Shockingly simple? Should you use manual or automated defibrillation in out of hospital cardiac arrest?

Abstract

In their EMJ paper, Derkenne and colleagues present an interesting study on the accuracy and speed of Emergency Physicians in assessing whether a defibrillator trace is shockable or nonshockable. The study used a webbased application (https://simul-shock.firebaseapp.com) to present 60 ECG rhythms from reallife outofhospital cardiac arrest (OHCA) cases to prehospital emergency physicians and compared their responses with a gold standard interpretation defined by three experts. In total, 190 complete responses were included in the analysis which identified a median sensitivity of 0.91 [IQR 0.81–1.00] to deliver a shock for shockable rhythms and specificity of 0.91 [0.80–0.96] to withhold a shock for a nonshockable rhythm. Sensitivity was highest where the shockable rhythm was ventricular tachycardia or coarse ventricular fibrillation (VF) (1.0 [1.0–1.0]) but significantly lower for fine VF (0.6 [0.2– 1.0]). We would recommend that you test yourself on the simulator app to see how you would have scored! This study raises a valuable question: whether prehospital practitioners should use an automated external defibrillator (AED) or use manual mode for the interpretation of rhythm and need for shock delivery in patients with OHCA. Emergency Physicians (EPs) manage patients conveyed to the Emergency Department in cardiac arrest on a daily–weekly basis and defibrillators are used in manual mode— the same skills are likely to be transferrable to the prehospital setting with a smaller team. However, in many settings. the majority of OHCA management is provided by paramedics and some Emergency Medical Services (EMS) insist that the AED mode is used for all OHCA cases. This aims to reduce the cognitive burden for a small resuscitation team, ensures 2 minute cycle timings are maintained and eliminates human performance variability in rhythm interpretation. The latter is particularly important when attendance at cardiac arrests or opportunities for training may be infrequent for an individual paramedic. The major disadvantage is that most AEDs require chest compressions to pause for 5–20 seconds to allow the machine to provide rhythm analysis: in comparison, EPs in this study took a median of 2.0–2.8 seconds to identify each cardiac rhythm. Pauses in chest compressions (CPR) are associated with a reduced likelihood of return of spontaneous circulation and are a particular concern in the majority of cardiac arrests where the underlying rhythm will not benefit from defibrillation. The present study showed that EPs exceeded the performance goals set for artefactfree ECG analysis by AEDs for coarse VF (performance standard >90% sensitivity) and those observed when an AED is applied in reallife practice (sensitivity for coarse VF 99% [95% CI 98 to 99]). By contrast, performance for fine VF was lower than that observed for reallife performance for AEDs (sensitivity 88% [95% CI 81 to 97]) and for nonshockable rhythms (specificity performance standard >95% specificity) and reallife performance (specificity 98% [95% CI 97 to 99]). This pattern of findings—shorter time to shock decision, high sensitivity for coarse VF, lower sensitivity for fine VF and lower specificity for nonshockable rhythms—when comparing clinician performance with AED performance is not new 4 and represents the tradeoff made by EMS systems when prioritising automated over manual defibrillation. It is unsurprising that Derkenne’s simulated study found that clinicians diagnosed ventricular tachycardia and coarse VF quicker and more accurately for defibrillation compared with fine VF, pulseless electrical activity or asystole. Previous 2015 European Resuscitation Council guidelines advised that “very fine VF which is difficult to distinguish from asystole is unlikely to be shocked successfully into a perfusing rhythm”. Therefore, continuing goodquality CPR may improve the amplitude/frequency of VF, thereby improving the chances of subsequent successful defibrillation and avoiding myocardial injury from unnecessary shock delivery or interruption of chest compressions. The 2021 European Resuscitation Council Guidelines now state: “The 2015 ERC ALS Guideline stated that if there is doubt about whether the rhythm is asystole or extremely fine VF, do not attempt defibrillation; instead, continue chest compressions and ventilation. We wish to clarify that when the rhythm is clearly judged to be VF a shock should be given”. This challenges the previously adopted concept that very fine VF should not be defibrillated and simplifies the decisionmaking to “any VF=shock”. Not all prehospital practitioners may be current with this development. It is also important to recognise that the progression from coarse VF to fine VF to asystole is a continuum. While coarse VF is easy to recognise, as it becomes finer, visual differentiation between fine VF and asystole is more difficult and likely to be subject to individual variation between clinicians. AEDs traditionally defibrillate fine VF with a measured amplitude typically <0.2 mV but not asystole typically defined as having an amplitude <0.1 mV. International guidelines do not state amplitude size to define each rhythm, and it is unlikely to be feasible to calculate on a defibrillator monitor screen, with a moving rhythm, and while providing all elements of resuscitation during a cardiac arrest. AED use has demonstrated a clear survival benefit when used by lay bystanders through reducing the time from onset of cardiac arrest to first shock, but there is no definitive evidence to date that automated waveform analysis increases survival when compared with manual defibrillation by trained advanced life support providers. In 2017, a study conducted in a single Australian highperforming, paramedicbased EMS compared the outcomes of cardiac arrest patients using manual mode with the results after 3 years of using AED rhythm analysis and defibrillation. The study found a significant increase in the proportion of patients with an initial shockable rhythm receiving the first shock within 2 min of arrival. However, alarmingly the revised protocol also resulted in a reduction in ROSC of −5.7% (95% CI −1.9% to −9.4%) and overall survival of −6.7% (95% CI −3.3% to −10.1%). The authors concluded these data supported the use of a manual defibrillation protocol with regular team training in rhythm recognition and simulated practice to ensure CPR continues during defibrillator charging. Lower rates of survival have also Emergency Department, University Hospitals Coventry & Warwickshire NHS Trust, Coventry, UK Warwick Clinical Trials Unit, Warwick Medical School, University of Warwick, Coventry, UK