Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Characterizing inflammatory biomarkers in post-stroke seizure risk and outcome prognostication

  • Ethan Y. Wang,

    Roles Data curation, Formal analysis, Methodology, Software, Writing – original draft, Writing – review & editing

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  • Shubham Misra,

    Roles Data curation, Formal analysis, Software, Writing – original draft, Writing – review & editing

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  • Jennifer Yan,

    Roles Data curation, Writing – review & editing

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  • Pei Yi Chook,

    Roles Writing – review & editing

    Affiliation Universiti Malaya, Kuala Lumpur, Malaysia,

    ⨯
  • Yuki Kawamura,

    Roles Writing – review & editing

    Affiliation University of Cambridge, Cambridge, United Kingdom

    ⨯
  • Rachel Kitagawa,

    Roles Writing – review & editing

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  • Jennifer A. Kim,

    Roles Writing – original draft, Writing – review & editing

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  • Emily J. Gilmore,

    Roles Writing – review & editing

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  • Adam de Havenon,

    Roles Writing – review & editing

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  • Adithya Sivaraju,

    Roles Writing – review & editing

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  • Lawrence J. Hirsch,

    Roles Writing – review & editing

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  • Guido J. Falcone,

    Roles Writing – review & editing

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  • Srikant Rangaraju,

    Roles Writing – review & editing

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  • Lauren H. Sansing,

    Roles Writing – review & editing

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  • Jessica Magid-Bernstein,

    Roles Data curation, Writing – review & editing

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  •  [ ... ],
  • Nishant K. Mishra

    Roles Conceptualization, Investigation, Methodology, Supervision, Writing – original draft, Writing – review & editing

    nishant.mishra@yale.edu

    Affiliation Department of Neurology, Yale School of Medicine, New Haven, Connecticut, United States of America

    ⨯
  • [ view all ]
  • [ view less ]

Abstract

Post-stroke seizures (PSS) are sequelae of intracerebral hemorrhage (ICH) that may negatively impact patient outcomes. Available literature suggests that inflammatory biomarkers contribute to epileptogenesis as well as mortality. This retrospective cohort study aimed to elucidate the prognostic ability of CCL2, IL6, and IL8 in PSS development. ICH data were collected at Yale New Haven Hospital from 2014 to 2021. Plasma biomarker levels were measured via cytometric bead array. Patients with EEG-recorded epileptiform discharges, EEG-recorded seizures, or clinical seizures after ICH were defined as seizures and epileptiform discharges (SED), while those without were defined as non-SED. SED was further divided into early and late, occurring before and 7 days post-ICH, respectively. Additionally, we examined if biomarkers were associated with poor outcome (modified Rankin Scale score of 3–6) and mortality at 90, 180, and 365 days post-ICH. We conducted univariable and multivariable logistic regression analyses and reported the findings as Odds Ratio (OR) and 95% CI. We examined 172 patients with ICH, of whom 33 had early SED, 29 had late SED, and 110 had no SED. In univariable analyses, CCL2, ICH volume, diabetes, and lobar ICH were significantly associated with late SED, whereas NIHSS score at admission, ICH volume, and lobar ICH were significantly associated with early SED (p < 0.05). Lower CCL2 levels were independent predictors of late SED in multivariable analyses (OR 0.58; 95% CI 0.41–0.80, p < 0.001), but not of early SED. Additionally, we identified an independent association between higher CCL2 levels and 90-day mortality in multivariable analysis for the combined early and late SED cohorts (OR 1.87; 95% CI 1.03–3.37, p = 0.038). Additional studies investigating additional aspects of biomarkers, such as their temporal profile post-stroke within 24 hours or beyond 72 hours, are needed.

Citation: Wang EY, Misra S, Yan J, Chook PY, Kawamura Y, Kitagawa R, et al. (2026) Characterizing inflammatory biomarkers in post-stroke seizure risk and outcome prognostication. PLoS One 21(3): e0345752. https://doi.org/10.1371/journal.pone.0345752

Editor: Giuseppe Biagini, University of Modena and Reggio Emilia, ITALY

Received: December 29, 2024; Accepted: March 9, 2026; Published: March 31, 2026

Copyright: © 2026 Wang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Data cannot be shared publicly because it is confidential patient data. Data are available from the Yale School of Medicine Department of Neurology (contact via victor.torreslopez@yale.edu, Data Manager/Data Analyst for Clinical, Neuroimaging, and Genetic Data) for researchers who meet the criteria for access to confidential data.

