J. Maxwell Chamberlain Paper for Adult Cardiac Surgery: Valve-in-Valve Transcatheter Aortic Valve Replacement vs. Redo Surgical Aortic Valve Replacement for Degenerated Bioprosthetic Valves in Medicare Patients

J. Awtry¹, T. Faggion Vinholo², M. Cho³, P. Allen⁴, R. Semco⁵, S. McGurk⁶, P. Newell⁷, M. Cunningham⁷, A. Sabe⁷, K. De La Cruz^8
1Brigham and Women's Hospital, Jamaica Plain, Massachusetts ²Brigham & Women’s, Boston, Massachusetts ³Center for Surgery and Public Health, Boston, Massachusetts ⁴Brigham and Women's Hospital, Brookline, Massachusetts ⁵Harvard Medical School, Boston, Massachusetts ⁶BWH, Boston, Massachusetts ⁷Brigham and Women's Hospital, Boston, Massachusetts ⁸Brigham and Women's Hospital, Newton, Massachusetts

Purpose: Valve-in-valve transcatheter aortic valve replacement (ViV-TAVR) is increasingly used to treat degenerated surgical bioprosthetic valves due to its perioperative safety compared to repeat surgical aortic valve replacement (redo-SAVR), but long-term outcomes remain largely unexplored. Therefore, this study examines long-term outcomes of ViV-TAVR and redo-SAVR in the Medicare population.

Methods: This retrospective study included Medicare beneficiaries who underwent aortic valve re-intervention after index SAVR between 2012 and 2021, either via ViV-TAVR or redo-SAVR. Patients requiring re-intervention due to endocarditis or during the same hospitalization as the index SAVR were excluded. Propensity score matching adjusted for differences in patient comorbidities including age, congestive heart failure, prior myocardial infarction, chronic kidney disease, prior CABG, COPD, emergency case status, Charlson comorbidity index, diabetes, and history of stroke/transient ischemic attack. Patients were also stratified into low- and intermediate or high-risk groups based on Charlson score ( < 8=low, ≥8=intermediate or high). The primary outcome was 5-year mortality. Secondary outcomes included 30-day, 1-year, and 3-year mortality, as well as major adverse cardiovascular events (MACE) including mortality, stroke, or acute MI within 30-days of surgery. Survival differences were assessed via Kaplan-Meier analysis before and after propensity score matching, including with landmark analysis at one year.

Results: Overall, 4,465 patients were included: 1,761 (39.4%) who underwent repeat SAVR and 2,704 (60.6%) who underwent ViV-TAVR. Annual ViV-TAVR volume increased substantially (59 in 2012 vs 686 in 2020) compared to a modest rise in redo-SAVR volume (126 in 2012 vs 225 in 2020). Patients who underwent redo-SAVR had more frequent emergency case status (17% vs 12%, p< 0.001) but were younger (median[IQR] 72[[68-77] vs 79[ 73-84]) with fewer comorbidities including CHF (40% vs. 71%), prior CABG (18% vs. 33%), and CKD (29% vs. 40%) (all p< 0.005) compared to ViV-TAVR patients. Before adjustment, redo-SAVR had higher perioperative MACE (11% vs 5.5%, p< 0.001) and 30-day mortality (8.9% vs 5.0%, p< 0.001), but similar 1-year mortality (17% vs. 15%, p=0.089) and reduced 3-year (25%% vs 32%, p< 0.001) and 5-year mortality (32% vs 40%, p< 0.001; Figure 1A) compared to ViV-TAVR. After matching, redo-SAVR had higher MACE (11.4% vs 5.4%, p< 0.001) and 30-day mortality (9.8% vs 4.7%, p< 0.001), but continued to have improved long-term survival (Figure 1B, p< 0.001). Landmark analysis of the propensity-matched cohorts highlighted a survival benefit for redo-SAVR after one year (Figure 1C). The long-term survival benefit was also present in propensity-matched low- (Figure 1D, p< 0.001) and intermediate or high-risk patients (Figure 1E, p=0.043).

Conclusion: Although use of ViV-TAVR is expanding relative to redo-SAVR and has more favorable short-term outcomes, redo-SAVR patients have better long-term survival. Even in Medicare patients surgical reintervention for degenerated bioprosthetic valves should be considered for those with reasonable life expectancy. Clinical trial data is needed to inform future treatment choices.

Identify the source of the funding for this research project: None