Abstract 9724: The Use of the Model for End-Stage Liver Disease Score Predicts 90-Day Mortality in Patients Undergoing HeartMate II Implantation
Purpose: Careful patient selection for left ventricular assist device (LVAD) therapy is essential to ensure optimal outcomes after device implantation. Although several risk assessment scores are available, there is no consensus on the most accurate score to use in patients undergoing HeartMate II (HMII) implantation. The Lietz-Miller Destination Therapy Risk Score (DTRS) was derived from analysis of patients undergoing HeartMate XVE implantation for destination therapy (Lietz et al. Circulation 2007). The Model for End-Stage Liver Disease (MELD) was shown to predict outcome in patients undergoing primarily HeartMate I implantation (Matthews et al. Circulation 2010). We evaluated the utility of these scores in predicting outcomes following HMII implantation.
Methods: Retrospective analysis was performed on 48 patients undergoing HMII implantation at our institution between 2009 and 2011. The DTRS and MELD scores were calculated from clinical data on day of device implantation and outcomes assessed.
Results: Device implantation was for bridge to transplant (69%) or destination therapy (31%). 14 patients (29%) underwent heart transplant and 17 patients died (35%), with a mean LVAD support time of 301 days and 85 days, respectively. At end of the study period 17 patients were alive (mean LVAD support time 389 days). The majority of deaths (71%) occurred within 90 days of device implantation. Mortality at 90 days was higher in patients with INTERMACS class 1 (class 1 29.4% vs. class 2 to 4 22.6%), high-risk DTRS values (high 33.3% vs. low-medium 22.2%), and high MELD values (MELD 20 50% vs. MELD < 20 16.6%). Only a MELD score 20 was statistically associated with outcome at 90 days with a hazard ratio of 3.5 (p=0.03) using a Cox proportional hazards model.
Conclusions: A MELD score 20 predicts mortality at 90 days in patients undergoing HMII implantation. Further validation of the MELD score in a larger population is needed. Elaboration of the score using cardiac-specific factors may further improve its predictive ability.
- © 2012 by American Heart Association, Inc.