Risk of Poststroke Epilepsy Among Young Adults With Ischemic Stroke or Intracerebral Hemorrhage

Esmée Verburgt, MSc1; Lina Fellah, MSc1; Merel S. Ekker, MD, PhD1; et al

2025년04월 14일

JAMA Neurol.  2025;82(6):597-604

DOI: 10.1001/jamaneurol.2025.0465

Abstract

Importance   Poststroke epilepsy (PSE) is a major complication among young adults and is associated with problems with functional recovery and daily life. Although scores have been developed to predict risk of PSE, they have not been validated among patients with stroke at a young age.

Objective  To investigate both the risk of and risk factors for PSE at a young age and validate current PSE risk scores among a cohort of young adults.

Design, Setting, and Participants  This cohort study used data from ODYSSEY (Observational Dutch Young Symptomatic Stroke Study), a prospective cohort study conducted among 17 hospitals in the Netherlands between May 27, 2013, and March 3, 2021, with follow-up until February 28, 2024. Participants included 1388 consecutive patients aged 18 to 49 years with neuroimaging-proven ischemic stroke or intracerebral hemorrhage (ICH) and without a history of epilepsy. Statistical analysis took place between June and August 2024.

Exposure  First-ever neuroimaging-proven ischemic stroke or ICH.

Main Outcome and Measures  Poststroke epilepsy was defined as at least 1 remote symptomatic seizure (>7 days). Cumulative incidence functions were used to calculate the 5-year risk of PSE. Fine-Gray regression models were used to identify risk factors associated with PSE (age, sex, clinical stroke, and neuroimaging variables). The performances of the SeLECT (severity of stroke, large-artery atherosclerosis, early seizure, cortical involvement, and territory of middle cerebral artery) 2.0 risk score (for ischemic stroke) and the CAVE (cortical involvement, age, bleeding volume, and early seizure) risk score (for ICH) were assessed with C statistics and calibration bar plots.

Results   This study included 1388 patients (ischemic stroke, 1231 [88.7%]; ICH, 157 [11.3%]; median age, 44.1 years [IQR, 38.0-47.4 years]; 736 men [53.0%]; median follow-up, 5.3 years [IQR, 3.4-7.4 years]), of whom 57 (4.1%) developed PSE. The 5-year cumulative risk of PSE was 3.7% (95% CI, 0.2%-4.8%) after ischemic stroke and 7.6% (95% CI, 3.5%-11.8%) after ICH. Factors associated with PSE after ischemic stroke were an acute symptomatic seizure (<7 days) (hazard ratio [HR], 10.83 [95% CI, 2.05-57.07]; P = .005) and cortical involvement (HR, 5.35 [95% CI, 1.85-15.49]; P = .002). The only factor associated with PSE after ICH was cortical involvement (HR, 8.20 [95% CI, 2.22-30.25]; P = .002). The C statistic was 0.78 (95% CI, 0.71-0.84) for the SeLECT 2.0 risk score and 0.83 (95% CI, 0.76-0.90) for the CAVE risk score, and calibration was good for both scores.

Conclusions This study suggests that the risk of PSE among young adults is relatively low and that the factors that were associated with PSE were similar to variables included in the existing risk scores, which can therefore also be applied for young adults after stroke. Future clinical trials should investigate the optimal primary and secondary prophylaxis for patients at high risk.

The incidence rate per 100 person-years was calculated by dividing the number of events by the total person-years at risk per 1-year intervals. The middle of those intervals was chosen as the point estimator in the graph. The shaded areas indicate 95% CIs.

Figure 2.  Five-Year Cumulative Risk and Incidence Per 100 Person-Years of Poststroke Epilepsy (PSE) After an Ischemic Stroke or Intracerebral Hemorrhage

A, Risk after ischemic stroke according to the SeLECT (severity of stroke, large-artery atherosclerosis, early seizure, cortical involvement, and territory of middle cerebral artery) 2.0 score. B, Risk after intracerebral hemorrhage according to the CAVE (cortical involvement, age, bleeding volume, and early seizure) score. The observed risk was calculated only for risk scores with at least 30 patients. Error bars indicate 95% CIs, which were not reported for the predicted risk per CAVE score.

