hyperextension injury. The proximal phalanx will come to rest in a position dorsal to the first metacarpal (Fig. 49–31). There are two basic types of MCP dislocations (this applies to the fingers also): simple and complex. In a complex MCP dislocation, the volar plate becomes entrapped dorsal to the metacarpal head (Fig. 49–32) with the flexor tendons and lumbricals acting to completely entrap the metacarpal head.75 The simple type is amenable to closed reduction, whereas the complex type requires operative reduction owing to interposed soft tissue.74,75 A simple MCP dislocation can be converted into a complex one during reduction.74 Clinical features that suggest a complex MCP dislocation include a proximal phalanx that is less acutely angulated than with a simple dislocation (i.e., 50,000

>90

VIII 

● 

MUSCULOSKELETAL PROCEDURES

TABLE 53–2  Characteristics of Synovial Fluid

Hemarthrosis

Low

Crystals Negative Occasional calcium pyrophospate and hydroxyapatite crystals Negative Negative Negative Negative Needle-shaped, positive birefringent monosodium urate monohydrate crystals Rhomboid, negative birefringent calcium pyrophosphate crystals Negative Negative Negative

PMN, polymorphic nuclear cell; PVNS, pigmented villonodular synovitis. Adapted from Harris ED, Budd RC, Genovese MC, et al (eds): Kelley’s Textbook of Rheumatology, 7th ed., Section VI, Table 46–1. Philadelphia, Elsevier, 2005.

984

Figure 53–18  This periarticular laceration raises the question of knee joint penetration. A plain radiograph may demonstrate air in the joint space, but a saline arthrogram may also be used. Methylene blue alone is not generally required and it can cause an inflammatory reaction and obscure arthroscopy. A small amount of methylene blue may be added to color saline.

Figure 53–19  Saline arthrogram.  Using a stopcock and a 18-gauge needle, enter the joint in a manner identical to that for arthrocentesis. Note that a bag of intravenous saline, or additional vials of saline, introduced into the syringe may be required to provide enough saline to distend the joint properly. Unless the joint is markedly distended, a false-negative test may result. When completed, drain the injected saline via the original needle. A positive test is saline egress into the original wound or a slow loss of saline from the joint. A small amount of methylene blue may be added to the saline.

46% sensitive; injecting 100 mL, 75% sensitive; to achieve 95% sensitivity required an average of 194 mL of saline.41,42 The recommended volume of injection per joint is ●

Knee—100–200 mL Elbow—20–30 mL ● Ankle—20–30 mL ● Wrist—5 mL ● Shoulder—40–60 mL ●

Once the injection is complete, do not remove the needle, but close the stopcock to avoid backflow. Examine the joint

for evidence for leakage of fluid from the wound. This is performed in a static position, but if negative, also with some gentle passive movement of the joint. Visible leakage of fluid into the laceration confirms the diagnosis of joint space violation. A negative test is defined as absence of evidence of leakage after an appropriate amount of saline has been injected. A slow loss of fluid may indicate a small insult to the joint, and saline can be left in the joint for a few minutes to observe

Small traumatic joint penetration can be difficult to diagnose clinically. The SA test can help confirm the diagnosis. To be a sensitive test, it must be performed using an adequate amount of saline infusion to truly “load” the joint. If the test is positive, orthopedic consultation is indicated. REFERENCES  c a n

be found on

Arthrocentesis

Complications associated with performing SAs are essentially the same as those for performing arthrocentesis. In addition, some temporary patient discomfort should be assumed, due to joint distention.

Conclusion

● 

Complications

53 

for this. After completion, the fluid should be evacuated for patient comfort. This is generally performed by leaving the original needle in place with a closed stopcock attached, which is then used to aspirate joint saline.

E x p e rt C o n s u lt

985

54

Compartment Syndrome Evaluation Merle A. Carter

VIII 

● 

MUSCULOSKELETAL PROCEDURES

C H A P T E R

986

Open fractures, dislocations, and exposed joints are true orthopaedic emergencies that must be managed aggressively to prevent morbidity and mortality. Even when managed appropriately, these injuries may be further complicated by compartment syndrome, a condition of increased pressure within a limited space resulting in compromised tissue perfusion and, ultimately, dysfunction of neural and muscular structures contained within that space.1 The magnitude of the trauma is usually significant, but compartment syndrome may also develop after seemingly minor injuries, prolonged proximal arterial occlusion, or prolonged external pressure in the absence of acute injury. Causes of compartment syndrome are categorized into those that decrease compartment volume capacity, those that increase compartment contents, and those that create externally applied pressure1 (Table 54–1). Subtleties in the early presentation of compartment syndrome or other clinical priorities render some cases simply impossible to recognize and treat early enough to thwart ultimate disability. This is particularly true in uncooperative, comatose, or critically injured patients. Unfortunately, the vagaries of the clinical scenario result in failure to recognize the early signs and symptoms of compartment syndrome might have severe and irreversible limb- or life-threatening consequences. Numerous drugs and toxins have been reported to cause rhabdomyolysis, possibly owing to a direct effect or secondary to agitation and exertion, with the theoretical potential for the development of compartment syndrome (Fig. 54–1). This list is exhaustive but includes heroin, various hydrocarbons, cocaine, amphetamines, antidepressants, antipsychotics, salicylates, propoxyphene, nonsteroidal anti-inflammatory drugs (NSAIDs), succinylcholine, human immunodeficiency virus (HIV) medications, antimetabolites/cancer drugs, antimalarials, diphenhydramine, baclofen, ecstasy, ethanol, anticoagulants/thrombolytics, strychnine, statins, and phenothiazines.2,3 Although compartment syndrome is essentially a clinical diagnosis, objective measurement of compartment tissue pressure may help confirm the diagnosis and determine whether operative treatment is required. This chapter discusses the indications, complications, and interpretation of compartment pressure monitoring. The equipment and techniques required to perform compartment pressure monitoring are described.

BACKGROUND Although recognized as a clinical syndrome in the mid-19th century, the pathophysiology of extremity ischemia was not fully described until more than a century later. Postischemic myoneural dysfunction and its associated contractures were

first described in the 1870s by German surgeon Richard von Volkmann4 who recognized the effects of increased pressure causing vascular compromise of the limb. In 1935, Henderson and coworkers5 developed a simple technique for measuring muscle “tonus” involving a syringe, a three-way connection, a mercury manometer, and a straight large-bore needle placed directly into the muscle. Forty years later, Whitesides and colleagues6 refined this technique to accurately reflect muscle compartment pressures and delineated the threshold at which fasciotomy is indicated. To improve the accuracy and reproducibility of intermittent pressure measurement, Matsen7 refined Whitesides’ simple needle technique by adding a constant-infusion pump, which permitted continuous pressure monitoring. Whereas this appeared to provide more accurate measurements, concerns were raised over the injection of fluid into an already compromised compartment. The wick catheter8,9 and the slit catheter10,11 were later introduced to reduce the risk of increasing compartment contents while providing a mechanism for continuous pressure monitoring. Various needles and equipment have been developed to measure compartment pressure (Fig. 54–2). The wick catheter, originally developed to measure subcutaneous and brain tissue pressures, was modified during the mid-1970s to provide continuous compartment pressure measurements. This catheter is rarely used today because of fears of catheter breakdown leading to measurement errors and retained foreign bodies, prompting the development of the slit catheter in 198010,11 (see Fig. 54–2C). This catheter has slits at the end being inserted into the tissue. The proximal end of the catheter is connected to a transducer and infusion system permitting continuous monitoring. The slits help prevent catheter clogging. Early reports found both methods to have similar accuracy and reproducibility as long as the patency of the catheter was ensured.11,12 In emergency medicine, the Stryker 295-2 Intracompartmental Compartment Pressure Monitoring System (Kalamazoo, MI) has become the most commonly used commercially available handheld device to measure compartment pressures. This device uses a fluid-filled pressure measurement catheter, a pressure monitor, and a fluid infusion mechanism that maintains catheter patency ensuring measurement accuracy. In contrast to earlier devices that utilized fluid injection to measure compartment pressures, the Stryker system uses a minimal amount of saline (>0.3 mL). This helps ensure accurate measurements and reduces the chance of causing a further increase in compartment pressure. The Stryker system also has the ability to record a single measurement or provide continuous compartment pressure recordings when required (Fig. 54–3). Noninfusion systems like the transducer-tipped fiberoptic system offer a distinct advantage over the conventional fluid-filled systems because they do not produce hydrostatic pressure artifacts or require injections of fluid for long-term or continuous measurements. However, the fiberoptic transducer is relatively large and must be attached to a sheath 2.1 mm in diameter likely to cause pain during measurements.13 In recent years, noninvasive, less painful methods for measuring compartment pressures have been studied in both acute and chronic exertional compartment syndromes. Investigations of magnetic resonance imaging (MRI), single-photon emission computed tomography (SPECT), myotonometry, electromyography, near-infrared spectroscopy, and ultrasound have provided encouraging results in the evaluation of

Increased Compartmental Content Bleeding Major vascular injury Coagulation defect Bleeding disorder Anticoagulation therapy Thrombolytic therapy Postarterial line placement Increased Capillary Filtration Reperfusion after ischemia Arterial bypass grafting Embolectomy Ergotamine ingestion Cardiac catheterization Lying on limb Trauma Fracture Contusion Intensive use of muscles Exercise Seizures Eclampsia Tetany Burns Thermal Electric Intra-arterial drug injection Cold Orthopaedic surgery Tibial osteotomy Hauser procedure Reduction and internal fixation of fractures Snakebite Increased Capillary Pressure Intensive use of muscles Venous obstruction Phlegmasia cerullea dolens Ill-fitting leg brace Venous ligation Diminished Serum Osmolarity, Nephrotic Syndrome Other Causes of Increased Compartmental Contents Infiltrated infusion Pressure transfusion Leaky dialysis cannula Muscle hypertrophy Popliteal cyst Carbon monoxide poisoning Externally Applied Pressure Tight casts, dressings, or air splints Lying on limb Pneumatic antishock garment Congenital bands Modified from Matsen FA: Compartmental Syndromes. New York, Grune & Stratton, 1980.

Compartment syndrome evaluation

Closure of fascial defects Application of excessive traction to fractured limbs

● 

Decreased Compartmental Volume

compartment syndrome.14–22 In addition, externally applied devices, which measure muscle tissue “hardness,” are under investigation as an economic alternative to these modalities, although support of their use has been mixed.23–25 Although promising, these evolving noninvasive methods have not yet replaced the needle-driven techniques as the standard for measuring intracompartmental pressures. The remainder of this chapter describes the most commonly employed measurement techniques used in the acute setting in which compartment syndrome of an extremity is suspected. Each method described provides rapid measurements with reasonable accuracy. The method chosen will depend upon the availability of the supplies and equipment necessary for the procedure and the experience of the operator.

