Abstract
Objective
Cognitive disorder is common after stroke at a young age, especially in patients with poststroke epilepsy (PSE). Whether the causative mechanism is direct (due to epilepsy‐related network alterations) or indirect (due to effect‐modifiers such as stroke severity) is not fully understood. We assessed the role of PSE in cognitive disorder in young stroke patients by investigating the association between vascular cognitive disorder (VCD) and PSE in young stroke patients, and by investigating the association between cognitive impairment per cognitive domain and PSE.
Methods
In this multicenter prospective cohort study, we investigated the occurrence of PSE in patients aged 18–49 years presenting with a first‐ever transient ischemic attack, ischemic stroke, or primary intracerebral hemorrhage between 2013 and 2021 and assessed the cognitive function 1 year after the event. We calculated composite z‐scores for seven cognitive domains. VCD was categorized as a composite z‐score in any domain between −2.0 SD and −1.5 SD (mild) and <−2.0 SD (major). We performed multivariable regression analyses to examine the association between PSE and VCD and between PSE and cognitive impairment per domain.
Results
Eight (median age = 38.6 years, median NIHSS score = 2, 30% male) of 20 patients with PSE (40.0%) had major VCD, compared to 93 (median age = 44.2 years, median NIHSS = 2, 50% male) of 426 patients (20.1%) without PSE (adjusted odds ratio [OR] = 3.96, 95% confidence interval [CI] = 1.24–12.65). Additionally, PSE was independently associated with cognitive impairment in the domain attention and working memory (adjusted OR = 5.09, 95% CI = 1.15–22.59).
Significance
We found independent associations between PSE and major VCD, and between PSE and cognitive impairment in the attention and working memory domain. This could support the “second‐hit hypothesis,” in which epilepsy following a primary injury relates directly to increased cognitive impairment with a subcortical neuropsychological profile.
Keywords: poststroke epilepsy, vascular cognitive disorder, young stroke
Key points.
In our prospective young stroke cohort study, PSE is independently associated with major VCD.
PSE is also independently associated with cognitive impairment in the attention and working memory domain.
This could support the “second‐hit hypothesis,” which assumes epileptogenesis following a primary injury adds to the development of cognitive impairment.
1. INTRODUCTION
Cognitive impairment is a common and serious consequence of stroke, especially at a young age due to its major effects on patients' personal lives and social environment. 1 Patients with poststroke epilepsy (PSE) are more often affected by cognitive impairment than patients without epilepsy after stroke. 2 , 3 This is also the case for young stroke patients. 4 This association could be indirect, confounded by antiseizure medication (ASM), or mediated by stroke severity or the occurrence of poststroke cognitive depression. 5 , 6 , 7 , 8 In addition, the neuropsychological profile of cognitive disorder in late onset epilepsy patients and patients suffering from a stroke at a young age shows overlap; in both patient groups, attention and working memory, processing speed, and episodic memory are affected domains. 1 , 9 This might substantiate such indirect association between PSE and cognitive disorder. On the other hand, there could also be a direct association, as chronic epilepsy in itself can lead to loss of neural network efficiency, which is associated with cognitive disorder and might support an independent association between PSE and cognitive disorder. 10 Hence, it is at present not fully understood whether the association between PSE and cognitive disorder in young stroke patients is direct or indirect.
Therefore, we aimed to investigate the role of PSE in cognitive functioning in young patients with a stroke in a longitudinal follow‐up study on cognition after stroke. We will determine the incidence of vascular cognitive disorder (VCD) after 1 year of follow‐up between young stroke patients with and without PSE and compare cognitive function per cognitive domain after 1 year of follow‐up between these groups, adjusted for baseline differences.
