Femoro-axillary versus femoro-femoral veno-arterial extracorporeal membrane oxygenation for refractory cardiogenic shock: A monocentric retrospective study

Abstract

Rationale For veno-arterial extracorporeal membrane oxygenation (ECMO), the femoral artery is the preferred cannulation site (femoro-femoral: Vf-Af). This results in retrograde aortic flow, which increases the left ventricular afterload and can lead to severe pulmonary edema and thrombosis of the cardiac chambers. Right axillary artery cannulation (femoral-axillary: Vf-Aa) provides partial anterograde aortic flow, which may prevent some complications. This study aimed to compare the 90-day mortality and complication rates between VF-AA and VF-AF. Methods Consecutive adult patients with cardiogenic shock who received peripheral VA-ECMO between 2013 and 2019 at our institution were retrospectively included. The exclusion criteria were refractory cardiac arrest, multiple VA-ECMO implantations due to vascular access changes, weaning failure, or ICU readmission. A statistical approach using inverse probability of treatment weighting was used to estimate the effect of the cannulation site on the outcomes. The primary endpoint was the 90-day mortality. The secondary endpoints were vascular access complications, stroke, and other complications related to retrograde blood flow. Outcomes were estimated using logistic regression analysis. Results VA-ECMO was performed on 534 patients. Patients with refractory cardiac arrest ( n = 77 (14%)) and those supported by multiple VA-ECMO ( n = 92, (17%)) were excluded. Out of the 333 patients studied ( n = 209 Vf-Aa; n = 124 VF-AF), the main indications for VA-ECMO implantation were post-cardiotomy (33%, n = 109), dilated cardiomyopathy (20%, n = 66), post-cardiac transplantation (15%, n = 50), acute myocardial infarction (14%, n = 46) and other etiologies (18%, n = 62). The median SOFA score was 9 [7-11], and the crude 90-day mortality rate was 53% ( n = 175). After IPTW, the 90-day mortality was similar in the Vf-Aa and VF-AF groups (54% vs 58%, IPTW-OR = 0.84 [0.54-1.29]). Axillary artery cannulation was associated with significantly fewer local infections (OR = 0.21, 95% CI:0.09-0.51), limb ischemia (OR = 0.37, 95% CI:0.17-0.84), bowel ischemia (OR = 0.16, 95% CI:0.05-0.51) and pulmonary edema (OR = 0.52, 95% CI:0.29-0.92) episodes, but with a higher rate of stroke (OR = 2.87, 95% CI:1.08-7.62) than femoral artery cannulation. Conclusion Compared to VF-AF, axillary cannulation was associated with similar 90-day mortality rates. The high rate of stroke associated with axillary artery cannulation requires further investigation.