Hemodynamic Changes With Left Ventricular Venting In Patients Supported With Va-ecmo For Cardiogenic Shock

Abstract

Introduction Despite the ability to provide full pulmonary and biventricular support for severe or refractory cardiogenic shock (CS), a recognized limitation of veno-arterial extracorporeal membrane oxygenation (VA-ECMO) is the increase in left ventricular afterload and its associated adverse effects on the myocardium. Limited data exists regarding the relative efficacy of typical left ventricular venting (LVV) strategies to address this issue. Methods We retrospectively reviewed 31 consecutive patients at our institution initially treated for severe or refractory CS with VA-ECMO who then underwent LVV from 01/01/17 - 12/31/19. Patients with alternate MCS devices implanted prior to VA-ECMO and those with post-cardiotomy CS were excluded. Clinical characteristics, baseline risk stratification scores and serial hemodynamic assessments were collected at 0, 24 and 48 hours. Results Twenty-six (84%) patients underwent LVV with Impella (E-I) and 5 (16%) with atrial septostomy (E-S). Patients in the E-S cohort were younger (mean age 37 vs. 52, p=0.02).There were no significant differences between the E-S and E-I cohorts with regards to baseline CARDShock scores (5 vs. 4; p=0.06), SOFA scores (10 vs. 9; p=0.89), index hemodynamic assessments, and mean time to LVV (32 vs. 31 hours; p=0.60). We also did not detect a significant difference in total time on the mechanical ventilator between the E-S and E-I patients (19 vs. 10 days; p=0.19). At 48 hours following LVV, there was no significant difference in changes in mean lactate, pulmonary artery diastolic pressure (PA-D), right atrial pressure (RA), cardiac power output (CPO) or pulmonary arterial pulsatility index (PAPi) between the E-S and E-I cohorts. Thirty-day survival in the E-S and E-I cohorts were 80% (4/5) and 50% (13/26), respectively (p=0.35). Conclusions Changes in hemo-metabolic profiles at 48 hours were similar between VA-ECMO patients who underwent LVV with E-S vs. E-I. There was a trend toward a greater reduction in PA-D with E-S vs. E-I. Larger prospective studies are needed to identify effective LVV strategies that optimize hemodynamics and improve long-term outcomes in these complex patients.