Pre-stroke seizures: A nationwide register-based investigation

Open ArchivePublished:May 16, 2017DOI:https://doi.org/10.1016/j.seizure.2017.05.010

      Highlights

      • Late-onset seizures increase the risk of later stroke.
      • The proportion of strokes that are preceded by seizures is not known.
      • We found seizures in 1.5% of all Swedish patients >60y with first stroke 2005–2010.
      • This represents 5–20% of all cases of late-onset seizures, depending on age group.
      • Studies on how to reduce stroke risk after late-onset seizures are needed.

      Abstract

      Purpose

      The relationship between cerebrovascular disease and seizures is clearly illustrated by poststroke epilepsy. Seizures can also be the first manifestation of cerebrovascular disease and case-control studies have demonstrated that seizures carry an increased risk of subsequent stroke. Thus, seizures could serve as a marker for vascular risk that merits intervention, but more data is needed before proper trials can be conducted. The occurrence of pre-stroke seizures has not been assessed on a national scale. We asked what proportion of strokes in middle-aged and elderly patients was preceded by seizures.

      Methods

      All patients over 60 years of age with first-ever stroke in 2005-2010 (n = 92,596) were identified in the Swedish stroke register (Riksstroke) and cross-sectional data on a history of a first seizure or epilepsy diagnosis in the ten years preceding stroke were collected from national patient registers with mandatory reporting.

      Results

      1372 patients (1.48%) had a first seizure or epilepsy diagnosis registered less than ten years prior to the index stroke. The mean latency between seizure and stroke was 1474 days (SD 1029 days).

      Conclusions

      Seizures or epilepsy preceded 1.48% of strokes in patients > 60 years of age. Based on recent national incidence figures, 5-20% of incident cases of seizures or epilepsy after 60 years of age could herald stroke, depending on age group. These proportions are of a magnitude that merit further study on how to reduce the risk of stroke in patients with late-onset seizures or epilepsy.