2. MATERIALS AND METHODS
2.1. Participants
This study is part of the Observational Dutch Young Symptomatic Stroke Study (ODYSSEY), a multicenter prospective cohort study on the risk factors and prognosis of patients with a stroke in young adults. 11 In short, we prospectively included patients with a first‐ever transient ischemic attack (TIA) or ischemic stroke with radiological evidence of cerebral ischemia, and patients with a primary intracerebral hemorrhage, aged between 18 and 49 years, presenting between May 2013 and March 2021 in participating hospitals. We excluded patients with epilepsy prior to stroke and patients who died during follow‐up. We also excluded patients who did not have a cognitive assessment at follow‐up or whose results were unreliable. The Medical Review Ethics Committee for the region Arnhem‐Nijmegen approved the study. Further information on data collection and informed consent is reported elsewhere. 1 , 11
2.2. Assessment of PSE
We defined PSE in the case that a single seizure occurred >7 days after the event, as one remote symptomatic seizure is sufficient for the diagnosis of epilepsy according to the present International League Against Epilepsy definition, in contrast to acute symptomatic seizures (occurring within the first 7 days after the event), which are insufficient for such diagnosis due to their lower risk of recurrence. 12 We defined seizures as either focal, with or without impaired awareness, or generalized. We assessed the occurrence of remote symptomatic seizures by contacting patients semiannually by phone or through online survey until the end of follow‐up using a standardized structured questionnaire in Dutch. We then validated all reported seizures for which the patient sought medical attention through medical records from the including hospital or general practitioner. As a cross‐checking measure, we checked the medical records of 64 patients from several hospitals who did not report seizures in the telephonic questionnaire or online survey and found no disparities.
2.3. Cognitive assessments
We administered a neuropsychological assessment 1 year after the index event, which concerned the following cognitive domains: episodic memory (test: the three‐trial version of the Rey Auditory Verbal Learning Test), processing speed (tests: written version of the Symbol‐Digit Modalities Test, abbreviated Stroop Color Word Test parts I and II), visuoconstruction (test: Rey–Osterrieth Complex Figure copy trial), executive functioning (tests: Fluency Test, Stroop Interference Score, Brixton Spatial Anticipation Test), visual neglect (test: Star Cancellation subtest of the Behavioral Inattention Test), language deficits (Short Token Test), and attention and working memory (Digit Span subtest from the Wechsler adult Intelligence Scale–Fourth Edition). 1 We converted raw test scores into z‐scores per test for each participant based on normative data from the Advanced Neuropsychological Diagnostics Infrastructure, which includes data from up to 26 000 healthy individuals from all ages, for a majority of the tests, or from other control studies for the rest. 1 , 13 We averaged z‐scores of neuropsychological tests that assessed the same cognitive domain to a composite z‐score per cognitive domain. We defined cognitive impairment on a specific cognitive domain as a z‐score below −1.5. We categorized VCD as mild VCD (composite z‐score between −1.5 and −2.0 in any cognitive domain) and major VCD (composite z‐score below −2.0 in any cognitive domain) in accordance with previous studies from the ODYSSEY cohort. 1 , 14 Because almost all of the patients had a (near) ceiling score on the test of visual neglect, these skewed results are not reported as composite z‐scores. Rather, an absolute cutoff score of <44 on the Star Cancellation test was used for cognitive impairment. 1
2.4. Additional assessments
We dichotomized the level of education as a higher level of education (a level of 6 or higher in the Dutch educational system, which ranges from 1 to 7) and a lower level of education (a level of 5 or lower in the Dutch system). 15 For stroke severity, we used the National Institutes of Health Stroke Scale (NIHSS) at admission. We assessed the presence of depressive symptoms and fatigue using the Mini International Neuropsychiatric Interview and the subscale Subjective Fatigue of the revised Checklist Individual Strength, respectively. 16 , 17 Risk factors for stroke were assessed during admission. We assessed hypertension by either antihypertensive use prior to event, a systolic pressure above 140 mmHg, or a diastolic pressure above 90 mmHg at least 24 h after the event. We assessed dyslipidemia by a prior history of dyslipidemia, statin use prior to event, or a total cholesterol value above 5.0 mmol·L−1 and/or low‐density lipoprotein cholesterol above 3.0 mmol·L−1 and/or high‐density lipoprotein cholesterol below 1.0 mmol·L−1. We assessed diabetes mellitus by a prior history of diabetes, by the use of antidiabetic medication prior to event, or by measuring either twice a fasting glucose level above 7.0 mmol·L−1 or once Hba1c above 48 mmol·L−1 following the event. We assessed smoking by the use of at least one cigarette 1 year prior to the event. We assessed alcohol abuse by the use of more than 20 units of alcohol per week. Cortical involvement of stroke was assessed either on imaging (computed tomography or magnetic resonance imaging) or on clinical symptoms (e.g., neglect, hemianopia, aphasia).
