Risk factors for seizures after intracerebral hemorrhage: Ethnic/Racial Variations of Intracerebral Hemorrhage (ERICH) Study

Can machine learning predict late seizures after intracerebral hemorrhages? Evidence from real-world data

INTRODUCTION

Intracerebral hemorrhage represents 15 % of all strokes and it is associated with a high risk of post-stroke epilepsy. However, there are no reliable methods to accurately predict those at higher risk for developing seizures despite their importance in planning treatments, allocating resources, and advancing post-stroke seizure research. Existing risk models have limitations and have not taken advantage of readily available real-world data and artificial intelligence. This study aims to evaluate the performance of Machine-learning-based models to predict post-stroke seizures at 1 year and 5 years after an intracerebral hemorrhage in unselected patients across multiple healthcare organizations.

DESIGN/METHODS

We identified patients with intracerebral hemorrhage (ICH) without a prior diagnosis of seizures from 2015 until inception (11/01/22) in the TriNetX Diamond Network, using the International Classification of Diseases, Tenth Revision (ICD-10) I61 (I61.0, I61.1, I61.2, I61.3, I61.4, I61.5, I61.6, I61.8, and I61.9). The outcome of interest was any ICD-10 diagnosis of seizures (G40/G41) at 1 year and 5 years following the first occurrence of the diagnosis of intracerebral hemorrhage. We applied a conventional logistic regression and a Light Gradient Boosted Machine (LGBM) algorithm, and the performance of the model was assessed using the area under the receiver operating characteristics (AUROC), the area under the precision-recall curve (AUPRC), the F1 statistic, model accuracy, balanced-accuracy, precision, and recall, with and without seizure medication use in the models.

RESULTS

A total of 85,679 patients had an ICD-10 code of intracerebral hemorrhage and no prior diagnosis of seizures, constituting our study cohort. Seizures were present in 4.57 % and 6.27 % of patients within 1 and 5 years after ICH, respectively. At 1-year, the AUROC, AUPRC, F1 statistic, accuracy, balanced-accuracy, precision, and recall were respectively 0.7051 (standard error: 0.0132), 0.1143 (0.0068), 0.1479 (0.0055), 0.6708 (0.0076), 0.6491 (0.0114), 0.0839 (0.0032), and 0.6253 (0.0216). Corresponding metrics at 5 years were 0.694 (0.009), 0.1431 (0.0039), 0.1859 (0.0064), 0.6603 (0.0059), 0.6408 (0.0119), 0.1094 (0.0037) and 0.6186 (0.0264). These numerical values indicate that the statistical models fit the data very well.

CONCLUSION

Machine learning models applied to electronic health records can improve the prediction of post-hemorrhagic stroke epilepsy, presenting a real opportunity to incorporate risk assessments into clinical decision-making in post-stroke care clinical care and improve patients' selection for post-stroke epilepsy research.

Risk Factors Associated with Epilepsy Related to Cerebrovascular Disease: A Systematic Review and Meta-Analysis

Background and Objective

Stroke is one of the most frequent neurological syndromes in the adult population and the cause of 10% of all diagnosed epilepsies. It is attributed to the origin of up to 50% of them in adults >60 years of age. Although a few risk factors have been described and considered when modeling predictive tools, this aspect is still clinically complex. The objective of this study is to describe and compare predictor scales of post stroke epilepsy (PSE) in adult patients with better performance.

Methods

A systematic review and meta-analysis were performed of studies published between 2010 and 2020 and found in PubMed, Scopus, EMBASE, LILACS, BVS, Google Scholar, and CENTRAL databases. Sixteen studies were included with a total of 298,694 patients with a diagnosis of stroke, 5590 presented late seizures (LS).

Results

Hemorrhage, cortical involvement, and early seizure were the elements most associated with the risk of presenting late seizures. The SeLECT score demonstrated a low risk of bias with a high predictive ability in patients with ischemic stroke (AUC: 0.77 [95% CI: 0.71-0.82]). In patients with hemorrhagic stroke, the CAVE score demonstrated adequate predictive ability (AUC: 0.81 [95% CI: 0.76-0.86]), but an uncertain risk of bias. Research has established risk factors for post ictal epilepsy; however, the numerous ways of assessing data in studies and the difference in their designs make the task of producing a predictive scale that covers the most important risk factors and is reliable for application in the clinical setting, regardless of stroke etiology, very arduous.

