Critical Care Handbook of the Massachusetts General Hospital PDF

Senior Acquisitions Editor: Keith Donnellan Senior Development Editor: Ashley Fischer Editorial Coordinator: Sunmerrilika Baskar Marketing Manager: Kirsten Watrud Production Project Manager: Bridgett Dougherty Design Coordinator: Stephen Druding Manufacturing Coordinator: Beth Welsh Prepress Vendor: S4Carlisle Publishing Services Seventh Edition Copyright © 2024 Wolters Kluwer. Copyright © 2016 Wolters Kluwer. Copyright © 2010 Lippincott Williams & Wilkins, a Wolters Kluwer business. © 2006 by Department of Anesthesia and Critical Care Massachusetts General Hospital. Published by LIPPINCOTT WILLIAMS & WILKINS. All rights reserved. This book is protected by copyright. No part of this book may be reproduced or transmitted in any form or by any means, including as photocopies or scanned-in or other electronic copies, or utilized by any information storage and retrieval system without written permission from the copyright owner, except for brief quotations embodied in critical articles and reviews. Materials appearing in this book prepared by individuals as part of their official duties as U.S. government employees are not covered by the above- mentioned copyright. To request permission, please contact Wolters Kluwer at Two Commerce Square, 2001 Market Street, Philadelphia, PA 19103, via email at [email protected], or via our website at shop.lww.com (products and services). 987654321 Printed in the United States of America Library of Congress Cataloging-in-Publication Data ISBN-13: 978-1-975183-79-0 ISBN-10: 1-975183-79-7 Library of Congress Control Number: 2023904026 This work is provided “as is,” and the publisher disclaims any and all warranties, express or implied, including any warranties as to accuracy, comprehensiveness, or currency of the content of this work. This work is no substitute for individual patient assessment based upon healthcare professionals’ examination of each patient and consideration of, among other things, age, weight, gender, current or prior medical conditions, medication history, laboratory data and other factors unique to the patient. The publisher does not provide medical advice or guidance and this work is merely a reference tool. Healthcare professionals, and not the publisher, are solely responsible for the use of this work including all medical judgments and for any resulting diagnosis and treatments. Given continuous, rapid advances in medical science and health information, independent professional verification of medical diagnoses, indications, appropriate pharmaceutical selections and dosages, and treatment options should be made and healthcare professionals should consult a variety of sources. When prescribing medication, healthcare professionals are advised to consult the product information sheet (the manufacturer’s package insert) accompanying each drug to verify, among other things, conditions of use, warnings and side effects and identify any changes in dosage schedule or contraindications, particularly if the medication to be administered is new, infrequently used or has a narrow therapeutic range. To the maximum extent permitted under applicable law, no responsibility is assumed by the publisher for any injury and/or damage to persons or property, as a matter of products liability, negligence law or otherwise, or from any reference to or use by any person of this work. shop.lww.com To our patients, teachers, and trainees—past, present, and future. CONTRIBUTORS Chiaka O. V. Akarichi, MD Assistant Professor Department of Surgery University of Texas Southwestern Medical School Associate Medical Director of Parkland Burn Center Department of Surgery Parkland Hospital Dallas, Texas Katherine H. Albutt, MD, MPH Assistant Professor Department of Surgery Harvard Medical School Attending Surgeon Departments of Trauma, Emergency Surgery, and Surgical Critical Care Massachusetts General Hospital Boston, Massachusetts Clíodhna Ashe, BA, BDentSc, BMBS Fellow in Critical Care Medicine Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Aranya Bagchi, MBBS Assistant Professor of Anesthesia Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School Staff Anesthesiologist and Intensivist Heart Center ICU, Corrigan Minehan Heart Center and Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Lisa M. Bebell, MD Associate Professor Harvard Medical School Associate Physician Division of Infectious Diseases Department of Medicine Massachusetts General Hospital Boston, Massachusetts William J. Benedetto, MD Assistant Professor Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School Massachusetts General Hospital Boston, Massachusetts Lorenzo Berra, MD Reginald Jenney Associate Professor Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School Staff Anesthesiologist and Critical Care Physician Medical Director, Respiratory Care Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Edward A. Bittner, MD, PhD, MSEd Associate Professor Department of Anesthesia Harvard Medical School Program Director, Critical Care Anesthesiology Fellowship Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Sharon E. Brackett, RN, BSN Registered Nurse Surgical Intensive Care Unit Massachusetts General Hospital Boston, Massachusetts Diana Barragan Bradford, MD Staff Anesthesiologist and Critical Care Intensivist Instructor, Harvard Medical School Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Benjamin R. M. Brush, MD Assistant Professor Department of Neurology New York University Attending Physician Department of Neurology NYU Langone Health Tisch Hospital New York, New York Ryan W. Carroll, MD, MPH Assistant Professor Department of Pediatrics Harvard Medical School Attending Faculty Division of Pediatric Critical Care Medicine Massachusetts General Hospital Boston, Massachusetts Marvin G. Chang, MD, PhD Assistant Professor Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School Associate Program Director, Critical Care Medicine Fellowship Director, Perioperative Transthoracic Echocardiography (Point of Care Ultrasound) Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Hovig V. Chitilian, MD Anesthesiologist Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Devan Cote, MD Instructor in Anesthesia Harvard Medical School Anesthesiologist Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Jerome Crowley, MD, MPH Clinical Instructor Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School ECMO Director Heart Center ICU Massachusetts General Hospital Boston, Massachusetts Brian M. Cummings, MD Assistant Professor Department of Pediatrics Harvard Medical School Pediatric Intensive Care Department of Pediatrics Massachusetts General Hospital Boston, Massachusetts Roberta Ribeiro De Santis Santiago, MD, PhD, RRT Instructor in Anesthesia Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School Research Staff Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Paige McLean Diaz, MD Gastroenterology Fellow Department of Medicine Massachusetts General Hospital Boston, Massachusetts Ander Dorken-Gallastegi, MD Post-Doctoral Research Fellow Department of Surgery Harvard Medical School Post-Doctoral Research Fellow Division of Trauma, Emergency Surgery, and Surgical Critical Care Massachusetts General Hospital Boston, Massachusetts David M. Dudzinski, MD Assistant Professor Cardiology and Critical Care Harvard Medical School Director, Cardiac Intensive Care Unit Department of Cardiology, Cardiac Intensive Care, Echocardiography Massachusetts General Hospital Boston, Massachusetts Walter (Sunny) Dzik, MD Associate Professor Department of Pathology Harvard Medical School Consulting Hematologist Department of Medicine Massachusetts General Hospital Boston, Massachusetts Michael G. Fitzsimons, MD Associate Professor Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School Director, Division of Cardiac Anesthesia Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Rachel C. Frank, MD Fellow Division of Cardiology Department of Medicine Harvard Medical School Fellow Division of Cardiology Department of Medicine Massachusetts General Hospital Boston, Massachusetts Manolo Rubio Garcia, MD Interventional Cardiology, Vascular Medicine and Endovascular Intervention Specialist Interventional Cardiology Cardiovascular Institute of San Diego Chula Vista, California Ed George, MD, PhD Assistant Professor Department of Anesthesiology Harvard Medical School Director of Disaster Planning and Resource Management Perioperative Services Department of Anesthesia and Critical Care Massachusetts General Hospital Boston, Massachusetts Lauren E. Gibson, MD Clinical Fellow Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Dusan Hanidziar, MD, PhD Instructor in Anesthesia Harvard Medical School Anesthesiologist and Intensivist Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Bryan D. Hayes, PharmD Associate Professor Division of Medical Toxicology, Department of Emergency Medicine Attending Pharmacist Department of Pharmacy Massachusetts General Hospital Boston, Massachusetts Dean R. Hess, PhD, RRT Lecturer Respiratory Care, College of Professional Studies Northeastern University Respiratory Care Services Massachusetts General Hospital Boston, Massachusetts Ronald E. Hirschberg, MD Assistant Professor Director, Physical Medicine and Rehabilitation Consultation Service Department of Physical Medicine and Rehabilitation Massachusetts General Hospital Boston, Massachusetts Ryan J. Horvath, MD, PhD Instructor in Anesthesia Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School Post Anesthesia Care Unit (PACU) Director Staff Anesthesiologist and Critical Care Intensivist Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Joanne C. Huang, PharmD, BCIDP Infectious Diseases Clinical Pharmacist Department of Pharmacy Massachusetts General Hospital Boston, Massachusetts Paul S. Jansson, MD, MS Instructor Departments of Surgery and Emergency Medicine Harvard Medical School Emergency and Critical Care Physician Departments of Surgery and Emergency Medicine Mass General Brigham Boston, Massachusetts Kristin Kennedy, DO Clinical Fellow Harvard Medical School Surgical Critical Care Fellow Department of Surgery Massachusetts General Hospital Boston, Massachusetts Emmett Alexander Kistler, MD Fellow Harvard Medical School Division of Pulmonary and Critical Care Medicine Massachusetts General Hospital Boston, Massachusetts Alexander S. Kuo, MS, MD Assistant Professor Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Assistant Professor Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School Boston, Massachusetts Carolyn J. La Vita, MHA, RRT Director Department of Respiratory Care Massachusetts General Hospital Boston, Massachusetts Yvonne Lai, MD Assistant Professor Harvard Medical School Associate Program Director, Anesthesiology Residency Department of Anesthesiology, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Sameer Lakha, MD Assistant Professor Anesthesiology, Perioperative and Pain Medicine Institute for Critical Care Medicine Icahn School of Medicine at Mount Sinai The Mount Sinai Hospital New York, New York Benjamin Levi, MD Associate Professor Division Chair for General Surgery Department