2.5. Statistical analysis
We used IBM SPSS Statistics 25 to perform all statistical analyses. We reported baseline characteristics as mean with SD, as median with interquartile range in the case of nonnormal distribution, or as frequency and percentage. Baseline characteristics were compared between patients with and without PSE using the independent samples t‐test (or Mann–Whitney U‐test in the case of nonnormal distribution) for continuous variables and Pearson chi‐squared test (or Fisher exact test if appropriate) for categorical and dichotomous variables. The occurrence of mild VCD and of major VCD after 1 year of follow‐up between patients with and without PSE was compared using Pearson chi‐squared test (or Fisher exact test if appropriate). We used logistic regression to examine the association between PSE and mild VCD or major VCD (depending on the results of the chi‐squared test) corrected for baseline differences in severity of stroke, depression, level of education, age, and fatigue. In an exploratory analysis, we analyzed which domain was most affected by PSE by performing a linear regression analysis to investigate the association between PSE and the composite z‐score per domain, corrected for baseline variables. Subsequently, we performed a logistic regression analysis to investigate the association between PSE and cognitive impairment per domain corrected for baseline variables. We performed sensitivity analyses by individually adding stroke recurrence, cortical involvement, and stroke subtype to the primary models. To account for a low event‐per‐variable ratio, we performed a Firth penalized logistic regression analysis. To adjust for attrition bias, we performed inverse probability of censoring weighting (IPCW) in our primary analysis. As a robustness check, we performed an overlap weighted analysis on a propensity score for PSE. We grouped ischemic stroke and TIA patients together because of the same underlying pathophysiological mechanism and the insignificant chances of misdiagnosing TIA due to the presence of radiological ischemia as inclusion criterium. We categorized NIHSS into three categories (<8, 8–14, and >14) in the regression analyses because of a nonlinear distribution. 18 Due to a relatively low amount of missing data, imputation was not performed.
3. RESULTS
3.1. Group characteristics
We included 1492 patients with a first‐ever TIA, ischemic stroke, or primary intracerebral hemorrhage. We excluded patients with epilepsy prior to stroke (n = 30) and patients who died during follow‐up (n = 189). We also excluded patients in whom no cognitive assessment was administered at follow‐up (n = 778) or whose results were invalid (n = 13; language barrier n = 4, unwillingness to participate n = 4, unknown n = 5). After exclusion criteria were met, 482 patients remained with a total of 1228.8 person‐years of follow‐up (Figure 1); 20 of these had PSE, seven of whom (35.0%) were diagnosed with PSE after cognitive assessment was performed, mostly (4, 20.0%) within 1 year after cognitive assessment. Seventeen of 20 (85%) PSE patients used ASM (sodium valproate n = 3, levetiracetam n = 12, lamotrigine n = 1, unknown n = 1), of whom 13 used during cognitive testing. Baseline characteristics are shown in Table 1. There were no significant baseline differences between patients with PSE and without PSE, besides the PSE groups having more intracerebral hemorrhages and more cortical involvement of stroke. Baseline characteristics of patients without cognitive assessment are shown in Table S1.
FIGURE 1.
Study population flowchart. TIA, transient ischemic attack.
TABLE 1.
Baseline characteristics.