Conclusion

Hemorrhage, cortical involvement, and early seizure are associated with an increased risk of post ictal epilepsy. Also, elements such as age, traditional vascular risk factors, and functional assessment failed to reflect statistical significance. Finally, further research is required to refine the available predictive tools.

Incidence and risk factors associated with seizures in cerebral amyloid angiopathy

BACKGROUND AND PURPOSE

Cerebral amyloid angiopathy (CAA) is a common cause of intracranial hemorrhage (ICH), which is a risk factor for seizures. The incidence and risk factors of seizures associated with a heterogeneous cohort of CAA patients have not been studied.

METHODS

We conducted a retrospective study of patients with CAA treated at Mayo Clinic Florida between 1 January 2015 and 1 January 2021. CAA was defined using the modified Boston criteria version 2.0. We analyzed electrophysiological and clinical features, and comorbidities including lobar ICH, nontraumatic cortical/convexity subarachnoid hemorrhage (cSAH), superficial siderosis, and inflammation (CAA with inflammation [CAA-ri]). Cognition and mortality were secondary outcomes. Univariate and multivariate analyses were performed to determine risk of seizures relative to clinical presentation.

RESULTS

Two hundred eighty-four patients with CAA were identified, with median follow-up of 35.7 months (interquartile range = 13.5-61.3 months). Fifty-six patients (19.7%) had seizures; in 21 (37.5%) patients, seizures were the index feature leading to CAA diagnosis. Seizures were more frequent in females (p = 0.032) and patients with lobar ICH (p = 0.002), cSAH (p = 0.030), superficial siderosis (p < 0.001), and CAA-ri (p = 0.005), and less common in patients with microhemorrhage (p = 0.006). After controlling for age and sex, lobar ICH (odds ratio [OR] = 2.1, 95% confidence interval [CI] = 1.1-4.2), CAA-ri (OR = 3.8, 95% CI = 1.4-10.3), and superficial siderosis (OR = 3.7, 95% CI = 1.9-7.0) were independently associated with higher odds of incident seizures.

CONCLUSIONS

Seizures are common in patients with CAA and are independently associated with lobar ICH, CAA-ri, and superficial siderosis. Our results may be applied to optimize clinical monitoring and management for patients with CAA.

Predictors of unprovoked seizures in intracerebral hemorrhages

BACKGROUND

Seizures are a common complication of intracerebral hemorrhage (ICH). We aimed to identify predictors of unprovoked seizures (US) after ICH in a Chinese cohort.

METHODS

We retrospectively included patients with ICH admitted in the Second Hospital of Hebei Medical University between November 2018 and December 2020. Incidence and risk factors of US were identified with univariate and then multiple Cox regression analysis. We used χ2 test to compare incidence of US between groups with or without prophylactic anti-seizure medications (ASM) in patients with craniotomy.

RESULTS

A total of 488 patients were included in the cohort, 58 (11.9%) patients developed US within 3 years after ICH. Analysis on the 362 patients without prophylactic ASM showed that craniotomy (HR 8.35, 95% CI 3.80-18.31) and acute symptomatic seizures (ASS) (HR 13.76, 95% CI 3.56-53.17) are independent predictors of US. No significant effect of prophylactic ASM use was found on incidence of US in ICH patients with craniotomy (P = 0.369).

CONCLUSIONS

Craniotomy and acute symptomatic seizures were independent predictors for unprovoked seizures after ICH, suggesting that more attention should be paid for such patients during follow-up. Whether prophylactic ASM treatment benefits ICH patients underwent craniotomy remains uncertain.

Incidence and risk factors of post-stroke seizures and epilepsy: systematic review and meta-analysis

OBJECTIVE

Due to variability in reports, the aim of this meta-analysis was to evaluate the incidence and risk factors of post-stroke early seizures (ES) and post-stroke epilepsy (PSE).

