Can Anesthesiologists Help? Analysis of the American Heart Association–Emergency Cardiac Care Recommendations for Cardiopulmonary Resuscitation, with a Focus on Management of Airway and Breathing

THE CHALLENGE OF GUIDELINE CREATION Clinical practice guidelines were originally developed to assist physicians with improving the quality and consistency of clinical decision-making and ultimately improving patient outcomes.1 Guidelines are intended to synthesize a complex literature and support clinical decision-making. However, many problems may arise both from the methods used to derive the guidelines and from their application.2 Both the literature and expert opinion may be swayed by economic incentives to provide inappropriate care.3,4 Guideline recommendations may become rapidly outdated with new evidence. Failure to identify academic misconduct or to appreciate potential harm5–7 may prevent guidelines from improving care. Challenges in guideline creation are compounded by their increasing use in an age of pay-for-performance as arbiters of medical quality or as tools for litigation.8,9 These issues, in addition to failure to understand new evidence or appreciate its potential effect on current care paradigms,6 may be additional sources of guideline failure. Under extreme circumstances, care that deviates from the guidelines may actually produce better outcomes than guideline-compliant care.10 Anesthesiology was the first specialty to make patient safety a primary focus. The development of anesthesiology guidelines, standards, and practice criteria has been credited with a large reduction in perioperative mortality. Furthermore, this reduction has occurred in parallel to an increasing burden of chronic disease and greater surgical complexity.11 Despite this proven skill in improving patient care through rigorous research in anesthesia,12–14 and despite the extensive experience of anesthesiologists in guideline development and implementation, the American Heart Association (AHA) continues to publish and promote resuscitation guidelines (including airway management recommendations) with little anesthesia contribution. HOW ARE AHA CARDIOPULMONARY RESUSCITATION GUIDELINES DEVELOPED? Current guidelines for cardiopulmonary resuscitation (CPR) and emergency cardiac care (ECC) constitute one of the most comprehensive efforts to synthesize existing literature into specific practice recommendations. The opening chapter of these guidelines, entitled “International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science With Treatment Recommendations,” describes this process. The International Liaison Committee for Cardiopulmonary Resuscitation (ILCOR) is an international consortium of many of the resuscitation councils of the world. AHA is a member of this consortium. ILCOR has developed a process in which international experts “reviewed, analyzed, evaluated, debated, and discussed research and hypotheses through in-person meetings, teleconferences, and online sessions (“webinars”) during the 36-month period before the 2010 Consensus Conference.”15 During the last guideline update, this effort produced >400 scientific evidence reviews that form the basis of the 2010 AHA-ECC guidelines for CPR.16 The evidence underlying therapeutic interventions was organized using standard evidence tables of PICO questions (Population, Intervention, Comparison, Outcome).16,17 The quality of evidence for studies related to therapeutic interventions was assessed on a scale ranging from human randomized controlled trials (Level of Evidence [LOE] 1), through trials with concurrent controls (LOE 2) or retrospective controls (LOE 3), uncontrolled human studies (LOE 4), to related studies in alternative populations (e.g., nonarrested humans, animal studies), and bench and mathematical models (LOE 5).18 These LOEs were then grouped into 3 levels of recommendation (A–C), in which studies with LOE 1 were designated as A, studies with LOE 2 or 3 were designated as B, and those with LOE 4 or 5 were designated as C.19 The potential benefit (or harm) stemming from implementation of each recommendation was then classified as follows: Class I—“intervention should be performed”; Class IIa—“intervention is reasonable to perform”; Class IIb—“intervention may be considered”; and Class III—“intervention is not helpful, may be harmful.” Guidelines before 2010 also had a “Class Indeterminate.” After 2010, this term was eliminated; when AHA writing groups felt that the evidence was insufficient to offer a recommendation either for or against the use of a drug or intervention, no recommendation was given. These standardized classifications were intended to ensure consistency across the guidelines.20 What is the level of evidence underlying the American College of Cardiology--ECC guidelines, and do higher levels of recommendation correlate with potential benefit for our patients? To study this question, we tabulated both the level of recommendation and the judgment of benefit for all recommendations contained in chapters 5 (Basic Life Support [BLS])21 and 8 (Advanced Life Support [ALS])22 of the most recent AHA-ECC recommendations. We excluded 7 recommendations for management of acute coronary syndromes and 5 for management of stroke. We also excluded 3 recommendations that had inconsistent judgments of benefit and levels of recommendation. One of these (“Lay rescuers should continue CPR until an automatic external defibrillator arrives/emergency medical service providers take over care”) was assigned 2 different judgments of benefit (I and IIa) despite having the same level of recommendation