ECMO in AMICS: Amid Few Prospective Trials, a Multicenter Registry Yields Insights
Mortality is high, especially for patients with cardiac arrest, but ECMO can still play “an important early role,” researchers say.
Registry data on nearly 7,000 patients with cardiogenic shock who received venoarterial extracorporeal membrane oxygenation (VA-ECMO) provide a peek into the therapy’s real-world results across a wide range of conditions, according to the authors of a newly published research letter.
In a key finding, patients receiving VA-ECMO support for cardiogenic shock in the setting of acute MI did just as well as those who got it for other indications, but their likelihood of death rose when cardiac arrest entered the picture.
Overall, “our results support an important early role for VA-ECMO in this vulnerable population,” Deepak Acharya, MD (University of Arizona, Tucson), and colleagues write in the Journal of the American College of Cardiology.
Acharya told TCTMD that there were “no major surprises” from their analysis of the Extracorporeal Life Support Organization (ELSO) registry, with mortality rates in line with previous reports. “But we had a large number of patients and centers and events, so we were able to do more-robust analyses,” he explained. “This provides a basis for discussion with the patients, families, and referring physicians regarding what the options are and what the expected outcomes are when somebody requires ECMO or is being considered for ECMO after having acute MI shock.”
Cardiogenic shock is notoriously difficult to address, partly due to its heterogeneity, Acharya said. Only last year did the Society for Cardiovascular Angiography and Interventions release the first-ever set of standardized definitions for the disease, he pointed out.
Peter Eckman, MD (Minneapolis Heart Institute, MN), commenting on the paper, agreed that it clarifies how well ECMO performs in acute MI complicated by cardiogenic shock (AMICS). “In many ways, they have surprisingly good outcomes despite how sick they are. I think it supports that ECMO is a reasonable modality for many of these patients,” he said.
A key unanswered question for Eckman is how ECMO compares to Impella (Abiomed), since many hospitals favor one mechanical circulatory support (MCS) strategy over another.
Much like with Impella, it has been hard to conduct randomized controlled trials of ECMO in cardiogenic shock, where patients are suffering from a rapidly shifting, severe disease. Acharya said observational studies can help by identifying patterns that can be used to improve care.
Acharya’s team looked at 6,646 patients with cardiogenic shock who underwent VA-ECMO between 2014 and 2018, of whom 756 (11.4%) had acute MI. Compared with those who underwent VA-ECMO for other cardiac indications related to shock, the AMICS group tended to have lower pH, blood pressure, and cardiac index, as well as a higher prevalence of cardiac arrest and mechanical ventilation at the time of cannulation.
Mean hospital stay was 18.5 days. Survival to hospital discharge was similar regardless of whether patients did or didn’t have AMICS (40.2% vs 42.1%; P = 0.32) and didn’t change over the course of the study period.
Multivariable analysis identified older age, higher weight, race, lower pH and diastolic BP levels before VA-ECMO, initiation during cardiac arrest, and shorter time on VA-ECMO—but not AMICS—as independent predictors of in-hospital mortality.
This provides a basis for discussion with the patients, families, and referring physicians regarding what the options are and what the expected outcomes are. Deepak Acharya
The non-AMICS group included diverse presentations like nonischemic cardiomyopathy, myocarditis, peripartum refractory ventricular arrhythmia, among others. “There were just too many groups there to compare individually,” Acharya said. As such, he emphasized the within-AMICS patterns as more important.
Around half of the AMICS patients (51.2%) were decannulated with the expectation that they would recover, though fewer than three-quarters (71.6%) were still alive at 30 days. Cardiac arrest was particularly risky. Among the 58.2% of AMICS patients who experienced arrest within 24 hours ahead of VA-ECMO, survival was 36.8%. It dropped further to 29.2% for the 23.5% of AMICS patients in whom ECMO was initiated for CPR during cardiac arrest.
The takeaway is that “measures of more-profound or late shock are predictors of poor outcomes. . . . So earlier institution of MCS may improve outcomes,” Acharya suggested.
The researchers point to several possible reasons why AMICS didn’t raise mortality risk, despite the fact that these patients presented with higher clinical acuity. Myocardial injury may be reversed in AMICS, they suggest, and the hemodynamic stability achieved with MCS may facilitate revascularization. Moreover, AMICS may carry fewer long-term sequelae than other chronic heart diseases managed with VA-ECMO.
“Decision-making for ECMO initiation for AMICS remains similar to other cardiac indications, to ensure adequate perfusion and prevent or reverse end-organ damage,” they conclude. “ECMO should be offered for those patients with AMICS in whom the hemodynamic benefits of ECMO are considered to outweigh the inherent risks, then instituted rapidly before development of multisystem organ dysfunction or prolonged CPR.”
Survival rates are still quite low, Acharya acknowledged. “But given how sick these patients were at the time of implant, including a high proportion who had cardiac arrest preimplant, the benefit of ECMO may still be substantial,” he said.
Eckman, too, emphasized that “this is a pool of patients that are probably all going to die if you do nothing, so even if there’s 40% survival, that’s actually pretty good.”
Although these data solidify ECMO’s role, one caveat is that they are derived from the ELSO registry and thus represent centers already familiar with the therapy, Eckman pointed out. “Starting an ECMO program takes a ton of work and a lot of people are involved—neurologists, vascular surgeons, nurses, perfusionists. So just because these are the outcomes seen in this registry doesn’t mean that every hospital should start an ECMO program,” he cautioned.
Acharya D, Torabi M, Borgstrom M, et al. Extracorporeal membrane oxygenation in myocardial infarction complicated by cardiogenic shock: analysis of the ELSO registry. J Am Coll Cardiol. 2020;76:1001-1002.
- Acharya and Eckman report no relevant conflicts of interest.