Abstract Body: Introduction: High-risk pulmonary embolism (PE) can lead to cardiac arrest, for which veno-arterial extracorporeal membrane oxygenation (VA-ECMO) may offer life-saving cardiopulmonary support. We evaluated clinical features associated with mortality in this population.
Methods: We conducted a single-center retrospective study at a high-volume VA-ECMO and resuscitation center. We included adult patients cannulated for VA-ECMO following cardiac arrest due to acute PE between 2015 and 2024. Demographic, imaging, cardiac arrest, and outcome data were collected. Univariate and multivariate logistic regression analysis were performed to identify predictors of mortality. The multivariate model adjusted for key covariates, including location of arrest (out-of-hospital vs. in-hospital), duration of CPR, and ROSC pattern (non-sustained vs. sustained).
Results: Among 1350 patients screened, 40 (3.0%) were cannulated after cardiac arrest secondary to acute PE. Non-survivors were more frequently associated with higher initial lactic acid level (5.8 vs. 9.5, p=0.04), out-of-hospital cardiac arrest (OHCA, 3 vs. 21, p=0.009), longer CPR duration (25.9 vs. 44.2, p=0.020), and non-sustained ROSC (1 vs. 14, p=0.022). There was higher mortality in an ECMO-alone strategy, although this was not significant (1 vs. 11, p=0.076). Eleven patients (27.5%) survived to discharge; of these, eight (72.7%) had a favorable neurologic outcome (CPC 1 or 2). Univariate analysis identified OHCA (OR 7.00, 95% CI 1.48–33.21, p=0.014) and non-sustained ROSC (OR 9.33, 95% CI 1.05–82.64, p=0.045) as predictors of mortality. Reperfusion strategy, time to cannulation, and bleeding complications were not associated with increased risk of mortality. In multivariate analysis, OHCA remained independently associated with mortality (OR 6.67, 95% CI 1.15–38.7, p=0.035).
Conclusions: Cardiac arrest due to PE is associated with high mortality despite VA-ECMO support. In this study, neither reperfusion strategy, time to cannulation, nor bleeding complications were associated with mortality. OHCA remained an independent predictor of mortality after adjusting for clinical covariates.