of Surgery University of Texas Southwestern Medical Center Dallas, Texas Ying Hui Low, MD, FASA, FASE Staff Physician Department of Anesthesiology Massachusetts General Hospital Boston, Massachusetts Shu Yang Lu, MD Clinical Instructor Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Robert S. Makar, MD, PhD Associate Professor Department of Pathology Harvard Medical School Director, Blood Transfusion Service Department of Pathology Massachusetts General Hospital Boston, Massachusetts Christopher J. Mariani, MD, PhD Fellow Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School Massachusetts General Hospital Boston, Massachusetts Laurie O. Mark, MD Attending Anesthesiologist and Critical Care Physician Department of Anesthesiology Jesse Brown VA Medical Center Chicago, Illinois Lukas H. Matern, MD Chief Resident Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts April E. Mendoza, MD, MPH Assistant Professor Department of Surgery Massachusetts General Hospital Boston, Massachusetts Ilan Mizrahi, MD Instructor Department of Anesthesia Harvard Medical School Anesthetist Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Catherine E. Naber, MD Clinical Instructor in Pediatrics Harvard Medical School Pediatric Intensivist Massachusetts General Hospital Boston, Massachusetts Emily E. Naoum, MD Instructor in Anesthesia Harvard Medical School Clinical Instructor Program Director, Obstetric Anesthesia Fellowship Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Michael O’Brien, PharmD Emergency Medicine Clinical Pharmacist Department of Pharmacy Massachusetts General Hospital Boston, Massachusetts Peter O. Ochieng, MD Instructor in Anesthesia Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School Staff Anesthesiologist Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Jonathan J. Parks, MD, FACS Clinical Instructor in Surgery Department of Surgery Harvard Medical School Instructor in Surgery Department of Surgery Massachusetts General Hospital Boston, Massachusetts Kristin Parlman, PT, DPT, NCS Physical Therapy Clinical Specialist Department of Physical Therapy Massachusetts General Hospital Boston, Massachusetts Sylvia Ranjeva, MD, PhD Resident Physician Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Benjamin Christian Renne, MD Instructor Harvard Medical School Critical Care Attending Physician Division of Trauma, Emergency Surgery, and Surgical Critical Care Massachusetts General Hospital Boston, Massachusetts Josanna Rodriguez-Lopez, MD Assistant Professor Department of Medicine Harvard Medical School Assistant Physician Division of Pulmonary and Critical Care Department of Medicine Massachusetts General Hospital Boston, Massachusetts Katarina J. Ruscic, MD, PhD Instructor in Anesthesia Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School Director of Anesthesia for Plastic, Reconstructive and Breast Oncology Surgery Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Daniel Saddawi-Konefka, MD, MBA Assistant Professor Harvard Medical School Program Director, Anesthesiology Residency Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Kyan C. Safavi, MD, MBA Assistant Professor Department of Anesthesia Harvard Medical School Anesthesiologist and Critical Care Physician Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Noelle N. Saillant, MD, FACS Assistant Professor Department of Surgery Boston University School of Medicine Surgeon Department of Surgery Trauma, Acute Care and Surgical Critical Care Boston Medical Center Boston, Massachusetts Takashi Sakano, MD Instructor Department of Anesthesia, Critical Care and Pain Medicine Harvard Medical School Anesthesiologist Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Esperance A. K. Schaefer, MD, MPH Instructor Department of Medicine Harvard Medical School Assistant in Medicine Gastroenterology Unit, Department of Medicine Massachusetts General Hospital Boston, Massachusetts Kenneth Shelton, MD Assistant Professor Harvard Medical School Chief of the Critical Care Division Department of Anesthesiology, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Robert Sheridan, MD Professor of Surgery Harvard Medical School Chief of Staff Department of Surgery Shriners Hospital for Children Boston, Massachusetts Robert D. Sinyard, III, MD, MBA Post-doctoral Research Fellow Ariadne Labs Harvard TH Chan School of Public Health Surgical Resident Department of Surgery Massachusetts General Hospital Boston, Massachusetts Jamie L. Sparling, MD Instructor of Anesthesia Department of Anesthesia Harvard Medical School Staff Anesthesiologist and Intensivist Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Zachary P. Sullivan, MD, MS Fellow, Critical Care Medicine Department of Anesthesiology, Perioperative and Pain Medicine Stanford University Stanford, California Resident-Anesthesiology Department of Anesthesiology, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Gina H. Sun, MD Resident Physician Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Alan J. Sutton, MD Staff Physiatrist Department of Physical Medicine & Rehabilitation Encompass Health Rehabilitation Hospital of Braintree Braintree, Massachusetts Lauren A. Sweetser, MD, MA Clinical and Research Fellow Department of Pediatric Critical Care Harvard Medical School Clinical and Research Fellow Department of Pediatric Critical Care Massachusetts General Hospital for Children Boston, Massachusetts Parsia Vagefi, MD, FACS Professor of Surgery Executive Vice Chair of Strategy and Finance Chief, Division of Surgical Transplantation Ernest Poulos, M.D. Distinguished Chair in Surgery UT Southwestern Medical Center Dallas, Texas Karen Waak, PT, DPT, CCS Physical Therapy Clinical Specialist Department of Physical Therapy Massachusetts General Hospital Boston, Massachusetts Elisa C. Walsh, MD Instructor Department of Anesthesia Harvard Medical School Staff Anesthesiologist and Intensivist Department of Anesthesia, Critical Care and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Alison S. Witkin, MD Assistant Professor of Medicine Department of Medicine Harvard Medical School Assistant in Medicine Department of Medicine Massachusetts General Hospital Boston, Massachusetts Victor W. Wong, MD Staff Surgeon Plastic and Reconstructive Surgery The Permanente Medical Group Santa Rosa, California Amanda S. Xi, MD, MSE Instructor Harvard Medical School Director of Pre-Procedure Evaluation and Procedural Sedation Department of Anesthesia, Critical Care, and Pain Medicine Massachusetts General Hospital Boston, Massachusetts Michael J. Young, MD, MPhil Associate Director, MGH NeuroRecovery Clinic Center for Neurotechnology and NeuroRecovery (CNTR) Division of Neurocritical Care, Department of Neurology Massachusetts General Hospital and Harvard Medical School Boston, Massachusetts Adil Yunis, MD Physician Department of Cardiovascular Medicine Massachusetts General Hospital Boston, Massachusetts Sahar F. Zafar, MD, MSc Neurointensivist Department of Neurology Massachusetts General Hospital Boston, Massachusetts Hilary L. Zetlen, MD, MPH Fellow Division of Pulmonary and Critical Care Massachusetts General Hospital Boston, Massachusetts PREFACE The Critical Care Handbook of the Massachusetts General Hospital, Seventh Edition is designed to be a concise, didactic, and practical guide for health care providers caring for patients with life-threatening illness and injuries. Care for the critically ill patient is broad in scope and practice. It can take place in a variety of settings including prehospital situations, the emergency department, hospital ward, operating room, and intensive care unit. The critical care provider needs to be knowledgeable not only about a broad range of medical and surgical conditions causing critical illness but also with the technological procedures and devices used in support of organ system dysfunction. To that end, this handbook includes basic principles of diagnosis and management of the critically ill patient including fundamental physiology, monitoring, diagnostic and therapeutic procedures, as well as more advanced knowledge of specific patient conditions and treatments. The handbook also includes chapters devoted to many of the complicated ethical and social issues encountered in critical care practice such as end-of-life decision-making, the economics of care, quality improvement, and understanding of the ongoing challenges faced by survivors of critical illness. While endeavoring to maintain the structure and style of successful prior editions, the handbook chapters have been updated to reflect the rapid changes in critical care management. In addition, new sections and chapters have been added to describe recent innovations in thought and practice. As with prior editions, the handbook reflects the current clinical practice at the Massachusetts General Hospital that is the foundation of our critical care training programs. I am indebted to a number of people who were essential in the publication of the Seventh Edition of The Critical Care Handbook of the Massachusetts General Hospital. First, I wish to gratefully acknowledge the past editors and contributors to the previous editions of this handbook. They established the solid foundation upon which our clinical practice and this current edition are based. My coeditors of this edition deserve special acknowledgment for their dedication and diligent editing efforts despite the unrelenting demands of post-COVID-19 health care practice. The professional staff at Wolters Kluwer including Keith Donnellan, Ashley Fischer, and Sunmerrilika Baskar have been indispensable throughout the planning and organizing stages of this book, and by providing “gentle reminders” to submit our chapters on time. Finally, I am grateful to my wife Karen and sons Daniel and Andrew for their ongoing love and support. Edward A. Bittner, MD, PhD, MSEd CONTENTS Contributors Preface PART 1: GENERAL PRINCIPLES 1 Hemodynamic Monitoring Christopher J. Mariani and Daniel Saddawi-Konefka 2 Respiratory Monitoring Catherine E. Naber and Ryan W. Carroll 3 Use of Ultrasound in Critical Illness Lauren E. Gibson and Marvin G. Chang 4 Airway Management Takashi Sakano and Edward A. Bittner 5 Mechanical Ventilation Carolyn J. La Vita 6 Hemodynamic Management Diana Barragan Bradford and Aranya Bagchi 7 Sedation and Analgesia Lukas H. Matern and Ryan J. Horvath 8 Fluids, Electrolytes, and Acid-Base Management Gina H. Sun and Yvonne Lai 9 Trauma Katherine H. Albutt and Jonathan J. Parks 10 Critical Care of the Neurologic Patient Benjamin R. M. Brush and Sahar F. Zafar 11 Nutrition in Critical Illness Ander Dorken-Gallastegi and April E. Mendoza 12 Infectious Disease Joanne C. Huang and Lisa M. Bebell 13 Critical Care Management of COVID-19 Rachel C. Frank and Dusan Hanidziar 14 Transporting the Patient Who Is Critically Ill Zachary P. Sullivan and Ilan Mizrahi 15 Coronary Artery Disease Ying Hui Low and Michael G. Fitzsimons 16 Valvular Heart Disease Sameer Lakha and Alexander S. Kuo PART 2: SPECIFIC CONSIDERATIONS 17 Cardiac Dysrhythmias Peter O. Ochieng and Kenneth Shelton 18 Heart Failure Rachel C. Frank, Adil Yunis, and David M. Dudzinski 19 ECMO and Ventricular Assist Devices Jerome Crowley 