ECLS-SHOCK: venoarterial membrane oxygenation in cardiogenic shock

Reported from ESC Congress 2023

Panos Xaplanteris & Leonidas Koliastasis provide their take on the final results of the ECLS-SHOCK (Extracorporeal Life Support in Infarct-Related Cardiogenic Shock) which were presented by Holger Thiele during the ESC 2023 congress in Amsterdam.

Why this study – the rationale/objective?

Cardiogenic shock (CS) complicating acute coronary syndromes (ACS) reaches mortality rates of 40% - 50% in 30 days. Mechanical circulatory support (MCS) is used to achieve adequate hemodynamic stabilization, but so far randomized controlled trials have failed to prove any benefit regarding mortality. The IABP-SHOCK II trial has downgraded the use of intra-aortic balloon pump to class III in the latest ESC guidelines and randomized data regarding other MCS systems (Impella and VA-ECMO) are scarce.

The ECLS-SHOCK trial was presented in a Hot Line Session at ESC Congress 2023 with simultaneous publication in the NEJM and sought to clarify the role of VA-ECMO in ACS complicated by CS.

How was it executed?

417 patients presenting with CS due to ASC were randomized in a 1:1 manner to either VA-ECMO in addition to usual medical therapy or usual medical therapy alone. Inclusion criteria ensured that patients with true CS were included with arterial lactate concentration of at least 3mmol/L. Patients who had undergone cardiopulmonary resuscitation for more than 45 minutes and those who had an obvious mechanical cause of CS or unfavourable vessel anatomy for VA-ECMO placement were excluded.

The primary endpoint was all-cause mortality at 30 days. Key secondary outcomes were time until hemodynamic stabilization, duration of intensive care unit hospitalization, acute kidney failure demanding renal-replacement therapy, recurrent myocardial infarction, rehospitalization due to heart failure, neurologic performance, disease severity measured by lactate levels, stroke-systemic embolization and vascular complications.

The median age of participants was 63 years; 81.3% were men. Two-thirds presented with ST-segment elevation ACS and 77.7% underwent cardiopulmonary resuscitation before randomization. Percutaneous coronary angioplasty was performed in 96.6% of the patients.

What is the main result?

100 patients from the ECLS group and 102 from the control group died at 30 days demonstrating no statistically significant difference between the groups (p=0.81).
No significant difference was observed regarding renal replacement therapy, repeat revascularization, myocardial reinfarction, rehospitalization due to heart failure, neurologic performance and stroke.
Moderate/severe bleeding rate was significantly higher in the VA-ECMO group with RR 2.44 (95%CI: 1.50-3.95).
Peripheral ischemic vascular complications warranting surgical or interventional therapy were more common in the VA-ECMO group with RR 2.86 (95% CI: 1.31-6.25).
No difference was found in subgroup analyses in terms of sex, age, diabetes, type of ACS, lactate levels and cardiopulmonary resuscitation.

Critical reading and the relevance for clinical practice

In the ECLS-SHOCK trial, VA-ECMO on top of standard medical care failed to provide a survival benefit at 30 days. This finding is in line with the other already published RCTs (ECMO-CS, EURO SHOCK and a smaller RCT by Brunner et al.). VA-ECMO also did not improve any of the secondary outcomes and was related to increased bleeding risk and vascular complications.

The negative findings could be attributed to several causes:

Bleeding and vascular complications may counterbalance any potential benefit of VA-ECMO.
Prolonged mechanical ventilation was needed in the VA-ECMO group.
VA-ECMO is associated with increased LV afterload; this may affect outcomes.
26 patients of the control group crossed over due to refractory circulatory collapse and received VA-ECMO; on top of that, 6 of the aforementioned patients received additional active LV unloading support. Moreover, 28 control patients received other MCS. Therefore, 54 out of the 208 patients of the control group (26% of the control group) received at least one MCS system added to the standard medical treatment; this may have diluted any potential benefit for VA-ECMO.

In conclusion, the ECLS-SHOCK trial showed no improvement in mortality and secondary outcomes in 30 days with the use of VA-ECMO for ACS patients presenting with CS, with more vascular and bleeding complications. Of note, the findings were confirmed by an individual patient data meta-analysis of the 4 published RCTs.

In the steps of the IABP-SHOCK II study that demonstrated the futility of intraaortic balloon counterpulsation (IABP), the present study adds VA-ECMO to the list of MCS systems that are of no benefit for patients in CS secondary to an ACS. The question of whether this is a class (lack of) effect of all MCS systems or the use of a microaxial transvalvular flow pump (such as Impella) will result in favourable outcomes remains to be answered by the currently ongoing DanGer Shock study.