54 

TABLE 54–1  Etiologies of Compartment Syndrome

PATHOPHYSIOLOGY Several theories have been proposed to account for the tissue ischemia associated with compartment syndrome. These include the “arteriovenous (AV) gradient” theory, which suggests that reduced AV pressure or perfusion gradients prevents adequate blood supply;26 the “critical closure” theory, in which blood flow arrests well before the AV perfusion gradient declines to zero;27,28 and the “venous occlusion” theory,29 which states that externally applied pressure, thrombotic events, and reperfusion contribute to increased compartment pressures and, ultimately, tissue ischemia. Although the exact mechanism has not been agreed upon, inherent in each of these theories is a decrease in blood supply insufficient to meet the metabolic demands of the involved tissues. Adequate blood flow to tissues is a function of AV gradients across capillary beds. Once reduced below a critical level, oxygen delivery to these structures is impaired and aerobic cellular metabolism is no longer possible. Anaerobic metabolism then ensues until these energy stores become depleted. Muscles then become ischemic, and reduction in venous and lymphatic drainage creates increased pressure within this confined space. Note that ischemia and necrosis of the musculature can occur despite an arterial pressure high enough to produce pulses so merely assessing distal pulses is insufficient.30 A drop in blood pressure, an increase in compartment pressure, or a combination of the two can reduce AV gradients and lead to insufficient blood flow to tissues. Hypotension can occur from a variety of problems including hypovolemia, acute blood loss, cardiac disease states (e.g., ischemia), and sepsis. An increase in the contents of a compartment, a decrease in its volume capacity, or an increase in the external pressure applied to the compartment will increase the pressure within it, also reducing the AV gradient. Thus, the relationship between the intracompartmental pressure and the circulatory status of the extremity is an important factor in the development of compartment syndrome.31 Compartment syndrome may develop in an extremity in the absence of direct trauma from prolonged ischemia associated with acute arterial occlusion by thrombus or proximal arterial injury. The pressure of normal skeletal muscle at rest is typically below 10 mm Hg. However, deviations of 2 to 6 mm Hg have been reported.1,8–10,12 The perfusion pressure of a compartment is defined as the difference between the diastolic blood pressure and the intracompartmental pressure.32 A model using legs of normal volunteers has shown that a progression of neuromuscular deficits occurs when intracompartmental

987

MUSCULOSKELETAL PROCEDURES ● 

VIII 

A

B

D 988

C

Figure 54–1  The diagnosis of compartment syndrome is not always straightforward. A, This man presented in a coma from a heroin overdose and had been lying on his arm for a number of hours. He was hypotensive, in renal failure, comatose, and on a ventilator. The entire arm was swollen and rhabdomyolysis was correctly suspected. B, Clear urine strongly positive dipstick for blood (arrow) and no red blood cells (RBCs) by microscopy equates to myoglobinuria. Because of the coma, he was unable to voice any complaint of pain. C, When he awakened 20 hr later, the pain was severe, and compartment pressures demonstrated the need for fasciotomy. Heroin can cause rhabdomyolysis, and hypotension/reperfusion, and certainly prolonged pressure on the muscles may have exacerbated the condition. D, The classic wringer washer injury predisposes to compartment syndrome, but industrial rollers are now usually the culprit.

pressure rises to within 35 to 40 mm Hg of the diastolic blood pressure.33 Above this level, tissue perfusion is interrupted. Studies of neuromuscular tissue ischemia have demonstrated that inflammatory necrosis can occur at pressures between 40 and 60 mm Hg.34 Whitesides and colleagues6 demonstrated that when tissue pressure within a closed compartment rises to within 10 to 30 mm Hg of the patient’s diastolic blood pressure, inadequate perfusion ensues resulting in relative ischemia of the involved limb. Heppenstall and associates35 further clarified this relationship demonstrating that the difference (ΔP)

between the mean arterial pressure (MAP) and the measured compartment pressure is directly related to blood flow to the tissue. They noted that as compartment pressure increases to approach the MAP, the ΔP decreases. Once ΔP falls below 30 mm Hg, tissue ischemia becomes more likely. In normal musculature, a ΔP of less than 30 mm Hg results in the loss of normal aerobic cellular metabolism.35 In traumatized muscle, a ΔP of less than 40 mm Hg was associated with abnormal cellular function, highlighting the importance of maintaining an adequate systemic blood pressure in the setting of neuromuscular injury.35

CLINICAL PRESENTATION

B

C Figure 54–2  A, An 18-gauge straight needle. B, An 18-gauge side-port needle. C, A slit catheter. (A–C, Boody AR, Wongworawat MD: Accuracy in the measurement of compartment pressures: A comparison of three commonly used devices. J Bone Joint Surg Am 87:2415, 2005.)

B A Figure 54–3  The Stryker 295 intracompartmental pressure monitor system. (From Custalow CB. Color Atlas of Emergency Department Procedures. Philadelphia: Elsevier Saunders, 2005; p. 28.)

For years, conventional wisdom maintained that immediate reperfusion of traumatized tissue would provide improved motor and neurologic function after injury. In the last decade, research suggests that muscle tissue may remain viable even after prolonged periods of ischemia and that a substantial percentage of the injury is generated upon reperfusion.36 Tissue acidosis, intra- and extracellular edema, free radical– mediated injury, loss of adenine nucleotide precursors, and interruption of mitochondrial oxidative rephosphorylation by increased intracellular calcium have been implicated in the development of reperfusion-associated compartment syndromes.36–39 Even in the absence of local trauma, ischemia followed by reperfusion has been shown to increase compartment pressures in canine models of hypovolemic shock.40 Evidence also suggests that elevated compartment pressure itself (in addition to causing ischemia) plays a role in the cellular deterioration seen with compartment syndrome.41 In a study comparing muscle ischemia caused by tourniquet placement with an experimentally derived high-pressure compartment syndrome, there was no difference in the degree to which phosphocreatine levels fell between groups. However, levels of adenosine triphosphate (ATP) diminished rapidly in the compartment syndrome group compared with the tour-

Any compartment limited by fascial planes is potentially at risk for compartment syndrome. However, because of their propensity for injury and the presence of several low-volume compartments, the lower extremities are most commonly affected. In the leg, the anterior compartment is involved most often,42 whereas the posterior compartment is a site frequently missed. The hands, feet, forearms, upper arms, thighs, abdomen, gluteal musculature, and back are other locations where compartment syndrome is known to occur.1 Increased compartment pressures may be caused by a variety of conditions (see Table 54–1). Risk factors for developing a compartment syndrome include recent extremity trauma (including acupuncture,43 venipuncture, intravenous infusions, or intravenous drug use), bleeding within an extremity, a restrictive cast or splint, a crush or compression injury, prolonged lithotomy position,44–47 tourniquet placement during an operative procedure, or a circumferential burn. In addition, some evidence suggests that compartment syndrome may occur in the setting of chronic exertion and overuse.48,49 Although the exact etiology remains elusive, studies have demonstrated elevated lactate concentration and water levels in the tibialis anterior muscle after exercise with a reduction in these after fasciotomy.50,51 Increases in muscle mass (related to a rise in blood volume during exertion) and hypertrophy of muscle and fascia with chronic use have also been reported.52–56 Clinical hallmarks of compartment syndrome include pallor of the extremity, a pulse deficit compared with that in the opposite limb, paresis or paralysis of the involved extremity, paresthesias in the distribution of the involved nerves, and pain on passive stretch of the involved musculature (note: the 5 Ps). These signs and symptoms may be unreliable in pediatric populations.57 In addition, alhtough commonly seen, pain and paralysis are late findings. Early, more subtle, signs of compartment syndrome include a burning sensation over the involved compartment, nonspecific sensory deficits, or poorly localized deep muscular pain that seems out of proportion to clinical examination and that intensifies when the musculature is passively stretched. The period between the injury and the onset of symptoms can be as short as 2 hours and as long as 6 days.58 The peak interval appears to be 15 to 30 hours. Often, the first symptom described by patients is pain greater than expected given the clinical scenario. Although pain out of proportion to the visible injury may raise the question of drug-seeking behavior, a focused evaluation for the possibility of limbthreatening disorders must precede this diagnosis of exclusion. Physical examination may reveal muscles that are weak and tense, with hypesthesia in the distribution of the nerves involved. Sensory deficits are often an indicator of the involved compartment, including loss of two-point discrimination and decreased vibratory sensation.59–61 The presence or absence of a palpable arterial pulse is not an accurate indicator of relative tissue pressure or the relative risk of developing compartment syndrome. Pulses may be present in a severely compromised

Compartment syndrome evaluation

A

● 

niquet group. Moreover, phosphocreatine levels, ATP, and tissue pH normalized within 15 minutes of releasing the tourniquet. In the compartment syndrome group, these levels remained low even after fasciotomy. These results suggest that elevated tissue pressure plays a synergistic role with ischemia in cellular deterioration.41

54 

Transducer tips

989

Compartment

Sensory Loss

Muscles Weakened

Painful Passive Motion

Tenseness Location

Forearm Dorsal Volar

— Ulnar/median nerves

Digital extensors Digital flexors

Digital flexion Digital extension

Dorsal forearm Volar forearm

—

Interosseus

Abduct/adduct (metacarpophalangeal joints)

Dorsum of hand between metacarpals

Anterior

Deep peroneal nerve

Toe flexion

Anterior aspect leg

Superficial posterior Deep posterior

— Posterior tibial nerve

Foot dorsiflexion Toe extension

Gluteal

(Rarely sciatic)

Toe extensors Tibialis anterior Soleus and gastrocnemius Toe flexors Tibialis posterior Gluteals, piriformis, or tensor fascia lata

Hip flexion

Calf Distal medial leg, between Achilles tendon and tibia Buttock

Biceps and distal flexors Triceps and forearm extensors Foot intrinsics Erector spinae

Elbow extension Elbow flexion Toe flexion/extension Lumbar flexion

Anterior upper arm Posterior upper arm Dorsal/plantar foot Paraspinous

Hand Interosseus

VIII 

● 

MUSCULOSKELETAL PROCEDURES

TABLE 54–2  Compartment Syndromes and Associated Physical Signs

Leg

Upper Arm Flexor Extensor Foot Lumbar

Ulnar/median nerves Radial nerves Digital nerves —

extremity.62 Table 54–2 lists the signs and symptoms of compartment syndrome specific to each compartment. 990

