Work

Bleeding and survival outcomes with Impella versus IABP for left ventricular unloading during VA-ECMO: a systematic review and meta-analysis

Year: 2026
Type: article
Abstract: Veno-arterial extracorporeal membrane oxygenation (VA-ECMO) is used for refractory cardiogenic shock, but retrograde aortic flow increases left ventricular (LV) afterload, impairing myocardial recovery. Adjunctive unloading strategies, including intra-aortic balloon pump (IABP) and Impella, are commonly employed, though comparative outcomes remain unclear. We performed a systematic review and meta-analysis of studies comparing Impella versus IABP for LV unloading during VA-ECMO in adults. Databases were searched through January 15, 2026. The primary outcome was clinically significant bleeding. Secondary outcomes included CNS hemorrhage, access-site bleeding, renal replacement therapy (RRT), and mortality. Random-effects models were used to calculate pooled risk ratios (RR) with 95% confidence intervals (CI). Seven retrospective studies ( n = 592; Impella 250, IABP 342) were included. 6 studies contributed data to the primary bleeding analysis (Impella N = 221; IABP n = 314). Clinically significant bleeding or author reported bleeding occurred in 58.8% of Impella-treated patients and 37.6% of IABP treated patients (RR 1.44, 95% CI 0.93–2.24; I²=90%), though not statistically significant. Impella was associated with increased CNS hemorrhage (RR 3.64, 95% CI 1.32–10.07) and access-site bleeding (RR 1.94, 95% CI 1.11–3.41). Gastrointestinal bleeding did not differ significantly between groups (RR 1.58, 95% CI 0.65–3.84), nor did Renal Replacement Therapy (RR 1.27, 95% CI 0.89–1.80). In-hospital mortality did not differ (RR 0.98, 95% CI 0.71–1.37), while pooled 30-day mortality was lower in the Impella cohorts (RR 0.79, 95% CI 0.64–0.99). Available retrospective comparative data suggest that Impella based LV unloading during VA-ECMO is associated with higher rates of selected bleeding complications, particularly CNS and access-site hemorrhage, while pooled 30-day mortality favored Impella use. All studies were retrospective with nonrandomized treatment allocation and therefore findings should be interpreted as hypothesis generating rather than causal. Prospective studies with standardized bleeding and anticoagulation reporting are needed to define the optimal unloading strategy during VA-ECMO.
Language: English
Field: Medicine
Open Access status: gold
Estimated APC paid: $2,890
OpenAlex ID: W7235070392
FWCI:
Cites:
Cited by:
Related to: 10