Funding: The author(s) received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Seizures are potential complications in patients who experience intracerebral hemorrhages (ICH), which can adversely affect their health outcomes [1,2]. These seizures, or post-stroke seizures (PSS), are divided into early PSS, which occur within seven days of stroke occurrence, and late PSS, which occur after seven days [3,4]. While their mechanisms likely exist within a shared spectrum [3,5], early seizures are understood to mainly result from acute inflammation due to insult, whereas late seizures likely result from longer-lasting tissue disruption, which can have chronic consequences like post-stroke epilepsy (PSE) [6]. Regardless of the timing of PSS, there is evidence of a high mortality risk by ten years [7].

Due to the disparate outcomes in PSS patients, there is a need to enhance our ability to identify patients at risk of PSS and to elucidate underlying biological mechanisms. This is especially true for late seizures, as they may signal chronic illness and morbidity such as PSE. Besides those with overt PSS, there is also evidence that patients with EEG-captured epileptiform discharges have an increased risk of PSS in ischemic and hemorrhagic stroke, possibly reflecting increased cortical hyperexcitability contributing to epileptogenesis [8–11]. By identifying risk factors leading to seizures and epileptiform discharges (SED) post-stroke, we can better identify the patient population that could benefit from potential therapeutics and other interventions.

Existing risk factors for PSS development have been reported in this pursuit. The most well-known are the cortical location of the lesion, younger patient age, hematoma size, and presence of early seizures [3,12,13]. However, additional factors may also predispose patients to experiencing PSS. Recent research has pointed to biomarkers, such as interleukin (IL)-6 and IL-1β, as possible predictors [14]. This is likely due to their involvement in inflammatory responses, alteration of the blood-brain barrier, and other mechanisms of seizure initiation and propagation in post-stroke patients [15–17].

Objectives

We conducted this retrospective cohort study to:

  1. Determine whether inflammatory plasma biomarkers are independently associated with SED development, with a focus on late SED.
  2. Determine whether inflammatory plasma biomarkers are predictors of poor functional outcome or mortality in late SED patients at 90, 180, or 365 days post-stroke.

Methods

Study sample

This was a retrospective study of data of ICH patients admitted to the Yale New Haven Hospital system from 2014 to 2021. Data were accessed on 18th July 2023, and all data were anonymized before access. Patients were included if they were over the age of 18 years, had a diagnosis of ICH confirmed by neuroimaging, and had biomarker data collected during the first 24–72 hours of admission. Biomarkers included monocyte chemoattractant protein-1 (CCL2), IL6, and CXCL8 (IL8). We excluded patients with ischemic stroke/transient ischemic attack (TIA) and patients with a history of cocaine use, seizures, or epilepsy before ICH.

Patients were classified as SED if they had a documented history of clinical seizures, EEG-recorded seizures, or EEG-recorded epileptiform discharges during and after hospitalization. These SED patients were further classified as early or late. Early is defined as SED occurring within 7 days after ICH, while late is defined as 7 or more days after ICH. Those who did not have these findings were deemed non-SED. Early and late SED patients were also divided based on poor functional outcomes defined as modified Rankin Scale (mRS) scores of 3–6 and mortality at 90-, 180-, and 365-day follow-up. Consent was collected verbally or through writing, which was documented in the patient’s electronic medical record.

Sample collection and biomarker measurement

Plasma samples were collected and analyzed from ICH patients within 24 and 72 hours of hospital admission from 15th August 2014–28th July 2021. The biomarker levels in plasma were measured using a cytometric bead array at 24 hours or 72 hours after admission [18]. For patients with both 24-hour and 72-hour biomarker data (38/172), biomarker values were averaged, as values were not significantly different between timepoints (see Table S1 in S1 File). The overall proportion of patients with samples collected at 24 hours, 72 hours, or both is included in Table S2 in S1 File.

Data collection

Patient characteristics were collected through their electronic health records, including demographics (age, gender, race, body mass index), vascular risk factors (hypertension, diabetes, hyperlipidemia, atrial fibrillation, coagulopathy, coronary artery disease, prior TIA or myocardial infarction), clinical scores (National Institutes of Health Stroke Scale (NIHSS), ICH score), radiological parameters (ICH volume, ICH type), and inflammatory biomarker levels. Functional outcomes and mortality at 90-, 180-, and 365-day follow-up were also collected.