      Keywords

      1. Introduction

      In developed countries, cerebrovascular disease is the most common cause of epilepsy in the middle-aged and elderly [
      • Forsgren L.
      • Beghi E.
      • Oun A.
      • Sillanpaa M.
      The epidemiology of epilepsy in Europe − a systematic review.
      ]. The risk of poststroke epilepsy varies across studies, but is typically reported to develop in 2–10% of stroke patients, depending on stroke type, stroke severity, and study methodology[
      • Graham N.S.
      • Crichton S.
      • Koutroumanidis M.
      • Wolfe C.D.
      • Rudd A.G.
      Incidence and associations of poststroke epilepsy: the prospective South London Stroke Register.
      ,
      • Jungehulsing G.J.
      • Heuschmann P.U.
      • Holtkamp M.
      • Schwab S.
      • Kolominsky-Rabas P.L.
      Incidence and predictors of post-stroke epilepsy.
      ,
      • Zelano J.
      • Redfors P.
      • Asberg S.
      • Kumlien E.
      Association between poststroke epilepsy and death: a nationwide cohort study.
      ]. The opposite relationship between stroke and epilepsy – that late-onset seizures could reflect cerebrovascular disease that has not yet resulted in TIA or stroke – attracts increasing interest. Could seizures, after a certain age, serve as a TIA-equivalent and identify more patients that may benefit from treatment of vascular risk? For primary prevention of stroke to be motivated after seizures, the risk of stroke must be sufficient to warrant the risks associated with treatment, which must be effective. Authorities recommend aspirin when the risk of a vascular event exceeds 6–20% in a 10-year period [
      • Goldstein L.B.
      • Bushnell C.D.
      • Adams R.J.
      • et al.
      Guidelines for the primary prevention of stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association.
      ,
      • Williams B.
      • Poulter N.R.
      • Brown M.J.
      • et al.
      British hypertension society guidelines for hypertension management 2004 (BHS-IV): summary.
      ]. Whether late-onset seizures infer such a risk of stroke is not known.
      The observation that seizures can precede stroke has been reported in several trials. Early examples are the Oxfordshire community study, where 19 (3%) of 675 patients with stroke had previous seizures and seven patients had a first seizure in the year before their stroke [
      • Burn J.
      • Dennis M.
      • Bamford J.
      • Sandercock P.
      • Wade D.
      • Warlow C.
      Epileptic seizures after a first stroke: the Oxfordshire Community Stroke Project.
      ], and a case-control study on 230 patients with stroke where 4.5% of patients had previous epilepsy compared to 0.6% of controls [
      • Shinton R.A.
      • Gill J.S.
      • Zezulka A.V.
      • Beevers D.G.
      The frequency of epilepsy preceding stroke. Case-control study in 230 patients.
      ]. A subsequent larger study, in which over 4700 patients with epilepsy onset after the age of 60 and matched controls were prospectively assessed, found a hazard ratio for stroke of 2.89 (95% CI 2.45–3.91) [
      • Cleary P.
      • Shorvon S.
      • Tallis R.
      Late-onset seizures as a predictor of subsequent stroke.
      ]. In addition to evidence for a higher prevalence of epilepsy in stroke patients, a recent systematic review described a higher prevalence of occult (subclinical) cerebrovascular lesions in patients with late-onset epilepsy and suggested the term “heraldic seizures” for seizures that subsequently turn out to have preceded a stroke [
      • Brigo F.
      • Tezzon F.
      • Nardone R.
      Late-onset seizures and risk of subsequent stroke: a systematic review.
      ].
      In summary, the concept of heraldic seizures is attracting interest. Randomized controlled trials are probably needed to inform clinical practice, but such trials cannot be undertaken before more robust epidemiological data are provided for power calculations. We have previously identified a nation-wide cohort of stroke patients and assessed development of poststroke epilepsy by cross-referencing several Swedish national registers with excellent coverage [
      • Zelano J.
      • Redfors P.
      • Asberg S.
      • Kumlien E.
      Association between poststroke epilepsy and death: a nationwide cohort study.
      ]. In the present study, we investigated how often first-ever stroke in the cohort was preceded by seizures. To our knowledge, this is the first such investigation on a national scale. Furthermore, by using age-adjusted incidence rates from a recent surveillance register, we calculated estimates of what proportion of all expected incident cases of seizures or epilepsy that could herald stroke.

      2. Materials and methods

      The study population was obtained from the Swedish stroke register (Riksstroke), a national quality register established in 1994 [
      • Asplund K.
      • Hulter Asberg K.
      • Appelros P.
      • et al.
      The Riks-Stroke story: building a sustainable national register for quality assessment of stroke care.
      ]. All hospitals admitting acute stroke patients participate in the register, the details of which are available at (http://www.riksstroke.org/eng/). The Riksstroke register has an estimated coverage of approximately 95% of acute stroke in adults.
      Patients with a stroke during the years 2005–2010 were identified in Riksstroke. The register includes patients with intracerebral haemorrhage (ICH), ischemic stroke, and unspecified stroke (International Classification of Diseases 10th revision (ICD-10) codes I61, I63 and I64, respectively). The number of unclassified strokes is low (n = 3325, 3.1%), and since these patients had similar basic characteristics as patients with ischemic stroke, they have been treated as having ischemic strokes for the purpose of this study.
      For information on epilepsy the Riksstroke output was cross-linked to the National Patient Register (NPR) [
      • Ludvigsson J.F.
      • Andersson E.
      • Ekbom A.
      • et al.
      External review and validation of the Swedish national inpatient register.
      ]. The NPR is based on personal identification numbers and run by the National board of Health and Welfare. NPR contains information on all diagnoses registered at hospital-based care from 1987 and some hospital-based outpatient care since 2001, and has been used to identify cases of epilepsy in other studies [
      • Mattsson P.
      • Tomson T.
      • Eriksson O.
      • Brannstrom L.
      • Weitoft G.R.
      Sociodemographic differences in antiepileptic drug prescriptions to adult epilepsy patients.
      ]. NPR was the source of information on seizure-related diagnoses; epilepsy, status epilepticus, and seizures.