in the same chapter. The 2 other excluded recommendations were assigned different levels of recommendation despite similar judgment of benefit (“Routine use of cricoid pressure is not recommended” was assigned a stronger recommendation in the BLS than in the ALS section despite similar judgment of benefit [III] in both chapters, and “Precordial thump may be considered for patients with witnessed, monitored, unstable Ventricular Tachycardia if a defibrillator is not immediately ready for use” was assigned as both a grade B and a grade C recommendation despite similar judgment of benefit [IIb] within the ALS section) (Appendix in Supplemental Digital Content,. Only 16% of the recommendations (23 of 147) suggested that an “intervention should be performed” (Fig. 1), and only 8% of the recommendations (n = 12 of 147) were based on randomized controlled trials or meta-analyses (Fig. 2). The proportion of recommendations based on case series, extrapolation of related literature, and expert opinion was the largest within the topic of monitoring during resuscitation (which made its debut in the 2010 guidelines) and second largest within the topic of airway and breathing (Fig. 3). Conversely, the second-largest proportion of Class I and Class IIa recommendations pertain to management of airway and breathing (Fig. 4). Overall, assessments of potential benefit for the recommendations correlated poorly with the strength of evidence underlying them (Gamma correlation value 0.150, P = 0.194)Figure 1: Distribution of the judgments given regarding the potential benefit of basic and advanced life support recommendations provided in the 2010 American Heart Association Emergency Cardiac Care (AHA-ECC) Guidelines for Cardiopulmonary Resuscitation.Figure 2: Distribution of the overall level of the evidence underlying basic and advanced life support recommendations provided in the 2010 American Heart Association Emergency Cardiac Care (AHA-ECC) Guidelines for Cardiopulmonary Resuscitation.Figure 3: The distribution of the levels of the recommendations provided in the 2010 American Heart Association Emergency Cardiac Care (AHA-ECC) Guidelines for Cardiopulmonary Resuscitation per topic. Basic life support (BLS) and advanced life support (ALS) recommendations can be classified into >1 topic category (e.g., the “30 compression: 2 breath” rule pertains both to airway and breathing as well as to circulation). Upon categorizing the recommendations into the 6 topics, 29 recommendations fell into 2 categories. After 3 repeated recommendations were eliminated, 174 recommendations were sorted into the 6 topic categories (Appendix in Supplemental Digital Content, https://links.lww.com/AA/B41).Figure 4: Distribution of the judgments of benefit of the recommendations provided in the 2010 American Heart Association Emergency Cardiac Care (AHA-ECC) Guidelines for Cardiopulmonary Resuscitation Guidelines per topic. The sum total of recommendations is 174, as elucidated in the legend for Figure 3.What Changes Have Occurred in the Recommendations for the Management of Airway and Breathing in the Guidelines During the Past 20 Years? We reviewed airway and breathing management guidelines published in the BLS and ALS sections of the AHA resuscitation guidelines during the past 20 years (1992,23 2000,24–26 2005,27,28 and 201021,22) and compared the judgment of benefit of both BLS and ALS recommendations (e.g., tidal volume, respiratory rate, % FIO2) and the level of the evidence of the most recent recommendations. BLS Airway and Breathing Recommendations Judgments of benefit for early (1992) BLS airway and breathing recommendations (Table 1) were often classified as I (“intervention should be performed”) (n = 7) or “no judgment of benefit” (n = 4). In 2000, most airway and breathing BLS recommendations became either “reasonable to perform” (Class IIa) or “intervention may be considered” (Class IIb) (n = 7) unless they continued to be unclassified (n = 6) or were categorized as having indeterminate benefit (n = 2). Seven of the 1992 BLS recommendations are still in the 2010 guidelines, but none are now considered more than Class II recommendations. Three previous recommendations are now considered possibly damaging (100% oxygen, checking for a pulse, and ventilation at rates >10 bpm), and 2 prior Class I recommendations have become either obsolete (peak inspiratory pressure limit) or left with no specific recommendation (airway insertion).Table 1: Changes in Basic Life Support Major Airway and Breathing Recommendations Over TimeALS Airway and Breathing Recommendations Early ALS airway and breathing recommendations (Table 2) were few and were often classified as “intervention should be performed” (insertion of a tracheal tube, respiratory rate 10–12 bpm, and tidal volume 10–15 mL/kg) or unclassified (cricoid pressure and clinical validation of tube location). None of these early recommendations have survived unchanged. Between 1992 and 2010, additional recommendations have appeared and disappeared (e.g., the stylet, the gum elastic bougie, tube holders, respiratory rates of 12–15 bpm, and tidal volumes of 10 mL/kg after intubation). Of those still in existence, only 18% (8 of 44 airway recommendations in the 2010 guidelines) are currently classified as Class I. Four of the remaining recommendations deal with assessment of breathing and the importance of ventilation. One addresses the need to secure the tube in place, one explains the priority of defibrillation over airway placement (Appendix in Supplemental Digital Content, https://links.lww.com/AA/B41), one addresses the need for