VA-ECMO is the most widely used form of mechanical circulatory support, providing rapid systemic flow of 4-6 L/min in cases of severe CS. However, its typical configuration with retrograde flow through femoral access frequently results in left ventricular (LV) overload, potentially leading to distention, blood stasis, thrombosis, and pulmonary congestion. Various methods of LV unloading to mitigate these complications have been widely discussed this year in a prior commentary on this blog.
The present study evaluated transseptal left atrial (LA) cannulation as a modality for LV mechanical unloading (LV-MU), contrasting it with more common options such as intra-aortic balloon pump (IABP) or peripherally inserted ventricular assist devices like the Impella CP® (commonly known as “ECMELLA” or “ECPELLA”). The EARLY-UNLOAD trial represents a prospective, single-center clinical trial conducted in CS patients requiring peripheral VA-ECMO. Its primary objective was to assess the efficacy of early routine LV-MU through transseptal LA cannulation within 12 hours of VA-ECMO initiation compared with conventional management.
A total of 116 patients, with a mean age of 67.6 years and 29.3% women, were included and placed on VA-ECMO for CS treatment. Of these, 58 (50%) were randomized to receive early routine LV-MU, while the other 58 (50%) followed conventional management. Primary outcomes showed no statistically significant difference in 30-day mortality (primary endpoint), with a rate of 46.6% in the early LV-MU group and 44.8% in the conventional management group (hazard ratio [HR] 1.02). However, 50% of the conventional group required rescue LV-MU due to refractory LV overload (distention with blood stasis, minimal aortic valve opening, and refractory pulmonary congestion). The average time to transseptal LA cannulation in the early LV-MU group was 1.1 hours, versus 21.8 hours in the conventional group who switched to rescue LV-MU.
Regarding the primary secondary endpoint, which included all-cause mortality or need for rescue transseptal LA cannulation, a statistically significant benefit was observed in the early LV-MU group (HR 0.44; 95% CI: 0.27-0.72; p = 0.001), largely driven by the need for rescue LV-MU. However, most secondary endpoints, such as in-hospital mortality, lactate levels, ECMO weaning, mechanical ventilation duration, renal replacement therapy, strokes, bleeding, or limb ischemia, showed no statistically significant differences between groups, with the exception of a shorter time to pulmonary congestion resolution in the experimental group.
In summary, the study’s findings suggest that routine early LV-MU through transseptal LA cannulation does not reduce 30-day mortality compared to conventional management in VA-ECMO-supported CS patients.
COMMENTARY:
This study is a milestone as the first randomized trial addressing early LV-MU therapy (using transseptal LA cannulation) compared to conventional management in CS patients requiring VA-ECMO. While awaiting further trials like the REVERSE study (using Impella CP® for LV-MU), these results suggest against the routine early application of an LV-MU method.
In CS secondary to complicated acute myocardial infarction (AMI), at least as indicated by the single weighty trial to date (ECLS-SHOCK, discussed in a recent commentary here), early VA-ECMO initiation versus standard medical therapy does not appear to benefit 30-day survival. However, our clinical experience and the results from multiple retrospective studies indicate that the rising use of these devices is crucial in certain CS cases to save lives. Notably, most VA-ECMO-supported CS cases in this study also involved AMI patients (66.4%), followed by decompensated heart failure (13.8%), fulminant myocarditis (8.6%), and other causes comprising the remaining 11.2%. Therefore, the most common VA-ECMO use in CS cases here aligned with the complicated AMI context observed in routine clinical practice and the ECLS-SHOCK study.
It’s worth noting that nearly all study patients were in severely critical states of CS, specifically in SCAI stages D or E, with 77.6% in stage D and 22.4% in stage E. They presented mean arterial lactate levels of 7.1 mmol/L, an average LV ejection fraction of 16%, and 45% had experienced cardiopulmonary arrest (CPA), though only those with witnessed CPA and responsive to commands were included. In Kim MC et al., the indication and need for ECMO are clear, enhancing the study’s value and applicability.
At first glance, transseptal cannulation might appear less invasive than other LV-MU methods like the Impella CP®, as it utilizes venous access rather than arterial. However, this approach is not without risks. Challenges include potential cardiac perforation and tamponade, as seen in two cases in the experimental group. Furthermore, this method requires an additional cannula in the VA-ECMO drainage circuit and iatrogenic interatrial communication with its hemodynamic implications. The feasibility of percutaneous atrial septostomy as a valid technique hinges on operator expertise, institutional infrastructure, and interdisciplinary collaboration in patient management.
No significant differences were found regarding 30-day mortality or other secondary endpoints, including weaning, ECMO duration, mechanical ventilation duration, or renal replacement therapy. In other words, results were comparable across both groups, with the only significant difference being a shorter time to resolve pulmonary congestion in the early decompression group (3 days vs. 5 days). These results alone hardly justify routine early LV-MU.
Nevertheless, the observed 50% crossover rate (with conventional management patients eventually requiring rescue LV-MU) underscores that, on average, half of the patients require some form of LV-MU within 24 hours. Therefore, at least half of VA-ECMO patients initially deemed unnecessary for LV-MU meet criteria for its initiation within 24 hours.
These findings suggest avoiding routine LV-MU in all patients receiving peripheral VA-ECMO from the outset. Instead, it seems prudent to conclude that individualized evaluation of LV unloading is warranted, especially during the initial 24-48 hours, paying particular attention to signs of refractory congestion and LV overload.
REFERENCE:
Kim MC, Lim Y, Lee SH, Shin Y, Ahn JH, et al. Early Left Ventricular Unloading or Conventional Approach After Venoarterial Extracorporeal Membrane Oxygenation: The EARLY-UNLOAD Randomized Clinical Trial. Circulation. 2023 Nov 14;148(20):1570-1581. doi: 10.1161/CIRCULATIONAHA.123.066179.