Outcomes

Our primary outcome was the prognostic value of inflammatory biomarkers in predicting early and late SED development. Our secondary outcome was the prognostic value of inflammatory biomarkers in predicting poor functional outcome and mortality in early, late, and early + late SED patients at 90-, 180- and 365-day follow-up.

Ethics approval

The plasma samples were collected under IRB 1405014045 and analyzed under IRB 1506016023, both through approval by the Institutional Review Board of Yale University.

Statistical analysis

The data normality was checked using the Shapiro-Wilk test. Biomarker values were log-transformed. Biomarkers with ≤50% values below the limit of detection were imputed from a normal distribution using Perseus 2.0.11. We conducted univariable analysis for early and late SED using Mann-Whitney U tests for continuous variables except the log-transformed biomarkers, which were analyzed via independent-sample t-tests, and the Pearson chi-square test for categorical variables.

Variables significant in univariable analysis (p < 0.05) were included in a backward stepwise multivariable logistic regression analysis to assess the independent association of biomarkers with early or late SED. We considered p-values <0.05 as statistically significant in the multivariable analyses. We report Odds Ratio (OR) and 95% Confidence Intervals (CI). Additionally, we conducted univariable (p < 0.1) and multivariable (p < 0.05) logistic regression analyses to examine the association of biomarkers with poor functional outcomes and mortality in SED groups at 90, 180, and 365 days. Analyses were done using SPSS (Version 28.0, IBM Corp, Armonk, NY) and STATA (Version 18, Stata Corp, College Station, TX).

Results

We included 172 ICH patients, 33 with early SED, 29 with late SED, and 110 with non-SED. Baseline characteristics of all groups with p-values of univariable analyses are shown in Table 1. All biomarkers had > 50% of values that were above the limit of detection before imputation during plasma sample analysis: 94% of CCL2 values, 60% of IL6 values, and 74% IL8 values were above the limit of detection. We classified patients into two groups based on SED subtypes: early SED (N = 33) vs. late + non-SED (N = 139) and late SED (N = 29) vs. non-SED (N = 110) to identify predictors of early and late SED, respectively.

thumbnail
Download:
Table 1. Baseline Characteristics of ICH patients included in the study.

https://doi.org/10.1371/journal.pone.0345752.t001

Predictors of early SED: In the early SED group, NIHSS score at admission, ICH volume, and lobar ICH were significant in the univariable analysis (p < 0.05), whereas no biomarkers were associated with early SED (Table 1). Lobar ICH (OR 3.50; 95% CI 1.54–7.98) and NIHSS score at admission (OR 1.08; 95% CI 1.03–1.13) were independent predictors of early SED compared to late + non-SED in the multivariable analysis (Table 2).

thumbnail
Download:
Table 2. Multivariable prediction models for prognosticating Early SED and Late SED.

https://doi.org/10.1371/journal.pone.0345752.t002

Predictors of late SED: In the late SED group, diabetes, ICH volume, lobar ICH, and CCL2 levels were significant in the univariable analysis (p < 0.05, Table 1). Lower CCL2 levels (OR 0.58; 95% CI 0.41–0.80) and lobar ICH (OR 10.5; 95% CI 3.79–29.16) were independent predictors of late SED compared to non-SED in the multivariable analysis (Table 2).

Poor outcomes and mortality after early and late SED: We identified age, diabetes, NIHSS score at admission, ICH score, ICH volume, CCL2, IL6, and IL8 as significant predictors of 90-day mortality in the univariable analysis (p < 0.1, Table 3). Higher CCL2 levels (OR 1.87; 95% CI 1.03–3.37) and NIHSS score at admission (OR 1.41; 95% CI 1.13–1.76) were independent predictors of 90-day mortality in patients with early and late SED combined after multivariable analysis (Table 4). A similar analysis was conducted to assess mortality and poor outcomes at 90, 180, and 365 days for patients with early SED, late SED, and early and late SED combined (apart from the 90-day mortality analysis of combined early and late SED described here), but no biomarker was found to be an independent predictor (Supplementary Tables S3-S10 in S1 File).

thumbnail
Download:
Table 3. Baseline Characteristics of Early and Late SED (Combined) by 90-Day Mortality.

