Poststroke Epilepsy Is Associated With a High Mortality After a Stroke at Young Age
Follow-Up of Transient Ischemic Attack and Stroke Patients and Unelucidated Risk Factor Evaluation Study
Background and Purpose—Poststroke epilepsy is a common complication after a young stroke. We investigated the association between poststroke epilepsy and mortality.
Methods—We performed a prospective cohort study among 631 patients with a first-ever transient ischemic attack or ischemic stroke, aged 18 to 50 years. Survival analysis and Cox proportional hazard analysis were used to estimate cumulative mortality and hazard ratios for patients with and without epilepsy.
Results—After mean follow-up of 12.5 years (SD 8.6), 76 (12.0%) developed poststroke epilepsy. Case fatality was 27.4% for patients with poststroke epilepsy and 2.1% for those without. Poststroke epilepsy was associated with 30-day mortality (hazard ratio, 4.8; 95% confidence interval, 1.7–14.0) and long-term mortality (hazard ratio, 1.8; 95% confidence interval, 1.2–2.9).
Conclusions—Epilepsy is a common problem after a young stroke and is associated with an increased short-term and long-term mortality.
Epilepsy is a common complication after stroke in young adults. Those with poststroke epilepsy have an even more impaired functional outcome in their already compromised life than patients without epilepsy.1 There are no studies on the association between poststroke epilepsy and mortality in young adults. A large cohort study in elderly patients with stroke showed that seizures were associated with case fatality and mortality at 1 year, in particular, in patients with a less severe stroke,2 whereas other showed conflicting results and usually did not take stroke severity into account and had a short follow-up.3–5 Long-term prognosis is of utmost importance, especially in young patients who usually still have a life expectancy of decades ahead.
The aim of this study was therefore to investigate the relationship between poststroke epilepsy and the risk of short- and long-term mortality in patients with a transient ischemic attack (TIA) or ischemic stroke at young age.
This study is part of the Follow-Up of Transient Ischemic Attack and Stroke Patients and Unelucidated Risk Factor Evaluation (FUTURE) study. The Medical Review Ethics Committee region Arnhem-Nijmegen approved the study. Participants provided written informed consent.
The FUTURE study comprises all consecutive patients with a TIA, ischemic stroke or intracerebral hemorrhage, aged 18 to 50 years, admitted to the Radboud University Nijmegen Medical Center from January 1980 until November 2010.6 To minimize bias because of changing diagnostic techniques, the World Health Organization definition for TIA and stroke was used.7 TIA was defined as a rapidly evolving focal neurological deficit with no other than a vascular cause lasting <24 hours. Stroke was defined similarly, but with symptoms lasting >24 hours. For the present study, only patients with a first-ever TIA or ischemic stroke were included. Exclusion criteria were cerebral venous sinus thrombosis and retinal infarction. A history seizure was an additional exclusion criteria.
At baseline the assessment of stroke pathogenesis (Trial of Org 10172 in Acute Stroke Treatment) and severity National Institutes of Health Stroke Scale (NIHSS) was done for all cases retrospectively,8 as these scales did not exist when a substantial proportion of our patients experienced their qualifying event.
Follow-up took place between November 2009 and January 2012 and subsequently between August 2014 and January 2015. During both follow-up assessments, patients underwent standardized, structured questionnaires to evaluate the occurrence of seizures. In case a patient had died, information on the occurrence of seizures was retrieved from the general practitioner by the same structured questionnaires. When patients (or their general practitioners) reported seizures, medical records were retrieved from their treating physician. This information was subsequently verified and adjudicated by an experienced neurologist who was blinded for the index event and outcome.
Poststroke epilepsy was defined according to the International League Against Epilepsy as a paroxysmal disorder of the central nervous system with our without the loss of consciousness and possible motor involvement.9 Early seizures were defined as those occurring within 1 week after the index event, late seizures thereafter.
Primary outcomes were all-cause long-term mortality and case fatality (death within 30 days after the index event). Information on the vital status was retrieved from the Dutch Municipal Personal Records database. Deaths were classified according to the International Classification of Diseases.10 Causes were categorized into stroke, cardiac, other vascular, malignancies, infections, and miscellaneous.
Cumulative case fatality and long-term mortality were estimated with Kaplan–Meier analysis for patients with and without epilepsy. In all analyses a time varying variable for poststroke epilepsy was created, as patients develop poststroke epilepsy during follow-up (and not fixed at baseline). For long-term mortality, survival plots were curtailed at 20 years to ensure reliable plots for all subgroups.
Cox proportional hazard models were used to calculate hazard ratios for death with poststroke epilepsy as a time varying variable. For case fatality, covariates in the model were sex, age at event, and NIHSS. For long-term mortality, covariates included sex, age at index event, history of diabetes mellitus, history of hypertension, Trial of Org 10172 in Acute Stroke Treatment classification, and recurrent stroke. In addition, analyses were stratified by stroke severity (mild, NIHSS 0–4; moderate, NIHSS 5–11; severe, NIHSS≥12).