20 Acute Respiratory Distress Syndrome Sylvia Ranjeva, Roberta Ribeiro De Santis Santiago, and Lorenzo Berra 21 Asthma and Chronic Obstructive Pulmonary Disease Lauren E. Gibson and Lorenzo Berra 22 Deep Venous Thrombosis and Pulmonary Embolism in the Intensive Care Unit Manolo Rubio Garcia and Alison S. Witkin 23 Pulmonary Hypertension Hilary L. Zetlen and Josanna Rodriguez-Lopez 24 Discontinuation of Mechanical Ventilation Dean R. Hess 25 Acute Kidney Injury William J. Benedetto 26 Critical Care of Patients With Liver Disease Paige McLean Diaz, Hovig V. Chitilian, Parsia Vagefi, and Esperance A. K. Schaefer 27 Coagulopathy and Hypercoagulability Robert D. Sinyard, III, Noelle N. Saillant, and Katherine H. Albutt 28 Acute Gastrointestinal Diseases Katherine H. Albutt and Kristin Kennedy 29 Endocrine Disorders and Glucose Management Devan Cote and Katarina J. Ruscic 30 Acute Weakness Kristin Parlman and Karen Waak 31 Drug Overdose, Poisoning, and Adverse Drug Reactions Michael O’Brien and Bryan D. Hayes 32 Adult Resuscitation Paul S. Jansson 33 Burn Critical Care Chiaka O. V. Akarichi, Benjamin Levi, Victor W. Wong, and Robert Sheridan 34 Transfusion Medicine Walter (Sunny) Dzik and Robert S. Makar 35 Obstetric Critical Care Emily E. Naoum and Elisa C. Walsh PART 3: HEALTH CARE SERVICES 36 Intensive Care Unit Handoffs and Transitions Clíodhna Ashe and Jamie L. Sparling 37 Recovery After Critical Illness Alan J. Sutton, Ronald E. Hirschberg, and Laurie O. Mark 38 The Economics of Critical Care: Measuring and Improving Value in the Intensive Care Unit Kyan C. Safavi 39 Telemedicine and Remote Electronic Monitoring Systems in the Intensive Care Unit Shu Yang Lu and Benjamin Christian Renne 40 Quality Improvement and Standardization of Practice Lauren A. Sweetser and Brian M. Cummings 41 Ethical, Legal, and End-of-Life Issues in Intensive Care Unit Practice Emmett Alexander Kistler, Sharon E. Brackett, and Michael J. Young 42 Intensive Care Unit Care After Organ Transplant Amanda S. Xi 43 Disaster Management Ed George Index PART 1: GENERAL PRINCIPLES 1 Hemodynamic Monitoring Christopher J. Mariani and Daniel Saddawi-Konefka I. HEMODYNAMIC MONITORING Hemodynamic monitoring is one of the cornerstones of patient evaluation in the intensive care unit (ICU) and provides diagnostic and prognostic value. The choice of monitoring depends on the diagnostic needs of the patient and the risk-benefit balance of monitor placement and maintenance and complications associated with its use. This chapter outlines an approach to the assessment of hemodynamics and perfusion in patients who are critically ill and the technical principles of commonly used monitoring methods. A. Perfusion: The goal of hemodynamic monitoring is to ensure adequate tissue perfusion for gas, nutrient, and waste exchange, with the goal of decreasing morbidity and mortality. It is difficult to link optimization of a single hemodynamic parameter to an improvement in morbidity and mortality. For this reason, intensivists should not rely solely on any one physical monitor or parameter but should evaluate multiple potential signs of adequate perfusion such as mental status, urine output, or laboratory findings (eg, central venous oxygen saturation, base deficit, and lactate). B. Optimizing Perfusion: Hemodynamic monitors are not therapeutic. Hemodynamic data, however, can be used to guide therapy. Optimizing perfusion may require fluid administration, diuresis, administration of pharmacologic agents (eg, vasoconstrictors, inotropic agents), or interventions (eg, intra-aortic balloon pumps, ventricular assist devices, extracorporeal membrane oxygenation). With this in mind, any monitor must be used dynamically to ensure that employed therapies are optimizing perfusion over time. 1. Fluid challenge: Much discussion around optimization of hemodynamics revolves around fluid status. The “fluid challenge” is a time-honored test to aid in determining whether fluid administration could be beneficial. With a fluid challenge test, intravenous (IV) fluid is rapidly administered while hemodynamics are monitored to determine whether the administration improves hemodynamic parameters (and thereby be of benefit to the patient). Although the fluid challenge is not standardized, it most commonly involves the rapid administration of 500 mL of fluid, with a positive test defined as an increase in cardiac output (CO) greater than 10% to 15%. A “passive leg raise” test can be used to provide similar information. To perform this test, a clinician passively elevates a supine patient’s legs. Blood moves to the central veins from the elevated limbs, providing an “autotransfusion” of approximately 150 to 300 mL. An improvement in stroke volume suggests fluid responsiveness, whereas deterioration in hemodynamics can be quickly reversed by lowering the legs. A passive leg raise test may also be performed using pulse pressure as a surrogate for stroke volume. The sensitivity and specificity of a passive leg raise are reduced when changes in pulse pressure are measured in place of stroke volume. II. ARTERIAL BLOOD PRESSURE MONITORING A. General Principles 1. Blood pressure describes the pressure exerted by circulating blood within the blood vessels. Because this pressure drives flow, it is used as a surrogate measure of blood flow and, in turn, organ perfusion (Figure 1.1). This simplified view has limitations and notably poor correlation with CO in some situations, such as emergency resuscitation of the patient with hypovolemia who is critically ill. Nonetheless, arterial blood pressure monitoring as a target for perfusion is used in almost all critical care settings. FIGURE 1.1 The assumptions when extrapolating mean arterial pressure (MAP) to a morbidity and mortality benefit. 2. Under normal circumstances, tissue perfusion is maintained across a range of pressures by autoregulation, which describes the intrinsic capacity of vascular beds to maintain flow by adjusting local vascular resistance. However, pathologic conditions common in the ICU such as chronic hypertension, trauma, and sepsis may impair autoregulation, resulting in blood flow that may depend directly on perfusion pressure. 3. The “gold standard” for blood pressure measurement is aortic root pressure, which is representative of the pressures received by the major organs (eg, heart, brain, kidneys). As the pressure wave travels distally from the aorta, the measured mean pressure decreases while the measured pulse pressure (systolic pressure minus diastolic pressure) is increased owing to pulse wave reflection from the high- resistant distal arterioles. In addition to being amplified, as one progresses distally, the arterial waveform is slightly delayed (Figure 1.2). FIGURE 1.2 Arterial waveforms as one travels distally along the arterial tree. B. Noninvasive Blood Pressure Monitoring: Various techniques can be used to measure blood pressure noninvasively, including manual palpation, determination of Korotkoff sounds with a sphygmomanometer and stethoscope or Doppler ultrasound, and automated oscillometric methods, which are most common in the ICU. 1. Function: The oscillometric method uses a pneumatic cuff with an electric pressure sensor, most commonly over the brachial artery. The cuff is inflated to high pressure and then slowly deflated. Arterial pulsations are recorded as oscillations. The pressure that produces the greatest oscillation recording is closely associated with mean arterial pressure (MAP). Systolic and diastolic pressures are then calculated, often using proprietary algorithms. 2. Technique: Accurate noninvasive blood pressure measurement requires appropriate cuff sizing and placement. Most blood pressure cuffs display reference lines for an acceptable arm length, and the cuffs should be sized as recommended in relationship to arm circumference. Cuffs that are too small may overestimate blood pressure, whereas cuffs that are too large may underestimate blood pressure. 3. Limitations and risks a. The pressure measured by a noninvasive cuff is the pressure at the cuff site. When an extremity pressure is measured to estimate coronary perfusion, either the extremity should be placed at the level of the heart or the extremity level relative to the heart should be accounted for (eg, a pressure measured at a site 10 cm below the heart will be 10 cm H2O or ~7.4 mm Hg greater than the pressure at the heart). b. Tissues, including vessels and nerves, can be damaged by cyclical compression of pneumatic cuffs with frequent cycling. As such, automatic methods may not be appropriate in rapidly changing situations. They may also be less accurate with extremes of blood pressure or in patients with dysrhythmias. C. Invasive Arterial Blood Pressure Monitoring provides beat-to-beat pressure transduction and allows for convenient blood sampling. 1. Indications for placement of an arterial line include hemodynamic instability, need for tight blood pressure control, and need for frequent blood sampling. 2. Site: The radial artery is frequently selected given the existence of collateral circulation to distal tissues, convenient access for ongoing care, and patient comfort. Alternative sites in adults include the brachial, axillary, femoral, and dorsalis pedis arteries. Under most physiologic conditions, the MAP should be similar between these sites. In patients with profound hypotension or high-dose vasopressor requirements, central MAP measurements (eg, at the femoral artery) are frequently higher than the MAP measured at the radial artery. 