Improved survival with extracorporeal cardiopulmonary resuscitation despite progressive metabolic derangement associated with prolonged resuscitation
Bartos JA, Grunau B, Carlson C, et al. Circulation. 2020;141(11):877-886.
SUMMARY:
Extracorporeal membrane oxygenation (ECMO) has become increasingly common over the past 5-10 years. Many small studies have shown better-than-expected results from this hemodynamic-support strategy for a myriad of conditions. Most studies have been small and uncontrolled, thus leading to great uncertainty. Large health systems have been implementing ECMO programs for selected patients, particularly those with out-of-hospital cardiac arrest and shock-refractory ventricular tachycardia (VT) or ventricular fibrillation (VF). The logic is to keep patients alive until the arteries can be opened.
In fact, the desire to use ECMO on these patients is a prime reason for the substantial interest in the LUCAS device for prehospital care. Transportation with EMS delivering CPR is potentially ineffective and dangerous for providers. This study conducted at the University of Minnesota describes broad experience with this ECMO approach based on integrated prehospital and emergent cardiac catheterization-laboratory care, as compared with a historical control.
This was a retrospective analysis of 160 consecutive patients brought to the University of Minnesota for their ECPR program.
The inclusion criteria were as follows: (1) adults with out-of-hospital cardiac arrest, (2) shock-refractory (initial rhythm) VF/VT, (3) amiodarone administration, (4) an ability to place the LUCAS device, and (5) an expected time to catheterization laboratory <30 minutes. Patients were brought to the catheterization laboratory, where they were pronounced dead if their end-tidal CO2 was <10, O2 partial pressure was <50, O2 saturation was <85%, and lactate was >18. The others received ECMO, targeted temperature management, percutaneous coronary intervention, intraaortic balloon pump, and other procedures as needed.
The control group was nonconcurrent and came from a publicly available data set from the Amiodarone, Lidocaine, or Placebo Study (ALPS) trial, which examined amiodarone, lidocaine, and placebo for shock-refractory VF/VT (and did not show a benefit of amiodarone). The ALPS trial was performed across the country and was published in 2016.
Results: the authors found survival with good neurologic outcomes in 33% of the ECPR group compared with 23% of the control group. The number needed to treat was 10. The purported benefits of ECPR were primarily in patients who were connected very quickly (20-29 minutes; 100% survival, or 8/8) vs 24% in the control group. Survival with good neurologic outcomes was reported to be at 19% in the ECPR group, which had a CPR time >60 minutes, compared with 0% in the ALPS group.
The results may have been influenced by selection bias. Patients who died in the field were extremely likely to not be transported and therefore to not be included in the study, but they would have been included in the control group. In fact, evidence supports this possibility, because the ALPS group was older, less likely to have experienced arrest, and less likely to have received bystander CPR. Moreover, how the authors determined favorable neurologic outcomes in the ALPS cohort is unknown. In the ALPS trial, the amiodarone group was assessed by an expert blinded to treatment, but this was not the case in the experimental arm. The assessment was performed by the clinical team, which would have had a tendency to find better outcomes. The extent of this assessment bias cannot be assessed. Interestingly, there is 1 really objective outcome, death, that is much less subject to biased assessment but was not reported.
This study should serve as a basis for an RCT of this aggressive prehospital and ECPR strategy, but is far from convincing on its own.
EDITOR’S COMMENTARY: This relatively small single-center, nonrandomized evaluation of ECMO for shock-refractory VT/VF in the out-of-hospital setting demonstrated an association with improved survival and good neurologic outcomes, as compared with the results in a historical control. The data are interesting and should serve as a motivation for a well-conducted RCT rather than enthusiastic adoption of this invasive and expensive modality.
Copyright 2020 by Emergency Medical Abstracts – All Rights Reserved 05/20 - #02