METHODS

The MEDLINE, EMBASE and Web of Science databases were searched for post-stroke ES/PSE articles published on any date up to November 2020. Post-stroke ES included seizures occurring within 7 days of stroke, and PSE included at least one unprovoked seizure. Using random effects models, the incidence and risk factors of post-stroke ES and PSE were evaluated. The study was retrospectively registered with INPLASY (INPLASY2023100008).

RESULTS

Of 128 included studies in total, the incidence of post-stroke ES was 0.07 (95% confidence interval [CI] 0.05, 0.10) and PSE was 0.10 (95% CI 0.08, 0.13). The rates were higher in children than adults. Risk factors for post-stroke ES included hemorrhagic stroke (odds ratio [OR] 2.14, 95% CI 1.44, 3.18), severe strokes (OR 2.68, 95% CI 1.73, 4.14), cortical involvement (OR 3.09, 95% CI 2.11, 4.51) and hemorrhagic transformation (OR 2.70, 95% CI 1.58, 4.60). Risk factors for PSE included severe strokes (OR 4.92, 95% CI 3.43, 7.06), cortical involvement (OR 3.20, 95% CI 2.13, 4.81), anterior circulation infarcts (OR 3.28, 95% CI 1.34, 8.03), hemorrhagic transformation (OR 2.81, 95% CI 1.25, 6.30) and post-stroke ES (OR 7.24, 95% CI 3.73, 14.06).

CONCLUSION

Understanding the risk factors of post-stroke ES/PSE may identify high-risk individuals who might benefit from prophylactic treatment.

Outcomes in Patients With Poststroke Seizures: A Systematic Review and Meta-Analysis

Importance

Published data about the impact of poststroke seizures (PSSs) on the outcomes of patients with stroke are inconsistent and have not been systematically evaluated, to the authors' knowledge.

Objective

To investigate outcomes in people with PSS compared with people without PSS.

Data Sources

MEDLINE, Embase, PsycInfo, Cochrane, LILACS, LIPECS, and Web of Science, with years searched from 1951 to January 30, 2023.

Study Selection

Observational studies that reported PSS outcomes.

Data Extraction and Synthesis

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses checklist was used for abstracting data, and the Joanna Briggs Institute tool was used for risk-of-bias assessment. Data were reported as odds ratio (OR) and standardized mean difference (SMD) with a 95% CI using a random-effects meta-analysis. Publication bias was assessed using funnel plots and the Egger test. Outlier and meta-regression analyses were performed to explore the source of heterogeneity. Data were analyzed from November 2022 to January 2023.

Main Outcomes and Measures

Measured outcomes were mortality, poor functional outcome (modified Rankin scale [mRS] score 3-6), disability (mean mRS score), recurrent stroke, and dementia at patient follow-up.

Results

The search yielded 71 eligible articles, including 20 110 patients with PSS and 1 166 085 patients without PSS. Of the participants with PSS, 1967 (9.8%) had early seizures, and 10 605 (52.7%) had late seizures. The risk of bias was high in 5 studies (7.0%), moderate in 35 (49.3%), and low in 31 (43.7%). PSSs were associated with mortality risk (OR, 2.1; 95% CI, 1.8-2.4), poor functional outcome (OR, 2.2; 95% CI, 1.8-2.8), greater disability (SMD, 0.6; 95% CI, 0.4-0.7), and increased dementia risk (OR, 3.1; 95% CI, 1.3-7.7) compared with patients without PSS. In subgroup analyses, early seizures but not late seizures were associated with mortality (OR, 2.4; 95% CI, 1.9-2.9 vs OR, 1.2; 95% CI, 0.8-2.0) and both ischemic and hemorrhagic stroke subtypes were associated with mortality (OR, 2.2; 95% CI, 1.8-2.7 vs OR, 1.4; 95% CI, 1.0-1.8). In addition, early and late seizures (OR, 2.4; 95% CI, 1.6-3.4 vs OR, 2.7; 95% CI, 1.8-4.1) and stroke subtypes were associated with poor outcomes (OR, 2.6; 95% CI, 1.9-3.7 vs OR, 1.9; 95% CI, 1.0-3.6).