DIAGNOSIS Even experienced clinicians find it difficult to evaluate a potential compartment syndrome, and no specific standard of care exists with regard to a time interval from injury to definitive treatment. In the unconscious patient or for those with other life-threatening conditions that mandate other priorities, the clinical scenario simply does not allow for a diagnosis to be made in a timely fashion. In the nontrauma case or for the patient unable to voice pain or cooperate with an examination, this diagnosis is often not considered or the diagnosis is delayed. Regional nerve blocks or epidural anesthesia might obscure signs or symptoms of increased compartment pressure, causing further delays in diagnosis. The difficulty in diagnosing an acute compartment syndrome was highlighted in a report by Vaillancourt and coworkers.63 In a retrospective review of 76 patients who underwent fasciotomy at major university trauma centers/ teaching hospitals, the interval from initial patient assessment to diagnosis of a compartment syndrome was up to 8 hours. As one would intuit, delay was most common in nontraumatic cases. The interval from the precipitating event to definitive surgery was up to 35 hours, reflecting the difficulty in suspecting this diagnosis and instituting definitive therapy in clinical practice. Such statistics describe actual care that may be less than ideal with regard to theoretical benchmarks. Not withstanding the difficulty described previously, the diagnosis of compartment syndrome is primarily a clinical one, supplemented by direct measurement of compartment pressures. In a study evaluating of the utility of clinical findings in the diagnosis of compartment syndrome, Ulmer64 noted that the sensitivity and positive predictive value of clini-

cal findings (see later) are low, whereas the specificity and negative predictive value of these findings are high. Never­ theless, the study found that whereas the sensitivity of an individual clinical finding may be low, the probability of compartment syndrome rises considerably when more than one clinical hallmark is present.64 The differential diagnosis of compartment syndrome is extensive and includes primary vascular, nerve, or muscle injuries that produce similar findings. Acute arterial occlusion, cellulitis, osteomyelitis, neuropraxia, reflex sympathetic dystrophy, synovitis, tenosynovitis, stress fractures, envenomations, necrotizing fasciitis, deep vein thrombosis, and thrombophlebitis are additional diseases that should be considered. Differentiating compartment syndrome from these and other orthopaedic disorders requires a detailed history and thorough physical examination with a high index of suspicion (Table 54–3).

ANCILLARY STUDIES In general, laboratory and radiographic studies are not help­­ ful in confirming the diagnosis of compartment syndrome. However, they might be useful in identifying other diagnoses, associated conditions, and complications. Table 54–4 lists useful studies for patients with the possibility of a compartment syndrome.

INVASIVE COMPARTMENT PRESSURE MONITORING Indications and Contraindications The earliest objective manifestation of acute compartment syndrome is an elevation in the tissue pressure of one or more compartments. However, signs and symptoms generally do

54 

TABLE 54–3  Clinical Findings of Compartment Syndrome, Arterial Occlusion, and Neuropraxia

● 

Arterial Occlusion

Neuropraxia

+ + + + +

− + + + −

− − + + +

Pressure increased in the compartment Pain with stretch Paresthesia or anesthesia Paresis or paralysis Pulses intact

From Mubarak S, Carroll N: Volkman’s contracture in children: Etiology and prevention. J Bone Joint Surg Br 61:290, 1979.

TABLE 54–4  Ancillary Studies That May Be Helpful in Identifying Other Diagnoses, Associated Conditions, and Complications in Patients Suspected of Having a Compartment Syndrome Laboratory Studies • Complete metabolic profile (including electrolytes and renal function testing) • Complete blood count with differential • Serum and urine myoglobin • Creatine phosphokinase • Urinalysis to evaluate for concurrent rhabdomyolysis • Coagulation studies Imaging Studies • Radiography of the affected limb to evaluate for fracture or foreign body • Ultrasonography to rule out deep vein thrombosis or Doppler ultrasonography to evaluate blood flow to the extremity

not occur until the tissue pressure has reached a critical level (see “Pathophysiology”). In some patients, the diagnosis of compartment syndrome is clinically obvious, and one can proceed directly to fasciotomy. However, when clinical findings are equivocal or difficult to interpret, tissue pressure measurement may help guide treatment (Fig. 54–4). It is important to remember that whereas tissue pressure measurements may suggest a compartment syndrome, equivocal measurements will still require clinical judgment. There are several groups of patients in whom clinical findings are difficult to interpret and would benefit from compartment pressure measurement. These include unresponsive patients, uncooperative patients, children, patients with multiple or distracting injuries, those with peripheral nerve deficits attributable to other causes (e.g., fracture-associated nerve injuries, diabetic peripheral neuropathies), and those whose clinical findings are equivocal. There are no absolute contraindications to performing compartment pressure measurements or continuous pressure monitoring. Caution should be taken when performing these procedures on patients with platelet dysfunction or other coagulation disorders. If possible, avoid needle insertion through areas of cellulitis, infection, or burns.

Patient Preparation and Positioning Explain the procedure to the patient or surrogate. Written informed consent is not a universal standard, but is suggested when possible. Patient and extremity positioning for compartment pressure measurement depends upon the extremity and

SUSPECTED COMPARTMENT SYNDROME

Unequivocally positive clinical findings

Compartment syndrome evaluation

Compartmental Syndrome

Patient not alert/ unreliable polytrauma victim; inconclusive clinical findings

Compartment pressure measurement

≥30 mm Hg*

10 mL) indicates a probable ruptured ovarian cyst, aspiration of an intact corpus luteal cyst, ascites, or possibly, carcinoma. The significance of these fluids and the interpretation of results are outlined in Tables 57–4 and 57–5. Elliot and colleagues22 cautioned that obtaining greater than 10 mL of clear fluid should not automatically rule out an ectopic pregnancy because the latter may coexist with other pathologic conditions. A “positive tap” is one in which nonclotting blood is obtained, although the presence of nonclotted blood does not confirm a tubal pregnancy. Intraperitoneal blood from any source (ectopic pregnancy, ovarian cyst, ruptured spleen) may remain unclotted after aspiration for days in the syringe as

TABLE 57–5  Interpretation of Culdocentesis

Aspirated Fluid

Condition and Suggested Differential Diagnosis

Positive

Clear, serous, strawcolored (usually only a few milliliters) Large amount of clear fluid

Normal peritoneal fluid

Exudate with polymorphonuclear leukocytes Purulent fluid

Bright red blood*

IX 

● 

GENITOURINARY, OBSTETRIC, AND GYNECOLOGIC PROCEDURES

TABLE 57–4  Interpretation of Culdocentesis Fluid

Old, brown, nonclotting blood

1068

Ruptured or large ovarian cyst (fluid may be serosanguineous); pregnancy may be coexistent Ascites Carcinoma Pelvic inflammatory disease Gonococcal salpingitis Chronic salpingitis Bacterial infection Tubo-ovarian abscess with rupture Appendicitis with rupture Diverticulitis with perforation Ruptured viscus or vascular injury Recently bleeding ectopic pregnancy* (ruptured or unruptured) Bleeding corpus luteum Intra-abdominal injury Liver Spleen Other organs Ruptured aortic aneurysm Ruptured viscus Ectopic pregnancy with intraperitoneal bleeding over a few days or weeks Old (days) intra-abdominal injury (e.g., delayed splenic rupture)

*Note: The hematocrit of blood from a ruptured ectopic pregnancy is usually 15% or greater (97.5% of cases), but some authors use greater than 3% as positive.

a result of the defibrination activity of the peritoneum. The return of a serosanguineous fluid also suggests a ruptured ovarian cyst. The hematocrit of blood from active intraperitoneal bleeding is greater than 10%. In one series, the hematocrit of blood from a ruptured ectopic pregnancy was 15% or greater in 97% of cases.6 It should be emphasized that a positive culdocentesis in the presence of a positive pregnancy test does not always prove an ectopic pregnancy.16 A ruptured corpus luteum cyst in the presence of an intrauterine pregnancy test is probably the most common cause of a false-positive scenario. When possible, ultrasound may help corroborate the culdocentesis findings.

>0.5 mL nonclotting, bloody fluid (hematocrit > 12%) Indicates hemoperitoneum When β-hCG also positive, ectopic pregnancy found in greater than 95% Nonspecific—can occur in intrauterine pregnancies and nonpregnant women (e.g., ruptured cyst, retrograde bleeding) Does not necessarily indicate tubal rupture 50%–62% of ectopic pregnancies with peritoneal blood may be unruptured Negative Serous fluid Excludes hemoperitoneum and tubal rupture False negative in 10%–15% of ectopic pregnancies (generally unruptured) Nondiagnostic Dry tap or clotting blood Excludes neither ectopic pregnancy nor hemoperitoneum 15% of procedures are nondiagnostic 16% of ectopic pregnancies have nondiagnostic study results β-hCG, β subunit human chorionic gonadotropin. From Brennan DF: Ectopic pregnancy: II. Diagnostic procedures and imaging. Acad Emerg Med 2:1090, 1995.

COMPLICATIONS Culdocentesis is one of the safest procedures performed in the emergency setting, and there are probably fewer complications with this technique than with peripheral venous cannulation. Complications have been reported, the most serious being rupture of an unsuspected tubo-ovarian abscess.20 Other complications include perforation of the bowel, perforation of a pelvic kidney, and bleeding from the puncture site in patients with clotting disorders. Because the most common complications result from the puncture of a pelvic mass, careful bimanual examination of the patient should help prevent this problem. Puncture of the bowel and the uterine wall occurs relatively frequently, but this does not generally result in serious morbidity. Obviously, penetration of the gravid uterus has greater potential for harm. Occasionally, one will aspirate air or fecal matter, confirming inadvertent puncture of the rectum. Although this may be disconcerting, it is seldom of serious clinical concern and requires no immediate change in therapy. REFERENCES  c a n

be found on

E x p e rt C o n s u lt

● 

58

58 

C H A P T E R

in improving the prosecution and conviction of sexual predators.