3. Function: Necessary equipment to monitor invasive arterial blood pressure includes an intra-arterial catheter, fluid-filled noncompliant tubing, transducer, continuous flush device, and electronic monitoring equipment. The flush device typically provides an infusion of plain or heparinized saline at a rate of 2 to 4 mL/h through the tubing and catheter to prevent thrombus formation. Commonly employed arterial line pressure bag/transducer systems provide this slow flush by design. The transducing sensor is “connected” to arterial blood by a continuous line of fluid and measures a pressure deflection in response to the transmitted pressure wave of each heartbeat. The accuracy of intra-arterial blood pressure measurement depends on the proper positioning and calibration of the catheter-transducer monitoring system. a. Positioning: The arterial pressure reflects the pressure at the level of the transducer, not the pressure at the level of the cannulation site. This is because of the fluid-filled tubing between the patient and transducer, which maintains energy by exchanging potential energy for pressure (Bernoulli’s equation). For example, if the transducer is lowered, the fluid in the tubing exerts an additional pressure on the transducer and the measured pressure will be higher. Therefore, the arterial transducer should be placed at the level of interest. For example, positioning at the fourth intercostal space on the midaxillary line (“the phlebostatic axis”) corresponds to the level of the aortic root, and positioning at the external acoustic meatus corresponds to the level of the circle of Willis. b. Static calibration zeros the system to atmospheric pressure. The transducer is opened to air and the recorded pressure (the atmospheric pressure) is set to zero. c. Dynamic response is important to account for because it affects the pressure readings in ways that are unrelated to the actual pressure wave at the cannulation site (eg, the reading may appear flattened [damped] or stretched [underdamped]). Dynamic response occurs as a result of the natural frequency of the arterial line system as well as the damping coefficient. The natural frequency is the frequency at which a pressure pulse oscillates within the system. The damping coefficient is a measure of how quickly an oscillating waveform decays in the system. These two characteristics of an arterial line system can be illustrated using a fast flush (also known as a square wave) test. In this test, the transducer and tubing system is transiently flushed with fluid at 300 mm Hg. Following the flush, the arterial pressure trace will exhibit a series of oscillations before returning to baseline. The frequency of these oscillations (calculated by dividing the monitor speed by the wavelength of the oscillating waves) represents the natural frequency of the system. The damping coefficient is related to the amplitude ratio of successive oscillations (eg, the height of the first oscillation relative to the second oscillation). To have an adequate dynamic response, and thereby accurately measure arterial blood pressure, the correct combinations of natural frequencies and damping coefficients are required. When a system has an inadequate dynamic response, over- or underdamping will be observed. Generally, return of the pressure signal to baseline after one to two oscillations following a fast flush test is indicative of a system with an adequate dynamic response. 1. Overdampened systems will underestimate systolic pressure and overestimate diastolic pressure (because the waveforms will tend to collapse around the MAP). Overdamped systems will return to baseline following a fast flush after one or less oscillations. Overdampening may be caused by air bubbles, clots, kinking of the catheter or tubing, or loose connections. 2. Underdampened systems will overestimate the systolic blood pressure and underestimate the diastolic pressure. Underdampened systems will have greater than two oscillations following the fast flush before returning to the baseline. Underdamping may be caused by excessive tubing length, stiff tubing, or tachycardia. 3. In both over- and underdampened systems, the measured mean pressure will still be accurate. 4. Complications: Arterial cannulation is relatively safe. Risks depend on the site of cannulation. For radial cannulation, reported serious risks include permanent ischemic damage (0.09%), local infection and sepsis (0.72% and 0.13%, respectively), and pseudoaneurysm (0.09%). Fastidious attention to the adequacy of distal perfusion is important. Thrombotic sequelae are associated with larger catheters, smaller arterial size, administration of vasopressors, duration of cannulation, and multiple arterial cannulation attempts. With regard to infectious risk, the aseptic technique was not standardized in the studies that yielded the aforementioned percentage of risk and longer duration of cannulation increased risk. Less serious risks include temporary occlusion (19.7%) and hematoma (14.4%). Axillary and femoral sites have been associated with higher risks of infection. Brachial cannulation has been associated with median nerve injury (0.2%-1.4%). However, in a center with standardized use of brachial catheters for cardiac surgery, no nerve injuries were reported following 21 000 brachial artery catheterizations, suggesting that previous risk of nerve injury may overestimate complications in centers where this approach is standard. 5. Respiratory variation: Increased intrathoracic pressure decreases preload, arterial pressure, and pulse pressure. This effect is marked in patients with hypovolemia who are more susceptible to increased intrathoracic pressures. Variation of more than 10% to 12% in systolic pressure or pulse pressure is suggestive of fluid responsiveness. Importantly, this has been validated in patients with regular cardiac rhythm and breathing pattern, and it is dependent on the ventilatory pressure delivered. 6. Arterial waveform analysis has been used to gauge stroke volume and is discussed later in this chapter. D. Discrepancies Between Noninvasive and Invasive Arterial Blood Pressure Measurements exist, with noninvasive measurements tending to yield higher measurements during hypotension and lower measurements during hypertension. These discrepancies persist even with appropriate cuff sizing in patients who are critically ill. Retrospective data suggest that the higher noninvasive systolic blood pressures during hypotension are overestimates of perfusion because the incidence of acute kidney injury and mortality is higher with noninvasive versus invasive systolic blood pressures. There was no difference in acute kidney injury or mortality when mean pressures were compared, which suggests that mean pressures should be targeted when the patient who is hypotensive and critically ill is treated with a noninvasive cuff. III. CENTRAL VENOUS PRESSURE MONITORING A. Indications for placement of a central venous catheter (CVC) include administration of certain medications, concentrated vasopressors, or total parenteral nutrition (TPN); need for dialysis; need for long-term medication administration such as chemotherapy or IV antibiotics; need for IV access in patients with difficult peripheral access; need for sampling central venous blood; or need for central venous pressure (CVP) hemodynamic data to guide management efforts. B. Site: Common central venous cannulation sites are the internal jugular, subclavian, and femoral veins. The ideal site of cannulation varies with the characteristics of the patient and the indications for insertion. For example, the subclavian site is relatively contraindicated in patients with coagulopathies because it is not directly compressible, and the femoral vein may be ideal in emergency situations because of ease of cannulation. Table 1.1 summarizes the advantages and disadvantages of the most commonly used sites for venous access. T A B L E 1.1 Risks and Benefits of Different Central Line Access Approaches Infection Bleeding Thrombotic Patient Site risk risk risk comfort Comments Internal ++ + ++ ++ Compressible, low jugular pneumothorax risk Subclavian + +++ + +++ Higher risk of pneumothorax, difficult to compress in case of bleeding Femoral +++ + +++ + Often easiest place in an emergency, high infection rate PICC + – ++ +++ Good for long-term access, low flow rate PICC, peripherally inserted central catheter. -, insignificant; +, to a less extent; ++, to a moderate extent; +++, to a substantial extent. C. CVP Waveform: CVP provides an estimate of right ventricular preload and should be measured with the transducer positioned at the phlebostatic axis. The CVP tracing contains three positive deflections (Figure 1.3). The a wave corresponds to atrial contraction and correlates with the P wave on the electrocardiogram (ECG). The c wave corresponds to ventricular contraction (and bulging of the tricuspid valve into the right atrium) and correlates with the end of the QRS complex on the ECG. The v wave corresponds to atrial filling against a closed tricuspid valve and occurs with the end of the T wave on ECG. The x descent after the c wave is thought to be due to the downward displacement of the atrium during ventricular systole and the y descent by the tricuspid valve opening during diastole. FIGURE 1.3 The central venous pressure (CVP) waveform. 1. Abnormal waveforms: Arrhythmias and valvular abnormalities can result in different central venous waveform morphologies. For example, the loss of atrial contraction that occurs with atrial fibrillation results in the loss of a waves on CVP tracings. Large a waves (“cannon a waves”) may occur when the atrium contracts against a closed valve, as occurs during atrioventricular dissociation or ventricular pacing. Abnormally large v waves that begin immediately after the QRS complex and often incorporate the c wave are associated with tricuspid regurgitation. Abnormally large v waves may also be observed during right ventricular failure or ischemia, constrictive pericarditis, or cardiac tamponade due to the volume and/or pressure overload seen by the right atrium. Tricuspid stenosis is a diastolic defect in atrial emptying and can result in an attenuated y descent with a prominent a wave. D. Interpretation of CVP 1. Measurement: CVP, when measured as a surrogate for end-diastolic filling, should be measured at the valley just before the c wave (ie, at the end of the diastole, just before ventricular contraction). Because CVP changes with respiration (owing to changes in intrathoracic pressure), it should be measured at end expiration, when the lung is closest to functional residual capacity and confounding intrathoracic pressure influences are minimized. 2. Utility and controversy: CVP has been used clinically to assess fluid status for decades. The physiologic determinants of CVP include patient position, the circulating volume status, interactions between systemic and pulmonary circulations, alterations in rhythm or valvular function (see Section III.C.1), and the dynamic changes in the respiratory system over the breathing cycle. Not surprisingly, the accurate interpretation of CVP can be difficult and a number of studies have challenged its use. Systematic review has suggested poor correlation of CVP with circulating blood volume as well as poor correlation of CVP (or trend in CVP) with fluid responsiveness. 3. Clinical confounders: When used clinically, CVP measurements are used to estimate end-diastolic volume, given in the following relationship: VRV = CRV ⋅ (CVP − Pextracardiac) where VRV is end-diastolic volume, CRV is compliance of the right ventricle, CVP approximates the pressure inside the ventricle, and Pextracardiac is extracardiac pressure. Given this relationship, the general categories of physiologic perturbation that alter the direct relation between CVP and volume are as follows: a. Abnormal cardiac compliance, as in concentric hypertrophy b. Altered extracardiac pressure, as with high positive end- expiratory pressure (PEEP), abdominal compartment syndrome, or tamponade, for example c. Valvular abnormalities, as with tricuspid insufficiency or stenosis (where CVP will no longer approximate right ventricular pressures) d. Ventricular interdependence E. Complications: Central venous cannulation complications vary depending on the selected anatomic site (see Table 1.1) and operator experience. Serious immediate complications include catheter malposition, pneumo- or hemothorax, arterial puncture, bleeding, air or wire embolism, arrhythmia, and thoracic duct injury (with left subclavian or left internal jugular approach). Delayed serious complications include infection, thrombosis and pulmonary emboli, catheter migration, catheter embolization, myocardial injury or perforation, and nerve injury. Recommendations for placement technique and avoidance of infection are described in Clinical Anesthesia Procedures of the Massachusetts General Hospital, 7th Edition, chapter 10. IV. PULMONARY ARTERY CATHETERS A. Indications for placement of a pulmonary artery catheter (PAC) include need for monitoring PA pressures, measuring CO with thermodilution, assessing left ventricular filling pressures, and sampling of mixed venous blood. In addition, some PACs have pacing ports and can be used for temporary transvenous pacing. B. Technique: PACs are positioned by floating a distally inflated balloon through the right atrium and right ventricle (RV) into the PA. Figure 1.4 shows the characteristic pressure waveforms seen as the PAC is advanced. During placement, attention to the pressure tracing, electrocardiogram, systemic blood pressure, and oxygen saturation is essential to ensure proper placement of the catheter and to minimize known complications. FIGURE 1.4 Characteristic pressure waves seen during insertion of a pulmonary artery catheter. CVP, central venous pressure; IJ, internal jugular; PA, pulmonary artery; PCW, pulmonary capillary wedge; RA, right atrium; RV, right ventricle. 