Outcomes After Transcatheter Aortic Valve Replacement Among Medicare Beneficiaries: The Impact of Frailty and Social Vulnerability

Abstract

Background

Transcatheter aortic valve replacement (TAVR) is an accepted alternative to surgery in many patients with severe aortic stenosis. Clinical trials have evaluated early and late outcomes in selected TAVR patients, but predictors of late mortality have been less well studied in a broadly inclusive, national patient cohort undergoing TAVR. We sought to characterize 5-year outcomes after TAVR in Medicare beneficiaries and to evaluate the incremental predictive value of demographics, comorbidities, procedural factors, frailty, and social vulnerability in determining late mortality risk.

Methods

We studied the fee-for-service Centers for Medicare & Medicaid Services MedPAR database that includes patients aged ≥65 years undergoing TAVR between 2017 and 2022. The primary endpoint was 5-year mortality. Sequential multivariable Cox models were constructed, incrementally adjusting for demographics, comorbidities, procedural and hospital characteristics, and frailty and social vulnerability. Model performance was assessed using C-statistics and integrated discrimination improvement (IDI).

Results

A total of 371,248 TAVR patients were included in the analysis. The baseline model, including only demographic factors (age, sex, and race), yielded modest model performance (C = 0.589). Inclusion of comorbidities improved the model discrimination substantially (C = 0.684; IDI +6.9%, p < 0.001), and adding hospital and procedural characteristics yielded additional gains (C = 0.695; IDI +0.9%, p < 0.001). The final model integrated frailty and social vulnerability and achieved the highest predictive accuracy (C = 0.705; IDI +1.0%, p < 0.001).

Conclusions

In this large national cohort, frailty and social vulnerability significantly improved risk prediction for long-term mortality after TAVR. We conclude that sociodemographic and frailty-related factors are important components for prediction of 5-year mortality after TAVR.