| Characteristic | No PSE, n = 462 | PSE, n = 20 | p |
|---|---|---|---|
| Median age, years (IQR) | 44.2 (37.7–47.3) | 38.6 (29.8–46.4) | .057 |
| Gender, male | 231 (50.0%) | 6 (30.0%) | .080 |
| Hypertension prior to event | 178 (38.5%) | 4 (20.0%) | .094 |
| Dyslipidemia prior to event | 277 (60.0%) | 8 (40.0%) | .075 |
| Diabetes prior to event | 33 (7.1%) | 1 (5.0%) | 1.000 |
| Smoking prior to event | 193 (41.8%) | 7 (35.0%) | .547 |
| Alcohol abuse prior to event | 21 (4.5%) | 2 (10.0%) | .322 |
| Stroke type | .008 | ||
| TIA/ischemic stroke | 421 (91.1%) | 14 (70.0%) | |
| Intracerebral hemorrhage | 41 (8.9%) | 6 (30.0%) | |
| Stroke recurrence prior to cognitive assessment | 25 (5.4%) | 2 (10%) | .310 |
| Median NIHSS at admission (IQR) | 2 (1–5) | 2 (1–8) | .608 |
| NIHSS, categorized | .196 | ||
| NIHSS < 8 | 394 (85.3%) | 14 (70.0%) | |
| NIHSS = 8–14 | 39 (8.4%) | 2 (10.0%) | |
| NIHSS > 14 | 27 (5.8%) | 3 (15.0%) | |
| Missing | 2 (.4%) | 1 (5.0%) | |
| Cortical involvement of stroke | 200 (43.5%) | 16 (80%) | .002 |
| Missing | 2 (.4%) | 0 (.0%) | |
| Median days until cognitive assessment (IQR) | 509 (441–579) | 458 (405–644) | .239 |
| Higher level of education | 182 (39.4%) | 7 (35.0%) | .443 |
| Mean fatigue subscore on CIS20‐R at follow‐up (SD) | 28.5 (12.4) | 33.2 (12.4) | .135 |
| Depressive symptoms at follow‐up (%) | 31 (6.7) | 2 (10.0) | .636 |
| Missing | 14 (3.0%) | 1 (5.0%) |
Note: Significant associations are shown in bold. Values represent n (%) unless otherwise indicated.
Abbreviations: CIS20‐R, Checklist Individual Strength–Revised; IQR, interquartile range; NIHSS, National Institutes of Health Stroke Scale; PSE, poststroke epilepsy; TIA, transient ischemic attack.
3.2. VCD at 1 year
Of the 482 patients who underwent cognitive assessment after 1 year, 144 (29.9%) had mild VCD and 101 (21.0%) had major VCD. Patients with PSE were more affected by major VCD (8/20, 40.0%) than patients without PSE (93/462, 20.1%; p = .018; see Table 2).
TABLE 2.
Cross‐tabulation between VCD and PSE.
| No VCD, n (%) | Mild VCD, n (%) | Major VCD, n (%) | |
|---|---|---|---|
| No PSE | 226 (48.9) | 143 (31.0) | 93 (20.1) |
| PSE | 11 (55.0) | 1 (5.0) | 8 (40.0) |
| Total | 237 (49.2) | 144 (29.9) | 101 (21.0) |
Note: Chi‐squared test, p = .018.
PSE was associated with major VCD (adjusted odds ratio = 3.96, 95% confidence interval = 1.24–12.65), adjusted for stroke severity, age, level of education, fatigue, and depressive symptoms. A higher NIHSS and a lower level of education were also independently associated with major VCD. Multiple logistic regression models are shown in Table 3. Results after penalized regression analysis, IPCW analysis, and overlap weighting remained similar and are shown in Table S2. In overlap weighted analysis, effective sample size was 54.5, with a standardized mean difference postweighting of all covariates below .1. Sensitivity analyses also remained similar and are shown in Table S3.
TABLE 3.
Logistic regression analysis for the occurrence of major VCD.
| OR | 95% CI | p | |
|---|---|---|---|
| PSE | 3.96 | 1.24–12.65 | .021 |
| Age | 1.03 | 1.00–1,07 | .051 |
| High level of education | .51 | .30–.88 | .016 |
| Stroke severity (NIHSS) | |||
| <7 | 1 (ref) | N/A | N/A |
| 8–14 | 2.25 | 1.04–4.89 | .040 |
| >14 | 3.07 | 1.25–7.55 | .015 |
| Fatigue subscore of CIS20‐R questionnaire | 1.01 | .99–1.03 | .421 |
| Depression | 2.00 | .78–5.15 | .152 |
Note: Significant associations are shown in bold.
Abbreviations: CI, confidence interval; CIS20‐R, Checklist Individual Strength–Revised; N/A, not applicable; NIHSS, National Institutes of Health Stroke Scale; OR, odds ratio; PSE, poststroke epilepsy; ref, reference.