1. Fluoroscopic guidance may be useful in certain situations, such as in the presence of a recently placed permanent pacemaker (generally within 6 weeks), the need for selective PAC placement (eg, following pneumonectomy), and the presence of significant structural or physiologic abnormalities (eg, severe RV dilation, large intracardiac shunts, or severe pulmonary hypertension). C. Waveforms During Placement: The right atrial pressure waveform is the same as the CVP waveform previously described. The pressure waveform in the RV has a systolic upstroke (in phase with the systemic arterial upstroke) with low diastolic pressures that increase during diastole, owing to ventricular filling. The PA pressure waveform will also be in phase with systemic pressures during systole but will differ from the RV tracing as the pressure decreases during diastole. Often, the diastolic pressure will increase when the balloon enters the PA, but the better marker of this advancement is the transition to a downward slope during diastole. The pulmonary artery occlusion pressure (PAOP) or wedge pressure waveform will resemble the CVP trace with a, c, and v waves, although these are often difficult to distinguish clinically. D. Physiologic Data 1. Thermodilution CO a. Method: A rapid bolus of cold saline is injected proximal to the right heart and temperature is monitored at the distal tip of the PAC. With higher CO, more blood is mixed with the cold fluid bolus and the temperature recorded over time will be attenuated, as described by the modified Stewart-Hamilton equation: where CO is cardiac output, Tbody is the temperature of the body, Tinjectate is the temperature of the saline bolus, V is the volume of the bolus, K reflects properties of the catheter system, and AUC is the area under the curve of temperature change. b. Reliability: Averaging of serial measurements is recommended for each CO determination because the calculated CO may vary by as much as 10% without a change in clinical condition. It is important to minimize variations in the rate and volume of injection, which also introduce error. Colder solutions (ie, increased Tbody − Tinjectate) decrease error, although attention should be paid to potential tachy- or bradyarrhythmias. Tricuspid regurgitation may affect calculations as a result of recirculated blood between the right atrium and RV. Intracardiac shunt can likewise introduce error. 2. Pulmonary artery occlusion pressure: Occluding the PA recreates a static fluid column between the distal tip of the catheter and the left atrium, allowing for equilibration of pressures between the two sites. In this manner, PAOP approximates left atrial pressure, a surrogate for left ventricular end-diastolic volume. For an accurate measure of PAOP, the proper atrial trace, similar to the “a-c-v” trace of the CVP waveform, should be visualized. As highlighted in the discussion of CVP measurements, volume is only one parameter that influences the PAOP measurements. Other variables include cardiac compliance, intrathoracic pressures, valvular lesions, ventricular interdependence, and so forth. 3. Mixed venous oxygen saturation (SvO2): As CO increases, tissue oxygen demand is met with less per-unit oxygen extraction and SvO2 increases. This is a loose correlation because SvO2 also depends on hemoglobin concentration and oxygen consumption, as outlined by the Fick equation: · where CO is cardiac output in liters per minute, V O2 is oxygen consumption in millimeters per minute, SaO2 and SvO2 are arterial and mixed venous oxygen saturation, and hgb is hemoglobin in grams per deciliter. SvO2 correlates with CO, and a low SvO2 suggests low CO (assuming adequate oxygen extraction by the tissues and an adequate hemoglobin). ScvO2, an oxygen saturation drawn from a non-PA central line, can act as a surrogate for SvO2. It is typically higher than the SvO2 is by about 5% because it does not include the oxygen- depleted blood from the heart itself (returned to the right atrium from the coronary sinus), although the exact difference between ScvO2 and SvO2 is not predictable. E. Complications: In addition to complications associated with central venous access, PAC placement is associated with an increased risk of arrhythmia including right heart block (especially in patients with recent myocardial infarction [MI] or pericarditis) and PA rupture. PA rupture risk is increased with pulmonary hypertension, advanced age, mitral valve disease, hypothermia, and anticoagulant therapy. It requires emergent thoracotomy. Catheter-related complications, including knotting or balloon rupture with subsequent air or balloon fragment emboli, have also been reported. F. Relative Contraindications to PA placement include left heart block (because a superimposed right heart block would lead to a complete block), presence of a transvenous pacer or recently placed pacemaker or implantable cardioverter-defibrillator (ICD) leads, tricuspid or pulmonary stenosis or prosthetic tricuspid or pulmonary valves (given the associated difficulty in passing the catheter and balloon), and patient predisposition for arrhythmia, coagulopathy, or severe pulmonary hypertension. Need for magnetic resonance imaging (MRI) is also a contraindication because most PA lines contain ferromagnetic material. G. Controversy Over PACs: Although hemodynamic data derived from PACs enhance understanding of cardiopulmonary physiology, the risk- to-benefit profile has been questioned. Since the mid-1990s, several large outcome studies assessing the benefit of PACs have been conducted without clear evidence of mortality benefit. Moreover, PAC- guided therapy has been associated with more complications than has CVC-guided therapy. Although these results are not sufficiently convincing to completely discourage the use of PACs, they underscore the importance of using PACs only when the benefit of management guidance derived from PA data is strongly believed to outweigh the associated risks. V. ALTERNATIVES TO BLOOD PRESSURE MONITORING Numerous alternative hemodynamic monitors have been developed to assess either cardiovascular function or tissue perfusion. A. Transthoracic Ultrasound: The use of ultrasound for hemodynamic assessment of cardiac function and fluid status has been growing in critical care and is discussed further in Chapter 3. B. Continuous Esophageal Doppler: Blood velocity in the descending aorta can be measured with a transesophageal Doppler ultrasound. The average velocity over one heartbeat is multiplied by the cross-sectional area of the aorta (either estimated on the basis of patient data or measured by the probe) to calculate stroke volume. Stroke volume is multiplied by heart rate (HR) to obtain CO. Notably, because this is only the flow in the descending aorta, a certain percentage (typically around 30%) is added to estimate total CO. Modern probes are roughly the size of nasogastric tubes, much smaller than are ordinary transesophageal echocardiography probes, and can provide “corrected flow time” and stroke volume variation in addition to CO, which can be used to gauge fluid responsiveness. 1. Advantages: Esophageal Doppler monitoring allows for continuous measurement with minimal risk of infection, is simple to use with a short setup time, and has a low incidence of iatrogenic complications. 2. Disadvantages: Esophageal Doppler monitoring can only be performed in intubated patients, requires frequent repositioning if the patient is moved, is operator dependent, and is not widely available. C. Partial CO2 Rebreathing Method: The partial CO2 rebreathing method is based on the Fick principle: where CO is cardiac output, CaO2 is the arterial blood oxygen content of · oxygen, CvO2 is the venous blood oxygen content, and V O2 is oxygen consumption. · Clinical measurement of V O2 is challenging, so this technique is based on a restatement of the Fick equation for carbon dioxide elimination rather than for oxygen consumption: Using an intermittent partial rebreathing circuit, the change in CO2 production and end-tidal CO2 concentration in response to a brief, sudden change in minute ventilation is measured. The changes in end- tidal CO2 are used to calculate CO. 1. Advantages: This method is low risk, noninvasive, and can be performed every few minutes. The partial rebreathing CO2 CO method has also shown reasonably good agreement with gold standard thermodilution in clinical trials in some settings. 2. Disadvantages: As currently designed, this method requires tracheal intubation for measurement of exhaled gases. Measurements can be affected by changing patterns of ventilation and intrapulmonary shunting. Furthermore, this technique has a relatively long response time. D. Transpulmonary Thermodilution and Transpulmonary Indicator Dilution: With these techniques, the same principles for PAC thermodilution are employed but only a CVC and an arterial line are required. A bolus of either cold saline (with the PiCCO device) or lithium chloride (with the LiDCO device) is injected into the central line and the dilution over time in a peripheral artery is used to derive the CO. These are commonly used in conjunction with pulse contour analysis (see Section V.E) to provide continuous assessment of CO. 1. Advantages: Both of these methods have been shown to correlate reasonably well with PAC thermodilution. Transpulmonary thermodilution has the added benefit of providing an assessment of extravascular lung water and intrathoracic blood volume. 2. Disadvantages: Both methods require repetitive blood draws, and calibration may be affected by neuromuscular blocking agents. Notably, the PiCCO system typically requires the placement of an axillary or femoral arterial catheter. E. Pulse Contour Analysis: This modality for measuring CO relies on the principle that stroke volume and CO can be gauged from characteristics of the arterial waveform, using calculations that are based on estimates of compliance of the arterial tree. Commercially available devices require calibration, typically against thermodilution or indicator dilution methods. 