Conclusions and Relevance

Results of this systematic review and meta-analysis suggest that PSSs were associated with significantly increased mortality and severe disability in patients with history of stroke. Unraveling these associations is a high clinical and research priority. Trials of interventions to prevent seizures may be warranted.

External validation and comparison of clinical scores for predicting late seizures after intracerebral hemorrhage in Chinese patients

PURPOSE

Clinical scores have been established to predict the probability of late seizures following intracerebral hemorrhage (ICH) for individual patients, including the CAVE, CAVS and LANE scores. The purpose of this study was to compare these prediction scores in the Chinese population and undertake an independent external validation on them.

METHODS

At one tertiary hospital in China, we retrospectively recruited consecutive inpatients who had been diagnosed with ICH. Medical records and tele interviews with a modified standardized questionnaire were used to identify late seizures. All the predictors of the prediction scores were collected from patient charts and databases by a standardized data collection protocol. The external validation of the prediction scores was quantified by the area under the curve (AUC), sensitivity, specificity, Youden index (YI), positive predictive value (PPV), and negative predictive value (NPV).

RESULTS

69 (5.4%) of 1276 patients experienced late seizures after ICH. There was no significant difference in the CAVE, CAVS, and LANE scores, which had AUCs of 0.75 (95% CI = 0.70-0.81), 0.74 (95% CI = 0.68-0.80), and 0.76 (95% CI = 0.70-0.82), respectively. At the optimal cutoff score, the LANE score had a lower sensitivity but a higher specificity than the CAVE and CAVS scores. Among the three prediction scores, the LANE score had a higher PPV than the others (0.145 vs. 0.088, 0.083), while the NPV was similar among the three prediction scores (0.989, 0.989, and 0.972).

CONCLUSION

Our study showed that the CAVE, CAVS and LANE scores had similar AUCs for the occurrence of late seizures, but the LANE score had a relatively high PPV at the optimal cutoff score. Due to low evidence for using prophylactic antiseizure medications (ASM) in patients with ICH and poor availability of specialist stroke care in China, the LANE score with a cutoff score of 3 could be an applicable prediction tool in Chinese patients with ICH.

Acute Spontaneous Lobar Cerebral Hemorrhages Present a Different Clinical Profile and a More Severe Early Prognosis than Deep Subcortical Intracerebral Hemorrhages-A Hospital-Based Stroke Registry Study

Acute spontaneous intracerebral hemorrhage (ICH) is the most severe stroke subtype, with a high risk of death, dependence, and dementia. Knowledge about the clinical profile and early outcomes of ICH patients with lobar versus deep subcortical brain topography remains limited. In this study, we investigated the effects of ICH topography on demographics, cerebrovascular risk factors, clinical characteristics, and early outcomes in a sample of 298 consecutive acute ICH patients (165 with lobar and 133 with subcortical hemorrhagic stroke) available in a single-center-based stroke registry over 24 years. The multiple logistic regression analysis shows that variables independently associated with lobar ICH were early seizures (OR 6.81, CI 95% 1.27−5.15), chronic liver disease (OR 4.55, 95% CI 1.03−20.15), hemianopia (OR 2.55, 95% CI 1.26−5.15), headaches (OR 1.90, 95% CI 1.90, 95% IC 1.06−3.41), alcohol abuse (>80 gr/day) (OR 0−10, 95% CI 0.02−0,53), hypertension (OR 0,41, 95% CI 0.23−0−70), sensory deficit (OR 0.43, 95% CI 0.25−0.75), and limb weakness (OR: 0.47, 95% CI 0.24−0.93). The in-hospital mortality was 26.7% for lobar and 16.5% for subcortical ICH. The study confirmed that the clinical spectrum, prognosis, and early mortality of patients with ICH depend on the site of bleeding, with a more severe early prognosis in lobar intracerebral hemorrhage.

Association between minimally invasive surgery and late seizures in patients with intracerebral hemorrhage: A propensity score matching study

Purpose

The association between minimally invasive surgery (MIS) for hematoma evacuation and late seizures after intracerebral hemorrhage (ICH) remains uncertain. We aimed to investigate whether MIS increases the risk of late seizures after ICH and identify the risk factors for late seizures in this patient subgroup.

