OBJECTIVE To systematically review available evidence regarding whether second-generation antiepileptic drugs (AEDs) contribute to the risk of balance disorders. METHODS We systematically evaluated data from randomized controlled trials that compared adjunctive therapy with a second-generation AED (gabapentin, lamotrigine, levetiracetam, oxcarbazepine, pregabalin, tiagabine, topiramate, or zonisamide) vs placebo for partial epilepsy and that reported dose-specific rates of ataxia or imbalance for each group. Random-effects meta-analysis was used to pool ratios (risk ratio [RR]) and associated 95% confidence intervals to determine whether there was evidence of an overall AED class effect or a dose-response effect and whether there were differences between individual AEDs. RESULTS Sixteen studies met inclusion criteria, representing 4279 individuals randomized to a second-generation AED and 1830 patients to placebo. Pooled analyses of all AEDs demonstrated that they increase imbalance risk at any dose (RR, 2.73; 95% confidence interval, 2.07-3.61) and at lowest dose (RR, 1.76; 95% confidence interval, 1.26-2.46). The highest dose analysis showed heterogeneity; evaluation of individual AEDs revealed that oxcarbamazepine and topiramate increased imbalance risk at all doses, whereas gabapentin and levetiracetam did not increase imbalance risk at any dose. A dose-response effect was observed for most AEDs. CONCLUSION Second-generation AEDs at standard dosages, except for gabapentin and levetiracetam, increase the imbalance risk, and evidence exists for a dose-response effect. The mechanisms, risk factors, and consequences of this risk for individual AEDs warrant further study. To systematically review available evidence regarding whether second-generation antiepileptic drugs (AEDs) contribute to the risk of balance disorders. We systematically evaluated data from randomized controlled trials that compared adjunctive therapy with a second-generation AED (gabapentin, lamotrigine, levetiracetam, oxcarbazepine, pregabalin, tiagabine, topiramate, or zonisamide) vs placebo for partial epilepsy and that reported dose-specific rates of ataxia or imbalance for each group. Random-effects meta-analysis was used to pool ratios (risk ratio [RR]) and associated 95% confidence intervals to determine whether there was evidence of an overall AED class effect or a dose-response effect and whether there were differences between individual AEDs. Sixteen studies met inclusion criteria, representing 4279 individuals randomized to a second-generation AED and 1830 patients to placebo. Pooled analyses of all AEDs demonstrated that they increase imbalance risk at any dose (RR, 2.73; 95% confidence interval, 2.07-3.61) and at lowest dose (RR, 1.76; 95% confidence interval, 1.26-2.46). The highest dose analysis showed heterogeneity; evaluation of individual AEDs revealed that oxcarbamazepine and topiramate increased imbalance risk at all doses, whereas gabapentin and levetiracetam did not increase imbalance risk at any dose. A dose-response effect was observed for most AEDs. Second-generation AEDs at standard dosages, except for gabapentin and levetiracetam, increase the imbalance risk, and evidence exists for a dose-response effect. The mechanisms, risk factors, and consequences of this risk for individual AEDs warrant further study.