부산대학교 도서관

OBJECTIVE: To examine the incidence, characteristics, treatment, and predictors of late seizures (LS) after cerebral venous thrombosis (CVT), we described these features in a registry of 1,127 patients with CVT. METHODS: We included consecutive adult patients from an international consortium of 12 hospital-based CVT registries. We excluded patients with a history of epilepsy or with <8 days of follow-up. We defined LS as seizures occurring >7 days after diagnosis of CVT. We used multivariable Cox regression to identify predictors of LS. RESULTS: We included 1,127 patients with CVT. During a median follow-up of 2.0 years (interquartile range [IQR] 1.0–6.3), 123 patients (11%) experienced ≥1 LS (incidence rate for first LS 30 per 1,000 person-years, 95% confidence interval [CI] 25–35). Median time to first LS was 5 months (IQR 1–16 months). Baseline predictors of LS included status epilepticus in the acute phase (hazard ratio [HR] 7.0, 95% CI 3.9–12.6), decompressive hemicraniectomy (HR 4.2, 95% CI 2.4–7.3), acute seizure(s) without status epilepticus (HR 4.1, 95% CI 2.5–6.5), subdural hematoma (HR 2.3, 95% CI 1.1–4.9), and intracerebral hemorrhage (HR 1.9, 95% CI 1.1–3.1). Eighty-five patients (70% of patients with LS) experienced a recurrent seizure during follow-up, despite the fact that 94% received antiepileptic drug treatment after the first LS. CONCLUSION: During a median follow-up of 2 years, ≈1 in 10 patients with CVT had LS. Patients with baseline intracranial bleeding, patients with acute symptomatic seizures, and those who underwent decompressive hemicraniectomy were at increased risk of developing LS. The high recurrence risk of LS justifies epilepsy diagnosis after a first LS.