Early pulmonary congestion as an independent predictor of mortality in cardiogenic shock supported with venoarterial extracorporeal membrane oxygenation

Venoarterial extracorporeal oxygenation (VA-ECMO) offers respiratory and cardiovascular support in acute critically ill patients. Its emergent and invasive nature implies multiple changing conditions such as bleeding and transfusion, congestion, left ventricular failure, diuretic use, and renal impairment, among others. Its support may last days or even weeks. Whether early congestion affects clinical prognosis is unknown.

To assess the prognostic relevance of early pulmonary congestion after cannulation in patients supported by VA-ECMO.

This was a retrospective study, between January 2021 and August 2025, in which we consecutively enrolled 65 patients requiring VA-ECMO for refractory cardiogenic shock. Post cardiotomy VA-ECMO patients were excluded from the analysis. Early pulmonary congestion (CONG-24) was defined by Weinberg radiological score assessment on chest radiography 24 hours after cannulation by two independent investigators.

The mean age was 55.5 ± 2.1 years, 73.8 % (n=48) were male and 61.5% (n=40) had cardiogenic shock complicating acute myocardial infarction, being 56.9% (n=37) placed under cardiopulmonary resuscitation (ECPR). Overall mortality was 52.3% (n=34). The median CONG-24 was 4.0 (2.0-6.0). Higher CONG-24 scores were associated with higher needs of noradrenaline (mcg/kg/min) during the first 24 hours (r = 0.435, p = 0.001), ECMO blood flow (liters/minute) (r = 0.34, p = 0.007) and higher post membrane pressure P3 (r = 0.46, p = 0.012). Moreover, higher CONG-24 scores were associated with acute pulmonary edema at any moment of the support (7.8 ± 4.0 vs 3.1 ± 2.2; p < 0.001), severe bleeding (BARC ≥ 3) (5.7 ± 4.6 vs 3.8 ± 2.5; p = 0.041), overall mortality (5.6 ± 4.1 vs 3.3 ± 2.4; p = 0.011) and cardiovascular mortality (7.1 ± 4.9 vs 3.6 ± 2.5; p = 0.001). Other demographic and clinical characteristics are listed in Table 1. In the multivariate analysis, after adjustment for age, sex, etiology of shock, severe bleeding (BARC ≥ 3) and ECPR, CONG-24 remained independently associated with mortality (OR 1.26 [95 % CI 1.02–1.55]; p = 0.036), as well as older age (OR 1.06 [95 % CI 1.00–1.13]; p = 0.037). (Figure 1)

In our cohort, pulmonary congestion assessed by Weinberg score at 24 hours after VA-ECMO was independently associated with all-cause mortality. Early recognition and prompt achievement of euvolemia may be key to improving survival outcomes.Table 1