Examination of the sexual assault victim

Examination of the Sexual Assault Victim

EVALUATION AND TREATMENT OF PATIENTS SUFFERING FROM SEXUAL ASSAULT Preparation

Carolyn Sachs and Malinda Wheeler

In most places, local jurisdictions or hospitals provide clinicians with detailed forms and instructions for the examination and documentation of sexual assault. This chapter is meant to supplement these instructions and forms. Clinicians should familiarize themselves with such local documents before performing a sexual assault examination. Careful step-by-step planning, using written protocols to guide the way in which a victim is handled in the ED and in follow-up, helps both to ensure the best care for the victim and to aid in the prosecution and conviction of assailants. The ED must secure patient privacy and designate a separate area for the care of sexually assaulted patients. If medically and logistically possible, interview the victim in a private room separate from the examination room. EDs often have such an area frequently called the “grieving room” or the “family room.” Many legal jurisdictions provide examination kits for the collection of forensic evidence from the victims. These kits should be available in the ED, and the staff should be familiar with them. If such kits are not provided by local jurisdiction, hospital staff may need to assemble their own kits from materials found in most EDs. Alter­ natively, private companies assemble and sell such kits (www.lynnpeavey.com or The Lynn Peavey Company, PO Box 14100, Lenexa, KS 66285-4100). Prepared kits save a 1069 tremendous amount of nursing and clinician time when a victim comes to the ED. A checklist for local requirements for sexual assault examinations should be included in the kits and serves as a reminder for all of the medicolegal procedures to be completed. Although this chapter is primarily devoted to the evaluation of the adult female sexual assault victim, guidelines for the evaluation of the adult male sexual assault victim, the female child victim, the male child victim, and the accused assailant are provided in separate sections of this chapter. The same examiners designated to perform adult female examinations may easily perform male victim and assailant examinations; however, the examination of the child sexual assault victim often requires considerable expertise and training. When possible, medical staff with extra training in the examination of the child sexual assault victim should perform these examinations. If this is not possible, the special section of this chapter should provide emergency medical personnel with a framework to perform an initial examination.

Despite a decrease in violent crime overall, sexual assault remains a significant societal problem that affects hundreds of thousands of American men and women yearly, and millions worldwide. Approximately 18% of U.S. women and 3% of U.S. men experience attempted or completed rape sometime in their lives. This correlates to a 1-year incidence of rape of 876,100 for women/yr and 111,300 men/yr.1 The majority of victims do not report the assault to anyone. Victims reported an average of one third of sexual assaults to law enforcement. After reporting to law enforcement, sexual assault victims may be transported to the emergency department (ED) for evaluation, examination, and treatment. Sexual assault victims may also present to the ED for treatment without prior contact with law enforcement. Victims are usually willing to cooperate with police investigation; others are not. Many states have laws that require medical personnel treating sexual assault victims to report the assault to local law enforcement. Clinicians must know their own state laws regarding this.

DEFINITIONS Although many use the term synonymously with rape, sexual assault more accurately refers to any sexual contact of one person with another without appropriate legal consent.2 Physical force may be used to overcome the victim’s lack of consent, but this is not mandatory to prove assault. Lack of consent for sexual contact by intimidation, threats, or fear equals sexual assault. State law differs slightly on the definition of exact acts that constitute sexual contact and on which populations are unable to give legal consent. In general, persons under the influence of drugs or alcohol, minors, and persons who are mentally incapacitated are deemed unable to give consent for sexual contact. Clinicians who treat sexual assault victims have a professional, ethical, and moral responsibility to provide the best medical and psychological care possible. At the same time, they must collect and preserve the proper medicolegal evidence that is unique to the evaluation of sexual assault cases. Many hospitals and jurisdictions are affiliated with designated sexual assault examination teams that provide specialized evaluation and treatment for victims. These sexual assault response teams (SARTs) provide clear advantages, which are outlined toward the end of the chapter. However, victims may be brought to an ED that does not routinely provide specialized care for sexual assault. This chapter is designed to aid clinicians in such a general care location. Prepared emergency personnel can help attenuate the psychological and physical impact of sexual assault. Through proper care of the victim and careful acquisition of evidence, ED staff can help the victim to recover from the assault and can aid society

Consent Consent for the treatment of a sexual assault victim is mandatory. The victim has undergone an experience in which her right to grant or deny consent was taken from her, and obtaining consent for medical treatment and for the gathering of evidence has important psychological and legal implications. The victim has the right to decline medicolegal examination and even medical treatment. Before beginning evaluation and treatment, obtain witnessed, written, informed consent. If there are no local forensic examination forms, use the standard ED “consent to treat” forms, but make sure that the

GENITOURINARY, OBSTETRIC, AND GYNECOLOGIC PROCEDURES ● 

IX 

patient is well informed and gives her verbal consent to each step of the examination. Although some states mandate that medical personnel report sexual assaults to law enforcement, victims may decline to discuss the event with police. If the victim cannot give consent for a forensic examination owing to a reversible process (e.g., intoxication, an acute psychological reaction), wait several hours for the victim’s mental status to improve to a reasonable level for consent to be obtained. When victims cannot give consent owing to minor status or developmental disability, the person authorized to give medical consent for the patient may consent for the examination unless he or she is a suspect in the assault. Many states allow an adolescent victim of a certain age (e.g., >12–14 yr) to consent to an examination for conditions related to sexually transmitted diseases (STDs), sexual assault, and pregnancy. State laws also differ in examiners’ requirements to make an attempt to contact the legal guardian (unless he or she is a suspected perpetrator). Clearly, emergency personnel must be informed regarding their local laws concerning these requirements. In the rare case that a victim cannot give consent owing to a potentially irreversible medical condition, such as severe head trauma and coma, seek the advice of institutional legal council before proceeding with a forensic examination. In some cases, the next of kin may provide the needed consent, whereas in others, it may be necessary to obtain a court order to proceed.

History The history of the event should include only those elements necessary to complete required forms, to perform a focused 1070 physical examination, and to collect evidence. Questions beyond this, such as the details leading up to the assault, should be left to the police investigators. Avoid the urge to “help” the alleged victim by unduly embellishing or detailing uncorroborated or nonmedical information supplied during the examination. Limiting the history not only shortens the evaluation in the ED but also helps to prevent discrepancies between the ED history and the official police investigation report, which could weaken the victim’s case in court. Document pertinent medical history including last menstrual period, current contraception, recent anal-genital injuries or surgeries, and preexisting injuries. The history of the event required by legal forms and/or protocol usually includes the time, date, and place of the alleged assault and a description of the use of force, threats of force, and the type of assault. Elements of force may include the type of violence used (e.g., grabbing, hitting, kicking, strangling, weapon use), threats of violence, the use of restraints, the number of assailants, the use of alcohol or drugs (forcibly or willingly) by the victim, and any loss of consciousness experienced by the victim. Sexually assaultive acts may include fondling (of breasts and/or genitalia); vaginal, oral, or anal penetration or attempted penetration (with fingers, penis, and/or objects); ejaculation on or in the body; and the use of a condom. The use of physical force or violence is partly a police matter, but from a medical standpoint, it is desirable to correlate positive findings on the physical examination (e.g., abrasions, ecchymosis, scratches) with a description of any force, restraint, or violence. Document postassault activity commonly requested by forms including douching, bathing, urinating, defecating, gargling, or brushing teeth. These can alter the recovery of seminal specimens and other sexual assault evidence. In

TABLE 58–1  Maximal Reported Time Intervals for Sperm Recovery Body Cavity Vagina Cervix Mouth Rectum Anus

Motile Sperm

Nonmotile Sperm

6–28 hr 3–7 days — — —

14 hr–10 days 7.5–19 days 2–31 hr 4–113 hr 2–44 hr

From Marx J (ed): Rosen’s Emergency Medicine: Concepts and Clinical Practice, 6th ed. Philadelphia, Elsevier, 2006.

addition, question victims about potential injuries from any preassault bodily trauma. Elements of the victim’s history should help in deciding which potential samples to collect. For example, sperm may be recovered after intercourse from the cervix for up to 12 days and from the vagina for 5 days (Table 58–1).3 If the victim had voluntary intercourse 48 hours before the examination and was sexually assaulted 3 hours before the examination, obtain samples from both the vagina and the cervix, and keep the two specimens separate. Taking a careful history makes it possible to perform an appropriate examination given these two separate events. In general, cervical swabs should be collected in addition to the usual vaginal swabs if the time between assault and examination is greater than 48 hours or if intercourse with a different person took place within a few days of the assault. Obtain a gynecologic history in preparation for injury documentation and treatment plans. From a medicolegal standpoint, question victims about any recent gynecologic surgical procedures or unintentional genital trauma that might alter the expected normal genital appearance. The history should also include the use of any method of birth control before the attack (with information regarding any missed birth control pills), last normal menstrual period, last voluntary intercourse, gravidity and parity, and recent STDs. As with all assaulted patients, the medical history should include current medications, tetanus immunization status, and allergies. While taking the history, observe the patient’s ability to understand and respond appropriately to questions. Victims of sexual assault may not possess the capacity to consent to intercourse because of developmental disability, young age, or intoxication with drugs or alcohol. Consider obtaining blood, urine, or both, and testing for drugs or alcohol when the history suggests lapses of (or impaired) consciousness. Most often, victims who lack consenting capacity owing to developmental disability will have sufficient prior documen­ tation of the condition. In the rare instance in which an examiner suspects previously undocumented developmental disability, formal examination of the patient’s mental capacity can be assessed at a later time by request of the district attorney.

Physical Examination The physical examination of the sexual assault victim differs from most other ED examinations because examiners are not only caring for a patient’s physical and mental well being but also investigating a crime scene and collecting specific evidence. Remember to explain every step of the examination to

Imaging Photographs can be a valuable addition to the documenta­­ tion of bodily injury. Medical institutions may employ pro­ fessional-quality photographic teams; others must rely on law enforcement for photo documentation. Most institutions

Examination of the sexual assault victim

General Body Examination After the patient disrobes and is placed in a gown or other suitable covering, examine her body for signs of trauma and foreign material. Uncover one part of the body at a time to examine and then carefully re-cover it. This allows the victim to retain some modesty during the examination. Important areas for evaluation are the back, the thighs, the breasts, the wrists, and the ankles (particularly if restraints were used). Even in the absence of ecchymosis, note tender areas during the examination. Leaves, grass, sand, and other materials can occasionally be found in the hair or on the skin. Retain these materials as evidence. Document areas of trauma and evaluate further (e.g., with radiographs) as indicated by the type and extent of injury. Approximately 10% to 67% of sexual assault victims display bodily injuries.4 Document these bodily injuries because they correlate significantly with successful perpetrator prosecution.5 Bodily evidence may range from abrasions to major blunt and penetrating trauma. If the victim has not bathed, bodily evidence in the form of dried semen stains may be visible on the hair or the skin of the victim. In a darkened room, dried semen (and, unfortunately, many other substances) on skin may fluoresce under examination with shortwave light, such as that produced by a Wood’s lamp or an alternative light source (ALS)6 but may be noticed equally well by its reflective appearance under regular room lighting.7 Use moistened swabs to collect potential dried secretions; then air-dry them thoroughly and preserve as evidence. The underside of fingernails can also contain evidence. Rape victims may have fragments of the assailant’s skin, blood, facial hair, or other foreign material from the rape site beneath their fingernails. Obtain fingernail scrapings by cleaning under a victim’s nails with a toothpick or small swab or by cutting the nails closely over a clean paper. Fold the toothpick and debris into the paper, place it in an envelope, and package it with the other specimens.