Methods

We retrospectively included consecutive inpatients diagnosed with ICH at two tertiary hospitals in China. The subjects were divided into the MIS group (ICH patients who received MIS including hematoma aspiration and thrombolysis) and conservative treatment group (ICH patients who received conservative medication). Propensity score matching was performed to balance possible risk factors for late seizures between the MIS and conservative treatment groups. Before and after matching, between-group comparisons of the incidence of late seizures were performed between the MIS and conservative treatment groups. Univariate and multivariate logistic regression analyses were used to identify independent risk factors for late seizures in MIS-treated patients.

Results

A total of 241 and 1,689 patients were eligible for the MIS and conservative treatment groups, respectively. After matching, 161 ICH patients from the MIS group were successfully matched with 161 ICH patients from the conservative treatment group (1:1). Significant differences (p < 0.001) were found between the MIS group (31/241, 12.9%) and conservative treatment group (69/1689, 4.1%) in the incidence of late seizures before matching. However, after matching, no significant differences (p = 0.854) were found between the MIS group (17/161, 10.6%) and conservative treatment group (16/161, 9.9%). Multivariate logistic regression analysis revealed that cortical involvement (OR = 2.547; 95% CI = 1.137-5.705; p value = 0.023) and higher National Institutes of Health Stroke Scale (NIHSS) scores (OR = 1.050; 95% CI = 1.008-1.094; p value = 0.019) were independent risk factors for late seizures.

Conclusion

Our study revealed that receiving MIS did not increase the incidence of late seizures after ICH. Additionally, cortical involvement and NIHSS scores were independent risk factors for late seizures in MIS-treated patients.

Effect of Primary Prophylactic Antiseizure Medication for Seizure Prevention Following Intracerebral Hemorrhage in the ERICH Study

OBJECTIVES

Intracerebral hemorrhage (ICH) has the highest morbidity and mortality rate of any stroke subtype and clinicians often administer prophylactic antiseizure medications (ASMs) as a means of preventing post-stroke seizures, particularly following lobar ICH. However, evidence for ASM efficacy in preventing seizures and reducing disability is lacking given limited randomized trials. Herein, we report analysis from a large prospective observational study that evaluates the effect of primary prophylactic ASM administration on seizure occurrence and disability following ICH.

MATERIALS AND METHODS

Primary analysis was performed on 1630 patients with ICH enrolled in the ERICH study. A propensity score for administration of prophylactic ASM was developed and patients were matched by the closest propensity score (difference < 0.1). McNemar's test was used to compare occurrence of in-hospital seizure and disability, defined by modified Rankin Score (mRS) ≥ 3 at 3 months post ICH.

RESULTS

Of the 815 matched pairs of patients treated with primary prophylactic ASM, there was no significant difference in seizure occurrence (p = 0.4631) or disability (p = 0.4653). Subset analysis of 280 matched pairs of patients with primary lobar ICH similarly revealed no significant difference in seizure occurrence (p = 0.1011) or disability (p = 1.00) between prophylactically treated and untreated patients.

CONCLUSIONS

Although current guidelines do not recommend primary prophylactic ASM following ICH, clinical use remains widespread. Data from the ERICH study did not find an association between administering primary prophylactic ASM and preventing seizures or reducing disability following ICH, thus providing evidence to influence clinical practice and patient care.

Risk models to predict late-onset seizures after stroke: A systematic review

BACKGROUND AND PURPOSE

We performed a systematic review to evaluate available risk models to predict late seizure onset among stroke survivors.

METHODS

We searched major databases (PubMed, SCOPUS, and Cochrane Library) from inception to October 2020 for articles on the development and/or validation of risk models to predict late seizures after a stroke. The impact of models to predict late-onset seizures was also assessed. We included seven articles in the final analysis. For each of these studies, we evaluated the study design and scope of predictors analyzed to derive each model. We assessed the performance of the models during internal and external validation in terms of discrimination and calibration.