      2.1 Study population and classification

      The Riksstroke output contained 131,453 unique individuals with valid personal identification numbers. Patients that had a prior stroke (n = 26980) were then excluded, followed by any patients younger than 60 years at the time of stroke (n = 11877), leaving 92596 patients over 60 years with first-ever stroke in 2005–2010. We next sought cross-sectional data on history of seizures or epilepsy. Patients were classified as having a potentially heraldic seizure if ICD-10-codes for epilepsy, seizures or status epilepticus (G40, G41, or R56.8) were detected in the NPR more than three days and less than ten years before the stroke. The three-day window was chosen to reduce the influence on the results of misdiagnosis or registration errors of acute symptomatic seizures at stroke onset. For calculations of the proportion of all incident seizures that could precede stroke, population numbers were obtained from Statistics Sweden and incidence rates were obtained from data published from the Stockholm Incidence Register of Epilepsy (SIRE) [
      • Adelow C.
      • Andell E.
      • Amark P.
      • et al.
      Newly diagnosed single unprovoked seizures and epilepsy in Stockholm, Sweden : First report from the Stockholm Incidence Registry of Epilepsy (SIRE).
      ]. Data were analysed using SPSS®, confidence intervals were computed in GraphPad Prism®., and calculations were performed in Excel®.

      2.2 Approvals and patient consent

      This study, performed in agreement with privacy legislation in Sweden, was approved by the regional ethics committee in Gothenburg (approval number 187-15). Data were obtained by linking the above registries using the unique personal identification number of every Swedish citizen. The National Board of Health and Welfare anonymised all data after linkage and before we were given access to them.

      3. Results

      A total of 92,596 patients with first-ever stroke after age 60 were identified in Riksstroke. Details on the cohort are given in Table 1. Out of these, 1372 patients (1.48%) had a first seizure or epilepsy diagnosis registered less than ten years prior to the index stroke. The mean latency was 1474 days (SD 1029 days). The distribution of latencies is illustrated in Fig. 1.
      Table 1Demographics and clinical characteristics.
      nMedian (range)
      Age9259679 (60–107)
      GenderProportion
      Male4451548.1%
      Female4808151.9%
      Stroke type
      Infarction8239989.0%
      ICH1019711.0%
      Diabetes1765319.1%
      Smoking1064612.9%
      Atrial fibrillation2656928,9%
      Fig. 1
      Fig. 1Latency from seizure to stroke. On the Y-axis are frequencies (number of strokes) and on the x-axis are days from seizure to stroke. Each bar represents 100 days.
      We next analysed data regarding age groups and stroke type (Table 2). Out of 82,399 patients with cerebral infarction, a first seizure or epilepsy diagnosis not more than ten years prior to the stroke was found in 1174 patients (1.42%). Out of 10,197 patients with ICH, a first seizure or epilepsy diagnosis less than ten years prior to the stroke was found in 198 (1.94%) of patients. The proportion of patients with a pre-stroke seizure was higher in younger age groups for both stroke types.
      Table 2Age at stroke and proportion of heraldic seizures in age strata per stroke type.
      AllAge at strokeN, strokeN,

      pre-stroke sz
      %Lower 95% CIUpper 95% CI
      60-69187983852,051,862,26
      70-79289614511,561,421,71
      80-89361784621,281,171,40
      90-998561740,860,691,08
      100-1099800,00
      total9259613721,481,411,56
      ICH
      60-692512752,992,393,73
      70-793335601,801,402,31
      80-893636561,541,191,99
      90-9970870,990,482,03
      100-109600,00
      total101971981,941,692,23
      Infarction
      60-69162863101,901,702,12
      70-79256263911,531,381,68
      80-89325424061,251,131,37
      90-997853670,850,671,08
      100-1099200,00
      total8239911741,421,351,51
      Finally, we calculated the expected number of incident cases of seizures or epilepsy in Sweden in ten years (the observation time of the cohort). We next determined what proportion the detected 1.48% potentially heraldic seizures represented in total and per age strata. The results of the calculations are shown in table S1. By the most conservative estimate, the 1372 potentially heraldic seizures represented 9.8% of all expected seizures in persons over 50 years of age.

