How should the left heart be unloaded when ventricular distension develops during venoarterial extracorporeal membrane oxygenation (VA-ECMO)? Moussa and colleagues compared Impella-supported unloading with selective biatrial drainage in a retrospective study of 57 adults at Lille University Hospitals.
Selective biatrial ECMO was associated with fewer serious bleeding events and lower transfusion requirements. The primary outcome—days alive without invasive ventilation—did not differ significantly, and the study did not establish a mortality benefit or equivalence between the techniques.
Two Approaches to Left-Heart Decompression
VA-ECMO can increase left ventricular afterload, contributing to ventricular distension, a persistently closed aortic valve, and pulmonary congestion. The patients in this study received additional unloading for symptomatic distension, rather than routine prophylactic unloading.
- ECPELLA: VA-ECMO combined with an Impella CP or 5.0 microaxial pump positioned across the aortic valve to move blood from the left ventricle into the aorta.
- Selective biatrial ECMO (SBA-ECMO): additional left atrial drainage through cannulated percutaneous atrioseptostomy, with the left atrial and conventional venous drainage lines joined upstream of the ECMO pump.
These approaches unload through different pathways. The study compares their observed clinical outcomes; it does not provide a universal device-selection rule.
Study Design and Population
The single-center study included patients treated between January 2014 and March 2023. Of 141 patients assessed, 57 met the inclusion criteria: 27 received SBA-ECMO and 30 received ECPELLA. All had pulmonary edema on the day of unloading; 54 were invasively ventilated.
Technique selection followed multidisciplinary discussion and depended on feasibility and contraindications. Treatment was not randomized. The primary endpoint was ventilation-free days, defined as days alive without invasive mechanical ventilation from unloading until day 28 or hospital discharge. Analyses adjusted for SAPS II, postcardiotomy status, and the interval between ECMO initiation and unloading.
Key Results
| Outcome | SBA-ECMO (n=27) | ECPELLA (n=30) |
|---|---|---|
| Ventilation-free days, median (IQR) | 10 (0–23) | 5 (0–23) |
| Blood products, median units (IQR) | 6 (3–25) | 19 (9–27) |
| Serious bleeding | 7 (26%) | 19 (63%) |
| Thrombotic complications | 6 (22%) | 11 (37%) |
| 28-day mortality | 13 (48%) | 14 (47%) |
Unadjusted results from Table 2. Blood products include red cells, platelets, and fresh frozen plasma. These are selected observational groups, not randomized treatment arms.
Primary endpoint: no statistically significant difference
The comparison of ventilation-free days was not significant before adjustment (P=0.61) or after adjustment (P=0.22). Although the group medians differed, the results do not demonstrate that SBA-ECMO improves respiratory recovery, nor that the two strategies are equivalent.
Bleeding and transfusion findings favored biatrial drainage
SBA-ECMO was associated with a lower adjusted cause-specific hazard of serious bleeding: hazard ratio 0.31 (95% CI 0.12–0.80). Access or cannulation-site bleeding was reported in 2 patients receiving SBA-ECMO and 10 receiving ECPELLA.
The full-text results report an adjusted blood-product consumption rate ratio of 0.54 (95% CI 0.34–0.85; P=0.009). The unadjusted rate ratio was 0.57 (95% CI 0.36–0.90); the abstract repeats this unadjusted estimate while referring to multivariable analyses. The adjusted estimate presented here follows the detailed Results section.
Transfusion modeling accounted for the duration of ECMO support after unloading. A lower transfusion rate is an association in this cohort and should not be interpreted as proof that choosing one technique will produce the same reduction in every patient.
Thrombosis and survival remain uncertain
No significant differences were detected in thrombotic complications or 28-day mortality. The small sample limits the ability to identify clinically important differences. Similar observed mortality percentages do not establish equal safety or equal effectiveness.
Relevance to Perfusion Practice
The study brings bleeding burden, vascular access, and transfusion exposure into the discussion of unloading strategy. It supports considering biatrial drainage as an option within an experienced multidisciplinary program when technically feasible.
Its findings apply to patients with established left-heart congestion during VA-ECMO. They do not establish a preferred prophylactic unloading strategy, apply automatically to other Impella models, or justify replacing an institution’s existing selection and monitoring protocols.
For perfusion teams, the comparison also emphasizes that unloading effectiveness and device-related complications need to be assessed together. A respiratory endpoint alone cannot confirm adequate ventricular decompression, and this study lacked pulmonary capillary wedge pressure data for outcome analysis.
Limitations and Evidence Assessment
The evidence is limited by retrospective data collection, a single-center setting, nonrandom treatment selection, and only 57 patients. Clinical feasibility and contraindications influenced allocation, creating potential selection bias. Adjustment for three prespecified covariates cannot eliminate residual confounding.
The long enrollment period also spans changes in practice. The authors noted a tendency toward higher anti-factor Xa values in the ECPELLA group, which is relevant when interpreting bleeding differences. Larger prospective comparative studies are needed to determine whether the observed reduction in bleeding translates into better patient-centered outcomes.
Source and Attribution
This original IPA educational summary is based on the open-access article by Moussa and colleagues, published under a Creative Commons Attribution license. Findings and selected numerical results have been summarized and their presentation adapted. The cover is conceptual artwork, not a device-placement or circuit-assembly guide.
Original Article
Moussa MD, Roux J, Jungling M, et al. Impella Versus Selective Biatrial Canulation for Left Ventricular Unloading During Extracorporeal Membrane Oxygenation. Cardiovascular Therapeutics. 2026;2026:3669575. Published February 8, 2026. doi:10.1155/cdr/3669575.