https://doi.org/10.1371/journal.pone.0345752.t003

thumbnail
Download:
Table 4. Multivariable prediction model for 90-day mortality in Early and Late SED (Combined).

https://doi.org/10.1371/journal.pone.0345752.t004

Discussion

Our study identified lobar ICH and lower CCL2 levels as independent predictors of late SED development. These findings support existing literature that having a lobar ICH may influence the development of seizures or epileptiform discharges after a hemorrhagic stroke. Many studies have illustrated the association between lobar hemorrhages and the development of seizures post-ICH [3,19–23]. The mechanism is thought to involve increased excitability of cortical neurons close to the lesion, triggering the likelihood of seizures [24,25]. The presence of cortical lesions can be used to help better identify patients who are at risk of PSS. Applications include serving as an enrichment criterion, like in the PEPSTEP trial, ensuring that potential therapies for PSS are tested on patients most likely to experience them [26].

CXCL8 (IL8) and IL6 have been associated with inflammation and seizures without directly examining post-hemorrhagic stroke seizures [27,28]. IL6 was also shown to be independently correlated with seizure recurrence in post-ischemic stroke patients in the cohort of Jia et al [29]. One major difference between ischemic stroke and hemorrhagic stroke is the presence of blood (and its various components) outside of the blood vessels, which can cause findings like cortical superficial siderosis in hemorrhagic stroke not seen in ischemic stroke [11].

CCL2 is a chemokine secreted by various immune cells, especially monocytes, to facilitate immune cell migration to areas of inflammation such as hemorrhagic lesions. Most literature supports its activity as a pro-inflammatory, epileptogenic biomarker [30–33]. CCL2 has also been linked to conditions such as childhood epilepsy and drug-resistant epilepsy [34,35]. Our findings appear to suggest that lower CCL2 levels are associated with late SED. Possible mechanisms include temporality-related effects, such that higher CCL2 levels within 24 hours may cause worsened outcomes, but between 24–72 hours of ICH may have protective effects against persistent hyperexcitability or lasting damage, causing late SEDs. Animal models have demonstrated that higher CCL2 levels or expression in ischemic stroke are associated with faster tissue recovery [36–38]. CCR2, the receptor for CCL2, inhibition leads to decreased anti-inflammatory response, increased mortality, and impaired behavior recovery in a mouse model, but after 72 hours post-insult [36,39]. We also identified an independent association of higher CCL2 levels and NIHSS score at admission with 90-day mortality in the combined early and late SED cohort. Our finding of increased 90-day mortality with higher CCL2 levels in the combined cohort may be reflective of a large vascular insult seen in the significance of the NIHSS score at admission for the early SED group and the combined SED (early and late) group. The insult could be significant to the point where lowered seizure and epileptiform discharge risk from possible neuroprotective effects is unable to compensate, leading to increased mortality at 90 days.

There are limitations to our study. The patient sample size was small (n = 172), including the early and late SED cohorts (n = 33 and n = 29, respectively) and selected through convenience sampling, decreasing the external validity and influencing our findings’ significance. EEGs were not collected on every patient, which may have led to undercounting of those with epileptiform discharges. We were also unable to verify that clinical seizures had corresponding changes on the EEG. The level of biomarkers in predicting outcomes may fluctuate and depend on the time it was collected, as differences between SED and non-SED groups may be in the temporal profile of the biomarkers themselves instead of differences in values at any given time. Research assessing the temporal profile of biomarkers in stroke has demonstrated many different variations. Most literature is limited to blood samples post-admission, leading to observation of broadly increasing/decreasing trends [40–42]. Ideally, biomarker levels are measured repeatedly, and frequently after the onset of symptoms, to closely assess temporal changes. This has led to findings of changes in trends within 24 hours in certain biomarkers [43].

Future research should include this distinction as well as serum samples at more frequent intervals to better track temporal trends of these biomarkers, especially CCL2. More studies should also focus on the effects of ICH specifically, given the differing mechanisms of ischemic stroke in causing hypoxia. Larger studies focused on patients not only with seizures, but post-stroke epilepsy are also needed to elucidate the relationship with inflammatory biomarkers. Future research could also incorporate current advancements in machine learning and AI algorithms. Current research in post-stroke seizure and epilepsy with machine learning has been done for both ischemic and hemorrhagic stroke, albeit with non-biomarker findings (NIHSS, cortical involvement) [44–46]. Our analysis relied on conventional methods, whereas the machine learning studies were able to use a larger sample of patients, which could greatly improve predictions and prognostication.

