Sudden unexplained death syndrome (SUDS) in epilepsy is identified as death in an epileptic individual with no anatomic cause found at autopsy. SUDS appears to be associated with subtherapeutic levels of anticonvulsants. Sudden death with no demonstrable cause at autopsy accounts for 5% to 30% of deaths in epileptic individuals. In the majority of cases, however, the cause of death in epileptic individuals can be demonstrated at autopsy. We examined the anti-epileptic drug concentrations in decedents who died as a direct result of epilepsy and compared these findings with those from a control population of epileptic patients who died suddenly due to some unrelated cause. This retrospective study was conducted on all deaths involving patients with epilepsy examined at the Jefferson County Coroner/Medical Examiner office from 1986-95. Out of 115 total cases the underlying cause of death was epilepsy in 60 cases - 52 cases of SUDS and 8 deaths caused by an accident precipitated by a seizure. In 44 cases death was unrelated to the decedent's epilepsy. In 11 cases the contribution of epilepsy to death could not be determined. Published articles on SUDS report subtherapeutic anti- epileptic medication levels in 63% to 94% of cases. We found subtherapeutic drug levels in 69% of the 52 cases of SUDS, in 75% of the 8 cases where a seizure precipitated an accident causing death, and in 34% of the control population. The incidence of subtherapeutic anticonvulsants is significantly greater in patients dying as a direct result of their epilepsy than in those dying of an unrelated cause.