Abstract 194: Pre-Operative Health Status is Only Weakly Associated With Mortality or Hospitalization Following Continuous-Flow Left Ventricular Assist Device Implantation

Background: Health status predicts death and hospitalization in heart failure and cardiac surgery, but its prognostic value in the setting of left ventricular assist device (LVAD) placement is unknown. We hypothesized that baseline health status could help identify patients at risk for adverse post-operative outcomes and improve patient selection for LVAD therapy. Methods: We examined the association of pre-operative health status, as measured by the Kansas City Cardiomyopathy Questionnaire (KCCQ) overall summary score (OSS), with overall mortality and hospitalization in 965 patients undergoing LVAD placement as part of two clinical trials of the HeartMate II device. Unadjusted statistical analyses were performed using Cox proportional hazard models and Kruskal-Wallis non-parametric tests. Results: Baseline OSS of survivors (median: 25.5, inter quartile range (IQR): 15.6-38.6) was similar to those who died (median: 23.4, IQR 12.9-39.2, p=0.06). Patients in the lowest quartile of KCCQ scores (OSS<14.8) had worse survival as compared with the upper 3 quartiles (unadjusted HR: 1.30. 95% CI 1.03-1.63, p=0.03). Patients in the lowest quartile of baseline OSS also spent more time in the hospital (median 25 days, IQR 19-42) as compared with the upper 3 quartiles (median 23 days, IQR 17-35, p=0.009). Outcomes for the 3 higher OSS quartiles were not significantly different. When comparing KCCQ sub-domain scores of patients who died during follow-up to those who did not, lower total symptom score was significantly associated with worse survival (p=0.009); all other KCCQ sub-domains were not. Conclusion: Only patients with very low baseline health status (KCCQ <15) were at increased risk for mortality and hospitalization following LVAD implantation, with the absolute differences being very small. This weak association between baseline KCCQ and post-operative outcomes suggests that LVAD implantation, unlike chronic heart failure and other cardiac surgeries, represents a unique clinical situation in which the benefits of the intervention may outweigh the anticipated risks of worse health status.