The primary aim of our study is to investigate the predictors of long-term anti-seizure medication (ASM) use in patients with acute stroke.

ASMs are commonly used for managing acute symptomatic seizures (ASyS) and epileptic abnormalities (EAs) after stroke. Many are discharged on ASMs. In absence of clear guidelines, patients may be continued on ASMs for a long period of time. Predictors of long-term ASM continuation in acute stroke patients remain unknown.

We performed a single-center, retrospective cohort study of patients fulfilling the following criteria from 2013 - 2018: adults (≥18 years), acute stroke [ischemic stroke (IS), intracerebral hemorrhage (ICH), or subarachnoid hemorrhage (SAH)], concern for ASyS [i.e. underwent continuous electroencephalogram (cEEG) monitoring], and attended out-patient clinic follow-up after hospitalization. The primary outcome was long-term ASM continuation, i.e. beyond the first post-hospitalization clinic visit. We performed a univariate, followed by multivariable logistic regression analysis.

A total of 504 patients were included and 290 (58%) were started on ASMs. The latter were included in the final analysis. Among them, 188 (37.3%) were discharged on ASMs. Primary outcome was met by 156 (30.9%) patients, accounting for 54% of those started on ASMs during admission. The multivariable model found that the odds of long-term ASM continuation were less if patients suffered SAH compared to IS [Odds ratio (OR) 0.38 [95% CI 0.15-0.97]) and were admitted to the neurological ICU (NICU) during admission (OR 0.37 [95% CI 0.15-0.92]). In contrast, history of an ASyS (clinical or electrographic) significantly increased the odds of ASM continuation (OR 19.73 [95% CI 9.16-47.16]) along with the presence of EAs on cEEG (OR 2.29 [95% CI 1.15-4.66]).

Close to one-third of acute stroke patients with ASyS concerns are continued on ASMs for a long-term after hospital discharge. This lack of deprescribing is primarily driven by ASyS and cEEG findings.