      4. Discussion

      We have here undertaken the first national-scale investigation of how often stroke is preceded by seizures. Out of 92,526 patients registered with a first-ever stroke, 1.48% had a first seizure or epilepsy diagnosis in the ten years before to the stroke.
      We detected a slightly higher proportion of patients with a history of seizures in ICH than infarctions and in younger patients. The higher proportions in younger age strata could have several explanations, for instance lower recognition of seizures in older patients or that occult cerebrovascular disease giving rise to seizures usually also manifests as stroke before patients reach a high age. The detected overall proportions are somewhat lower than those previously found in earlier studies [
      • Burn J.
      • Dennis M.
      • Bamford J.
      • Sandercock P.
      • Wade D.
      • Warlow C.
      Epileptic seizures after a first stroke: the Oxfordshire Community Stroke Project.
      ,
      • Shinton R.A.
      • Gill J.S.
      • Zezulka A.V.
      • Beevers D.G.
      The frequency of epilepsy preceding stroke. Case-control study in 230 patients.
      ]. This could at least partly reflect that the included patients in those studies were younger and perhaps also the development of stroke care leading to a relative increase in detection of stroke compared to that of seizures over the last twenty years.
      As all registry-based investigations, our study has limitations. The registries are very good in terms of coverage and frequently used to identify cases of stroke and seizure. Detection of seizures and epilepsy by an ICD code in NPR should be adequate, as seizures are dramatic symptoms often resulting in hospital visits. The NPR also contains hospital-based outpatient care. Most neurologists are hospital-based in Sweden, and very few cases of new onset-seizures are managed without involvement of hospital-based in- or outpatient care. Physician reluctance to diagnose seizures or epilepsy in order to reduce the impact of driving restrictions etc. is in our experience extremely rare in Sweden. Low detection of seizures would make our estimates of risk for subsequent stroke more conservative. Our data did not allow exclusion of patients with other causes of epilepsy or seizures, such as neurodegenerative disease. In addition, the patient register only has information from 1987, so patients with epilepsy onset before that date may have been erroneously classified as “new-onset” in the ten years before the index stroke. This risk is probably small, since cases of epilepsy are usually considered prevalent if a diagnosis is found in patient records within the last five years [
      • Christensen J.
      • Vestergaard M.
      • Pedersen M.G.
      • Pedersen C.B.
      • Olsen J.
      • Sidenius P.
      Incidence and prevalence of epilepsy in Denmark.
      ]. As long as these limitations are kept in mind − we believe that data on a national scale can add to the important discussion on heraldic seizures.
      Is 1.48% a high or low figure? Or differently put; are heraldic seizures common enough to warrant further studies in terms of cerebrovascular prophylaxis? To address this issue, we related the number of seizures that preceded stroke in our cohort to the expected number of incident seizures in Sweden over ten years. The proportions varied between 5 and 20%, depending on age group. Such calculations provide only estimates, but the figures are similar to those found in the major prospective study conducted to date, where 10% of patients with late-onset epilepsy suffered stroke over an average follow-up of 5 to 25 years (1919 days) [
      • Cleary P.
      • Shorvon S.
      • Tallis R.
      Late-onset seizures as a predictor of subsequent stroke.
      ]. Of note, our estimate may be conservative, since almost 20% of the patients in SIRE had an identifiable aetiology and heraldic seizures could represent an even greater proportion of unexplained late-onset seizure/epilepsy cases.