Conclusion

In our retrospective analyses, we found that lower CCL2 levels were independent predictors of late SED. We also found an independent association of higher CCL2 levels in SED patients (early and late) with 90-day mortality. Additional studies investigating further aspects of biomarkers, such as levels within 24 hours or those past 72 hours, are needed.

Supporting information

S1 File. S1-10 Supplementary Tables (includes paired T-test for patients with 24- and 72-hour values and Timing of sample collection for all patient groups).

https://doi.org/10.1371/journal.pone.0345752.s001

(DOCX)

Acknowledgments

Thank you to Dr. Kevin Sheth who led the collection of plasma samples.

References

  1. 1. Vespa PM, O’Phelan K, Shah M, Mirabelli J, Starkman S, Kidwell C, et al. Acute seizures after intracerebral hemorrhage. Neurology. 2003;60(9):1441–6.
  2. 2. 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):1155–65. pmid:37721736
  3. 3. Haapaniemi E, Strbian D, Rossi C, Putaala J, Sipi T, Mustanoja S, et al. The CAVE Score for Predicting Late Seizures After Intracerebral Hemorrhage. Stroke. 2014;45(7):1971–6.
  4. 4. Tanaka T, Ihara M, Fukuma K, Mishra NK, Koepp MJ, Guekht A. Pathophysiology, diagnosis, prognosis, and prevention of poststroke epilepsy: clinical and research implications. Neurology. 2024;102(11):e209450.
  5. 5. Hsu C-J, Weng W-C, Peng SS-F, Lee W-T. Early-onset seizures are correlated with late-onset seizures in children with arterial ischemic stroke. Stroke. 2014;45(4):1161–3. pmid:24595587
  6. 6. Camilo O, Goldstein LB. Seizures and epilepsy after ischemic stroke. Stroke. 2004;35(7):1769–75. pmid:15166395
  7. 7. Hesdorffer DC, Benn EKT, Cascino GD, Hauser WA. Is a first acute symptomatic seizure epilepsy? Mortality and risk for recurrent seizure. Epilepsia. 2009;50(5):1102–8. pmid:19374657
  8. 8. Schubert KM, Dasari V, Oliveira AL, Tatillo C, Naeije G, Strzelczyk A. The Role of Electroencephalography in Predicting Post-Stroke Seizures and an Updated Prognostic Model (SeLECT-EEG). Ann Neurol. 2025.
  9. 9. Punia V, Ellison L, Bena J, Chandan P, Sivaraju A, George P, et al. Acute epileptiform abnormalities are the primary predictors of post-stroke epilepsy: a matched, case-control study. Ann Clin Transl Neurol. 2022;9(4):558–63. pmid:35243824
  10. 10. Bentes C, Martins H, Peralta AR, Morgado C, Casimiro C, Franco AC, et al. Early EEG predicts poststroke epilepsy. Epilepsia Open. 2018;3(2):203–12. pmid:29881799
  11. 11. Tanaka T, Fukuma K, Abe S, Matsubara S, Ikeda S, Kamogawa N, et al. Association of Cortical Superficial Siderosis with Post-Stroke Epilepsy. Ann Neurol. 2023;93(2):357–70. pmid:36053955
  12. 12. Kwon SY, Obeidat AZ, Sekar P, Moomaw CJ, Osborne J, Testai FD, et al. Risk factors for seizures after intracerebral hemorrhage: Ethnic/Racial Variations of Intracerebral Hemorrhage (ERICH) Study. Clin Neurol Neurosurg. 2020;192:105731. pmid:32062309
  13. 13. Zhang C, Wang X, Wang Y, Zhang J, Hu W, Ge M, et al. Risk factors for post-stroke seizures: a systematic review and meta-analysis. Epilepsy Res. 2014;108(10):1806–16. pmid:25454500
  14. 14. Misra S, Khan EI, Lam TT, Mazumder R, Gururangan K, Hickman LB. Common pathways of epileptogenesis in patients with epilepsy post–brain injury. Neurology. 2023;101(22):e2243–56.
  15. 15. Stellwagen D, Beattie EC, Seo JY, Malenka RC. Differential regulation of AMPA receptor and GABA receptor trafficking by tumor necrosis factor-alpha. J Neurosci. 2005;25(12):3219–28. pmid:15788779