● 

Collection of Clothing If not already collected by law enforcement, collect the clothes that the victim wore during the assault for potential evidence. The victim should disrobe by dropping clothes onto a clean sheet or a large clean piece of paper. Using gloved hands, place each item of clothing in a separate paper bag. Label all collected material meticulously and describe it in the chart. Bundle the sheet or paper and any material that might have fallen during the victim’s disrobing and place it in a separate paper bag. When a victim’s clothing must be collected, be sure to provide suitable dress for the victim to wear home after release from the ED.

require patient consent for photographs taken by hospital personnel. Optimally, institutions should arrange a prior plan to handle film or digital media according to a written “chain of custody.” Alternatively, self-developing film (Polaroid) that can be permanently labeled (e.g., subject, date, details of pictured injury) may be used. The photographs should be labeled immediately and may be added to the legal evidence. In some jurisdictions, photographs of physical injuries will be taken and retained by an accompanying law officer. These photographs may serve as evidence or may simply refresh the examiner’s memory at the time of the trial.

58 

the victim. Remind the victim to communicate any discomfort or questions during the examination and to ask for a break from the examination if needed. In addition, remind the victim of her right to decline any portion of the examination and the ability to stop at any point. Each victim should have the opportunity to have a family member, friend, victim advocate, or a combination of these, in the room during all parts of the examination. In some jurisdictions, state law mandates that victims be informed of this right.

Oral Evaluation If indicated by the history, inspect the oral cavity closely for signs of trauma and collect evidence if indicated. Mouth injuries from forced oral copulation include lacerations of the labial or lingual frenulum, mucosal lacerations, and abrasions. Injury to the lips is often produced by the victim’s own teeth as her lips are forced inward with the perpetrator’s penis. Potential injuries to the posterior pharyngeal wall and soft palate include petechiae, contusions, and lacerations. Document these injuries at the time of initial examination because mucosal injuries heal quickly and may not be present hours or days later. Collect potential evidence with swabs rubbed between the teeth and the buccal mucosa on both upper and lower gingival surfaces bilaterally. Spermatozoa have been identified in oral smears for hours after the attack despite toothbrushing, using mouthwash, or drinking various fluids and may show valuable evidence up to 12 hours after examination.8 Collect any foreign material (e.g., hair) to include as potential evidence. During oral inspection, local law enforcement may request that examiners collect buccal cell swabs to 1071 provide the crime laboratory with a sample for victim DNA reference. Genital Examination Once the victim is in the lithotomy position, inspect the thighs and perineum for signs of trauma and for foreign materials, such as seminal stains. Use an ultraviolet light again to look at suspicious dried secretions. Many jurisdictions recommend routine collection of swabs from the external genital area owing to the high likelihood of evidence being present and the inconsistent fluorescence of seminal fluid with the Wood’s lamp. Pubic Hair Samples Before the pelvic examination, comb the victim’s pubic hair for foreign material (particularly pubic hair belonging to the assailant). Place a clean paper below the victim’s buttocks with the victim in the lithotomy position and comb the pubic hair onto the paper. Fold these hairs and the comb into the paper and place them directly into a large paper envelope to be given to law enforcement. Foreign pubic hairs can often provide enough cellular DNA material from the root to enable the crime laboratory to perform DNA analysis. In addition, specialized laboratories possess the capability of performing mitochondrial DNA analysis from the hair shaft in many cases. Significant hair transfer occurs in less than 5% of assaults.9 In the small minority of cases in which foreign suspect hairs must be compared with victim hairs, a pulled victim sample may be desired. Although the routine pulling of the patient’s hair from the roots may provide the best sample, the act of pulling the victim’s hair may be considered insensitive and

Glans penis

Urethral orifice

A

Mons pubis

IX 

Labium majora

Clitoral hood

Labium menora

1072

Scrotal sac

Figure 58–1  The hymen in a pre-pubertal female as seen with inferior labial traction.

● 

GENITOURINARY, OBSTETRIC, AND GYNECOLOGIC PROCEDURES

Penile shaft

Urethral meatus

Vaginal orifice with view of anterior vaginal wall Anus

Penile shaft

Hymen Fossa navicularis Scrotal sac

Posterior fourchette

Foreskin (covering the glans penis)

Figure 58–2  Female anatomy.

unnecessary during the initial evaluation. In addition, these hairs will rarely be needed because the vast majority of cases are never adjudicated, and those that are rarely concern this type of evidence. A victim can provide the hairs at a later time, if needed; often, the victim is willing to pluck the hairs herself at that time. The genital examination of the sexual assault victim differs considerably from most ED pelvic examinations. First, perform a careful evaluation of the vulva and the vaginal introitus for signs of trauma. The techniques of separation and traction move the tissues most likely to suffer injury into view. In performing separation, use both hands to separate the labia laterally in each direction and inspect the posterior fourchette and vaginal introitus. Similarly, in performing tractions, use both hands to grab each labia majora and apply gentle inferior labial traction (i.e., toward the examiner); this gives a much-improved view of the hymen, especially in prepubertal females (Fig. 58–1). If the examiner fails to perform these maneuvers, traumatic genital injuries may be missed. Be familiar with female (Fig. 58–2) and male (Fig. 58–3) genital anatomy, including all terms used to describe these areas. Although most novice examiners concern themselves with detecting injuries to the hymen, the majority of sexual

B Figure 58–3  Male anatomy. A, Circumcised male. B, Uncircumcised male.

assault–related vaginal injuries occur to the posterior fourchette10 (Fig. 58–4). In fact, hymenal injuries are rare in sexually active adult women and are more commonly observed in sexually inexperienced adolescents11,12 (Fig. 58–5). More uncommon injuries to the vaginal walls and cervix may be discovered during the speculum examination. Reported rates of genital injury among forensically examined victims range from 6% to 20% without colposcopy to 53% to 87% with colposcopy.10,12 Most importantly, examiners must be cognizant of the fact that a completely normal genital examination can still be consistent with forced sexual assault. In fact, a study of more than 1000 sexual assault victims found that almost half of all victims with forensic evidence positive for sperm had no genital injury.4

Examination of the sexual assault victim

A

● 

Teixeira13 first described the use of colposcopy for documentation in sexual assault in 1981. Although it is not readily available in most EDs, the use of colposcopy has revolutionized the documentation of injury. The colposcope provides magnification, a bright light source, and usually permanent documentation of injuries in the form of traditional film or digital pictures or video. In one small study, the colposcope increased the rate of genital injury detection from 6% to 53%.14 Colposcopes with photo or video attachments provide excellent photographic documentation for court and allow for expert practitioner review for court testimony without subjecting the victim to reexamination (Fig. 58–6). Experienced sexual assault examiners programs are increasingly using

58 

Colposcopy

B Figure 58–4  Posterior fourchette injuries are the most common site in the adult victim of sexual assault. A, Before toluidine blue application. B, After toluidine blue application.

1073 Figure 58–5  Hymenal injury at the 6 o’clock position (arrow), usually found in the adolescent female. Such injuries are uncommon in adults.

A

B

Figure 58–6  Colposcope (A) and method of examination (B). This technique is not a standard intervention by an emergency physician.

GENITOURINARY, OBSTETRIC, AND GYNECOLOGIC PROCEDURES

high-quality digital cameras mounted on a tripod to obtain excellent images that are indistinguishable from those obtained with colposcopy. ED practitioners often have access to such equipment. Colposcopically visible injuries have also been described in adolescent women after first consensual intercourse; hence, genital injury does not always correlate with nonconsensual vaginal penetration.15 Conversely, a totally normal genital examination by colposcopy is often found after sexual assault. Even in sexually inexperienced adolescents, forced penetration can occur without leaving discernible genital injury.16 Although prior victim sexual experience decreases the likelihood of finding genital injury, experts cannot fully explain the reasons why some rape victims sustain measurable genital injury when others do not.

Forensic Evidence Collection

IX 

● 

Protocols for evidence collection vary in different jurisdictions. Many rape evaluation centers have abandoned the cumbersome rape kits that have been used in the past, substituting simple collection methods that concentrate on important and usable legal evidence. The following discussion is patterned after the model protocol suggested by the state of California and the American College of Emergency Physicians (ACEP) manual.2 During inspection of the external genitalia, rectum, vagina, and cervix, standard forensic specimens should be obtained. Lubricate the speculum with warm water rather than lubricant owing to the potential spermicidal activity of lubricants. However, if lubricants are inadvertently used, the potential for corruption of DNA evidence should be negligi17 1074 ble. Generally, the specimens collected will be determined by victim history and local protocol, but they may include any of the items listed in Table 58–2. Some protocols recommend that examiners make a wet mount of one swab from the vaginal pool and look at it under the microscope for the presence of motile sperm. Because of rapid cell death, studies have shown a negligible chance of finding motile sperm from a vaginal wet mount more than 8 hours after intercourse.18 Furthermore, in complying with Clinical Laboratory Improvement Amendments (CLIA) of 1988, ED practitioners in the United States rarely have sufficient access or experience with microscopy to routinely recommend this step. Take several swabs from the vaginal pool (including the one used to make TABLE 58–2  Potential Evidence Collected Clothing (list and describe all clothing collected on chart and checklist form) Debris Dried secretions, swabs, and slides External genital swabs and slides Pubic hair combings Oral mucosal swabs and slides Rectal swabs and slides Vaginal pool swabs and slides Vaginal lavage fluid in test tube or urine container Tampon or condom present in vagina (dried or frozen) Urine or blood toxicology sample (timed collection) Reference blood, buccal mucosa, and/or hair sample Collect all evidence using gloved hands to avoid DNA contamination Change gloves when necessary to avoid interlocation contamination

the wet mount, if done) and the external genitalia and run them over clean slides for a dry mount. Air-dry, label, and package all swabs and slides in paper envelopes for the local crime laboratory. Some EDs maintain specific equipment (i.e., a Dry Box) to aid the drying of specimens; in others, the swabs and slides must be left out until completely dried. Many crime laboratories also request collection of a vaginal washing. For this procedure, insert 5 mL of sterile (but not bacteriostatic) water or saline into the vagina and then remove it. Place the washing in a sealed container (such as those for urine collection or a red-top blood tube) for later examination for evidence. In addition, collect cervical swabs if the time from assault to examination (the postcoital interval) is greater than 48 hours or if there is a history of recent consensual intercourse as well. The crime laboratory may recover sperm from cervical specimens up to 12 days after coitus.3 Label each sample separately. Record the area from which the specimen was collected in the chart.

