Extracorporeal CPR in Out-of-Hospital Cardiac Arrest — Still on Life Support?
John F. Keaney, Jr., Thomas Münzel
N Engl J Med 2023; 388:370-371
DOI: 10.1056/NEJMe2214116
Despite many advances in the treatment of heart disease over the past four decades, up to one third of all patients with acute manifestations of heart disease do not reach the hospital after cardiac arrest. Only 8 to 10% of patients with out-of-hospital cardiac arrest survive; the event is responsible for up to 570,000 deaths annually in North America and Europe. The patients who are most likely to survive cardiac arrest are those who present with ventricular fibrillation that is responsive to defibrillation. Yet even among patients with a shockable rhythm, up to half will have ventricular fibrillation that is refractory (unresponsive to three shocks), a factor that portends a poor prognosis. Undoubtedly, this poor prognosis is partly related to end-organ ischemia that is manifest despite the continuous application of cardiopulmonary resuscitation (CPR).
One strategy that has been proposed to limit end-organ ischemia in patients who have ventricular fibrillation that is refractory to CPR and defibrillation is extracorporeal life support, commonly known in this context as extracorporeal CPR.1 When this strategy is used, patients undergo cannulation of the venous and arterial system to provide perfusion support, extracorporeal oxygenation, and temperature management as a means of organ preservation while the return of spontaneous circulation is pursued, often with the use of coronary angiography and intervention.1 In geographic locations with developed emergency medical systems (EMS), this strategy has proven feasible. However, the safety and effectiveness of extracorporeal CPR is not yet clear, since most data are derived from case series or single-center studies. Indeed, the most recent guidelines of the European Resuscitation Council characterized the evidence supporting extracorporeal CPR as being of very low certainty.2
In this issue of the Journal, Suverein and colleagues3provide important additional data to address the usefulness of extracorporeal CPR in refractory out-of-hospital cardiac arrest. In the multicenter, controlled INCEPTION (Early Initiation of Extracorporeal Life Support in Refractory Out-of-Hospital Cardiac Arrest) trial, Suverein et al. randomly assigned 160 adults with witnessed out-of-hospital cardiac arrest to receive extracorporeal CPR or conventional CPR. Key inclusion criteria were a shockable rhythm (i.e., ventricular fibrillation or pulseless ventricular tachycardia) and no return of spontaneous circulation within 15 minutes after the initiation of CPR. The trial was conducted in the Netherlands, where EMS personnel were able to reach patients in approximately 8 minutes and hospital arrival averaged approximately 35 minutes after arrest. The primary outcome of the trial was 30-day survival with a favorable neurologic outcome that corresponded to a cerebral function that is at least sufficient for independent activities of daily life. In total, 134 of the patients who underwent randomization were successfully enrolled, and intention-to-treat and per-protocol analyses showed no significant difference in the primary outcome between extracorporeal CPR and conventional CPR.
These results are likely to be disappointing to proponents of extracorporeal CPR, particularly in light of the previously reported ARREST (Advanced Reperfusion Strategies for Refractory Cardiac Arrest) trial,4which compared the use of a highly standardized extracorporeal CPR pathway with the use of conventional CPR and was terminated early owing to a difference in survival of 36 percentage points in favor of patients who received extracorporeal CPR. The inclusion criteria and patient demographic characteristics in the ARREST trial were similar to those used by Suverein et al., but some important distinctions merit recognition. ARREST enrolled a very modest sample size — only 30 patients — before its premature termination. This small sample size and the early termination are both characteristics known to increase the risk that any finding would be attributable to the play of chance. In addition, ARREST was a single-center trial, whereas the INCEPTION trial3 involved multiple centers, much as would be expected under real-world conditions if extracorporeal CPR were to become more widely available.
A single-center, randomized trial conducted in Prague and recently reported by Belohlavek et al.5involved 264 patients with witnessed out-of-hospital cardiac arrest who did not have a return of spontaneous circulation within 5 minutes after the initiation of CPR. Patients were transported to the trial cardiac center and randomly assigned to receive extracorporeal CPR or conventional advanced cardiac life support. Similar to the INCEPTION trial, the Prague trial was conducted in a well-developed EMS environment, with response times and time to extracorporeal CPR that compared favorably with those in INCEPTION. The primary outcome was survival to 180 days with a favorable neurologic outcome (i.e., the same definition as that used by Suverein et al.).3 The Prague trial was terminated early for futility after assessment of 256 patients, with the primary outcome having been observed in 32% of patients in the extracorporeal CPR group as compared with 22% in the conventional group. Bleeding events were twice as common with extracorporeal CPR (31%) as with the conventional strategy (15%). Thus, similar to the trial by Suverein et al.,3 the Prague trial showed that extracorporeal CPR did not significantly improve outcomes as compared with conventional CPR.
Although the data from randomized trials that have evaluated extracorporeal CPR to date are mixed, these findings do not necessarily indicate that extracorporeal CPR is not useful in some cases. Collectively, results of all three randomized trials3-5 showed a numerical advantage in their respective primary outcomes in favor of extracorporeal CPR as compared with conventional therapy. Thus, it is possible that larger trials are needed for the statistical power necessary to capture any benefit associated with extracorporeal CPR as compared with conventional CPR. Similarly, we must consider that we have not adequately identified the specific subsets of patients who may benefit from extracorporeal CPR. This contention is supported by the fact that the two largest randomized trials of extracorporeal CPR only managed to enroll approximately 6 to 7% of the patients with out-of-hospital cardiac arrest who underwent screening.
Available data do not yet support the broad use of extracorporeal CPR in patients with witnessed out-of-hospital cardiac arrest. Definitive evaluation of extracorporeal CPR will probably require studies on a much larger scale with robust statistical power to estimate any potential benefit of this resource-intensive strategy. Until then, the future of extracorporeal CPR remains in question.
