Abstract 8: Health Status and Outcomes Following Left Ventricular Assist Device Placement

Abstract

Background: Health status predicts mortality and hospitalization in patients with heart failure, but its relation with such outcomes following left ventricular assist device (LVAD) placement is less clear. The current study describes the association of global and disease-specific health status before and 3 months after LVAD with subsequent mortality and hospitalization. Methods: We analyzed data from 3836 patients registered in the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) database. Global health status was measured using the visual analog scale (VAS) of the EQ-5D. Heart failure-specific health status was measured using the Kansas City Cardiomyopathy Questionnaire (KCCQ). Inverse propensity weighting was used to account for missing health status data. Weighted Kaplan-Meier curves were used to analyze the association of each health status measure with 2-year mortality and rehospitalization. The incremental prognostic value of health status beyond that of variables in a validated risk model (HeartMate II Risk Score (HMIIRS)) was measured using weighted Cox models. Results: Global and heart failure-specific health status were poor pre-operatively, (VAS median 43 (IQR 25-65); (KCCQ median 34.6 (IQR 21.4-50.5)), and improved 3 months after LVAD placement (VAS median 75 (IQR 60-85); KCCQ median 69.3 (IQR 54.2-82.3)). Prior to device implantation, patients missing pre-operative KCCQ data (42%) were significantly more likely to be in cardiogenic shock/worsening on inotropes or require extracorporeal membrane oxygenation, intra-aortic balloon pump, mechanical ventilation or dialysis. Neither pre-operative KCCQ nor VAS scores were associated with mortality or rehospitalization 2 years after LVAD. Only 3-month KCCQ was associated with 2-year mortality (p<0.001), whereas both 3-month KCCQ and VAS were associated with 2-year rehospitalization (p<0.001 and 0.01, respectively). Pre-operative health status and 3-month VAS did not add significantly to the HMIIRS in predicting 2-year mortality (HMIIRS c-stat 0.60) or rehospitalization (HMIIRS c-stat 0.51), however 3-month KCCQ did add incremental prognostic value to the HMIIRS in predicting 2-year mortality (c-stat increased to 0.66). Conclusions: Pre-operative health status among patients undergoing LVAD implantation is not associated with 2-year mortality or rehospitalization, suggesting that very poor health status alone should not be used to exclude patients from treatment. In contrast, health status measured 3 months after LVAD is associated with subsequent outcomes. Furthermore, the KCCQ significantly improves the discrimination of a previously validated risk model, suggesting that systematic health status assessment can improve prognostication and might be helpful in directing care following device implantation.