Stroke and Risk of Epilepsy: A Danish Nationwide Register-Based Study

BACKGROUND: Stroke is associated with a risk of epilepsy, but associations with age, sex, stroke type and severity, time trends, and mortality are uncertain. We studied the risk of epilepsy after stroke while accounting for sex, age, stroke types and severity, calendar time, and death. METHODS: This was a prospective nationwide register-based, matched cohort study of patients admitted with a validated first stroke in Denmark from April 1, 2004, to December 16, 2018, excluding those with prior epilepsy. Patients with stroke were matched 10:1 on age, sex, and calendar time with reference people without prior epilepsy or stroke. We estimated the cumulative incidence of an epilepsy diagnosis in the Danish National Patient Registry (International Classification of Diseases Tenth Revision: G40) with death as a competing risk using competing risk regression and estimated adjusted hazard ratios by Cox regression models. RESULTS: We identified 101 034 patients with stroke (46.5% female; mean age, 70.4 years) who survived 14 days after stroke along with 1 010 333 matched reference people. Two years after the stroke, the cumulative incidence of epilepsy was 3.0% (95% CI, 2.9-3.2) after ischemic stroke and 8.6% (95% CI, 8.0-9.2) after intracerebral hemorrhage versus 0.7% (95% CI, 0.7-0.7) in the matched references. Compared with the reference population, the 2-year hazard ratio of epilepsy was 21.7 (95% CI, 20.3-23.2) after ischemic stroke and 61.3 (95% CI, 51.1-73.4) after intracerebral hemorrhage. The risk of epilepsy increased with stroke severity; the 2-year cumulative incidence of epilepsy was 10.5% (95% CI, 9.5-11.4) for very severe ischemic stroke and 13.1% (95% CI, 11.1-15.1) after very severe intracerebral hemorrhage. CONCLUSIONS: In this population-based study of patients with validated stroke, the absolute and relative risk estimates of poststroke epilepsy were lower compared with previous studies. Reasons for the lower risk estimates include accounting for the high mortality associated with stroke, which had a significant impact on risk especially for severe stroke.