Abstract 17212: Arterial Cannula Size and 30-Day Mortality in Adults With Cardiogenic Shock Supported With Venoarterial Extracorporeal Membrane Oxygenation and Concomitant Left Ventricular Mechanical Unloading: An ELSO Registry Analysis

Introduction: VA-ECMO with left ventricular mechanical unloading (LVMU), including intra-aortic balloon pump (IABP) or percutaneous left ventricular assist device (pVAD), is commonly used to support patients with cardiogenic shock (CS). The impact of arterial return cannula size on short-term mortality in the presence of LVMU is unknown.

Hypothesis: The 30-day inpatient survival in patients with VA-ECMO and LVMU will be the same with a small vs large arterial cannula.

Methods: We queried the ELSO Registry from 2017-2022 for adults with CS receiving VA-ECMO and concomitant LVMU, with either IABP or pVAD placed within -24h to +2h of ECMO initiation. We restricted to patients with body surface area (BSA) 1.8-2.2 m2 and excluded those with central cannulation, multiple ECMO runs, and ECPR. The primary outcome of 30-day inpatient mortality was compared using time-to-event analysis. Secondary outcomes included 24h lactate clearance and rates of acute renal and hepatic injury.

Results: Among 2055 patients supported with VA-ECMO and LVMU (54.3% IABP; 45.7% pVAD), 292 (14.2%) patients received a small cannula and 1763 (86.8%) patients a large arterial cannula. Patients with a small cannula were less likely to be male, had smaller BSA, less often had a distal perfusion cannula, and had a lower incidence of pre-ECMO concomitant organ failures (Figure, A). There was no difference in 30-day inpatient mortality between the cannula size groups (Figure, B), which persisted in multivariable modeling with adjusted hazard ratio 0.94 (95% CI: 0.72-1.23), p=0.65. Patients with a small arterial cannula had no significant difference in 24h lactate clearance (Figure, C) or in the rates of acute renal injury (37.3% vs 39.8%, p = 0.42) or hyperbilirubinemia (5.5% vs 6.5%, p=0.52).

Conclusions: In adults with CS receiving VA-ECMO with LVMU, the use of a smaller arterial return cannula was associated with no difference in 30-day inpatient mortality or metrics of end-organ perfusion.