Long-Term Outcomes of Repeat TAVR vs. TAVR Explant: A Nationwide Medicare Analysis of Bioprosthetic Valve Degeneration in the TAVR Era

T. Faggion Vinholo¹, J. Awtry², M. Cho³, P. Allen⁴, R. Semco⁵, P. Newell⁶, S. McGurk⁷, S. Hirji⁸, M. Harloff⁶, A. Osho⁹, A. Sabe^6
1Brigham & Women’s, Boston, Massachusetts ²Brigham and Women's Hospital, Jamaica Plain, Massachusetts ³Center for Surgery and Public Health, Boston, Massachusetts ⁴Brigham and Women's Hospital, Brookline, Massachusetts ⁵Harvard Medical School, Boston, Massachusetts ⁶Brigham and Women's Hospital, Boston, Massachusetts ⁷BWH, Boston, Massachusetts ⁸Brigham and Women's Hospital, Allston, Massachusetts ⁹Massachusetts General Hospital - Division of Cardiac Surgery, Boston, Massachusetts

Purpose: This study aims to assess long-term comparative durability and efficacy of of re-interventions, namely repeat transcatheter aortic valve replacement (TAVR) or TAVR explantation, for bioprosthetic valve dysfunction following initial index TAVR. We seek to provide insights using longitudinal data from the Centers for Medicare and Medicaid Services (CMS) database.

Methods: Patients >65 years within the CMS database who underwent aortic valve re-intervention after index TAVR between January/2012 and December/2020 were included. Patients were categorized by reintervention type, Repeat TAVR vs. TAVR Explant, based on International Classification of Diseases Procedure Codes. Patients who underwent reintervention due to endocarditis or reintervention during the index TAVR hospitalization were excluded. To ensure comparability between groups, patients undergoing repeat TAVR who were older than the oldest patient in the TAVR Explant group were excluded. Primary outcomes included short- and long-term mortality, and major adverse cardiovascular events(MACE; composite of death, pacemaker placement, major bleeding, acute kidney injury, cardiac arrest). Survival was calculated in days from the reintervention date until event or end of study observation(July/2022). The difference between repeat TAVR and TAVR explant were assessed with Kaplan-Meier analysis over a 6-year period. Adjusted outcomes analyses included 1:1 propensity score matching using nearest neighbor with caliper method.

Results: A total of 318,620 patients underwent index TAVR with 1,181(0.37%) patients requiring re-intervention. Of those, 846(71.6%) underwent repeat TAVR and 335(28.4%) underwent TAVR Explant. Repeat TAVR patients were significantly older(80 vs 75 years), more likely to be female(47% vs 40%), had higher rates of prior percutaneous coronary intervention(7.1% vs. 3%) and coronary artery bypass grafting(22% vs. 14%), and had a higher comorbidity burden with higher Charlson Comorbidity Index scores compared to TAVR explant patients (all p< 0.05). Repeat TAVR patients showed lower unadjusted 30- and 90-day mortality (5.3% vs 11% and 9.6% vs 17%, p< 0.001), lower overall MACE (71% vs 85%, p< 0.001), but higher mortality at 1-year, which became significant at 3-years (43% vs 30%, p< 0.001) and persisted at 5-years after reintervention (54% vs 40%, p< 0.001); these findings remained after propensity score matching. Repeat TAVR patients had significantly worse unadjusted and propensity score matched adjusted long-term survival compared to TAVR explant patients (p < 0.01), which remained on landmark analysis at 1-year post-reintervention. TAVR explant patients demonstrated a significant survival advantage at 13 months post-reintervention, which remained on sensitivity analysis even after stratifying by Charlson comorbidity scores. Notably, female patients were less likely to undergo TAVR explant despite its survival benefits.

Conclusion: While repeat TAVR patients initially experienced lower mortality, this advantage diminishes over time, with long-term outcomes favoring TAVR explant. For patients with bioprosthetic valve dysfunction post-TAVR, TAVR explant should be strongly considered, particularly for those with a life expectancy exceeding one year, as it may provide superior long-term survival.

Identify the source of the funding for this research project: N/A