3.3. Cognitive function per domain at follow‐up
There were no differences in composite z‐score at follow‐up per domain between patients with PSE and patients without PSE (Table S4). In contrast, PSE was independently associated with cognitive impairment (composite z‐score < −1.5 SD) in the domains visual neglect and attention and working memory (Table 4). Penalized regression, IPCW analysis, and overlap weighting showed similar results (Table S2). Sensitivity analyses also showed similar results, except for the sensitivity analysis that included cortical involvement, in which no significant association was found between PSE and cognitive impairment in the attention and working memory domain (Table S3).
TABLE 4.
Risk of cognitive impairment (z‐score < 1.5 SD) on the different cognitive domains for PSE patients.
| Domain | OR (PSE) | 95% CI | p |
|---|---|---|---|
| Episodic memory | 3.29 | .89–12.10 | .074 |
| Processing speed | 1.65 | .48–5.64 | .426 |
| Visuoconstruction | .48 | .13–1.77 | .269 |
| Executive functioning | 3.76 | .76–18.54 | .103 |
| Visual neglect a | 4.83 | 1.14–20.53 | .033 |
| Language deficit | 1.61 | .32–7.99 | .561 |
| Attention and working memory | 5.09 | 1.15–22.59 | .032 |
Note: Significant associations are shown in bold.
Abbreviations: CI, confidence interval; OR, odds ratio; PSE, poststroke epilepsy.
An absolute threshold of <44 stars on the Star Cancellation test was chosen for cognitive impairment.
4. DISCUSSION
We investigated the role of PSE on cognitive functioning in young stroke patients and found PSE was associated with major VCD, independent of stroke severity, depressive symptoms, level of education, age, and fatigue. Furthermore, PSE was associated with cognitive impairment in the attention and working memory and visual neglect domains.
Our study provides new evidence for the association between PSE and cognitive impairment in young stroke patients. Previous studies investigating an association between PSE and cognitive disorder found similar results but consisted of smaller study populations and had retrospective designs. 2 , 3 Additionally, a recent meta‐analysis, which included mostly retrospective studies, showed an association between PSE and dementia (which overlaps with our main outcome parameter of major VCD). 19 In young stroke patients specifically, an independent relation between PSE and dementia after several years of follow‐up has also been demonstrated in a large retrospective cohort study. 4 Additionally, an association between PSE and cognitive impairment in the processing speed domain has been shown in a prospective cohort study of young stroke patients, but there were no associations between PSE and VCD (only mild VCD was assessed), or between PSE and cognitive impairment in the attention and working memory domain. 20 In the same study population, PSE was independently associated with a worse functional outcome, but it was not investigated whether this is due to VCD. 21 One possible explanation for the association between PSE and cognitive disorder is the concept of a “second‐hit model.” This model assumes that the development of epilepsy (epileptogenesis) in itself adds to the development of cognitive impairment in patients with epilepsy secondary to a primary injury, possibly with a different, more subcortical, neuropsychological profile compared to cognitive impairment in other types of epilepsy. 9 , 22 Alterations in neuronal networks as a causative mechanism underlying both cognitive disorder and epileptogenesis is a plausible pathophysiological explanation for the second hit hypothesis, as neuronal injury leads to astrocyte activation, which leads to increased neural network hyperexcitability, which subsequently leads to epilepsy, but is also associated with cognitive disorder. 20 , 23 , 24 Further research should investigate whether the associations found between VCD and PSE, as well as the possible associations between cognitive impairment in subcortical domains and PSE, substantiate the second‐hit model.