  16. 16. Vezzani A, Balosso S, Ravizza T. Neuroinflammatory pathways as treatment targets and biomarkers in epilepsy. Nat Rev Neurol. 2019;15(8):459–72. pmid:31263255
  17. 17. Abraira L, Giannini N, Santamarina E, Cazorla S, Bustamante A, Quintana M, et al. Correlation of blood biomarkers with early-onset seizures after an acute stroke event. Epilepsy Behav. 2020;104(Pt B):106549. pmid:31677998
  18. 18. Morgan E, Varro R, Sepulveda H, Ember JA, Apgar J, Wilson J, et al. Cytometric bead array: a multiplexed assay platform with applications in various areas of biology. Clin Immunol. 2004;110(3):252–66. pmid:15047203
  19. 19. De Herdt V, Dumont F, Hénon H, Derambure P, Vonck K, Leys D, et al. Early seizures in intracerebral hemorrhage. Neurology. 2011;77(20):1794–800.
  20. 20. De Reuck J, Hemelsoet D, Van Maele G. Seizures and epilepsy in patients with a spontaneous intracerebral haematoma. Clin Neurol Neurosurg. 2007;109(6):501–4. pmid:17509750
  21. 21. Bladin CF, Alexandrov AV, Bellavance A, Bornstein N, Chambers B, Coté R, et al. Seizures after stroke: a prospective multicenter study. Arch Neurol. 2000;57(11):1617–22. pmid:11074794
  22. 22. Beghi E, D’Alessandro R, Beretta S, Consoli D, Crespi V, Delaj L, et al. Incidence and predictors of acute symptomatic seizures after stroke. Neurology. 2011;77(20):1785–93. pmid:21975208
  23. 23. Qian C, Löppönen P, Tetri S, Huhtakangas J, Juvela S, Turtiainen H-ME, et al. Immediate, early and late seizures after primary intracerebral hemorrhage. Epilepsy Res. 2014;108(4):732–9. pmid:24661429
  24. 24. Li H, Prince DA. Synaptic activity in chronically injured, epileptogenic sensory-motor neocortex. J Neurophysiol. 2002;88(1):2–12. pmid:12091528
  25. 25. Prince DA, Parada I, Scalise K, Graber K, Jin X, Shen F. Epilepsy following cortical injury: cellular and molecular mechanisms as targets for potential prophylaxis. Epilepsia. 2009;50 Suppl 2(Suppl 2):30–40. pmid:19187292
  26. 26. Nicolo J-P, Chen Z, Moffat B, Wright DK, Sinclair B, Glarin R, et al. Study protocol for a phase II randomised, double-blind, placebo-controlled trial of perampanel as an antiepileptogenic treatment following acute stroke. BMJ Open. 2021;11(5):e043488. pmid:33972334
  27. 27. Librizzi L, Vila Verde D, Colciaghi F, Deleo F, Regondi MC, Costanza M, et al. Peripheral blood mononuclear cell activation sustains seizure activity. Epilepsia. 2021;62(7):1715–28. pmid:34061984
  28. 28. Pernhorst K, Herms S, Hoffmann P, Cichon S, Schulz H, Sander T, et al. TLR4, ATF-3 and IL8 inflammation mediator expression correlates with seizure frequency in human epileptic brain tissue. Seizure. 2013;22(8):675–8. pmid:23706953
  29. 29. Jia Q, Jiang F, Ma D, Li J, Wang F, Wang Z. Association Between IL-6 and Seizure Recurrence in Patients with the First Post-Ischemic Stroke Seizure. Neuropsychiatr Dis Treat. 2020;16:1955–63. pmid:32848401
  30. 30. Cerri C, Genovesi S, Allegra M, Pistillo F, Püntener U, Guglielmotti A, et al. The Chemokine CCL2 Mediates the Seizure-Enhancing Effects of Systemic Inflammation. J Neurosci. 2016;36(13):3777–88.
  31. 31. Wu Y, Wang X, Mo X, Xi Z, Xiao F, Li J, et al. Expression of monocyte chemoattractant protein-1 in brain tissue of patients with intractable epilepsy. Clin Neuropathol. 2008;27(2):55–63. pmid:18402383
