Objectives: The general anesthetic sevoflurane is being repurposed as a topical analgesic for painful chronic wounds. We conducted a Bayesian cost-effectiveness analysis (CEA) comparing the addition of domiciliary topical sevoflurane to conventional analgesics (SEVOFLURANE, n = 38) versus conventional analgesics alone (CONVENTIONAL, n = 26) for the treatment of nonrevascularizable painful leg ulcers in an outpatient Pain Clinic of a Spanish tertiary hospital.
Methods: We used real-world data collected from charts to conduct this CEA from a public healthcare perspective and with a one-year time horizon. Costs of analgesics, visits and admissions were considered, expressed in €2016. Analgesic effectiveness was measured with SPID (Sum of Pain Intensity Difference). A Bayesian regression model was constructed, including "treatment" and baseline characteristics for patients ("arterial hypertension") and ulcers ("duration", "number", "depth", "pain") as covariates. The findings were summarized as a cost-effectiveness plane and a cost-effectiveness acceptability curve. One-way sensitivity analyses, a re-analysis excluding those patients who died or suffered from leg amputation, and an extreme scenario analysis were conducted to reduce uncertainty.
Results: Compared to CONVENTIONAL, SEVOFLURANE was associated with a 46% reduction in costs, and the mean incremental effectiveness (28.15±3.70 effectiveness units) was favorable to SEVOFLURANE. The estimated probability for SEVOFLURANE being dominant was 99%. The regression model showed that costs were barely influenced by any covariate, whereas effectiveness was noticeably influenced by "treatment". All sensitivity analyses showed the robustness of the model, even in the extreme scenario analysis against SEVOFLURANE.
Conclusions: SEVOFLURANE was dominant over CONVENTIONAL as it was less expensive and much more effective.