Abstracts

Rationale: Patients suffering from drug resistant epilepsy carry the highest risk of sudden unexpected death in epilepsy (SUDEP), with an average incidence estimated at 0,4%/year. The efficacy of new antiepileptic drugs (AEDs) is first evaluated in this population, using randomised placebo-controlled trials (RCTs). We investigate the incidence of SUDEP in patients included in such RCTs, and tested for the first time whether patients receiving placebo may be exposed to a higher risk of SUDEP than those receiving AEDs. Methods: Two electronic databases were searched for RCTs investigating any antiepileptic drug in the add-on treatment of drug-resistant epilepsy in adults. All deaths and their underlying cause were collected. The pharmaceutical companies involved in the development of licensed AEDs were contacted to confirm the occurrence and the cause of deaths. Patients were then separated in three groups: (i) those randomised to AED at efficacious dosage, (ii) those randomised to AED at non-efficacious dosage and (iii) those randomised to placebo. According to the duration of the whole treatment phase of each RCT, the incidence of SUDEP per 1000 patient-years was calculated for each group. Results: Our search yielded 101 RCTs representing 4626 patient/years of follow-up. 10592 patients were randomised to AEDs at efficacious dose, 769 to AED at non-efficacious dosage and 6520 to placebo. Overall 26 deaths were reported including 17 possible SUDEP, translating into a death rate of 0,56%/year and a SUDEP rate of 0,37%/year. The risk of SUDEP varied between the patients groups and was greater among those receiving placebo (0,64%/year) as compared to those allocated to AED at efficacious dosage (0,08%/year), yielding a relative risk of 8,15. Conclusions: The global incidence of SUDEP occurring during the double blind phase of add-on RCTs (0,37%/year) is comparable to that generally reported in the population of patients with drug resistant epilepsy (0,4%/year). However, this incidence appears to be five time lower in patients receiving an additional AED at efficacious dosage (0,08%/year), and about 50% higher in those receiving placebo (0,64%). This observation questions the safety of double-blind placebo controlled RCTs in patients with drug resistant epilepsy, and raises the issue of alternative strategy for an optimal development of new AEDs.