Genital Testing for STDs The Centers for Disease Control and Prevention (CDC) guidelines suggest obtaining a (cervical, rectal, or oral) culture or polymerase chain reaction (PCR) specimen, or both, for Chlamydia trachomatis and for Neisseria gonorrheae. However, the majority of SART programs in the United States do not routinely perform these tests.19 STD testing during sexual assault examination can detect only preassault infection and provides no meaningful information for the crime laboratory. In addition, the routine prophylactic treatment with antibiotics effective against N. gonorrheae and C. trachomatis makes detection of these preexisting infections superfluous; however, clinicians might want to consider obtaining cultures on child victims in whom the presence of an STD would be indicative of previous abuse.

Perineal Toluidine Blue Dye Staining Toluidine blue dye is a nuclear stain, often used in cancer detection, that adheres to areas of injury (subepithelial nucleated cells) but not to intact epithelial cells. It adheres to skin where the epidermal layer of non-nucleated cells has been removed (Fig. 58–7). The underlying nucleated cells take up the dye. Although it is not a uniform standard of care, the dye can enhance the examiner’s ability to visualize genital injuries (see Figs. 58–4 and 58–5). Genital lacerations may provide corroborating evidence of nonconsensual intercourse. The test is done before speculum examination or other instrumentation. To outline injuries, apply a 1% aqueous solution of toluidine blue dye to the perineum and wipe excess dye off with a cotton ball moistened with lubricating jelly. A swab containing the dye is commercially available. Some examiners use 1% acetic acid to remove the excess; however, acetic acid may produce pain when it contacts injured tissues. After the excess dye is removed, any areas that retain the stain signify injury. Separate any folds of the area and carefully examine them to avoid missing injuries. Ideally, apply the dye before speculum examination to eliminate the possibility of iatrogenic injury. The procedure is shown in Figure 58–7 and Table 58–3. In one study, the use of toluidine blue dye increased the injury detection rate from 16% to 40% in women without the use of colposcopy;20 however, injuries detected with the aid of toluidine blue dye are not

58 

12

● 

Labia minora

A

6

Fossa navicularis Posterior fourchette

Stain

B

Irregular stellate marks retain deep royal blue stain indicating zone of parakeratoses

C

Examination of the sexual assault victim

3

9 Hymen

Figure 58–7  A, During a sexual assault, injuries are often multiple and typically occur between the 3, 6, and 9 o’clock positions. B and C, Traumatic skin injury can be highlighted by applying toluidine blue to the perineum and vaginal area, then wipe it off to show the lesions.  (A, From Marx: Rosen’s Emergency Medicine: Concepts and Clinical Practice, 6th ed. Philadelphia, Elsevier, 2006.)

TABLE 58–3  Toluidine Blue* Staining of the Perineum to Detect Microabrasion 1. Collect all external genital specimens as indicated by examination before dye application.† 2. Before speculum examination or instrumentation, apply 1% toluidine blue to the entire vulva (labia majora, labia minora, posterior fourchette, perineal body, and perianal area). The anus may also be stained. Do not use dye in the vaginal vault or mucous membranes. 3. Allow to dry for approximately 1 min. 4. Remove excess dye with a spray of 1% acetic acid, irrigating the area until excess dye is removed. A water-soluble lubricant can also remove excess stain. 5. Gently blot the area with 4 × 4 gauze pads. DO NOT rub the area. 6. Photograph the area if indicated. 7. The dye will fade in 1–2 days.

Tear

1075 Thumbs separate tissue

A

100% specific for sexual assault because such injuries have also been found after consensual intercourse, especially in adolescents.21 If seminal stains are noted on the perineum, collect samples before toluidine blue application. Contrary to earlier thinking, the use of toluidine blue dye does not interfere with recovery of DNA evidence22 and has proved safe for mucosal application.23

Co

nt

us

ion

*A prefilled swab (T-Blue Swab/TBS, Tri-Tech, Inc.) is available through The National Forensic Nursing Institute). † This does not interfere with DNA or semen testing. Modified from The National Forensic Nursing Institute (NFNI.org/t-blueswab. html).

Tear

B

Anal Evaluation

Figure 58–8  Anal injury is best seen with separation of perianal tissues. A, Anal tear in a 13-year-old boy after forced penile-anal penetration. B, Anal contusion and tear (arrows) in an adult male after forced penile-anal penetration.

The anal examination follows the genital examination in most cases (Fig. 58–8). Because of a reluctance of some victims to admit to anal penetration, some clinicians recommend an anal examination in all cases. Documentation of anal penetration holds significant value because it is a separate crime in addition to vaginal penetration and can add years to the sentence of the alleged perpetrator. Insert anal swabs approximately 2 cm into the anus. Gently move them in a circular motion and then remove them. Use the swabs to make slides, air-dry

them, and include them in evidence sent to the crime laboratory. In some jurisdictions, rectal washings may also be requested. To do this, inject 5 to 10 mL of normal saline into the rectum with a syringe and a small plastic intravenous catheter. Then aspirate and preserve the fluid as evidence. Use toluidine blue dye, as described earlier, to better visualize injury. Anoscopy is not a routine part of the examination, but

Reference Samples Crime laboratories often request reference samples taken from various locations on the victim’s body to use in comparison testing with potential perpetrator evidence. These reference samples include blind swabs on the victim’s skin in a location complementary to suspicious skin samples. For example, if a suspicious discharge is swabbed from the victim’s right shoulder, take a control swab from the victim’s left shoulder as well. Crime laboratories may also request other control samples for victim DNA reference, such as blood, buccal mucosa cells, head hairs, or pubic hairs. The need for such samples will be determined by local protocol.

IX 

● 

GENITOURINARY, OBSTETRIC, AND GYNECOLOGIC PROCEDURES

it may be a useful adjunct in the documentation of rectal injuries. Perform this procedure in the same manner as diagnostic anoscopy done to evaluate other anal or rectal emergencies in the ED. In one retrospective observational study of male victims, the use of anoscopy and colposcopy provided superior documentation of injuries over colposcopy alone.24 The location of anoscopically detected injuries may be recorded geographically.

Blood Tests

Some crime laboratories request blood samples for DNA reference or toxicology analysis or both. Include these samples in the materials sent to law enforcement and not to the hospital laboratory unless a firm procedural “chain of custody” has been established for this purpose. When collecting blood for toxicology testing, record the exact time of collection on 1076 the specimen so that the criminologist may estimate dose and timing of substances used to perpetrate the assault.

Urine Tests Perform bedside urine β–human chorionic gonadotropin (βhCG) testing in all women victims of childbearing age to exclude preexisting pregnancy before giving pregnancy and STD prophylaxis. Collect urine as requested by local crime laboratories for toxicology testing. Collect the victim’s first available voided urine to optimize recovery of potential toxicologic evidence.

Spermatozoa, Semen, and DNA Testing Motile and immotile sperm may be found microscopically in wet mounts of vaginal aspirates and in vaginal, oral, and rectal swabs. If the examiner is formally trained, evaluate the slide microscopically immediately after the physical examination. Examiners find sperm in 13% to 26% of vaginal wet mount specimens.4,25 Early discovery of sperm may be helpful to law enforcement investigation. However, most ED examiners lack formal training in this process, and crime laboratories possess much higher sensitivity for sperm detection, making a negative initial wet mount unhelpful. For these reasons, many examiners do not routinely perform the wet mount examination.4 After consensual intercourse with normal ejaculate, laboratory testing of vaginal secretions will likely be able to detect sperm after 3 days and in 50% of cases at 4 days.26 However, despite penile penetration during sexual assault, the crime laboratories may fail to detect sperm. Reasons for this failure include inadequate specimen collection, degradation of ejaculate, azoospermia, failure of the per-

petrator to ejaculate, perpetrator vasectomy, victim washing, or condom use. A crime laboratory analyst initially looks for semen in a given sample by searching microscopically for sperm on a concentrated specimen and by testing for other components found in semen. Such seminal plasma components include p30 and acid phosphatase. p30 is a glycoprotein specific to the prostate27 and is regarded as conclusive evidence of semen (i.e., ejaculation within 48 hr), whereas acid phosphatase is presumptive evidence only because it can occur in other body fluids, such as vaginal secretions. Although this was a main component of crime laboratory investigation in the past, many laboratories have abandoned the acid phosphatase test in favor of the more specific p30 test.27,28 Despite negative testing for seminal plasma components, laboratories may be able to detect valuable DNA evidence from persistent sperm cells or perpetrator epithelial cells.29,30 As DNA testing technology rapidly changes, crime laboratory ability to perform a specific forensic test varies by location and over time. Most crime laboratories use PCR testing, which requires minimal material.

Chain of Custody Give samples and other evidence to the police, a crime laboratory, or a forensic pathologist. Label each sample with the patient’s name, hospital number, date, time of collection, area from which the specimen was collected, and collector’s name. Package these specimens according to local crime laboratory specifications and transfer them to the next appropriate official (police officer, pathologist, or other individual) along with a written chain of custody, including a list of the specimens, the signature of each person who provided them, and the signature of each person who received them. If this chain is broken, important evidence might be deemed inadmissible in court.

TREATMENT STD Prophylaxis Address the issues of STD, pregnancy, psychological distress, and follow-up in the treatment of a sexual assault victim. Because preassault infection rates are not known, the risk of contracting an STD as a consequence of a sexual assault has been difficult to determine, and estimates vary widely (Table 58–4). Jenny and colleagues31 found the postassault incidence of STDs to be 2% for chlamydia and 4% for gonorrhea. The reported rates of 12% for Trichomonas and 19% for bacterial vaginosis may reflect a preexposure infection because male

TABLE 58–4  Risk of Sexually Transmitted Disease after Sexual Assault Disease Gonorrhea Chlamydia Syphilis HIV

Risk (%) 6–18 4–17 0.5–3 102,9 rad >102,9 rad >109,8 rad None None

6 × 10−4 per rad 1 × 10−2 per rad

Whole pregnancy

Modified from Fattibene P, Mazzei F, Nuccetelli C, Risica S: Prenatal exposure to ionizing radiation: Sources, effects, and regulatory aspects. Acta Paediatr 88:693, 1999.