1. Advantages: Pulse contour analysis devices are continuous and employ catheters (CVCs and arterial lines) that are commonly employed in patients in the ICU. 2. Disadvantages: This technique requires patients who are mechanically ventilated and has shown reduced accuracy in patients who are hemodynamically labile or on vasoactive medications. The altered arterial waveform of patients with aortic insufficiency may also decrease the accuracy of this technique. F. Impedance Cardiography (Also Known as Electrical Impedance Plethysmography): With impedance cardiography, a high-frequency, low-magnitude current is applied to the chest, and impedance is measured. As the aorta fills with blood with each heartbeat, impedance decreases, and this change is used to determine stroke volume and CO. Advances in phased-array and signal processing technologies have improved impedance cardiography, largely overcoming artifact due to electrode placement, HR and rhythm disturbances, and differences in body habitus, although its use is still fairly limited in the ICU. Both electrical velocimetry and bioreactance employ similar principles. Electrical velocimetry relates the velocity of blood flow in the aorta to determine CO, whereas bioreactance uses changes in electrical current frequency (rather than in impedance) to measure changes in blood flow during the cardiac cycle. 1. Advantages: This method is noninvasive and continuous. 2. Disadvantages: Impedance cardiography is more time consuming to set up and its usefulness is limited, with noisy environments, extravascular fluid accumulation, or arrhythmias. G. Tissue Perfusion Monitors: Most hemodynamic monitors are surrogates for adequate perfusion, whereas a few aim to assess perfusion at the tissue level. Notably, these only assess tissue perfusion in the tissues where they are measured. Gastric tonometry measures gastric CO2, which decreases with low perfusion states. Tissue oxygenation (StO2) measures the percentage of oxygenated hemoglobin at the microcirculation/tissue level. Although there are many tissue perfusion monitors, most of these are still used primarily for research and not for clinical applications at this time. Selected Readings Bartels K, Esper S, Thiele R. Blood pressure monitoring for the anesthesiologist: a practical review. Anesth Analg. 2016;122(6):1866-1879. Cavallaro F, Sandroni C, Marano C, et al. Diagnostic accuracy of passive leg raising for prediction of fluid responsiveness in adults: systematic review and meta-analysis of clinical studies. Intensive Care Med. 2010;36(9):1475-1483. Galluccio ST, Finnis ME, Chapman MJ. Femoral-radial arterial pressure gradients in critically ill patients. Crit Care Resusc. 2009;11(1):34-38. Gardner RM. Direct blood pressure measurement—dynamic response requirements. Anesthesiology. 1981;54(3):227-236. Kaufmann T, Cox EGM, Wiersema R, et al. Non-invasive oscillometric versus invasive arterial blood pressure measurements in critically ill patients: a post hoc analysis of a prospective observational study. Crit Care. 2020;57:118-123. Kaur B, Kaur S, Yaddanapudi LN, et al. Comparison between invasive and noninvasive blood pressure measurements in critically ill patients receiving inotropes. Blood Press Monit. 2019;24(1):24-29. Kim WY, Jun JH, Huh JW, et al. Radial to femoral arterial blood pressure differences in septic shock patients receiving high-dose norepinephrine therapy. Shock. 2013;40(6):527-531. Kobe J, Mishra N, Arya VK, et al. Cardiac output monitoring: technology and choice. Ann Card Anaesth. 2019;22(1):6-17. Messina A, Longhini F, Coppo C, et al. Use of the fluid challenge in critically ill adult patients: a systematic review. Anesth Analg. 2017;125(5):1532-1543. Monnet X, Rienzo M, Osman D. Passive leg raise predicts fluid responsiveness in the critically ill. Crit Care Med. 2006;34(5):1402-1407. Riley LE, Chen GJ, Latham HE. Comparison of noninvasive blood pressure monitoring with invasive arterial pressure monitoring in medical ICU patients with septic shock. Blood Press Monit. 2017;22(4):202-207. Saugel B, Kouz K, Mediert A, et al. How to measure blood pressure using an arterial catheter: a systematic 5-step approach. Crit Care. 2020;24(1):172. Singh A, Bahadorani B, Wakefield BJ, et al. Brachial arterial pressure monitoring during cardiac surgery rarely causes complications. Anesthesiology. 2017;126(6):1065-1076. 2 Respiratory Monitoring Catherine E. Naber and Ryan W. Carroll INTRODUCTION Intensive care units (ICUs) provide high-resolution and high-frequency surveillance for patients who are critically ill to capture disease trajectory in real time. This intensive monitoring enables clinicians to alter, add, or remove interventions to match the progression of disease, thereby reducing mortality, morbidity, and length of stay. Intensive monitoring in the ICU promotes a culture of safety and provides the data needed to develop thresholds for initiating, escalating, deescalating, and ceasing treatment. Whether by invasive or noninvasive means, this monitoring aims to accurately capture disease progression. Respiratory monitoring, invasive and noninvasive, and subsequent treatment are the mainstays of critical care; it is what separates the subspecialty from all others. This chapter discusses the core aspects of respiratory monitoring in the ICU setting and how to troubleshoot a shift in a patient’s respiratory trajectory. I. BASIC CONCEPTS: The chief focus of respiratory monitoring is on the delivery and removal (exchange) of gases, oxygen (O2), and carbon dioxide (CO2), respectively. A. Hypoxia is the condition of inadequate oxygen. Hypoxia can occur in the whole body (generalized hypoxia) or regionally (tissue hypoxia). 1. Reasons for hypoxia include the following: a. Hypoxemia b. Ischemia: a deficiency in oxygen supply due to insufficient blood flow c. Histotoxic: inability of cells to use oxygen despite appropriate delivery (eg, cyanide toxicity) d. Anemia e. Inhibition of the function of hemoglobin such as in carbon monoxide poisoning or elevated methemoglobinemia B. Hypoxemia is low partial pressure of O2 in the arterial blood (PaO2) resulting in a reduction of oxygen in the arterial blood. The PaO2 should always be interpreted in relation to the level of supplemental oxygen or FIO2 (fraction of inspired oxygen). a. For example, a PaO2 of 95 mm Hg breathing 100% oxygen is quite different from a PaO2 of 95 mm Hg breathing air (21% oxygen). b. The relationship between PaO2 and FIO2 is often captured in the PaO2/FIO2 (P/F) ratio and is used to determine severity of lung injury, specifically acute respiratory distress syndrome (ARDS). A low P/F ratio reflects poor oxygen exchange, for example. 1. Causes of hypoxemia include the following: a. Pulmonary diseases or conditions resulting in increased passage of deoxygenated blood from the right side of the heart to the left without participating in gas exchange · · 1. Increased shunt (Q S/Q T) · · 2. Ventilation-perfusion (V /Q ) mismatch 3. A low mixed venous PO2 (eg, decreased cardiac output) will magnify the effect of shunt on PaO2. b. Hypoventilation c. Diffusion defect d. PaO2 is also decreased with decreased inspired oxygen (eg, at high altitude). C. Anoxia: complete deprivation of oxygen D. Hyperoxemia (increased PaO2) may occur when breathing supplemental oxygen. The PaO2 also increases with hyperventilation. E. Ventilation/CO2 removal: Arterial partial pressure of CO2 (PaCO2) is · the balance between carbon dioxide production (V CO2) and alveolar · ventilation (V A). · PaCO2 = K × V CO2/VA where K is a constant. F. Dead space: Portions of the airways that do not participate in gas exchange. Physiologic dead space (VDphys) is considered the TOTAL dead space and is the sum of the anatomic dead space (VDana) and the alveolar dead space (VDalv). VDphys = VDana + VDalv 1. Anatomic dead space comprises airways known to be devoid of gas exchange mechanisms—areas without alveoli (ie, trachea, bronchi). a. Anatomic dead space in an adult is usually considered 150 mL (and in children or small adults, ~2 mL/kg). 2. Alveolar dead space is ventilated alveoli that lack concomitant perfusion, essentially leading to ventilation (gas flow) with reduced CO2 removal. a. This occurs whenever the pulmonary vasculature blood flow is lower than the level of gas exchange in the corresponding alveolar · space. The extreme of dead space ventilation is cardiac arrest (all V · and no Q ); but a pulmonary embolism (PE) is the pathophysiology more commonly encountered. 3. Note: Be mindful of added dead space from inappropriate additions to the airway circuit, which manifests greater relative impact in patients with smaller lung volumes (eg, pediatric, patients with severe kyphoscoliosis and achondroplasia). · · 4. Dead space, total (V DS/V T) can also be calculated from the Bohr equation by calculating the discrepancy between arterial blood gas (ABG)-based PaCO2 and end-tidal CO2 (ETCO2), which measures the ratio of dead space to total ventilation: VDS/VT = (PaCO2 − ETCO2)/PaCO2 where VDS is the volume of dead space and VT is the total tidal volume. a. The normal VDS/VT is 0.3 to 0.4, so 30% to 40% of the total tidal volume. G. Minute ventilation: Minute ventilation (liters per minute, LPM) = Tidal volume (VT, in mL or L) × respiratory rate (breaths per minute) 1. Minute ventilation (LPM) can also be framed as the sum of physiologic dead-space minute ventilation + alveolar minute ventilation. a. Multiplying the respiratory rate (RR) by the dead-space volume determined by the Bohr equation results in the minute ventilation of the respective volume. 1. For example, a physiologic dead space (or VDS) of 200 mL (0.33 of a total VT of 600 mL) at an RR of 14 = 2.8 LPM of dead-space minute ventilation; and conversely, the alveolar minute ventilation of the same patient receiving 600 mL VT results from subtracting the dead-space volume from the total (600 mL − 200 mL = 400 mL) × RR of 14 = alveolar minute ventilation of 5.6 LPM. The total minute ventilation, as displayed on the ventilator, would be 8.4 LPM. II. METHODS OF MONITORING GAS EXCHANGE A. Arterial Blood Gas (ABG) 1. ABG analysis is considered the standard assessment of pulmonary gas exchange. ABGs provide the following values: a. Arterial partial pressure of oxygen (PaO2) 1. Normal PaO2 is 90 to 100 mm Hg breathing room air at sea level. b. Arterial partial pressure of carbon dioxide (PaCO2) 1. Normal PaCO2 is 35 to 45 mm Hg. c. pH d. Base excess e. Blood oxygen saturation (or saturation level of oxyhemoglobin, SaO2), which can help verify pulse oximetry f. HCO3− (bicarbonate) 1. A calculated value using pH and PaCO2 through the Henderson- Hasselbalch equation pH = 6.1 + log [HCO3−/(0.03 × PaCO2)] g. ABG testing can also report dyshemoglobin subtypes such as the following: 1. Carboxyhemoglobin (COHb): reflects carbon monoxide inhalation. Endogenous COHb levels are 1% to 2% and can be elevated in cigarette smokers and people living in polluted environments. Because carboxyhemoglobin does not transport oxygen, the SaO2 is proportionally reduced by the COHb level. 