  32. 32. Tian DS, Peng J, Murugan M, Feng LJ, Liu JL, Eyo UB. Chemokine CCL2–CCR2 signaling induces neuronal cell death via STAT3 activation and IL-1β production after status epilepticus. J Neurosci. 2017;37(33):7878–92.
  33. 33. Foresti ML, Arisi GM, Katki K, Montañez A, Sanchez RM, Shapiro LA. Chemokine CCL2 and its receptor CCR2 are increased in the hippocampus following pilocarpine-induced status epilepticus. J Neuroinflammation. 2009;6:40. pmid:20034406
  34. 34. Česká K, Papež J, Ošlejšková H, Slabý O, Radová L, Loja T, et al. CCL2/MCP-1, interleukin-8, and fractalkine/CXC3CL1: Potential biomarkers of epileptogenesis and pharmacoresistance in childhood epilepsy. Eur J Paediatr Neurol. 2023;46:48–54.
  35. 35. Milano C, Montali M, Barachini S, Burzi IS, Pratesi F, Petrozzi L, et al. Increased production of inflammatory cytokines by circulating monocytes in mesial temporal lobe epilepsy: A possible role in drug resistance. J Neuroimmunol. 2024;386:578272.
  36. 36. Geng H, Chen L, Tang J, Chen Y, Wang L. The Role of CCL2/CCR2 Axis in Cerebral Ischemia-Reperfusion Injury and Treatment: From Animal Experiments to Clinical Trials. Int J Mol Sci. 2022;23(7):3485. pmid:35408846
  37. 37. Stowe AM, Wacker BK, Cravens PD, Perfater JL, Li MK, Hu R, et al. CCL2 upregulation triggers hypoxic preconditioning-induced protection from stroke. J Neuroinflammation. 2012;9:33. pmid:22340958
  38. 38. Lee S, Kim OJ, Lee KO, Jung H, Oh S-H, Kim NK. Enhancing the Therapeutic Potential of CCL2-Overexpressing Mesenchymal Stem Cells in Acute Stroke. Int J Mol Sci. 2020;21(20):7795. pmid:33096826
  39. 39. Pimentel-Coelho PM, Michaud JP, Rivest S. C–C chemokine receptor type 2 (CCR2) signaling protects neonatal male mice with hypoxic–ischemic hippocampal damage from developing spatial learning deficits. Behav Brain Res. 2015;286:146–51.
  40. 40. Da Silva-Candal A, Dopico-López A, Pérez-Mato M, Rodríguez-Yáñez M, Pumar JM, Ávila-Gómez P, et al. Characterization of a Temporal Profile of Biomarkers as an Index for Ischemic Stroke Onset Definition. J Clin Med. 2021;10(14):3136. pmid:34300300
  41. 41. Zafar A, Farooqui M, Ikram A, Suriya S, Kempuraj D, Khan M, et al. Cytokines, brain proteins, and growth factors in acute stroke patients: A pilot study. Surg Neurol Int. 2021;12:366. pmid:34513133
  42. 42. Luheshi NM, Kovács KJ, Lopez-Castejon G, Brough D, Denes A. Interleukin-1α expression precedes IL-1β after ischemic brain injury and is localised to areas of focal neuronal loss and penumbral tissues. J Neuroinflammation. 2011;8:186. pmid:22206506
  43. 43. Wunderlich MT, Lins H, Skalej M, Wallesch C-W, Goertler M. Neuron-specific enolase and tau protein as neurobiochemical markers of neuronal damage are related to early clinical course and long-term outcome in acute ischemic stroke. Clin Neurol Neurosurg. 2006;108(6):558–63. pmid:16457947
  44. 44. Lekoubou A, Petucci J, Femi Ajala T, Katoch A, Hong J, Sen S, et al. Can machine learning predict late seizures after intracerebral hemorrhages? Evidence from real-world data. Epilepsy Behav. 2024;157:109835. pmid:38820686
  45. 45. Liu J, He H, Wang Y, Du J, Liang K, Xue J, et al. Predictive models for secondary epilepsy in patients with acute ischemic stroke within one year. eLife. 2024;13:RP98759.
  46. 46. Yu Y, Chen Z, Yang Y, Zhang J, Wang Y. Development and validation of an interpretable machine learning model for predicting post-stroke epilepsy. Epilepsy Res. 2024;205:107397. pmid:38976953