Abstract WP354: Prophylactic Anti-epileptic Drugs and Spontaneous Intracerebral Hemorrhage

    loading  Checking for direct PDF access through Ovid


Introduction: Seizure is a common complication after intracerebral hemorrhage (ICH) but there are conflicting data about the impact of seizure on ICH outcome and the clinical utility of prophylactic antiepileptic drugs (AEDs) in ICH treatment.

Hypothesis: We hypothesized that prophylactic AEDs would be associated with worse outcomes after ICH.

Methods: We queried the Virtual International Stroke Trials Archive (VISTA), and identified 802 ICH patients who were followed for 90 days in prior clinical trials. Uni- and multivariate logistic and Cox regression models were designed to evaluate the effect of prophylactic AEDs on ICH outcome reflected by mortality and modified Rankin Scale (mRS)>3 at 90 days.

Results: Eighty two (10.2%) subjects received prophylactic AEDs after ICH. Patients who received AED prophylaxis had higher ICH volume (median [IQR] 23.2 [10.5, 38.0] vs. 14.3 [7.1, 27.0] cm3, P=0.001), and ICH score (1 [0, 2] vs. 1 [0, 1], P=0.028). In univariate analysis (Table), patients who received prophylactic AEDs had higher mRS at 90 days (4 [2, 6] and 3 [2, 5], P=0.03), and numerically greater mortality (27% vs. 18%, P=0.06). Phenytoin was the most commonly used AED and appeared to have a larger effect size than other AEDs. The ICH score and its components were strongly associated with outcomes (Table). Multivariable models adjusting for ICH score components attenuated the effect of prophylactic AEDs overall (mRS OR 1.33, 95% CI: 0.78-2.26, P=0.29; mortality HR 1.14, 95% CI: 0.72-1.80, P=0.58). Similar results were obtained for mRS and mortality outcomes when adjusted for overall ICH score, and for all AEDs as well as specifically for phenytoin.

Conclusion: Patients with higher ICH scores and larger hemorrhages are more likely to receive prophylactic AEDs and more likely to have poor outcomes. In this large multitrial cohort from VISTA, prophylactic AEDs were not independently associated with outcome after adjustment for these confounders.

Related Topics

    loading  Loading Related Articles