OBJECTIVES

: To determine cost-effectiveness of epilepsy surgery as an additional treatment option for post-stroke epilepsy (PSE) patients with medication management from the National Health Service (NHS) England perspective.

METHODS

: A systematic literature review (SLR) was conducted on Medline and Google Scholar to provide an overview regarding methodologies used for cost-effectiveness analysis in epilepsy and to provide baseline assumptions for modelling. We opted to do a discrete-event simulation model in order to incorporate patient characteristics which may vary across contexts. This modelling uses a time-to-event analysis since there is no pattern in epilepsy occurrence time. Our model ran deterministic (base-case study) and probabilistic (stochastic error taken into account through Monte Carlo sampling) analyses over a 5-year horizon with a 3.5% annual discount rate. Input parameters were sampled from 748 UK patients (75-year-old in average, 49% male, 83% ischemic stroke) gathered through a secondary care real world dataset, Hospital Episode Statistics. Primary outcomes were cost, quality-adjusted life years (QALYs), and incremental cost-effectiveness ratio (ICER). Uncertainty was measured using univariate sensitivity analysis.

RESULTS

: Among the four studies included in the SLR, three used discrete event simulation. When surgery performed as an additional treatment option on a PSE patient on medical management is compared with a scenario without surgery, our model estimates an incremental cost of £3,960 (£3,587), with a 0.041 (0.043) increase of QALYs, resulting in an ICER of £96,585 (£83,751) per QALY for deterministic and probabilistic analysis, respectively.

CONCLUSIONS

: Our modelling suggests offering epilepsy surgery in addition to medication management for PSE patients is unlikely to be cost-effective. Given the heterogenous nature of epilepsy patients, this model is not intended to limit access to epilepsy surgery but illustrate the economic impact in the PSE sub-population. Future studies to further characterize epilepsy and determine optimal treatment pathways are encouraged.