RESULTS

Three studies focused on ischemic stroke alone, with c-statistic values ranging from 0.73 to 0.77. The SeLECT model from Switzerland was externally validated in Italian, German, and Austrian cohorts where c-statistics ranged from 0.69 to 0.81. This model along with the PSEiCARe model, were internally validated and calibration performance was provided for both models. The CAVS and CAVE models reported on the risk of late-onset seizures in patients with hemorrhagic stroke. The CAVS model derivation cohort was racially diverse. The CAVS model's c-statistic was 0.76, while the CAVE model had a c-statistic of 0.81. Calibration and internal validation were not performed for either study. The CAVS model, created from a Finnish population, was externally validated in American and French cohorts, with c-statistics of 0.73 and 0.69, respectively. Finally, the two studies focusing on both types of stroke came from the PoSERS and INPOSE models. Neither model provided c-statistics, calibration metrics, internal or external validation information. We found no evidence of the presence of impact studies to assess the effect of adopting late-onset seizure risk models after stroke on clinical outcomes.

CONCLUSION

The SeLECT model was the only model developed in line with proposed guidelines for appropriate model development. The model, which was externally validated in a very similar and homogeneous population, may need to be tested in a more racially/ethnic diverse and younger population; testing the SeLECT model, accounting for overall brain health is likely to improve the identification of high-risk patients for late post stroke seizures.

Risk factors for early-onset seizures after stroke: A systematicreview and meta-analysis of 18 observational studies

OBJECTIVES

To systematically evaluate the risk factors of early-onset seizures after stroke, in order to better provide evidence-based results for early detection, identification, targeted prevention, and treatment of this disease.

METHODS

PubMed, EMBASE, The Cochrane Library, CNKI, and WanFang databases were searched to collect relevant studies on the risk factors of early-onset seizures after stroke from January 2010 to January 2020. Meta-analysis of all included studies was performed by using RevMan version 5.3 and Stata version 14.0 software.

RESULTS

Eighteen case-control studies with a total sample size of 13,289 cases, including 813 cases with early-onset seizures after stroke, and 12,476 cases with non-early-onset seizures after stroke were included. The results of meta-analysis showed that cortical involvement [Odds Ratio (OR) = 5.00, 95%Confidence Interval (CI) (2.85, 8.74), p < .00001], cerebral infarction with hemorrhagic transformation [OR = 2.77, 95%CI (1.87, 4.11), p < .00001] and intracerebral hemorrhage [OR = 1.83, 95%CI (1.13, 2.97), p = .01]-related factors showed greater association with the occurrence of early-onset seizures after stroke.

CONCLUSIONS

These findings suggest that cortical involvement, intracerebral hemorrhage, and cerebral infarction with hemorrhagic transformation are important predictors and risk factors for early seizures after stroke, while the patient's gender, age, NHISS score, alcoholism, smoking, high blood pressure, diabetes, atrial fibrillation, dyslipidemia, receiving surgical treatment, and reperfusion therapy showed no association with the occurrence of early-onset seizures after stroke.

Multifactorial Predictors of Late Epileptic Seizures Related to Stroke: Evaluation of the Current Possibilities of Stratification Based on Existing Prognostic Models-A Comprehensive Review

Background: Epilepsy associated with strokes is a significant clinical and public health problem and has a negative impact on prognosis and clinical outcome. A late epileptic seizure occurring seven days after stroke is actually equated with poststroke epilepsy due to the high risk of recurrence. Predictive models evaluated in the acute phase of stroke would allow for the stratification and early selection of patients at higher risk of developing late seizures. Methods: The most relevant papers in this field were reviewed to establish multifactorial predictors of late seizures and attempt to standardize and unify them into a common prognostic model. Results: Clinical and radiological factors have become the most valuable and reproducible predictors in many reports, while data on electroencephalographic, genetic, and blood biomarkers were limited. The existing prognostic models, CAVE and SeLECT, based on relevant, readily available, and routinely assessed predictors, should be validated and improved in multicenter studies for widespread use in stroke units. Conclusions: Due to contradictory reports, a common and reliable model covering all factors is currently not available. Further research might refine forecasting models by incorporating advanced radiological neuroimaging or quantitative electroencephalographic analysis.