Background: Seizures are a common complication after intracerebral hemorrhage (ICH) but there is a substantial lack of information on the long-term incidence in ICH survivors and whether post-ICH seizures affect functional long-term outcome.
Methods: Over a five-year period 464 consecutive patients with spontaneous ICH were analyzed. Focussing on 1-year ICH survivors, clinical, and radiological parameters were retrieved from institutional prospective databases. The occurrence of seizures was categorized as early (≤7 days) or late (>7 days). Functional outcome was assessed by mailed questionnaires and telephone interviews, and was categorized into good vs. poor (mRS: 0-2 vs. 3-5) and favorable vs. unfavorable (mRS: 0-3 vs. 4-5). Multivariate regression models were calculated to investigate risk factors associated with post-ICH seizures including an a priori defined subgroup analysis of lobar ICH patients.
Results: Among 203 long-term ICH survivors, 19.7 % developed seizures of which 55 % occurred late. Factors associated with seizures were lobar location (OR 8.10; 95 % CI 3.04-21.59; p < 0.001), sepsis (OR 4.59; 95 % CI 1.20-17.53; p = 0.026), and history of alcohol abuse (OR 3.36; 95 % CI 1.25-9.06; p = 0.017). Subgroup analysis of lobar ICH patients revealed history of alcohol abuse as the only independent predictor of post-ICH seizures (OR 5.22; 95 % CI 1.25-21.78; p = 0.024). Functional long-term outcome among survivors was slightly worse in patients with post-ICH seizures (p = 0.059). In multivariate regression modeling for prediction of poor outcome, the parameter "post-ICH seizures" again reached a statistical trend (p = 0.065), and established parameters such as age, GCS, and hemorrhage volume were independently related to poor outcome.
Conclusions: Post-ICH seizures among long-term ICH survivors are common and may contribute to unfavorable functional outcome. Especially lobar ICH patients with a history of alcohol abuse are at risk to develop post-ICH seizures. Therefore, this subgroup may represent a target population for a prophylactic anticonvulsive treatment approach, preferably investigated in a prospective randomized trial.