Abstract 373: Unexpected Differences In Traditional Etiologies of Readmission and Mortality Post Left Ventricular Assist Device After Adjusting For Socioeconomic Status

Abstract

Background: Readmission rates after placement of left ventricular assist device (LVAD) remain high and are generally attributed to arrhythmia, bleeding, heart failure (HF), infection, and thrombosis. Readmission and mortality risks have not been clearly assessed after adjusting for socioeconomic status (SES). We hypothesized that etiologies unrelated to medication adherence and nutrition would contribute to worse outcomes. Methods: We retrospectively evaluated 101 patients who underwent HeartMate II LVAD between January 2006- December 2010 at our facility. All patients were followed until December 2011 for readmission and death. Multivariable analysis with adjustment for median household income, insurance status, age, gender, and race was performed. Results: The main causes of initial readmission and death respectively were arrhythmia (n=5, 5.4%; n=1, 3%), bleeding (n=13, 14.0%; n=9, 30%), device issue (n=20, 21.5%; n=6, 20%), HF (n=8, 8.6%; n=2, 7%), and infection (n=6, 6.5%; n=0, 0%). After adjusting for SES, bleeding alone remained a significant independent predictor for readmission [Hazard Ratio (HR) 2.57, p=0.003] and mortality (HR 4.73, p<0.0001). Standard HF therapy upon discharge or place of discharge (home versus facility) after LVAD provided no significant reduction in risk of readmission or mortality. Conclusions: Bleeding was associated with significant increased risk of readmission and mortality after LVAD even after adjusting for SES. This suggests that income levels and perhaps education do not provide protection from readmission and mortality after LVAD.