Why is this study important?

Mortality from cardiogenic shock has remained largely unchanged over the past 3 decades. Mechanical circulatory support may play a role in reducing mortality, but randomized trials evaluating mechanical circulatory support are limited. 

What question was this study supposed to answer?

Venoarterial (VA) extracorporeal membrane oxygenation (ECMO) also known as extracorporeal life support (ECLS) is available for circulatory support in patients with cardiogenic shock in many institutions worldwide and the utilization has been increasing annually. To date, there have been no randomized controlled trials evaluating the use of venoarterial ECMO in patients with cardiogenic shock. 

ECMO-CS was a modest-sized RCT including 117 cardiogenic shock patients at four centers in the Czech Republic between 2014 and 2022. Patients in severe cardiogenic shock (corresponding to SCAI stage D or E shock) were randomized to immediate VA ECMO cannulation versus early conservative therapy with inotropes and vasopressors. The primary endpoint was a composite of death from any cause, resuscitated arrest or implantation of another mechanical circulatory support device at 30 days.  

What did the study show?

There was no significant difference in the primary endpoint between the early ECMO cannulation group compared to the initial conservative therapy group at 30 days in the intention to treat analysis. There was also no difference in the secondary endpoint of serious adverse events, which included bleeding, limb ischemia, stroke, sepsis and pneumonia. There are several limitations to this study. First, the study took eight years to enroll 117 patients across four ECMO centers. Thus, there were likely many patients with cardiogenic shock who met the criteria for the study, but were not enrolled, suggesting that there was likely selection bias (as seen in other trials evaluating mechanical circulatory support devices). It is important to note that there was a 39% crossover rate from the conservative therapy arm into the ECMO arm as a “bailout”. The Kaplan-Meier curve for the primary endpoint shows an initial separation in the curves between the two arms. This gap, however, narrows starting at approximately 5 days from initial presentation, perhaps as the crossovers occurred. Additionally, pulmonary artery catheter hemodynamics were not required. Perhaps one of the most important limitations is the low rate of ventricular unloading, which has been shown in small RCT’s to improve outcomes in VA ECMO patients. 

Given the high rate of crossover, the conclusion that can be taken away from this study is that an early ECMO strategy is probably not better than an initial conservative strategy using VA ECMO as a bailout. The rate of adverse events, however, is similar regardless of the approach. Additional RCTs assessing the utility of VA ECMO in cardiogenic shock should include hemodynamic data and criteria for required LV unloading.