Cost-Effectiveness of Aortic Valve Replacement in Low- and Intermediate-Risk Chinese Patients With Severe Aortic Stenosis.

BACKGROUND

Transcatheter aortic valve replacement (TAVR) remains debated as an alternative to surgical aortic valve replacement (SAVR). We aimed to evaluate the cost-effectiveness of aortic valve replacement strategies in low- and intermediate-risk patients with severe aortic stenosis in China.

METHODS

A decision-analytic model combining decision tree and Markov model was developed to compare outcomes of universal SAVR, universal TAVR, and a risk-based strategy (SAVR in low-risk patients and TAVR in intermediate-risk patients) in a hypothetical cohort of 75-year-old patients with aortic stenosis within the perspective of the Chinese health care system. A meta-analysis was performed to derive the clinical inputs; the 2019 to 2021 claims data from Shaanxi Province were used for cost analysis, and quality of life was measured using EuroQoL-5D. One-way and probabilistic (10 000 Monte Carlo simulations) sensitivity analyses were conducted to examine the robustness of model results. Primary outcomes included total costs, quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratio (ICER).

RESULTS

Universal TAVR gained the most QALYs (6.76 QALYs) with the highest costs (USD 58 949). Compared with universal SAVR, the risk-based strategy gained 0.12 additional QALYs at higher costs (USD 14 046); the ICER (117 048 USD/QALY) exceeded the willingness-to-pay threshold (37 657 USD/QALY, 3-fold gross domestic product per capita in China). The ICER of universal TAVR versus universal SAVR (80 526 USD/QALY) also exceeded the willingness-to-pay threshold. Sensitivity analysis showed that universal TAVR would be cost-effective if TAVR valve costs were <USD 21 477 (>44.23% cost reduction). Subgroup analysis showed that universal TAVR and risk-based strategy remained not cost-effective compared with universal SAVR in both low-risk (ICER of 64 414 USD/QALY) and intermediate-risk (ICER of 124 851 USD/QALY) patients. In 10 000 Monte Carlo simulations, the probabilities of being cost-effective for universal SAVR, universal TAVR, and risk-based strategy were 89.81%, 10.14%, and 0.05%, respectively.

CONCLUSIONS

The risk-based strategy and universal TAVR appeared not to be cost-effective versus universal SAVR in low- and intermediate-risk patients with severe aortic stenosis in China.