Our study has some limitations. First, there is most likely a selection bias; more than half of the patients were excluded because cognitive assessment at follow‐up was unavailable due to either unreliable test results or inability or refusal to undergo the test. Those excluded had a significantly more severe stroke than the study population, which affects the generalizability of our results. Selection bias due to attrition leading to skewed results is a known issue in longitudinal studies related to cognition. 25 This is expected to lead to an underestimation of the association, however, as is demonstrated by our weighted analyses. Second, the time of neuropsychological evaluation at follow‐up varies in our sample, which might have led to an overestimation of the effect size, as cognitive impairment can change over time. 1 Third, some patients with PSE were diagnosed after cognitive assessment, raising reverse temporality concerns. Because cognitive disorder may precede a diagnosis of epilepsy, possibly as the result of epileptogenesis, we believe retaining these patients in our analysis is justified. 26 Additionally, the Star Cancellation test used for the visual neglect domain was performed very well overall, which hampers the clinical relevance of the association we found between cognitive impairment in this domain and PSE. Furthermore, the sample size of 20 patients with PSE who underwent neurocognitive assessment is small, which may affect the reliability of the estimation of the effect on cognitive function and cognitive impairment in the various cognitive domains, although both penalized regression models and overlap weighted regression analysis showed similar results. This small sample size also limited the factors we adjusted for in the regression models, although adjustment was performed for possible other confounders through sensitivity analyses. Next, we were unable to adjust for lesion volume as potential confounder due to a high rate of missing data. Furthermore, ASMs were frequently prescribed in patients with PSE, which may affect cognitive performance in PSE patients only and is therefore a serious potential confounder. 27 No adjustment for ASM use in the primary regression model was feasible due to collinearity. Other means to adjust for ASM use, such as inverse probability of treatment weighting or overlap weighting, was not possible due to our very small sample size. 28 Next, we measured stroke severity using NIHSS at admission instead of discharge due to better data availability at admission. Consequently, the potential effects of reperfusion therapies will not be captured. Nevertheless, we adjusted for the most important possible confounders in the regression models and sensitivity analyses. Finally, although the primary association between PSE and major VCD remained robust, findings regarding individual cognitive domains were less consistent following sensitivity and weighted analyses, which attenuates the subcortical profile observed in the primary analysis. Despite these limitations, the strengths of our study remain the large and multicenter prospective study cohort of well‐characterized young stroke patients in which both PSE and cognitive impairment were assessed as secondary outcome measures. Furthermore, all stroke diagnoses were proven by computed tomography or magnetic resonance imaging, minimizing the risk of including patients with a different diagnosis. Additionally, extensive neuropsychological evaluation was performed, which was compared with normative data from a large group of healthy controls.
In clinical practice, the results of this study may be used by health care professionals to inform patients about future perspectives regarding their cognition. Future longitudinal studies should investigate whether the cognitive impairment of young stroke patients with PSE affects their functional ability, such as societal participation. Additionally, further research is needed to confirm our hypothesis that network alterations due to stroke are the cause of the association we found between PSE and VCD in young stroke patients.
5. CONCLUSIONS
In conclusion, we showed that PSE is independently associated with a major VCD and possibly associated with cognitive impairment in the attention and working memory domain in patients with an ischemic or hemorrhagic stroke at a young age. These findings could substantiate the “second‐hit hypothesis” of cognitive impairment in adult onset epilepsy patients in a prospective cohort, with network alterations as a consequence of the stroke as a possible causative mechanism, although further research is necessary to further support these findings.
AUTHOR CONTRIBUTIONS
Frederik J. Reitsma researched literature, did data analysis, and wrote the first draft of the manuscript. Renate M. Arntz, Mayte E. van Alebeek, Frank‐Erik de Leeuw, and Roy P. C. Kessels were involved in protocol development. Frank‐Erik de Leeuw acquired ethical approval. All authors were involved in patient recruitment. All authors reviewed and edited the manuscript and approved the final version of the manuscript. Marieke J. H. Wermer has moved to a new institution since completing the research: Department of Neurology, University Medical Center Groningen, Groningen, the Netherlands.
FUNDING INFORMATION
This study was supported by the Radboud University Medical Center, Dutch Heart Foundation (grant 2016T86 [M.J.H.W.], grant 03‐005‐2022‐0031 [N.A.H.], grant 2016 T044 [A.M.T.], and grant 2014 T060 [F.‐E.d.L.]), and Vidi (grant 9171337 of the ZorgOnderzoek Nederland and the Medical Sciences/Nederlandse Organisatie voor Wetenschappelijk Onderzoek [M.J.H.W.] and grant 016‐126‐351 [F.‐E.d.L.]). The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
CONFLICT OF INTEREST STATEMENT
I.R.v.d.W. reports receiving personal fees from Philips and Medtronic outside of the submitted work. The remaining authors report no conflict of interest.
ETHICS STATEMENT
The Medical Review Ethics Committee of the region Arnhem‐Nijmegen approved the study. All participants have given written informed consent for this study. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
GUARANTOR
Frank‐Erik de Leeuw.
