Cost-Effectiveness of Left Ventricular Assist Device for Transplant-Ineligible Patients.

Abstract

Singapore is considering subsidizing left ventricular assist devices (LVADs) for end-stage heart failure (ESHF) and uses cost-effectiveness evidence to inform the determination. Yet, no economic evaluation has thus far been conducted. To estimate the lifetime cost-effectiveness of LVAD compared with optimal medical management for transplant-ineligible patients. This economic evaluation used a Markov model to simulate survival, stroke incidence, other adverse events, and heart failure hospitalizations for a cohort of adult patients in Singapore with transplant-ineligible ESHF, most of whom were inotrope dependent. Latest LVAD mortality data from a randomized clinical trial were age-adjusted, and an indirect comparison of prior trial results was performed to estimate survival for inotrope-dependent and inotrope-independent patients. Costs were estimated (in 2023 Singapore dollars [SGD]) using cohort billing data from 2017 to 2022 and National Heart Centre Singapore LVAD charges. Statistical analysis was performed from December 2023 to July 2024. HeartMate 3 LVAD (Abbott). Health care costs, quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratio (ICER) evaluated against a threshold of SGD 114 000 (US $85 075) per QALY gained. At model initiation, the cohort had a mean (SD) age of 64 (12) years, and 78% (range, 68%-86%) of patients were inotrope dependent. In the base case analysis, LVAD yielded an additional 3.45 QALYs at an incremental cost of SGD 404 678 (US $301 999), producing an ICER of SGD 117 370 (US $87 590) per QALY gained for the transplant-ineligible population. The ICER differed for inotrope-use subgroups at SGD 106 458 (US $79 446) per QALY gained for inotrope-dependent patients and SGD 174 798 (US $130 446) per QALY gained for inotrope-independent patients (with 59% and 19% probabilities, respectively, of attaining high value). The inotrope-dependent ICER was sensitive to model input changes and structural assumptions, whereas the inotrope-independent ICER consistently exceeded the high-value threshold in scenario analyses. In threshold analyses, a 44% reduction in the total implantation price or a 54% reduction in the all-cause mortality hazard were required for LVAD to be high value for inotrope-independent patients. Confidence that the inotrope-dependent ICER is high value increased to 75% and 85% with respective 20% and 33% reductions in total implantation price. In this economic evaluation comparing LVAD with optimal medical management, LVAD was potentially high value for most transplant-ineligible patients who are inotrope dependent. Confidence in this result was improved with plausible price reductions, yet only extreme changes rendered LVAD high value for inotrope-independent patients.