Abstract Background Bleeding remains one of the main complications during veno-arterial extracorporeal membrane oxygenation (VA-ECMO). In the setting of extracorporeal cardiopulmonary resuscitation (ECPR), urgent cannulation, coagulopathy secondary to systemic hypoperfusion and mechanical trauma from chest compressions may further increase the risk of haemorrhage. Despite these pathophysiological differences, anticoagulation management in ECPR remains largely similar to that of conventional VA-ECMO. Objective We aimed to compare the incidence and determinants of bleeding between ECPR and non-ECPR patients in our centre. Methods This is a retrospective, single-centre, observational study of 65 patients supported with VA-ECMO percutaneously cannulated between January 2021 and August 2025. Baseline characteristics, shock aetiology, hemodynamic status, biochemical parameters, ECMO-related parameters and clinical outcomes were collected. Patients were categorized according to ECMO cannulation during ongoing cardiopulmonary resuscitation (ECPR) or not (non-ECPR). Anticoagulation strategy did not differ between groups and was managed according to the same institutional protocol. Bleeding events were classified according to the Bleeding Academic Research Consortium (BARC) scale, with major bleeding defined as BARC ≥ 3a. Comparisons between groups were performed using the appropriate statistical test. Variables with clinical relevance or p0.10 in univariate analyses were entered into a multivariable logistic regression to identify independent predictors of major bleeding. Results A total of 65 patients were included, of whom 37 (56.9%) underwent ECPR. ECPR patients were more frequently male (83.8% vs 60.7%, p=0.036), had a similar age to non-ECPR patients and had a higher prevalence acute myocardial infarction cardiogenic shock (AMICS) (73% vs. 46.4%, p=0.007). Baseline hemoglobin and creatinine were comparable, though peak lactate was significantly higher in ECPR patients (110 vs 55, p0.001). Major bleeding occurred in 23 patients (35.4%), being more frequent in ECPR (48.6% vs 17.9%, p=0.006). Digestive bleeding also showed higher incidence in ECPR (30% vs. 3.5%, p=0.019), while arterial thrombosis and pulmonary embolism rates were similar between groups. Mortality was significantly higher in the ECPR group (67.6% vs 32.1%, p=0.005). In multivariable logistic regression, ECPR remained independently associated with major bleeding (OR 13.1; 95% CI 2.1–80.3; p=0.006) after adjustment for age, sex, shock etiology, baseline hemoglobin and diabetes. Conclusions In our cohort, ECPR showed a higher incidence of major bleeding, independent of baseline or etiological factors. This association does not appear to be explained by ischemic etiology or the potential use of antiplatelet therapy. These findings suggest that anticoagulation protocols in ECPR patients should be specifically adapted for this high-risk population.
Domenech et al. (2026) studied this question.