A Cox proportional hazard model was constructed with age, sex, and period of the qualifying event (1980–1989, 1990–1999, 2000–2010) to evaluate whether a cohort or period effect could have influenced or mortality results, because the present study features a long inclusion period.
Schoenfeld residuals from the Cox models were examined to assess possible departures from model assumptions. There were no indications that the proportional hazards assumption was violated.
Baseline characteristics of participants (n=631) are shown in the Table. Stroke pathogenesis and comparison with nonparticipants is shown in Tables I and II in the online-only Data Supplement. After mean follow-up of 12.5 (SD 8.6) years, 76 patients (12.0%) developed poststroke epilepsy (Table). In 27.6% the first seizure was an early seizure and in 72.4% the first seizure was a late seizure. In addition, 36 patients developed recurrent seizures.
Cox proportional hazard models of mortality revealed no significant effect of the period in which patients experienced their qualifying event (1980–1989, 1990–1999, and 2000–2010), adjusted for age and sex.
A total of 21 patients (3.3%) died within 30 days after the index event. Twenty-eight patients (4.4%) developed poststroke epilepsy within these first 30 days. Cumulative case fatality was higher among those with poststroke epilepsy than those without (27.4% versus 2.1%; P<0.0001; Figure 1A). After adjusting for confounders, poststroke epilepsy remained associated with case fatality (hazard ratio, 4.8; 95% confidence interval, 1.7–14.0; P=0.004). Stroke was the major cause of death (Table III in the online-only Data Supplement).
At the end of follow-up 153 (24.2%) of all patients had died, 32.9% of the patients with and 23.1% of the patients without poststroke epilepsy. Figure 1B shows 20-year cumulative mortality for patients with and without epilepsy (56.5% and 32.6%, respectively; P=0.007). There were no differences in cause of death (Table III in the online-only Data Supplement). Poststroke epilepsy remained significantly associated with mortality after adjustment for confounders (hazard ratio, 1.8; 95% confidence interval, 1.2–2.9; P=0.009). In patients with a mild stroke, poststroke epilepsy was associated with mortality, after adjustment for confounders (hazard ratio, 2.5; 95% confidence interval, 1.2–5.3; P=0.012). Poststroke epilepsy was not associated with mortality in the moderate and severe stroke groups.
Poststroke epilepsy was associated with long-term mortality and case fatality in patients with a stroke at young age.
Strengths of our study were the prospective and single center design. Furthermore, our study has one of the longest follow-up and largest cohort.
There are also a few limitations. Selection bias may have occurred because of selective loss to follow-up. However, there was no difference in NIHSS, making this less likely.
Because our study features a long inclusion period, recall bias might have influenced our findings, especially when patients died a long time ago. However, we did not find a difference in incidence of epilepsy between patients who died a long time ago and those who died more recently. Furthermore, imaging techniques have changed during the past 30 years, which may have resulted in misclassification. We have tried to overcome this using the clinical definitions of stroke and TIA according to the World Health Organization.
Poststroke epilepsy was associated with case fatality. An explanation might be that patients with poststroke epilepsy have a more severe stroke. However, after adjustment for NIHSS, epilepsy remained associated with mortality. Nevertheless, because NIHSS assessment was done retrospectively as this scale did not exist when a large proportion of our population experienced their qualifying event, measurement error in stroke severity during this long inclusion period cannot totally be excluded.
In addition, we found that patients with poststroke epilepsy had a higher long-term cumulative mortality, even after adjusting for confounders. Especially in patients with a less severe stroke epilepsy was associated with mortality. Probably in patients with more severe strokes, severity is more important for mortality than poststroke epilepsy. In a large study on patients with stroke for 50 years, it was also shown that in patients with a less severe stroke epilepsy was associated with mortality2; however, follow-up was limited.
In conclusion, we showed that poststroke epilepsy is associated with both case fatality and long-term mortality. Future studies should investigate the role of antiepileptic drugs in poststroke epilepsy and the association with outcome.
Sources of Funding
This study was supported by the Dutch Epilepsy Fund (10–18).
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.115.010115/-/DC1.
- Received May 19, 2015.
- Revision received May 21, 2015.
- Accepted May 22, 2015.
- © 2015 American Heart Association, Inc.
- Reith J,
- Jørgensen HS,
- Nakayama H,
- Raaschou HO,
- Olsen TS.
- Williams LS,
- Yilmaz EY,
- Lopez-Yunez AM.
- Fisher RS,
- van Emde Boas W,
- Blume W,
- Elger C,
- Genton P,
- Lee P,
- et al
- 10.↵World Health Organization. International Statistical Classification of Diseases and Related Health Problems. Geneva: WHO; 1992.