Supporting information
Table S1.
ACKNOWLEDGMENTS
N.A.H. received grant support from the Dutch Heart Foundation (grant number 03‐005‐2022‐0031). A.M.T. is a junior staff member of the Dutch Heart Foundation (grant number 2016T044). F.‐E.d.L. has received support as a Clinical Established Investigator of the Dutch Heart Foundation (2014T060).
Reitsma FJ, Verburgt E, Schellekens MMI, Hilkens NA, Verhoeven JI, van Alebeek ME, et al. Poststroke epilepsy is associated with vascular cognitive disorder in young stroke patients: The ODYSSEY study. Epilepsia. 2026;67:2855–2863. 10.1002/epi.70200
Description for social media: Poststroke epilepsy is independently associated with major vascular cognitive disorder in young stroke patients in a prospective multicenter cohort study.
DATA AVAILABILITY STATEMENT
With publication, collected deidentified participant data can be made available upon reasonable request, after approval of regulatory and medical ethics bodies.
REFERENCES
- 1. Schellekens MM, Boot EM, Verhoeven JI, Ekker MS, van Alebeek ME, Brouwers PJ, et al. Subacute cognitive impairment after first‐ever transient ischemic attack or ischemic stroke in young adults: the ODYSSEY study. Eur Stroke J. 2023;8(1):283–293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. van Tuijl JH, van Raak EPM, van Oostenbrugge RJ, Aldenkamp AP, Rouhl RPW. Cognition and quality of life in patients with poststroke epilepsy: a case‐control study. Epilepsy Behav. 2020;104(Pt B):106444. [DOI] [PubMed] [Google Scholar]
- 3. De Reuck J, De Clerck M, Van Maele G. Vascular cognitive impairment in patients with late‐onset seizures after an ischemic stroke. Clin Neurol Neurosurg. 2006;108(7):632–637. [DOI] [PubMed] [Google Scholar]
- 4. Lekoubou A, Ba DM, Nguyen C, Liu G, Leslie DL, Bonilha L, et al. Poststroke seizures and the risk of dementia among Young stroke survivors. Neurology. 2022;99(4):e385–e392. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Millan MJ, Agid Y, Brune M, Bullmore ET, Carter CS, Clayton NS, et al. Cognitive dysfunction in psychiatric disorders: characteristics, causes and the quest for improved therapy. Nat Rev Drug Discov. 2012;11(2):141–168. [DOI] [PubMed] [Google Scholar]
- 6. Kanner AM. Depression in epilepsy: prevalence, clinical semiology, pathogenic mechanisms, and treatment. Biol Psychiatry. 2003;54(3):388–398. [DOI] [PubMed] [Google Scholar]
- 7. Galovic M, Dohler N, Erdelyi‐Canavese B, Felbecker A, Siebel P, Conrad J, et al. Prediction of late seizures after ischaemic stroke with a novel prognostic model (the SeLECT score): a multivariable prediction model development and validation study. Lancet Neurol. 2018;17(2):143–152. [DOI] [PubMed] [Google Scholar]
- 8. Pendlebury ST, Rothwell PM. Prevalence, incidence, and factors associated with pre‐stroke and post‐stroke dementia: a systematic review and meta‐analysis. Lancet Neurol. 2009;8(11):1006–1018. [DOI] [PubMed] [Google Scholar]
- 9. Sen A, Capelli V, Husain M. Cognition and dementia in older patients with epilepsy. Brain. 2018;141(6):1592–1608. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Vlooswijk MC, Vaessen MJ, Jansen JF, de Krom MC, Majoie HJ, Hofman PA, et al. Loss of network efficiency associated with cognitive decline in chronic epilepsy. Neurology. 2011;77(10):938–944. [DOI] [PubMed] [Google Scholar]
- 11. Arntz RM, van Alebeek ME, Synhaeve NE, Brouwers PJ, van Dijk GW, Gons RA, et al. Observational Dutch Young symptomatic StrokE studY (ODYSSEY): study rationale and protocol of a multicentre prospective cohort study. BMC Neurol. 2014;14:55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Beghi E, Carpio A, Forsgren L, Hesdorffer DC, Malmgren K, Sander JW, et al. Recommendation for a definition of acute symptomatic seizure. Epilepsia. 2010;51(4):671–675. [DOI] [PubMed] [Google Scholar]