      What could mechanistically explain the relationship between seizures and subsequent stroke? Several possibilities exist. Seizures may be the first manifestation of occult cerebrovascular disease. Alternative or synergistic factors may be treatment with antiepileptic drugs with negative effects on vascular risk profile [
      • Renoux C.
      • Dell'Aniello S.
      • Saarela O.
      • Filion K.B.
      • Boivin J.F.
      Antiepileptic drugs and the risk of ischaemic stroke and myocardial infarction: a population-based cohort study.
      ,
      • Brodie M.J.
      • Mintzer S.
      • Pack A.M.
      • Gidal B.E.
      • Vecht C.J.
      • Schmidt D.
      Enzyme induction with antiepileptic drugs: cause for concern?.
      ,
      • Olesen J.B.
      • Abildstrom S.Z.
      • Erdal J.
      • et al.
      Effects of epilepsy and selected antiepileptic drugs on risk of myocardial infarction, stroke, and death in patients with or without previous stroke: a nationwide cohort study.
      ], lifestyle restrictions due to seizures, or comorbidities. Our dataset consists only of patients that have suffered stroke, but associations could perhaps be sought between such explanatory factors and latency from first seizure to stroke. In addition, such associations may be of interest in future cohort or case-control studies on patients with late-onset unexplained seizures and the risk of stroke.
      There is an on-going debate about when primary prophylaxis for cardiovascular disease is motivated. Recent America Heart Association recommendations state that 6–10% risk of any cardiovascular event within ten years motivates aspirin treatment [
      • Goldstein L.B.
      • Bushnell C.D.
      • Adams R.J.
      • et al.
      Guidelines for the primary prevention of stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association.
      ]. Older European recommendations have been more conservative, and suggest 20% ten-year risk of a cardiovascular event [
      • Williams B.
      • Poulter N.R.
      • Brown M.J.
      • et al.
      British hypertension society guidelines for hypertension management 2004 (BHS-IV): summary.
      ]. The matter is further complicated by a recent meta-analysis indicating that increases in the number of cerebral bleeds may outweigh benefits of reduction in ischemic stroke [
      • Lei H.
      • Gao Q.
      • Liu S.R.
      • Xu J.
      The benefit and safety of aspirin for primary prevention of ischemic stroke: a meta-Analysis of randomized trials.
      ].
      The retrospective cross-sectional nature of our investigation does not establish causal relationships, so the results are purely descriptive. Nonetheless, the findings support the emerging notion that a substantial proportion of all seizures in the investigated age groups seem to pre-date stroke. Our investigation is the first on a national scale and validates the observations made in previous studies. In a recent systematic review, the authors discuss that late-onset seizures in itself suffices as indication for work-up and treatment of vascular risk factors [
      • Brigo F.
      • Tezzon F.
      • Nardone R.
      Late-onset seizures and risk of subsequent stroke: a systematic review.
      ]. When it comes to anti-platelet therapy, which carries a risk of bleeding, some caution is probably needed. Future studies are warranted. A reasonable approach would be to identify subgroups where late-onset seizures are particularly likely to herald stroke − such groups could be patients with occult cerebrovascular disease on MRI, a family history of stroke, or other vascular risk factors as hypertension or hyperlipidemia. Clinical trials could then be performed to elucidate the risk-benefit ratio of vascular primary prophylaxis in cases of late-onset seizures.

