In-hospital outcomes and cost-effectiveness of transcatheter aortic valve replacement among younger patients: a double/debiased machine learning approach using electronic health records in Germany

BACKGROUND: The prevalence of severe symptomatic aortic stenosis is increasing with population aging. Although surgical aortic valve replacement (SAVR) has traditionally been the standard treatment, transfemoral transcatheter aortic valve replacement (TF-TAVR) is increasingly used. The optimal treatment for patients aged 60-75 remains debated. METHODS: This retrospective cohort study analyzed 28,805 German patients who underwent isolated SAVR or TF-TAVR (2018-2022). We applied double/debiased machine learning estimators that combined adaptive lasso variable selection with propensity score-based weighting across 21 baseline characteristics. Cost-effectiveness was assessed via incremental cost-effectiveness ratios (ICER) and cost-effectiveness acceptability curves from in-hospital and 1-year perspectives. RESULTS: Compared with SAVR, TF-TAVR was associated with a significant reduction in in-hospital mortality (causal risk ratio [RR] 0.65; p = 0.012), along with lower rates of bleeding (RR 0.29; p < 0.001), postoperative delirium (RR 0.32; p < 0.001), and mechanical ventilation > 48 h (RR 0.39; p < 0.001). No significant difference was observed in acute kidney injury rates (RR 0.89; p = 0.150). However, reimbursement was substantially higher for TF-TAVR (€7071 more per case, p < 0.001). A hypothetical shift from SAVR to TF-TAVR was associated with an ICER of €857,413 (95% CI €472,195-€4,310,651) from the in-hospital perspective and €196,422 (95% CI €123,873-€457,813) from the 1-year perspective. Notably, unadjusted analyses indicated a narrowing cost gap over time: Reimbursement for TF-TAVR decreased by approximately 12% between 2018 and 2022, while SAVR costs remained stable. Consequently, TF-TAVR is becoming increasingly cost-effective. CONCLUSION: Given an estimated life expectancy of 11 to 25 years in this population, the incremental costs per life saved associated with a hypothetical shift from SAVR to TF-TAVR appear justifiable. Nonetheless, individual patient circumstances must always be considered in decision-making.