2. Methemoglobin (metHb): The iron in the hemoglobin molecule can be oxidized to the ferric form in the presence of a number of oxidizing agents, the most notable of which are nitrates. Because methemoglobin (metHb) does not transport oxygen, the SaO2 is also proportionally reduced by the metHb level. 3. Fetal hemoglobin (HbF): has a higher affinity for oxygen than does adult hemoglobin and is involved in transportation of oxygen from the mother’s bloodstream to the fetus. Levels remain high in an infant until around 4 months of age, nearing 1% by 6 months of age. B. Peripheral venous blood gas (VBG): reflects PCO2 and PO2 at the local tissue level 1. Arterial PO2 (PaO2) versus venous PO2 (PvO2) a. PvO2 is affected by oxygen delivery and oxygen consumption at the tissue level, whereas PaO2 is affected by lung function. Thus, PvO2 should not be used as a surrogate for PaO2. 2. Venous pH versus arterial pH a. Venous pH is typically lower than arterial pH. 3. Venous PCO2 (PvCO2) versus arterial PaCO2 a. Venous PCO2 is typically higher than arterial. 4. Hemodynamic instability affects the difference between arterial and venous pH and PCO2. a. During cardiac arrest, for example, it has been shown that PvCO2 can be very high even when PaCO2 is low, a consequence of low cardiac output in the context of equivalent CO2 production. C. Mixed venous or central venous blood gas: preferred when using venous blood to assess acid-base status 1. Drawn from blood in the pulmonary artery that stems from blood flowing from the superior and inferior vena cavae and the coronary sinus. 2. Mixed venous oxygen (PvO2 and SvO2) provides an indication of global tissue oxygen extraction. a. Normal partial pressure of mixed venous oxygen (PvO2) is 35 to 45 mm Hg. b. Normal mixed venous oxygen saturation (SvO2) is 65% to 75%. c. Factors affecting mixed venous oxygen level can be illustrated by the following equation, a rearrangement of the Fick equation: SvO2 = SaO2 − [VO2 / (CO × Hb × 1.36)] where SvO2 is the mixed venous oxygen saturation; SaO2, arterial oxygen saturation; VO2, oxygen consumption; CO, cardiac output; and Hb is hemoglobin. 3. SvO2 is decreased if the following apply: a. Oxygen consumption is increased in the absence of increased delivery, such as in hyperthermia or pain. b. Cardiac output is decreased, such as in hypovolemia or shock. c. The patient is anemic. d. Oxygen saturation is decreased. 4. SvO2 is elevated under conditions of increased oxygen delivery: a. Increased inspired oxygen concentration b. Conditions of decreased oxygen utilization 1. Hypothermia 2. Sepsis 3. Extreme vasodilation D. Minimization of errors in blood gas sample collection 1. Care must be taken to avoid sample contamination with air because the PO2 and PCO2 of room air at sea level are approximately 155 and 0 mm Hg, respectively. 2. Care should also be taken to avoid contamination of the sample with saline from the sample line or venous blood. 3. Leukocyte larceny (spurious hypoxemia, pseudohypoxemia). The PaO2 in samples drawn from subjects with very high leukocyte counts can decrease rapidly. Immediate chilling and analysis are necessary. E. Laboratory analysis of blood gases 1. Blood gases and pH are measured at 37 °C. Using empiric equations, the blood gas analyzer can adjust the measured values to the patient’s body temperature. This feature is important to consider in the context of targeted hypothermia after cardiac arrest and focal cerebral ischemia. 2. Once the results of blood gases are available, they are often used to adjust the patient’s acid-base status through ventilation or medications. Two strategies for management exist. a. α-stat (“alpha-stat”) management is the acid-base adjustment method in which blood gas measurements (pH, PaCO2) obtained at 37 °C from the blood gas machine are directly used to reach the targets (PaCO2 = 40 mm Hg, pH = 7.4). Using empiric equations, the blood gas analyzer can adjust the measured values to the patient’s body temperature. 1. Most ICUs utilize the α-stat strategy. The choice of ventilation strategy is becoming increasingly important with the use of targeted hypothermia as a therapeutic tool. Because of increased gas solubility during hypothermia, the α-stat strategy results in relative hyperventilation (and, therefore, relative cerebral vasoconstriction and reduced cerebral blood flow). b. pH-stat management is an alternative method in which measurements obtained at 37 °C are corrected to the patient’s actual body temperature before use to achieve those same numerical targets. It may be necessary to enquire with your laboratory to determine whether the laboratory is reporting temperature-compensated values (ie, that may facilitate the use of the pH-stat strategy). c. There are different conditions under which each of these methods can be more or less advantageous. F. Pulse oximetry: Pulse oximetry has become a standard of care in the ICU. It is useful for titrating supplemental oxygen in patients with hypoxemia receiving noninvasive and invasive respiratory support. It displays the SpO2, or pulse oxygen saturation a. An SpO2 of 92%, or more, reliably predicts a PaO2 of 60 mm Hg or higher. SpO2 should be periodically confirmed by blood gas analysis of SaO2. 1. Principles of operation (Figure 2.1) FIGURE 2.1 Red and infrared wavelengths used in pulse oximetry and corresponding types of hemoglobin: oxyhemoglobin (O2Hb), reduced hemoglobin (sometimes labeled “deoxyhemoglobin,” HHb), methemoglobin (MetHb), and carboxyhemoglobin (COHb). Note where the lines of each hemoglobin subtype intersect with the corresponding red and infrared wavelength cutoffs. (Reprinted from Jubran A. Pulse oximetry. Crit Care. 2015;19(1):272. doi:10.1186/s13054-015-0984-8, with permission from Springer.) a. The commonly used pulse oximeter emits two wavelengths of light (infrared at 940 and red at 660 nm) from light-emitting diodes through a pulsating vascular bed to a photodetector. b. The ratio of absorption of red and infrared lights is used to determine the fraction of oxygenated hemoglobin (oxyhemoglobin). c. A variety of probes are available in disposable or reusable designs and include digital probes (finger or toe), ear probes, and nasal probes. 2. Accuracy a. Pulse oximeters use empiric calibration curves developed from studies of healthy volunteers. They are typically accurate within ±2% for SpO2 readings as low as 70%. Thus, their accuracy may be reduced in the clinical setting (see Limitations in Section II.F.6). b. As illustrated by the oxyhemoglobin dissociation curve (Figure 2.2), if the pulse oximeter displays an oxygen saturation (SpO2) of 95%, the true saturation could be as low as 93% or as high as 97%. This range of SpO2 translates to a PaO2 range from as low as about 60 mm Hg to greater than 150 mm Hg. FIGURE 2.2 Oxyhemoglobin dissociation curve. Note that small changes in oxygen saturation relate to large changes in partial pressure of oxygen (PO2) when the saturation is greater than 90%. Also note that the saturation can change without a change in PO2 if there is a shift in the oxyhemoglobin dissociation curve. 3. What do traditional pulse oximeters detect? a. Oxyhemoglobin and deoxyhemoglobin 4. What can multiple-wavelength pulse oximetry and co-oximetry detect? a. COHb, metHb, and total hemoglobin, in addition to SpO2. 5. Additional monitoring features of pulse oximetry a. Respiratory variation in the plethysmographic waveform 1. Photoplethysmography of peripheral perfusion is displayed by some pulse oximeters. The beat-to-beat plethysmogram displayed on the pulse oximeter reflects beat-to-beat changes in local blood volume. The cyclic changes in the blood pressure and plethysmographic waveform baseline can be caused by changes in intrathoracic pressure relative to the intravascular volume (likened to pulsus paradoxus). b. Perfusion index (PI) is a measurement displayed on many pulse oximeters. It is the ratio of the pulsatile blood flow to the nonpulsatile and thus represents a noninvasive measure of peripheral perfusion. c. Plethysmographic variability index (PVI) is a measure of the dynamic changes in the PI that occur during the respiratory cycle. The lower the number, the lesser the variability. 1. PVI may be increased in patients with severe airflow obstruction and/or in patients who are hypovolemic. It has been proposed as a predictor of fluid responsiveness in patients who are mechanically ventilated, although individual variability in waveform amplitude can be a limitation (Figure 2.3). FIGURE 2.3 Pulse oximeter waveform from a patient who responded to volume expansion and from another who did not respond to volume expansion. (Reprinted from Cannesson M, Desebbe O, Rosamel P, et al. Pleth variability index to monitor the respiratory variations in the pulse oximeter plethysmographic waveform amplitude and predict fluid responsiveness in the operating theatre. Br J Anaesth. 2008;101:200-206, with permission from Elsevier.) 6. Limitations of pulse oximetry a. Dyshemoglobinemia: Carboxyhemoglobinemia and methemoglobinemia (Figure 2.1) result in significant inaccuracy in dual-wavelength pulse oximeters. Multiple-wavelength pulse oximeters address these issues by measuring COHb% and metHb%. 1. Carboxyhemoglobinemia produces an SpO2 greater than the true oxygen saturation. 2. Methemoglobinemia causes the SpO2 to reduce commensurate with the rise in metHb but hovers around 85% when metHb levels rise beyond 15%. 3. HbF does not affect the accuracy of pulse oximeters. b. SpO2 (oxygen saturation) versus PO2: Because of the shape of the oxyhemoglobin dissociation curve, pulse oximetry is a poor indicator of hyperoxemia. It is also an insensitive indicator of hypoventilation. If the patient is breathing supplemental oxygen, significant hypoventilation can occur without desaturation measured by pulse oximetry. c. Ventilation versus oxygenation: Pulse oximetry provides no clinical information related to PaCO2 and acid-base balance. d. Differences between devices and probes: Calibration curves vary from manufacturer to manufacturer. The output of the light- emitting diodes of pulse oximeters varies from probe to probe. e. The penumbra effect occurs when the pulse oximeter probe does not fit correctly and light from the light-emitting diodes escapes the photodetector. It leads to an SpO2 reading that is erroneously lower than the actual value. f. Endogenous and exogenous dyes and pigments 1. Intravascular dyes (eg, methylene blue) and nail polish can also affect the accuracy of pulse oximetry. a. Although this issue may be less problematic in newer generations of pulse oximeters, it is nonetheless prudent to remove nail polish before applying the pulse oximetry probe. 2. Hyperbilirubinemia does not affect the accuracy of pulse oximetry. g. Skin pigmentation: The accuracy and performance of pulse oximetry may be affected by deeply pigmented skin, although this is debated by manufacturers. h. Perfusion 1. Pulse oximetry becomes unreliable during conditions of low flow such as low cardiac output or severe peripheral vasoconstriction. a. An ear probe may be more reliable than is a digital probe under these conditions. 