Despite significant advances in cardiology, the high early mortality in cardiogenic shock (CS) remains a significant challenge in clinical practice. Although various mechanical circulatory support (MCS) devices that enable hemodynamic stabilization in CS have emerged over the last decade, their impact on survival remains inconclusive. Among available MCS options, veno-arterial extracorporeal membrane oxygenation (VA-ECMO) is the most widely utilized due to its capacity to rapidly and effectively provide circulatory and respiratory support in cases of right, left, or biventricular failure.

Both the ESC and AHA current guidelines consider MCS for CS patients, particularly when hypoperfusion and hemodynamic deterioration persist despite inotropic and vasopressor treatment. However, they lack specificity on timing, target populations, or particularly beneficial cases, as most recommendations rely on retrospective studies, registries, and expert opinions. To date, no randomized clinical trial (RCT) had specifically examined VA-ECMO use in CS patients.

The ECMO-CS study under analysis today is a multicenter RCT including 117 patients with severe CS or rapidly deteriorating hemodynamics, randomly allocated to two groups: 58 patients received immediate VA-ECMO, while 59 initially received conservative therapy. The primary outcome was a composite of 30-day mortality, need for any MCS device, and cardiac arrest. No significant differences were found between the groups, with a 63.8% incidence in the ECMO group and 71.2% in the conservative group. Notably, 39% of the conservative group patients ultimately required VA-ECMO. There were no significant differences in complications, including 30-day cardiac arrest incidence (10.3% vs. 13.6%), all-cause mortality (50% vs. 47.5%), and serious adverse events (60.3% vs. 61%).

The authors concluded that immediate VA-ECMO in patients with severe or rapidly deteriorating CS did not improve clinical outcomes compared to an initial conservative approach with later VA-ECMO implantation in cases of hemodynamic deterioration.

COMMENTARY:

It is striking that, despite CS being the leading cause of death in patients with acute myocardial infarction (AMI), fewer than 20 RCTs, including fewer than 3000 patients in total, have been conducted since the first SHOCK trial in 1999. Almost all of these studies face repeated limitations. First, patient recruitment has proven challenging; second, there is vast design variability across studies. Another constant limitation is the lack of critical variables such as cause of death, right-heart catheterization data, details on inotropic and vasoactive drugs, or basic cardiac arrest information. In 2019, the Society for Cardiovascular Angiography and Intervention (SCAI) proposed a new CS classification to unify criteria, establishing five stages (A to E) to stratify mortality risk by severity.

This multicenter study presented by Ostadal et al. is noteworthy as the largest VA-ECMO RCT in CS patients to date. At first glance, its findings suggest limited benefit from early VA-ECMO implantation, but there are nuances to consider. Conducting this study required 8 years and collaboration among 4 centers to recruit just 117 patients, reinforcing the recruitment difficulty for RCTs in CS and underscoring the need for new research strategies. Registries could potentially offer valuable data. The baseline characteristics of both groups were comparable, with STEMI (50.4%) and heart failure exacerbation (23.1%) as the most common underlying causes, aligning with registered CS series data. Although this study did not use the new SCAI shock classification, the patients included roughly correspond to SCAI stages D (hypoperfusion with deterioration) and E (extreme, hypoperfusion with deterioration and refractory shock) of this classification. These are the most severe but least prevalent stages, with expected mortalities of 40% and 67%, respectively—figures in line with the final mortality rates in both study groups.

Notably, and perhaps most critically, in the conservative group, crossover to other MCS devices (including VA-ECMO) was permitted for patients with worsening hemodynamics, defined as a lactate increase ≥3 mmol/L in 24 hours. This led to 39% of this group eventually receiving VA-ECMO. Given the high crossover rate, results should be interpreted with caution. The study does not imply that MCS is ineffective in CS; rather, early VA-ECMO implantation does not impact outcomes if later MCS use is allowed in cases of deterioration. Another takeaway is that early, unnecessary VA-ECMO implantation may lead to avoidable complications, while delayed or excessively late implantation fails to improve outcomes and patient prognosis. This underscores the need for clear, standardized criteria to determine the optimal timing for MCS implantation in CS.

The study has limitations to consider when interpreting results. First, it lacks precise data on the definition and timing of resuscitated cardiac arrest, complicating interpretation of ECMO group results. Furthermore, no information is provided on the incidence and duration of cardiac arrest prior to randomization—a recognized independent risk factor in CS mortality. Importantly, out-of-hospital comatose patients were excluded from the study, limiting the generalizability of results to this high-risk population. Additionally, the median patient age exceeded 65 years, surpassing the inflection point indicating a poorer CS prognosis by 5 years. Other considerations, such as deciding on left ventricular drain cannulation or limb perfusion to prevent ischemia in ECMO patients, were left to the clinicians’ discretion and may reflect non-uniform criteria. Lastly, the study did not establish formal criteria for transitioning to more advanced destination therapies, weaning from other devices, or limiting life-support treatments.

Ultimately, the ECMO-CS study has highlighted certain limitations that future CS treatment research with VA-ECMO should consider. Nevertheless, there are still unanswered questions and advancements to be made in this complex and challenging field. Fortunately, two ongoing RCTs promise to shed more light on VA-ECMO use in AMI-CS patients. These are the ECLS-SHOCK trial, randomizing 420 AMI-CS patients to ECMO versus standard treatment, and the ANCHOR trial, in which 400 AMI-CS patients will be randomized to ECMO with intra-aortic balloon pump versus conventional treatment. We await their results, hoping to inch closer to the long-awaited solution for improving CS patient prognosis.

REFERENCE:

Ostadal P, Rokyta R, Karasek J, Kruger A, Vondrakova D, Janotka M, et al.; ECMO-CS Investigators. Extracorporeal membrane oxygenation in the therapy of cardiogenic shock: results of the ECMO-CS randomized clinical trial. Circulation. 2023 Feb 7;147(6):454-464. doi: 10.1161/CIRCULATIONAHA.122.062949.