Background and purpose Approximately 30% of epilepsy patients develop a drug‐refractory epilepsy, that is, seizures cannot be controlled with antiepileptic drugs. Surgery has been evaluated as an effective but costly form of treatment. The aim of this systematic review is to synthesize the available evidence on the cost‐effectiveness of surgical treatment compared to medical treatment for these patients. Method A systematic literature search was performed in MEDLINE, Embase, PsycINFO, Cochrane Library and the National Health Service Economic Evaluation Database until September 2022. Title, abstract and full‐text screening were conducted by two researchers. Original studies published in English or German analyzing the cost‐effectiveness of surgical compared to medical treatment were included. Study characteristics, effectiveness measures, costs and incremental cost‐effectiveness ratios (ICERs) were extracted. The quality of studies was assessed using the Drummond checklist. Results Fourteen studies were included. Most studies evaluated surgery as cost‐effective. The ICER per patient seizure free ranged from dominant to purchasing power parity US dollars (PPP‐USD) 479,275. The ICER per 1% seizure reduction ranged from PPP‐USD 227 to PPP‐USD 342. The ICER per year without seizures was PPP‐USD 4202 and the ICER per quality‐adjusted life‐year ranged from dominant to PPP‐USD 90,874. The studies varied greatly in their methodology and time horizon. Conclusion Surgical treatment is cost‐effective compared to medical treatment, especially when a lifetime horizon is adopted. It is concluded that all disease‐specific costs should be considered over a long period when assessing the cost‐effectiveness of epilepsy treatment. From an economic perspective, efforts should be made to improve access to surgical treatment for patients with drug‐refractory epilepsy.