- 13. de Vent NR, van Agelink Rentergem JA, Schmand BA, Murre JM, Consortium A , Huizenga HM. Advanced neuropsychological diagnostics infrastructure (ANDI): a normative database created from control datasets. Front Psychol. 2016;7:1601. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Sachdev P, Kalaria R, O'Brien J, Skoog I, Alladi S, Black SE, et al. Diagnostic criteria for vascular cognitive disorders: a VASCOG statement. Alzheimer Dis Assoc Disord. 2014;28(3):206–218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Verhage F. Intelligence and age: research among Dutch people aged twelve to seventy‐seven years. Assen: Van Gorcum; 1964. [Google Scholar]
- 16. Sheehan DV, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, et al. The Mini‐international neuropsychiatric interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM‐IV and ICD‐10. J Clin Psychiatry. 1998;59(Suppl 20):22–33. quiz 4–57. [PubMed] [Google Scholar]
- 17. Vercoulen JH, Swanink CM, Fennis JF, Galama JM, van der Meer JW, Bleijenberg G. Dimensional assessment of chronic fatigue syndrome. J Psychosom Res. 1994;38(5):383–392. [DOI] [PubMed] [Google Scholar]
- 18. Shuaib A, Schwab S, Rutledge JN, Starkman S, Liebeskind DS, Bernardini GL, et al. Importance of proper patient selection and endpoint selection in evaluation of new therapies in acute stroke: further analysis of the SENTIS trial. J Neurointerv Surg. 2013;5(Suppl 1):i21–i24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Misra S, Kasner SE, Dawson J, Tanaka T, Zhao Y, Zaveri HP, et al. Outcomes in patients with poststroke seizures: a systematic review and meta‐analysis. JAMA Neurol. 2023;80(11):1155–1165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Boot EM, Omes QPM, Maaijwee N, Schaapsmeerders P, Arntz RM, Rutten‐Jacobs LCA, et al. Functional brain connectivity in young adults with post‐stroke epilepsy. Brain Commun. 2023;5(6):fcad277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Arntz RM, Maaijwee NA, Rutten‐Jacobs LC, Schoonderwaldt HC, Dorresteijn LD, van Dijk EJ, et al. Epilepsy after TIA or stroke in young patients impairs long‐term functional outcome: the FUTURE study. Neurology. 2013;81(22):1907–1913. [DOI] [PubMed] [Google Scholar]
- 22. Breuer LEM, Grevers E, Boon P, Bernas A, Bergmans JWM, Besseling RMH, et al. Cognitive deterioration in adult epilepsy: clinical characteristics of “accelerated cognitive ageing”. Acta Neurol Scand. 2017;136(1):47–53. [DOI] [PubMed] [Google Scholar]
- 23. Holmes GL. Cognitive impairment in epilepsy: the role of network abnormalities. Epileptic Disord. 2015;17(2):101–116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Vezzani A, Pascente R, Ravizza T. Biomarkers of Epileptogenesis: the focus on glia and cognitive dysfunctions. Neurochem Res. 2017;42(7):2089–2098. [DOI] [PubMed] [Google Scholar]
- 25. Matthews FE, Chatfield M, Freeman C, McCracken C, Brayne C, Mrc C. Attrition and bias in the MRC cognitive function and ageing study: an epidemiological investigation. BMC Public Health. 2004;4:12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Witt JA, Helmstaedter C. Cognition in the early stages of adult epilepsy. Seizure. 2015;26:65–68. [DOI] [PubMed] [Google Scholar]
- 27. Kosmider K, Kamieniak M, Czuczwar SJ, Miziak B. Second generation of antiepileptic drugs and oxidative stress. Int J Mol Sci. 2023;24(4):3873. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Raad H, Cornelius V, Chan S, Williamson E, Cro S. An evaluation of inverse probability weighting using the propensity score for baseline covariate adjustment in smaller population randomised controlled trials with a continuous outcome. BMC Med Res Methodol. 2020;20(1):70. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1.
Data Availability Statement
With publication, collected deidentified participant data can be made available upon reasonable request, after approval of regulatory and medical ethics bodies.