frequent experience/retraining, and the final surviving Class I recommendation recommends the use of capnography to verify tube location (Table 2).Table 2: Changes in Advanced Life Support Major Airway and Breathing Recommendations Over TimeWHAT DO ANESTHESIOLOGISTS THINK OF THE GUIDELINES? In light of their role as specialists in airway management and resuscitation, anesthesiologists are extremely familiar with clinical resuscitation. To provide some perspective on whether anesthesiologists agree with the authors of the American College of Cardiology--ECC guidelines with respect to the evidence underlying the management of airway and breathing and the associated levels of recommendation, we performed a survey among attendees of the 2011 annual meeting of the American Society of Anesthesiologists. At a 90-minute pro–con session entitled “Should we be following ASA not AHA guidelines” delivered by 2 of the authors, attendees were issued an electronic response device and informed that their responses may be used for study purposes. All but a minority felt that AHA-ECC airway and breathing management guidelines were not up to par with anesthesia airway management guidelines (Table 3) and that anesthesia experts should be more involved in the writing of these guidelines.Table 3: Participant Responses to Questions Regarding the Level of the Evidence Underlying Current GuidelinesAHA-ECC GUIDELINES ARE A MOVING TARGET The paucity of evidentiary support for guideline recommendations is not a new finding,29,30 nor are mismatches between the strength of evidence and level of recommendation.31–33 The goal of evidence-based medicine is to “track down the best external evidence with which to answer our clinical questions.”34 As shown above, the body of literature that informs the current incarnation of ALS recommendations falls short of this ideal because few ALS recommendations are based on evidence from the highest quality of research (Fig. 2). In light of these observations, ongoing revision of the original CPR guidelines using contemporary evidence-based approaches35 is a positive development. In guidelines that have existed for decades, change is expected and likely reflects better science. Both animal studies and clinical trials during the past 15 years have produced fundamental changes in CPR. One recent example is the 2013 article by Hasegawa et al., which describes an increased likelihood of neurologically favorable survival of adult victims of cardiac arrest who received bag-valve-mask ventilation alone compared with those who underwent any type of advanced airway management.36 Such findings clearly require further assessment, which would benefit from involvement by experts in airway management. The exponential increase in the rate of clinical resuscitation research worldwide37 suggests that guidelines will likely continue to change as they become based on increasingly better evidence. However, as of now, changes are occurring with little anesthesia contribution or involvement. A CALL TO ACTION FOR ANESTHESIOLOGISTS The use of lipid emulsions for treatment of local anesthetic-induced cardiovascular collapse38–40 and the study of maternal arrest and resuscitation41–44 show that anesthesiologists can significantly contribute to resuscitation research. We should continue to contribute to future resuscitation guidelines with more well-performed clinical research in our areas of expertise. Alternative pressors (some of which are being used mainly by anesthesiologists), balanced drug use, early warning systems, monitoring (e.g., capnography and echocardiography for assessing return of spontaneous circulation), and differences between various types of cardiac arrest (e.g., respiratory, iatrogenic) are but a few such topics. We should collaborate more with the community of researchers in resuscitation. Much of what is known today about airway management is either taken from expert opinion or inferred from observational studies or “randomized controlled trials without pertinent controls.”45 This paucity of controlled data is most glaring in the uncontrolled environment of CPR. We should use our expertise in airway management to identify causes of failures in airway management both in the hands of less experienced providers in general and in the hands of highly trained providers in austere settings. We should also share our experience in managing critical situations unique to the perioperative and inpatient settings, which remain unaddressed in current guidelines. ILCOR is currently considering dynamic online guideline development using a “wiki” platform (methodologically similar to the development and maintenance of Wikipedia). Regardless of the platform to be used, anesthesiologists should become more involved in the development of resuscitation guidelines. Resuscitation is a multidisciplinary effort and can only benefit from the participation of anesthesiologists. We anesthesiologists have proved that we can change patient outcomes for the better. What better place to expand this contribution than in the area of medicine in which patients have the grimmest outcomes of all? DISCLOSURES Name: Sharon Einav, MD. Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript. Attestation: Sharon Einav approved the final manuscript. Name: Nechama Kaufman, MN. Contribution: This author helped write the manuscript. Attestation: Nechama Kaufman approved the final manuscript. Name: Michael O’Connor, MD. Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript. Attestation: Michael O’Connor approved the final manuscript. This manuscript was handled by: Avery Tung, MD.