Risk of sudden unexpected death in epilepsy in patients given adjunctive antiepileptic treatment for refractory seizures: a meta-analysis of placebo-controlled randomised trials.

BACKGROUND Sudden unexpected death in epilepsy (SUDEP) represents the main cause of death in patients with refractory epilepsy. No evidence-based intervention to prevent SUDEP exists. We postulated that pooling data from randomised placebo-controlled trials in patients with refractory epilepsy might show a lower incidence of SUDEP in patients receiving antiepileptic drugs (AEDs) at efficacious doses than in those receiving placebo. METHODS We searched Medline and the Cochrane Library for randomised trials investigating any AED in the add-on treatment of drug-resistant epilepsy in adults. We extracted the number and causes of death in patients allocated to AEDs at doses that were more efficacious than placebo against seizures, AEDs at non-efficacious doses, and placebo. In our primary analysis, we compared the occurrence of definite or probable SUDEP between patients given efficacious AED doses and those given placebo using the Mantel-Haenszel method, with exclusion of trials with no event. FINDINGS Data of 33 deaths, including 20 deemed as SUDEP, were extracted from 112 eligible randomised trials. 18 deaths were classified as definite or probable SUDEP and two as possible SUDEP. Definite or probable SUDEP, all SUDEP, and all causes of death were significantly less frequent in the efficacious AED group than in the placebo group, with odds ratios of 0·17 (95% CI 0·05-0·57, p=0·0046), 0·17 (0·05-0·57, p=0·0046), and 0·37 (0·17-0·81, p=0·0131), respectively. Rates of definite or probable SUDEP per 1000 person-years were 0·9 (95% CI 0·2-2·7) in patients who received efficacious AED doses and 6·9 (3·8-11·6) in those allocated to placebo. INTERPRETATION Treatment with adjunctive AEDs at efficacious doses may have reduced the incidence of definite or probable SUDEP by more than seven times compared with placebo in patients with previously uncontrolled seizures. This result provides evidence in favour of active treatment revision for patients with refractory epilepsy. FUNDING None.