RESUMO
BACKGROUND: Overall, up to one-third of epilepsy patients have drug-resistant epilepsy. However, there was previously no meta-analysis to support the guidelines for broad-spectrum antiseizure medication selection for the adjunctive treatment of refractory epilepsy. In the present meta-analysis, we assessed the efficacy and safety of three second-generation broad-spectrum antiseizure medications, lamotrigine (LTG), levetiracetam (LEV), and topiramate (TPM), and two third-generation broad-spectrum antiseizure medications, perampanel (PER) and lacosamide (LCM), for the adjunctive treatment of refractory epilepsy. METHODS: We systematically searched PubMed, Embase, and CENTRAL from inception to July 15, 2022. The studies included in the meta-analysis were required to meet the following criteria: (1) be randomized, double-blind clinical trials; (2) include patients aged >2 years with a clinical diagnosis of drug-resistant epilepsy; (3) have at least 8 weeks for the treatment period excluding the titration phase; and (4) report the outcomes of seizure response, seizure freedom and the withdrawal rate due to treatment-emergent adverse effects. Data were extracted, and the risk of bias for each study was assessed by two authors independently using RoB2 tools. We performed the network meta-analysis for each outcome through a group of programs in the mvmeta and network packages in Stata. Relative odds ratios with 95% confidence intervals were calculated as the result of the analyses. The surface under the cumulative ranking curve (SUCRA) and mean ranks were used to rank these treatments. RESULTS: Forty-two randomized controlled trials (RCTs) (LTG-placebo: n = 6, LEV-placebo: n = 13, TPM-placebo: n = 9, PER-placebo: n = 6, LCM-placebo: n = 7, LEV-TPM: n = 1) with 10257 participants (LTG = 569, LEV = 1626, TPM = 701, PER = 1734, LCM = 1908, placebo = 3719) were included. Levetiracetam had subequal efficacy in 50 % seizure frequency reduction to TPM [odds ratio (OR) 1.00, 95% confidence interval (CI) 0.73-1.38], and LEV had a higher rate of ≥ 50% seizure frequency reduction than LCM (OR 1.49, 95% CI 1.11-2.01) and PER (OR 1.68, 95% CI 1.24-2.29). Levetiracetam was also related to a higher proportion of seizure freedom participants than TPM (OR 1.87, 95% CI 1.20-2.89), PER (OR 2.23, 95% CI 1.12-4.43), and LCM (OR 2.97, 95% CI 1.46-6.05). In addition, LEV was associated with a lower risk of experiencing at least one treatment-emergent adverse event (TEAE) than PER (OR 0.63, 95% CI 0.46-0.85) and TPM (OR 0.51, 95 % CI 0.36-0.72) and a lower proportion of patients experiencing TEAEs leading to discontinuation than PER (OR 0.51, 95% CI 0.27-0.97) and TPM (OR 0.50, 95 % CI 0.27-0.93). CONCLUSIONS: Third-generation drugs (PER and LCM) had no advantages in terms of efficacy and safety for adjunctive treatment of refractory epilepsy compared with several second-generation drugs (LEV and LTG). Levetiracetam was the priority choice for adjunctive treatment of refractory epilepsy. Perampanel and LCM had no advantages in terms of efficacy and safety among the five drugs. REGISTRATION: PROSPERO registration number, CRD42022344153; last edited on December 23, 2022.