Radiation in pregnancy and clinical issues of radiocontrast agents

The magnitude of risk for carcinogenesis after low-dose radiation exposure and whether the risk changes throughout gestation have been the subject of many publications,15–17 yet interpretation of the data remains open to date. Numerous studies18–21 indicate a 1.3- to 3.0-fold higher incidence of leukemia in children exposed to diagnostic radiation in utero, although some studies fail to substantiate the association.14,22 Excess cancer as a result of in utero exposure has not been clearly demonstrated among Japanese atomic bomb survivor studies even though the population has been followed for about 50 years, but the number exposed is not large.9 Identification and control of confounding factors make interpretation of radiation carcinogenesis studies difficult, if not impossible, to interpret. Brent and coworkers14 noted that most investigators agree that low doses of radiation present a carcinogenic risk to the embryo; however, findings of an increased cancer risk among children exposed in utero to lowdose diagnostic radiation must be reconciled with the fact that high-dose animal and human studies have not found a marked increase in cancer incidence. Risk can be expressed in several ways, including as relative risk or absolute risk. Relative risk indicates the risk as a function of the “background” cancer risk. A relative risk of 1.0 indicates that there is no effect of irradiation, whereas a relative risk of 1.5 for a given dose indicates that the radiation is associated with a 50% increase in cancer above background rates. The absolute risk estimate simply indicates the excess number of cancer cases expected in a population due to a certain radiation dose.9 The International Commission on Radiological Protection Publication 849 noted that a recent analysis of many of the epidemiologic studies conducted on prenatal x-ray and childhood cancer are consistent with a relative risk of 1.4 (a 40% increase over the background risk) following a fetal dose of about 1 rad. The best methodological studies, however, suggest that the risk is probably lower than this. Even if the relative risk were as high as 1.4, the individual probability of childhood cancer after in utero irradiation would be very low (∼0.3%–0.4%) because the background incidence of child-

Mutagenesis

● 

Carcinogenesis

hood cancer is so low (∼0.2%–0.3%). Absolute risk estimates for cancer risk from ages 0 to 15 after in utero irradiation have been estimated to be in the range of 600/10,000 persons each exposed to 100 rad, or 0.06%/rad.9,11 If a fetus is exposed to 0.1 rad, the increased risk for carcinogenesis is 0.006% or 3/50,000, compared with the background incidence of 0.2% to 0.3% or 100 to 150/50,000. The increased carcinogenic risk from 0.1 rad exposure is approximately 50 times smaller than the already low natural incidence of cancer.

59 

10 rad, the spontaneous incidence of mental retardation is much larger than any potential radiation effect on IQ reduction.9 Regardless of the time of gestation, IQ reduction cannot be clinically identified at fetal doses of less than 10 rad (see Table 59–3).9 Growth Retardation.  The human data for Hiroshima and Nagasaki reveal that the major congenital anomaly observed was microencephaly.12 Studies have demonstrated no increased risk for microencephaly in the population exposed to less than 150 rad in Nagasaki; however, an increased risk in the Hiroshima population exposed to doses as low as 10 to 19 rad has been reported.14 It is possible that the difference between the two cities is secondary to other causes (e.g., trauma, stress, malnutrition) than radiation. In experimental animal data, a dose of 10 to 20 rad does not increase the incidence of microencephaly.14 A dose threshold for microencephaly, as well as other congenital anomalies, is generally accepted to be in the range of a few rad. Permanent growth retardation is not typically seen unless doses exceed 50 rad.12 Irradiation of the human fetus at doses below 10 rad has not been observed to cause congenital malformations or growth retardation (see Table 59–3).1,2,4

Investigating possible radiation-induced alteration to the human genome is exceedingly difficult. The geneticists who studied the radiated populations in Japan are convinced that there were radiation-induced mutations. However, the calculated and demonstrated risks were so small that the investigators were unable to demonstrate statistically significant genetic effects.23 The risk of radiation-induced hereditary disease in humans is reported to be around 1%/100 rad.12,14 If a fetus is exposed to 0.1 rad, then the increased risk is approximately 0.001% or 1/100,000. The natural frequency of genetic disease manifesting at birth is approximately 3%10 or 3000/100,000. For 0.1 rad, the increased genetic risk is minute compared with the natural incidence of genetic disease. In order to put all risks into proper perspective, the range of fetal absorbed doses for diagnostic imaging must be reviewed. The vast majority of diagnostic radiographic studies are markedly less than 5 rad. A comparison of fetal absorbed doses for the more common ED radiographic procedures follows. 1087

Radiation Exposure from Diagnostic Radiographs Table 59–4 lists estimated fetal exposure for various diagnostic imaging modalities.25 The number of examinations required to reach a cumulative dose of 5 rads is calculated in the second column to underscore the order-of-magnitude difference between the dose considered to have negligible risk (5 rad) and the actual exposed fetal dose. For example, one would require 5000 x-rays of an upper or lower extremity, 125 pelvic x-rays, or an impressive 70,000 two-view chest x-rays before the 5-rad limit is reached.

Radiation Exposure from CT Scans Many variables affect the calculation of fetal radiation dose from CT scans, especially slice thickness, number of cuts, distance of target organ from fetus, and gestational age. Table 59–4 summarizes estimated maximal fetal doses from CT scans. It should be noted that a CT of the lumbar spine delivers radiation to the fetus that approaches the safe cutoff range. A CT scan of the abdomen exposes the fetus to less radiation than the 5-rad cutoff, but alternative methods of investigation such as US or MRI should be considered in early pregnancy if the clinical condition warrants. The head CT is the most commonly requested CT scan in pregnancy. The expected fetal absorbed dose is less than 50 millirad (mrad), which is 100 times less than the dose with negligible risk. The estimated radiation dose to the fetus for CT of the chest is less than 0.100 rad. Spiral CT is common-

GENITOURINARY, OBSTETRIC, AND GYNECOLOGIC PROCEDURES ● 

IX 

TABLE 59–4  Estimated Fetal Exposure for Various Diagnostic Imaging Methods Examination Type

Estimated Fetal Dose per Examination (rad)

Number of Examinations Required for Cumulative 5-rad Dose

Plain Films Skull Dental Cervical spine Upper or lower extremity Chest (2 views) Mammogram Abdominal (multiple views) Thoracic spine Lumbosacral spine Intravenous pyelogram Pelvis Hip (single view)

0.004 0.0001 0.002 0.001 0.00007 0.020 0.245 0.009 0.359 1.398 0.040 0.213

1,250 50,000 2,500 5,000 71,429 250 20 555 13 3 125 23

CT Scans (Slice Thickness: 10 mm) Head (10 slices) Chest (10 slices) Abdomen (10 slices) Lumbar spine (multiple views) Pelvimetry (1 slice with scout film)

50 1–2 1–2 20

0.056 0.006 3.986

89 833 1

Fluoroscopic Studies

1088

Upper gastrointestinal series Barium swallow Barium enema Nuclear Medicine Studies Most studies using technetium (99mTc) Hepatobiliary technetium HIDA scan Ventilation-perfusion scan (total) Perfusion portion: technetium Ventilation portion: xenon (133Xe) Iodine (131I), at fetal thyroid tissue

10

0.150

33

0.215

23

0.175 0.040

28 125

590.000

Environmental Sources (for Comparison) Environmental background radiation (cumulative dose over 9 mo)

0.100

N/A

CT, computed tomography; HIDA, hepatobiliary iminodiacetic acid; rad, the unit of absorbed radiation. Reproduced from Toppenberg KS, Hill DA, Miller DP: Safety of radiographic imaging during pregnancy. Am Fam Physician 59:1813, 1999.

place in radiology departments and is a popular diagnostic tool used for suspected pulmonary embolism (PE) in the pregnant patient. The dose for a spiral CT of the chest is less because the duration of the procedure is much shorter.26 Van der Molen27 reported that using 16-slice versus 4-slice CT can equate to a radiation dose reduction of 20% to 30%.

Ordering CT scans of the head, chest, abdomen, and pelvis is a daily occurrence for emergency medicine clinicians. With that in mind, it is sobering to realize that the seventh National Academy of Science report on Biological Effects of Ionizing Radiation (BEIR VI) indicated that a 10-rad dose is associated with a lifetime attributable risk for developing a solid cancer or leukemia in 1 : 1000.28,29 As data on the effects of ionized radiation accumulate, and the technology of nonionizing techniques improves, our utilization of ionization based modalities will diminish. The American College of Radiology notes that iodinated low-osmolality contrast media (LOCM), most of which are nonionic agents, have been shown to be associated with less discomfort and have a lower incidence of minor (1% vs. 5% for high-osmolality contrast media [HOCM]) and severe reactions (0.015% vs. 0.1% for HOCM). Many Radiology departments routinely use LOCM.30 Although authors have expressed concern over the possibility that iodinated contrast may suppress fetal or neonatal thyroid function for a short period of time,7 the added benefit of a nonionic contrast material is that the intravascular use of nonionic contrast media has been reported to have no effect on neonatal thyroid function.31 To note, routine postnatal screening in the United States includes thyroid function tests. Iodinated contrast material that is injected intravenously for CT scans does not emit radiation and is classified as pregnancy category B. The product insert for barium sulfate suspension used as oral contrast for an abdominal CT scan (e.g., Redicat) notes no adverse fetal reactions under the heading “Usage in Pregnancy.” Barium preparations do not emit radiation.

NUCLEAR MEDICINE STUDIES A common nuclear medicine procedure ordered from the  scan. The perfusion  Q) ED is the ventilation perfusion ( V portion of the scan is performed by injecting a radioisotope intravenously. The isotope emits radiation and is detected by sensitive cameras. This requires that a radioisotope (e.g., 99 Tc) be tagged to a substrate, most commonly albumin. The albumin-technetium aggregate is temporarily trapped in the arterioles and capillaries in the lung and its distribution can be identified. The principal photon that is useful for detection and imaging with technetium studies is the γ-ray.32 When the radio-tagged substrate is excreted into the maternal bladder, the fetus will receive additional radiation exposure based on the proximity of the maternal bladder. Patient hydration and frequent voiding or bladder catheterization  scan will lessen the radiation exposure to the  Q after a V fetus. The measurement of radioactive substances is based on its decay, and the units are the Curie (Ci) or the Becquerel (Bq). Doses are usually expressed in milliCurie (mCi). The usual dose of technetium for the lung perfusion portion of the scan is 1 to 5 mCi of 99Tc. Reduced doses, as low as 1 mCi, are often used in pregnancy. Depending on the radioisotope and substrate employed, the average fetal exposures can be calculated. Commonly used  scans  Q radiopharmaceuticals and estimated fetal doses for V and other radionuclide studies are given in Table 59–5.13 A 5-mCi 99Tc albumin scan results in 175 mrad fetal exposure. Reducing the dose to 2 mCi results in a 70-mrad fetal exposure. 99Tc albumin is contraindicated in patients with severe pulmonary hypertension and is pregnancy category C.