      Disclosure of conflict of interest

      No authors have any conflict of interest to disclose. No funders had any influence over the study.

      Funding

      JZ receives funding from Swedish Society of Medicine , Jeanssons foundation , The Göteborg Medical Society , and Sahlgrenska university hospital . EK receives funding from The Selander and Thureus foundations . SÅ receives funding from The Swedish Research Council and (institutional research funding) AstraZeneca NordicBaltic .

      Acknowledgements

      The authors are grateful to Mattias Molin, Statistiska konsultgruppen, Gothenburg (statistical consultant) for data management and statistical services and to The Riksstroke Collaboration (www.riksstroke.org).

      Appendix A. Supplementary data

      The following are Supplementary data to this article:

      References

        • Forsgren L.
        • Beghi E.
        • Oun A.
        • Sillanpaa M.
        The epidemiology of epilepsy in Europe − a systematic review.
        Eur J Neurol. 2005; 12: 245-253
        • Graham N.S.
        • Crichton S.
        • Koutroumanidis M.
        • Wolfe C.D.
        • Rudd A.G.
        Incidence and associations of poststroke epilepsy: the prospective South London Stroke Register.
        Stroke. 2013; 44: 605-611
        • Jungehulsing G.J.
        • Heuschmann P.U.
        • Holtkamp M.
        • Schwab S.
        • Kolominsky-Rabas P.L.
        Incidence and predictors of post-stroke epilepsy.
        Acta Neurol Scand. 2013; 127: 427-430
        • Zelano J.
        • Redfors P.
        • Asberg S.
        • Kumlien E.
        Association between poststroke epilepsy and death: a nationwide cohort study.
        Eur Stroke J. 2016; 1: 272-278
        • Goldstein L.B.
        • Bushnell C.D.
        • Adams R.J.
        • et al.
        Guidelines for the primary prevention of stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association.
        Stroke. 2011; 42: 517-584
        • Williams B.
        • Poulter N.R.
        • Brown M.J.
        • et al.
        British hypertension society guidelines for hypertension management 2004 (BHS-IV): summary.
        BMJ. 2004; 328: 634-640
        • Burn J.
        • Dennis M.
        • Bamford J.
        • Sandercock P.
        • Wade D.
        • Warlow C.
        Epileptic seizures after a first stroke: the Oxfordshire Community Stroke Project.
        BMJ. 1997; 315: 1582-1587
        • Shinton R.A.
        • Gill J.S.
        • Zezulka A.V.
        • Beevers D.G.
        The frequency of epilepsy preceding stroke. Case-control study in 230 patients.
        Lancet. 1987; 1: 11-13
        • Cleary P.
        • Shorvon S.
        • Tallis R.
        Late-onset seizures as a predictor of subsequent stroke.
        Lancet. 2004; 363: 1184-1186
        • Brigo F.
        • Tezzon F.
        • Nardone R.
        Late-onset seizures and risk of subsequent stroke: a systematic review.
        Epilepsy Behav. 2014; 31: 9-12
        • Asplund K.
        • Hulter Asberg K.
        • Appelros P.
        • et al.
        The Riks-Stroke story: building a sustainable national register for quality assessment of stroke care.
        Int J Stroke. 2011; 6: 99-108
        • Ludvigsson J.F.
        • Andersson E.
        • Ekbom A.
        • et al.
        External review and validation of the Swedish national inpatient register.
        BMC Public Health. 2011; 11: 450
        • Mattsson P.
        • Tomson T.
        • Eriksson O.
        • Brannstrom L.
        • Weitoft G.R.
        Sociodemographic differences in antiepileptic drug prescriptions to adult epilepsy patients.
        Neurology. 2010; 74: 295-301
        • Adelow C.
        • Andell E.
        • Amark P.
        • et al.
        Newly diagnosed single unprovoked seizures and epilepsy in Stockholm, Sweden : First report from the Stockholm Incidence Registry of Epilepsy (SIRE).
        Epilepsia. 2009; 50: 1094-1101
        • Christensen J.
        • Vestergaard M.
        • Pedersen M.G.
        • Pedersen C.B.
        • Olsen J.
        • Sidenius P.
        Incidence and prevalence of epilepsy in Denmark.
        Epilepsy Res. 2007; 76: 60-65
        • Renoux C.
        • Dell'Aniello S.
        • Saarela O.
        • Filion K.B.
        • Boivin J.F.
        Antiepileptic drugs and the risk of ischaemic stroke and myocardial infarction: a population-based cohort study.
        BMJ Open. 2015; 5: e008365
        • Brodie M.J.
        • Mintzer S.
        • Pack A.M.
        • Gidal B.E.
        • Vecht C.J.
        • Schmidt D.
        Enzyme induction with antiepileptic drugs: cause for concern?.
        Epilepsia. 2013; 54: 11-27
        • Olesen J.B.
        • Abildstrom S.Z.
        • Erdal J.
        • et al.
        Effects of epilepsy and selected antiepileptic drugs on risk of myocardial infarction, stroke, and death in patients with or without previous stroke: a nationwide cohort study.
        Pharmacoepidemiol Drug Saf. 2011; 20: 964-971
        • Lei H.
        • Gao Q.
        • Liu S.R.
        • Xu J.
        The benefit and safety of aspirin for primary prevention of ischemic stroke: a meta-Analysis of randomized trials.
        Front Pharmacol. 2016; 7: 440