Figure 3.  Bar Plots Showing Predicted and Observed Risk of Poststroke Epilepsy (PSE)

Key Points

Question  What is the risk of poststroke epilepsy (PSE) among young adults (<50 years) with ischemic stroke or intracerebral hemorrhage?

Findings   In this cohort study including 1388 patients, the 5-year cumulative risk of PSE was 3.7% after ischemic stroke and 7.6% after intracerebral hemorrhage. Factors associated with PSE were similar to variables in current risk scores for PSE.

Meaning  This study suggests that the risk of PSE among young adults is relatively low and that current risk scores for PSE can be applied to identify young patients at high risk.

요약

·         중요성 :   뇌졸중 후 간질(Poststroke Epilepsy, PSE)은 젊은 성인에서 흔한 합병증으로, 기능 회복과 일상생활에 큰 영향을 미친다.  PSE의 위험을 예측하는 여러 점수가 개발되어 있지만, 젊은 환자군에서는 그 타당성이 검증되지 않았다.

·         목적 :   젊은 성인에서 뇌졸중 후 간질의 위험 및 위험 요인을 평가하고, 기존의 PSE 위험 예측 점수의 유효성을 검증하는 것.

·         설계, 환경 및 참가자 : 이 코호트 연구는 네덜란드 내 17개 병원에서 수행된 전향적 연구인ODYSSEY(Observational Dutch Young Symptomatic Stroke Study) 데이터를 사용했다.  2013년 5월 27일부터 2021년 3월 3일까지 등록된 18세 이상 49세 이하의 영상학적으로 허혈성 뇌졸중 또는 뇌내출혈(ICH) 이 확인된 1388명이 대상이며, 추적관찰은 2024년 2월 28일까지 진행되었고, 통계 분석은 2024년 6월부터 8월 사이에 수행되었다.

·         노출요인 : 첫 번째로 발생한 영상학적으로 확인된 허혈성 뇌졸중 또는 뇌내출혈(ICH)

·         주요 결과 및 측정 :   PSE는 7일 이후에 발생한 증상성 발작(원격 발작)으로 정의하였다. 5년 누적 위험은 누적발생함수(cumulative incidence function)를 통해 계산되었고, 위험 요인은 Fine-Gray 회귀모형을 사용해 분석하였다. 허혈성 뇌졸중 환자에는 SeLECT 2.0 점수, ICH 환자에는 CAVE 점수의 성능(C 통계, calibration bar plot)을 평가하였다.

·          결과 : 총 1388명(허혈성 뇌졸중 1231명[88.7%], ICH 157명[11.3%]; 중앙 나이 44.1세 [IQR 38.0–47.4세]; 남성 736명[53.0%]) 중 57명(4.1%)이 PSE가 발현되었다.

- 5년 누적 위험은 다음과 같이 나타났다.

• 허혈성 뇌졸중 후 3.7%(95% CI, 0.2%–4.8%)

• ICH 후 7.6%(95% CI, 3.5%–11.8%)

- 허혈성 뇌졸중 후 PSE와 유의하게 연관된 요인은 다음과 같다.

• 7일 이내 급성 증상성 발작 (HR 10.83 [95% CI, 2.05–57.07]; P = .005)

• 피질 침범 (HR 5.35 [95% CI, 1.85–15.49]; P = .002)

- ICH 후 PSE와 유의한 관련이 있는 유일한 요인은 피질 침범이었다.

 (HR 8.20 [95% CI, 2.22–30.25]; P = .002).

- 위험 예측 점수의 C 통계는 다음과 같았다. 두 점수 모두 예측 정확도(보정력)는 양호했다.

• SeLECT 2.0 점수: 0.78 (95% CI, 0.71–0.84)

• CAVE 점수: 0.83 (95% CI, 0.76–0.90)

·         결론  :   젊은 성인에서의 PSE 위험은 비교적 낮으나, 위험 요인은 기존의 예측 점수에 포함된 변수들과 유사하므로, 해당 점수는 젊은 뇌졸중 환자에게도 적용할 수 있다. 향후 연구에서는 고위험 환자에게 적절한 1차 및 2차 예방 전략을 탐색할 필요가 있다.