This test uses both intravenous (perfusion) and aerosolized (ventilation) agents.

Brain Hepatobiliary Bone Respiratory   Perfusion

  Ventilation Renal Abscess or tumor Cardiovascular

Estimated Activity Administered per Examination (mCi) 99m

Dose to Uterus/ Embryo per Pharmaceutical (mrad)

20 mCi Tc DTPA 20 mCi 99mTc O4 5 mCi 99mTc sulfur colloid 5 mCi 99mTc HIDA 20 mCi 99mTc phosphate

700 960 55 150 500

5 mCi 99mTcmacroaggregated albumin 10 mCi 133xenon gas 20 mCi 99mTc DTPA 3 mCi 67Ga citrate 20 mCi 99mTc-labeled red blood cells

175

40 700 840 120

DTPA, diethylenetriamine penta-acetic acid; HIDA, hepatobiliary iminodiacetic acid; mCi, millicurie; mrad, millirad. Reproduced from Cunningham GF [ed]: Williams Obstetrics, 21st ed. New York, McGraw-Hill, 2001.

Ten mCi of 133Xe is used for the ventilation portion of  scan. 133Xe has a short half-life and results in 40  Q the V mrad fetal exposure. Normal findings on the perfusion scan may obviate the need for the ventilation scan, and some centers routinely perform only the perfusion portion because most pregnant women have normal ventilation.

PE DIAGNOSIS The reported incidence of PE associated with pregnancy is equivalent to roughly 1 in every 2000 pregnancies.7 The mortality rate of untreated acute PE is about 30% compared with 3% in treated patients.33 Therefore, the potential morbidity of PE and the attendant risk of anticoagulant therapy in pregnant patients necessitate definitive diagnosis. The radiologic modalities of choice for definitive diag scan versus CT pulmonary angiography.  Q nosis are V Although conventional pulmonary angiography was long considered the “gold standard” against which other imaging techniques were compared, it is now thought to be no more accurate than well-performed CT pulmonary angiography.7 The calculated radiation exposure to the fetus from both  scanning and CT of the chest confers minimal, and  Q V essentially only theoretical, fetal risk.26 Fetal exposure from CT of the chest is less than 0.100 rad. Fetal exposure from CT has been reported as low as 0.026 rad using a single-detector row helical CT and 0.013 rad for a multidetector row helical machine.7 A 5-mCi 99Tc perfusion scan and 10 mCi 133 Xe ventilation scan summates to 0.225 rad.25 The dose of fetal radiation from the perfusion scan can be altered and is often lowered in the evaluation of pregnant patients (Table 59–6). Lowering the dose by 60%, a level that will usually produce a suitable study, results in lowering fetal exposure to 0.110 mrad. Pulmonary angiography results in an estimated fetal exposure of 0.22 to 0.37 rad when done via the femoral route but can be lowered to less than 0.05 rad using the brachial route.26

Perfusion 1. Before injection, prepare the IV technetium. Mix sodium pertechnetate 99m Tc with macroaggregated human albumin (MAA), forming 99m Tc-MAA, the substance that is injected intravenously to investigate blood flow in the lungs. If the preparation is not used within 8 hr, discard it. 2. The usual dose is 1–5 (mCi. Doses as low as 1 mCi are used in pregnancy. 3. Within 5 min of injection, more than 90% of the Tc albumin aggregate is trapped in the arterioles and capillaries of the lung. The particle size determines where the 99m Tc will be localized in the body. 4. The accumulation in the lung is temporary, and fragile albumin aggregate quickly breaks down, allowing the Tc to enter the general circulation. 5. Once in the body, the half-life of 99m Tc is 6 hr. 6. The majority of the 99m Tc is excreted in the urine. If it remains in the urinary bladder, it is in close proximity to the fetus. 7. Tc in the bladder also exposes the fetus to small amounts of radiation. 8. Frequent voiding or bladder catheterization after the study will lessen radiation exposure to the fetus. 9. Tc is relatively contraindicated in patients with severe pulmonary hypertension (because the 99m Tc-MAA temporarily blocks blood flow in the lungs). 10. Allergic reactions to Tc and human serum albumin are extremely rare. 11. The radiation exposure to the total body from 2.5 mCi is extremely low: 125 scans have been performed. 4. The estimated dose to the fetus of a standard 10 mCi of xenon  scan is 0.04 rad. used in a V Q 5. If Tc-based aerosol is the marker used for the ventilation portion, fetal exposure is higher than with xenon aerosol.

 as the standard diagnostic  Q CT scan has supplanted V test in ruling out PE. Because perfusion scan is being done on a relatively young healthy subset of the population, one would suspect that the percentage of nondiagnostic studies (low or intermediate probability) would be less. However, Chan and colleagues34 published that ventilation perfusion scintigraphy is nondiagnostic in 25% of patients, with 73.5% read as normal, and only 1.8% read as high probability (113 patients in the study). Interestingly, 86% (24/28 patients) that received the nondiagnostic finding were not anticoagulated and were found to be free of a thromboembolism event for the following 20.6 months. Recognizably, the utility of a test that does not answer your question 25% of the time is concerning but should be tempered with the fact that there is a 75% chance of getting a definitive diagnosis, along with published concerns that the higher level of radiation exposure to childbearing women’s

Radiation in pregnancy and clinical issues of radiocontrast agents

Examination

● 

TABLE 59–6  The Technique of Ventilation/Perfusion  ) Scanning  Q (V

59 

TABLE 59–5  Radiopharmaceuticals Used in Nuclear Medicine Studies

1089

GENITOURINARY, OBSTETRIC, AND GYNECOLOGIC PROCEDURES ● 

IX 

breasts via CT may cause cancer decades later.7,8 Remy-Jardin and associates35 and Scarsbrook and coworkers7 reported that an exposure of 10 rads to the breasts of a woman aged 35 years increases the risk of breast cancer by approximately 14% over the background rate for the general population. So which study does one order? Both modalities expose a fetus to low levels of radiation of similar magnitude (CT  with very low theoretical fetal risk. Scarsbrook  Q) less than V and coworkers7 presented an algorithm mindful of the radiation exposure to both fetus and mother and provided convincing evidence for the recommendation (Fig. 59–2). Scarsbrook and coworkers7 suggested that an echocardiogram is a good first step in critically ill pregnant patients in which one considers PE. All others should start with a chest x-ray with shielding of the fetus. If the x-ray is normal, then do an US of the lower extremities to evaluate for deep venous thrombosis. Although this will have low diagnostic yield, it exposes the mother and fetus to no risk. If deep venous thrombosis is present, then treat. If US is negative, then move onto a halfdose lung perfusion scan if there is no patient history of obstructive lung disease. The literature reports that up to 75% of these scans are normal in the pregnant population. An important stipulation for using this algorithm is that one’s hospital radiologist needs to be comfortable reporting a Chest radiography (ECHO if in extremis and expertise readily available)

Normal

1090

Lower limb ultrasound

Abnormal or history of COPD or asthma

Diagnostic of non-embolic disease, e.g., pneumothorax

No specific non-embolic diagnosis

Treat cause

Lower limb ultrasound

Positive

Negative

Positive

Negative

Treat

Q scan

Treat

CTPA

Normal

Nondiagnostic

Normal

Nondiagnostic

Stop

Serial ultrasound or CTPA

Positive

Treat

Stop

Serial ultrasound or repeat CTPA

Positive

Treat

Figure 59–2  Suggested imaging algorithm for investigation of suspected pulmonary embolism in pregnancy. COPD, chronic obstructive pulmonary disease; CTPA, computed tomography pulmonary angiography. (Modified from Scarsbrook AF, Evans AL, Owen AR, et al: Diagnosis of suspected venous thromboembolic disease in pregnancy. Clin Radiol 61:1, 2006.)

normal scan as opposed to low probability because all nondiagnostic tests would then go on for a CT pulmonary angiogram. The authors note that utilizing this algorithm allows a definitive diagnosis in the vast majority of cases while minimizing risk to both mother and fetus. Scarsbrook and coworkers7 recommended several dose reduction methods when using CT pulmonary angiography on pregnant patients. Although these do not fall into the realm of emergency medicine, it is worthwhile to raise these points with the radiologists when developing a protocol to lower the radiation exposure to your patients (Table 59–7). The role of d-dimer levels in the diagnosis of PE is evolving.36,37 During pregnancy, d-dimer levels increase and should be considered physiologic. d-Dimer levels are similar to those in nonpregnant patients up to around 20 weeks, then are noted to increase throughout pregnancy to three times higher than the mean of a healthy nonpregnant patient.38 Recently, attempts have been made to establish a range of normal d-dimer values throughout pregnancy, which may be of great value, but as of yet, have not been tested in clinical practice.7

DIAGNOSIS OF PREGNANCY AND CONSENT If exposure of less than 5 rad does not measurably affect the exposed embryo, then why should the clinician determine the pregnancy status of the patient? Brent2 reported sound reasoning for diagnosis of pregnancy before radiographic study. First, the diagnosis of pregnancy may be in the differential diagnosis for the patient’s presenting problem and may change or obviate any further need for imaging. Second, it is beneficial to have the patient informed of pregnancy status before the imaging, if possible. An informative discussion about the risk/benefit aspects of the test before the study conveys a concern for the patient and fetus. Discussing the risk/benefit aspects of imaging after the study may be misconstrued as “back-stepping” and make the patient upset. Many lawsuits are stimulated by the factor of surprise. A frank discussion before the imaging may prevent misguided litigation. Amenorrhea and physical changes in size and shape of the uterus may be consistent with pregnancy. However, menstrual history by itself may not be totally reliable in the determination of pregnancy. A history of recent menstruation, intrauterine device, tubal ligation, no coitus, or the proper use of birth control pills will result in the suggestion of pregnancy

TABLE 59–7  Dose Reduction Methods When Using Computed Tomography Pulmonary Angiography to Image Suspected Pulmonary Embolic Disease in Pregnancy Reduce milliampere-second (mAs) Reduce kilovoltage (kVp) Increase pitch Increase detector and beam collimation Reduce field of view Reduce z-axis scan volume (caudal extent limited to top of diaphragm) Eliminate frontal and lateral scout views Circumferential shielding of the abdomen and pelvis From Scarsbrook AF, Evans AL, Owen AR, et al: Diagnosis of suspected venous thromboembolic disease in pregnancy. Clin Radiol 61:1, 2006.

COMMON RADIOGRAPHIC STUDIES

Upper or lower extremity

0.001

Cervical spine

0.002

CT of head