Behavioral adverse events with brivaracetam, levetiracetam, perampanel, and topiramate: A systematic review.

Abstract Purpose To understand the currently available post-marketing real-world evidence of the incidences of and discontinuations due to the BAEs of irritability, anger, and aggression in people with epilepsy (PWE) treated with the anti-seizure medications (ASMs) brivaracetam (BRV), levetiracetam (LEV), perampanel (PER), and topiramate (TPM), as well as behavioral adverse events (BAEs) in PWE switching from LEV to BRV. Methods A systematic review of published literature using the Cochrane Library, PubMed/MEDLINE, and Embase was performed to identify retrospective and prospective observational studies reporting the incidence of irritability, anger, or aggression with BRV, LEV, PER, or TPM in PWE. The incidences of these BAEs and the rates of discontinuation due to each were categorized by ASM, and where possible, weighted means were calculated but not statistically assessed. Behavioral and psychiatric adverse events in PWE switching from LEV to BRV were summarized descriptively. Results A total of 1500 records were identified in the searches. Of these, 44 published articles reporting 42 studies met the study criteria and were included in the data synthesis, 7 studies were identified in the clinical trial database, and 5 studies included PWE switching from LEV to BRV. Studies included a variety of methods, study populations, and definitions of BAEs. While a wide range of results was reported across studies, weighted mean incidences were 5.6% for BRV, 9.9% for LEV, 12.3% for PER, and 3.1% for TPM for irritability; 3.3%* for BRV, 2.5% for LEV, 2.0% for PER, and 0.2%* for TPM for anger; and 2.5% for BRV, 2.6% for LEV, 4.4% for PER, and 0.5%* for TPM for aggression. Weighted mean discontinuation rates were 0.8%* for BRV, 3.4% for LEV, 3.0% for PER, and 2.2% for TPM for irritability and 0.8%* for BRV, 2.4% for LEV, 9.2% for PER, and 1.2%* for TPM for aggression. There were no discontinuations for anger. Switching from LEV to BRV led to improvement in BAEs in 33.3% to 83.0% of patients (weighted mean, 66.6%). *Denotes only 1 study. Conclusions This systematic review characterizes the incidences of irritability, anger, and aggression with BRV, LEV, PER, and TPM, and it provides robust real-world evidence demonstrating that switching from LEV to BRV may improve BAEs. While additional data remain valuable due to differences in methodology (which make comparisons difficult), these results improve understanding of the real-world incidences of discontinuations due to these BAEs in clinical practice and can aid in discussions and treatment decision-making with PWE.