2. A dampened plethysmographic waveform suggests poor signal quality. a. Newer technology uses signal processing software that improves the reliability of pulse oximetry with poor perfusion. i. Anemia: Although pulse oximetry is generally reliable over a wide range of hematocrits, it becomes less accurate under conditions of severe anemia. j. Motion of the oximeter probe can produce artifacts and inaccurate pulse oximetry readings. 1. Newer generation oximeters incorporate noise-canceling algorithms to lessen the effect of motion on signal interpretation. 2. Newer technology uses signal processing software that improves the reliability of pulse oximetry with motion of the probe. k. High-intensity ambient light can affect pulse oximeter performance and can be corrected by shielding the probe. l. Abnormal pulses: Venous pulsations and a large dicrotic notch can affect the accuracy of pulse oximetry. G. Capnometry is the measurement of CO2 in the airway. Capnography is the display of a CO2 waveform called the capnogram (Figure 2.4). FIGURE 2.4 Normal capnogram. Phase I, anatomic dead space; phase II, transition from dead space to alveolar gas; phase III, alveolar plateau. 1. The PCO2 measured at end exhalation is called the end-tidal PCO2 (ETCO2). a. TheETCO2 represents alveolar PCO2. It is a function of the rate at which CO2 is added to the alveoli and the rate at which CO2 is · · cleared from the alveoli. Thus, theETCO2 is a function of the (V /Q ). · · b. With a normal and homogeneously distributed (V /Q ), theETCO2 approximates the PaCO2. · · c. With a high V /Q ratio (dead-space effect), theETCO2 is lower than the PaCO2. · · d. With a low V /Q ratio (shunt effect), theETCO2 approximates the mixed venous PCO2. e. Changes inETCO2 can be due to changes in CO2 production, CO2 delivery to the lungs, or changes in the magnitude and distribution of the alveolar ventilation. 2. Types of capnometers a. Quantitative capnometers measure CO2 using the principles of infrared spectroscopy, Raman spectroscopy, or mass spectroscopy. b. Nonquantitative capnometers indicate CO2 by a color change of an indicator material, based on a shift in pH. c. Mainstream capnometer chambers are placed directly into the airway circuit. d. Sidestream capnometers aspirate gas from the airway circuit through tubing to a measurement chamber in the capnometer. 3. Uses of capnometers a. ETCO2 monitoring to confirm tracheal intubation is generally regarded as the standard of care for clinically confirming the correct placement of the endotracheal tube. Low-cost, disposable devices that produce a color change in the presence of CO2 are commercially available. b. Capnometers can aid in the detection of increases in dead space, prompting corrective interventions. This early dead-space detection is useful during resuscitation. 1. Be mindful of added dead space from inappropriate additions to the airway circuit. c. Capnometers can indicate obstructive respiratory mechanics via waveform changes. 1. Waveform: normal (Figure 2.4) and abnormal, as seen in a patient with obstructive lung disease when the waveform changes from box/plateau to saw-toothed (Figure 2.5). FIGURE 2.5 An increased phase III occurs in the capnogram in patients with obstructive lung disease. 4. Volumetric capnometry, also called volume-based capnometry, displays exhaled CO2 as a function of exhaled tidal volume (Figure 2.6). Note that the area under the volume-based capnogram is the volume of CO2 exhaled. Assuming steady-state conditions, this · · represents carbon dioxide production (V co2). Because V co2 is determined by metabolic rate, it can be used to estimate resting energy expenditure (REE): FIGURE 2.6 The volume-based capnogram. Note that the area under the curve represents carbon dioxide elimination, which equals carbon dioxide production during steady-state conditions. REE= CO2 (L/min) × 5.25 kcal/L × 1440 min/d a. Using volume-based capnometry and a partial-rebreathing circuit, pulmonary capillary blood flow can be measured by applying a modification of the Fick equation using a capnography tool called the NICO (noninvasive cardiac output monitor; Figure 2.7). With corrections for intrapulmonary shunt, this allows noninvasive estimation of cardiac output. There is significant variability in the clinical application of this method. FIGURE 2.7 Use of the partial carbon dioxide rebreathing method to measure cardiac output using capnometry. Assuming that changes in pulmonary capillary carbon dioxide content (Cc´CO2) are proportional to changes in end-tidal CO2 (ETCO2), we can use the following equation to calculate pulmonary capillary blood flow (PCBF): PCBF = ΔCO2/(S × ΔETCO2), where ΔCO2 is the change in CO2 output and S is the slope of the CO2 dissociation curve. Cardiac output is determined from PCBF and pulmonary shunt: CO = PCBF/(1 − s/t). Noninvasive estimation of pulmonary shunt (s/t) is adapted from Nunn’s isoshunt plots, which are a series of continuous curves for the relationship between partial pressure of oxygen (PaO2) and inspired oxygen (FIO2) for different levels of shunt. PaO2 is estimated using a pulse oximeter. PaCO2, arterial partial pressure of CO2. (Non invasive cardiac output (NICO) timing diagram courtesy of Novametrix, Wallingford, CT.) 5. Limitations of capnography a. There is considerable intra- and interpatient variability in the relationship between PaCO2 andETCO2. The difference between the two is often too variable in the critically ill to allow precise extrapolation of PaCO2 fromETCO2. III. INDICATORS OF OXYGEN DELIVERY, CONSUMPTION, AND GAS EXCHANGE A. Indices of Impaired Oxygenation 1. Shunt fraction: a measure of oxygen exchange inefficiency and is calculated by the shunt equation: · · Q s/Q t = (Cc´o2 − Cao2) / (Cc´o2 − Cvo2) a. Cc´o2 is the pulmonary capillary oxygen content. 1. To calculate Cc´o2, assume the pulmonary capillary oxygen content (PO2) is equal to the alveolar PO2 (PAO2) and the pulmonary capillary hemoglobin is 100% saturated with oxygen. b. CaO2 is the arterial oxygen content. c. CvO2 is the mixed venous oxygen content. CvO2 = (1.34 × Hb × HbO2) + (0.003 × Pao2) · · d. If measured when the patient is breathing 100% oxygen, the Q S/Q T represents shunt (ie, blood that flows from the right heart to the left heart without passing functional alveoli). If measured at FIO2 less · · · · than 1.0, the Q S/Q T represents shunt and V /Q mismatch (ie, venous admixture). 2. An increased P(A − a)O2 gradient is another measure of oxygen exchange inefficiency. The P(A − a)O2 is normally 10 mm Hg or less breathing room air and 50 mm Hg or less breathing 100% oxygen. The alveolar partial pressure of oxygen, PAO2, is calculated from the alveolar gas equation: PAO2 = [FIO2 × (PB – PH2O)] – [PaCO2/R] where PB is the barometric pressure, PH2O the water vapor pressure, and R the respiratory quotient. For calculation of PAO2, an R of 0.8 is commonly used. a. An increased difference the P(A − a)O2 gradient can be caused by 1. Shunt · · 2. V /Q mismatch 3. Diffusion defect 3. The ratio of the PaO2 to PAO2 (PaO2/PAO2) can also be calculated as an index of oxygenation and is normally greater than 0.75 at any FIO2. 4. PaO2/FIO2 is an easy index of oxygenation to calculate, correcting PaO2 to the FIO2 used. For example, it is used to classify ARDS as mild (201 – 300 mm Hg), moderate (101 – 200 mm Hg), or severe ( 300 mm Hg) should be avoided, given the association of hyperoxia with worse outcomes in many conditions including acute myocardial infarction and cardiac arrest. In practice, targeting SaO2 not higher than 92% to 94% and PaO2 not greater than 120 mm Hg seem to be safe. b. Circulation. Delivery of well-oxygenated blood to the tissues depends on an adequate cardiac output and driving pressure. Thus, fluid resuscitation plays an integral role in the treatment of shock. In the event that infusion of crystalloid, colloid, or blood products is insufficient to establish and maintain adequate systemic oxygen delivery, pharmacologic therapy with inotropes, and/or vasopressors may be required. 2. Fluid resuscitation is the cornerstone of the treatment of hypotension and shock. Its aim is both to increase effective circulating intravascular volume and, through the Frank-Starling mechanism, to increase cardiac ventricular preload and therefore cardiac output. Unfortunately, it is often difficult to predict whether, and by how much, the cardiac output will increase in response to volume loading. Although inadequate fluid replacement may result in continued tissue hypoperfusion and the progression of shock, overly aggressive resuscitation may result in heart failure and pulmonary and tissue edema, which in turn will further compromise tissue perfusion and possibly increase mortality. The fluids available for resuscitation include crystalloids, colloids, and blood products; however, the optimal choice of fluid remains controversial. In most ICU patients, fluid resuscitation should be limited to those who have been identified to be likely to be “responders” to the volume challenge. Recommendations for indiscriminate fluid administration, such as 30 mL/kg of crystalloids to all patients with septic shock, although part of some guidelines, should be discouraged in favor of a more thoughtful approach to resuscitation. a. Crystalloids. The most commonly used crystalloid solutions are lactated Ringer and 0.9% “normal” saline solutions, which are inexpensive, easily stored, and readily available. There are data to suggest that in most cases, lactated Ringer is a preferable crystalloid over 0.9% saline, as too much chloride may lead to hyperchloremic acidosis and risk increasing renal dysfunction. Of note, hyperkalemia or acute kidney injury are not contraindications to the administration of lactated Ringer or other balanced electrolyte solutions. Patients with traumatic brain injuries or those that need hyperosmolar therapies are an exception and should be resuscitated with 0.9% saline. b. Colloids include both natural and synthetic solutions. Because of their high molecular weight and increased osmotic activity, colloids remain in the intravascular space longer than crystalloids and thus require less volume to achieve the same hemodynamic goals. 1. Human albumin is derived from pooled human plasma and is available as 5% and 25% solutions in normal saline. Heat treatment eliminates the risk of transmission of viral infections. Although there is no evidence of harm from the use of albumin as a resuscitation fluid (except perhaps in head injury victims), no clear benefit has been shown either and its relatively high cost limits its widespread use. Relatively weak data support the use of albumin in patients with septic shock and patients with severe hypoalbuminemia. 2. Synthetic colloids include dextran and hydroxyethyl starch (HES). Because of their antigenicity and high incidence of anaphylactic and anaphylactoid reactions, the dextrans have largely been replaced by starch-based compounds. Hydroxyethyl starches are high-polymeric glucose compounds available with a variety of mean molecular weights and molar

Critical Care Handbook of the Massachusetts General Hospital PDF

Document Details

Tags

Related

Summary

Full Transcript