BackgroundDespite progress in stroke therapy (e.g., revascularisation interventions by thrombolysis and/or thrombectomy, organised stroke care), many stroke survivors will have impairment of neurological function. We aimed to compare the cost-effectiveness of an oral natural formulation, MLC601, versus placebo in functional recovery among subjects receiving standard of care after an ischemic stroke of intermediate severity assessed with NIH Stroke Scale at baseline (b-NIHSS 8–14).MethodsA Markov cohort model with a 2-year time horizon was developed to simulate patients from a published randomised placebo-controlled clinical trial of MLC601 in their post-stroke functional recovery assessed by modified Rankin Score (mRS), from a health system perspective. Transition probabilities were derived from a multi-centre clinical trial in South East Asia. As cost and utility data were not collected in the trial, therefore we extracted them from the published literature. The main outcomes were incremental cost, incremental quality-adjusted life-year (QALY) gained, and incremental cost-effectiveness ratio (ICER). Besides base-case and sensitivity analyses, we performed subgroup analyses to explore the heterogeneity of patients with poor-prognosis factors (b-NIHSS 10–14, stroke onset to treatment time > 48 h, rehabilitation during first 3 month). All costs are expressed in 2022 Euro and USD, with an annual discount rate of 3% applied to costs and QALYs.ResultsBase-case analysis showed that MLC601 was cost-effective compared with placebo, with €5,080 saved and 0.45 QALY gained, resulting in an ICER of -€11,352.50 per QALY gained. Similarly, results from subgroup analyses indicated that the use of MLC601 was a dominant strategy in all subgroups with poor-prognosis factors. Sensitivity analyses revealed the results were robust.ConclusionCompared with placebo on top of standard stroke care, MLC601 was cost-effective in post-stroke functional recovery over two years. Due to the lack of cost and utility data from the study population, the results might not be generalizable to other settings. Further studies with country-specific data are needed to confirm the results of this study.Trial registrationURL http://www.clinicaltrials.gov. Unique identifier NCT00554723 November 7, 2007.
Read full abstract