Transcatheter Aortic Valve Replacement in Hemodialysis-Dependent Patients
Katherine McDermott, Fenton McCarthy, Dale Kobrin, Wilson Szeto, Joseph Bavaria, Nimesh Desai, Hospital of the University of Pennsylvania, Philadelphia, PA
OBJECTIVE: Transcatheter aortic valve replacement (TAVR) trials generally excluded patients on dialysis, thus outcomes in this population are unknown. We aim to examine TAVR outcomes in dialysis-dependent patients.
METHODS: All Medicare fee-for-service patients undergoing TAVR (n=5003) or surgical aortic valve replacement (SAVR) (n=32915) between January 1, 2011 and November 30, 2012 were identified using procedural codes collected by the Centers for Medicare & Medicaid Services. Dialysis status and comorbidities were identified using diagnosis codes present on admission. TAVR patients on dialysis (4.5%, n=224) were compared to TAVR patients not on dialysis and a propensity matched group of SAVR dialysis patients (n=194 pairs).
RESULTS: TAVR dialysis patients were younger than non-dialysis TAVR patients (79.2 vs 84.1 years; p <0.01) with a higher prevalence of comorbidities (Table 1) and increased composite incidence of stroke, acute myocardial infarction, or mortality at 30 days (15% vs 9%, p = 0.01) and 90 days (23% vs 15%, p = 0.01). Kaplan-Meier analysis showed worse survival (Figure 1a, p < .01), which was independently associated with dialysis by multivariable regression (HR 1.74, 95% CI 1.33 – 2.27, p <0.01). Propensity-matched TAVR and SAVR patients had no significant differences in age (77.6 vs 77.9 years; p = 0.77), demographics, or risk factors. TAVR patients had shorter hospital stays (8.7 vs 15.9 days; p <0.01) and comparable survival (Figure 1b, p = 0.54).
CONCLUSIONS: TAVR in dialysis patients is associated with decreased survival compared to non-dialysis patients; however, survival is comparable in demographically and clinically similar patients undergoing SAVR.