OBJECTIVES: Transcatheter aortic valve replacement (TAVR) has been mainly advised for inoperable and high surgical risk patients for treating severe aortic stenosis (AS). We conducted cost-effectiveness analysis on TAVR compared with common practice, surgical AVR (SAVR), in intermediate-low risk patients with severe AS in Singapore.

METHODS: Due to transfemoral (TF) access being dominant in Singapore, TF-TAVR was used in current analysis. A two-phase economic model combining decision tree and Markov model was developed to assess the incremental cost-effectiveness (ICER) of TF-TAVR versus SAVR over18-year horizon from Singapore healthcare system perspective. Event rates of mortality and complications were estimated from the Placement of Aortic Transcatheter Valves (PARTNER) 2 RCT cohort A and Singapore life table. Cost data were retrieved from the National University Heart Centre database. Health utilities were obtained from Singapore population norm based on EQ-5D. Scenario analyses were conducted using shorter horizon and event rates from PARTNER 2 S3 propensity analysis that used newer TAVR systems. Both one-way and probabilistic sensitivity analyses were performed.

RESULTS: In base case scenario, the incremental effectiveness of TAVR versus SAVR is 0.30 QALYs. The ICER of TF-TAVR is S$21,062/QALY. In scenario analysis, where horizon was reduced to 10 years, ICER of TF-TAVR is S$27,060/QALY; where event rates from PARTNER 2 S3 propensity analysis was used, ICER of TF-TAVR is S$16,843/QALY and S$21,378/QALY over 18-year and 10-year horizon, respectively. At a willingness to pay threshold of S$65,000/QALY, TF-TAVR, compared with SAVR, has 99.87% probability of been cost-effective after 10,000 simulations. The model is the most sensitive to cost of TF-TAVI procedure, followed by cost of SAVR procedure.

CONCLUSIONS: TF-TAVR seems to be a highly cost-effective option compared to SAVR for intermediate-low surgical risk patients with severe AS in Singapore. Increased procedure experience and/or technology advance may have further increased the cost-effectiveness of TF-TAVR per scenario analysis.