Predictors of mid-term outcomes in patients undergoing implantation of a ventricular assist device directly after extracorporeal life support.

Abstract

Extracorporeal life support (ECLS) can be applied as a bridge to diagnosis and decision-making for further treatment with long-term left ventricular assist devices (LVADs). From January 2012 to January 2018, 714 adult patients were treated with ECLS in our institution. During the same period, 618 patients received an LVAD for long-term support. Of them, 100 patients were further supported with a long-term LVAD. We retrospectively analysed the datasets of these 100 consecutive patients with the goal of developing an algorithm to predict outcomes for a rational use of long-term ventricular assist device therapy in this setting. The mean age of the 100 patients was 54.1 ± 11.6 years, and 72 were men. Twenty-nine patients had a BMI of >30 kg/m2. In 33 patients, a temporary right ventricular assist device was necessary postoperatively. The 30-day, 1-year and 2-year survival after ventricular assist device implantation was 62.0% [95% confidence interval (CI) 53.2-72.3], 43.0% (95% CI 34.3-53.9) and 37.1% (95% CI 28.2-48.7%), respectively. Penalized multivariable logistic regression analysis showed following predictors for 1-year mortality: bilirubin increase per mg/dl [odds ratio (OR) 1.41, 95% CI 1.12-1.77], C-reactive protein increase per mg/dl (OR 1.11, 95% CI 1.05-1.19), ECLS duration >7 days (OR 4.90, 95% CI 1.66-14.41), BMI >30 kg/m2 (OR 1.41, 95% CI 1.05-8.52) and female gender (OR 3.06, 95% CI 1.02-9.23). On the basis of these data, a nomogram to estimate 1-year mortality after LVAD implantation was created. After stabilization of patients experiencing cardiogenic shock using ECLS, LVAD implantation can be performed with elevated mortality in an otherwise futile situation. Liver dysfunction, inflammatory status and obesity increase the risk for mid-term mortality.