Risk stratification in patients with advanced heart failure requiring biventricular assist device support as a bridge to cardiac transplantation

Background Prior studies have identified risk factors for survival in patients with end-stage heart failure (HF) requiring left ventricular assist device (LVAD) support. However, patients with biventricular HF may represent a unique cohort. Methods We retrospectively evaluated a consecutive cohort of 113 adult, end-stage HF patients at University of California Los Angeles Medical Center who required BIVAD support between 2000 and 2009. Results Survival to transplant was 66.4%, with 1-year actuarial survival of 62.8%. All patients were Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) Level 1 or 2 and received Thoratec (Pleasanton, CA) paracorporeal BIVAD as a bridge to transplant. Univariate analyses showed dialysis use, ventilator use, extracorporal membrane oxygenation use, low cardiac output, preserved LV ejection fraction (restrictive physiology), normal-to-high sodium, low platelet count, low total cholesterol, low high-density and high-density lipoprotein, low albumin, and elevated aspartate aminotransferase were associated with increased risk of death. We generated a scoring system for survival to transplant. Our final model, with age, sex, dialysis, cholesterol, ventilator, and albumin, gave a C-statistic of 0.870. A simplified system preserved a C-statistic of 0.844. Patients were divided into high-risk or highest-risk groups (median respective survival, 367 and 17 days), with strong discrimination between groups for death. Conclusions We have generated a scoring system that offers high prognostic ability for patients requiring BIVAD support and hope that it may assist in clinical decision making. Further studies are needed to prospectively validate our scoring system.