Abstract Background Venoarterial extracorporeal membrane oxygenation (VA-ECMO) is used as temporary circulatory support for postcardiotomy cardiogenic shock in adult cardiac surgery. Whether perioperative implantation offers advantages over postoperative rescue initiation remains uncertain. Methods This retrospective single-center study included adult patients receiving VA-ECMO after cardiac surgery between 2013 and 2025. Patients were categorized as perioperative ECMO (intraoperative implantation or failure to separate from cardiopulmonary bypass) or postoperative ECMO (implantation in the intensive care unit after hemodynamic or respiratory deterioration). Univariate, multivariable logistic regression, and propensity-adjusted analyses were performed for hospital mortality and successful ECMO weaning. Successful weaning was defined as survival for at least 24 hours after ECMO decannulation. Results Ninety-seven patients were included: 55 perioperative and 42 postoperative ECMO implantations. Surgical urgency, cardiopulmonary bypass duration, pre-ECMO vasoactive–inotropic score (VIS), ECMO configuration, and arterial cannula size differed significantly between groups. Hospital mortality and successful weaning were similar in unadjusted analyses. Higher pre-ECMO VIS was independently associated with hospital mortality (aOR 1.51 per 10-point increase; p=0.007). Postoperative timing was not independently associated with hospital mortality after adjustment. Higher VIS, longer cardiopulmonary bypass duration, and urgent or emergent surgery were associated with lower odds of successful weaning. Propensity-adjusted analyses showed similar findings. Conclusion Outcomes in postcardiotomy VA-ECMO patients appeared more strongly related to shock severity and operative complexity than implantation timing alone. Residual confounding and methodological limitations preclude causal conclusions.
Domonkos et al. (Wed,) studied this question.