Abstract 250: Does Driving Time to the Transcatheter Aortic Valve Replacement Site Impacts on Long-Term Mortality Following a Successful Procedure? Insights from the STS/ACC TVT Registry

Abstract

Background: The expansion of transcatheter aortic valve replacement (TAVR) centers in the United States has been controlled through the institution of center and provider volume requirements, but little is known about the impact of potential increased patient travel time resulting from this policy. We sought to assess the association between driving time and 1-year outcomes following TAVR. Methods: Patients ≥65 years registered in the TVT Registry who underwent a successful transfemoral TAVR implantation from June 2015 to June 2017, and were linked to Centers for Medicare and Medicaid Services data, were eligible. Driving time to the TAVR site was determined using Google Maps. The zip codes of patients’ homes were used to determine rural status, according to the Federal Office of Rural Health Policy data files. Multivariable Cox proportional hazards modelling was performed to assess the relationship between driving time and all-cause death (primary endpoint) prespecified secondary endpoints (all-cause hospitalization, heart failure hospitalization, stroke, and any bleeding). The presence of residual confounding was assessed with falsification analysis. Results: Among 31,098 patients, median driving time to the TAVR site was 0.58 hours (interquartile range [IQR]: 0.33-1.18). Median age was 83.0 years old (IQR: 78.0-88.0), and 47% were females. Mean STS score was 7.31±5.56%, and 7,422 patients (23.9%) lived in a rural area. Patients living in rural areas were younger than those who were not (median age: 82.0 and 84.0 years, respectively; p<0.0001). Cumulative incidence of all-cause mortality by quartiles, and unadjusted and adjusted hazard ratios for 1-year outcomes, are presented in Table 1. Driving time was not significantly associated with falsification endpoints, suggesting that there was no residual confounding in the adjusted model. Conclusion: Following successful transfemoral TAVR, longer driving times from the patients’ residence to the TAVR site are associated with statistically significant, yet non-clinically meaningful, lower adjusted hazards of one-year all-cause mortality. Longer driving distance resulting from control of expansion of TAVR centers in the United States to respect institutional and operators volume requirements do not negatively impact long-term clinical outcomes.