Network Meta-analysis of Antiepileptic Drugs in Focal Drug-resistant Epilepsy

Abstract Objective To compare and rank the efficacy and acceptability of new antiepileptic drugs (AEDs) for patients with focal drug-resistant epilepsy. Methods PubMed, EMBASE, Cochrane databases and Clinicaltrials.gov were systematically searched from their inception through January 1, 2020, to identify trials evaluating AEDs for focal drug-resistant epilepsy. We included randomized controlled clinical trials (RCTs) comparing new AEDs with placebo or with other AEDs as adjunctive therapy for focal drug-resistant epilepsy. A Bayesian network meta-analysis was performed to determine efficacy and acceptability, as reflected by odds ratios (ORs), 95% credible intervals (CrIs) with random-effects and consistent models. Results Sixty-two RCTs were included, involving 12,739 patients with focal drug-resistant epilepsy. Regarding the seizure-free rate (40 RCTs involving 9,136 patients), 8 AEDs were more efficacious than placebo, with lnORs ranging between 1.69 for brivaracetam (95% CrI, 0.56 to 2.81) and 0.72 for pregabalin (95% CrI, 0.12 to 1.32). Regarding the responder rate, all AEDs except oxcarbazepine were more efficacious than placebo, with lnORs ranging between 1.31 for levetiracetam (95% CrI, 0.92 to 1.71) and 0.66 for carisbamate (95% CrI, 0.17 to 1.14). Regarding acceptability (60 RCTs comprising 12,139 patients), 9 AEDs were inferior to placebo. Estimated from seizure-free rate, brivaracetam was ranked as the most efficacious AED based on cumulative probability plots and SUCRAs, with fatigue as the main