OBJECTIVE: To determine whether it is worth pursuing surgery for the treatment of epilepsy in patients with normal neuroimaging. METHODS: Two patient populations were studied: (1) 136 consecutive patients who were surgically treated; (2) 105 consecutive patients assessed with chronically implanted intracranial electrodes within the same period. Sixty patients belonged to both groups, and included all 21 patients who had normal neuroimaging. RESULTS: There were no differences in the proportion of patients with favourable outcome between those with normal and those with abnormal neuroimaging, irrespective of whether intracranial recordings were required. Among the 19 operated patients with normal neuroimaging, 74% had a favourable outcome (Engel's seizure outcome grades I and II), and among the 93 patients with abnormal neuroimaging, 73% had favourable outcome (p = 0.96). In patients with temporal resections, 92% of the 13 patients with normal neuroimaging had a favourable outcome, whereas among the 70 patients with abnormal neuroimaging, 80% had a favourable outcome (p = 0.44). In patients with extratemporal resections, two of the six patients with normal neuroimaging had a favourable outcome, while 12 of the 23 patients with abnormal neuroimaging had a favourable outcome (p = 0.65). Among the 105 patients studied with intracranial electrodes, five suffered transitory deficits as a result of implantation, and two suffered permanent deficits (one hemiplegia caused by haematoma and one mild dysphasia resulting from haemorrhage). CONCLUSIONS: It is worth pursuing surgery in patients with normal neuroimaging because it results in good seizure control and the incidence of permanent deficits associated with intracranial studies is low.