Abstract
Post-stroke epilepsy may occur after aneurysmal subarachnoid hemorrhage (aSAH). Both early and late seizures could cause severe neurocognitive deficits if administration of appropriate antiseizure medication is delayed. Therefore, it is important to elucidate the risk factors for early and late seizures, which could be shared with medical teams to promptly manage seizures. There are aspects of both hemorrhage and ischemia in aSAH, and thus, numerous risk factors are considered for early and late seizures. We examined factors associated with aSAH-related early and late seizures. Among 297 patients who had aSAH and underwent direct or endovascular surgery, 25 had early seizures and 20 had late seizures. Patients who did not experience any seizures in at least 2-years of follow-up (n = 81) were used as the control group. Early seizures were associated with older age and acute severe nonneurological infection, whereas late seizures were associated with intraparenchymal lesion volume >10 mL and shunt placement. In patients with late seizures, consistency was frequently observed between electroencephalogram and the presence of intraparenchymal lesions. The frontopolar electrode on electroencephalogram was highly sensitive to abnormality in early seizures. Early seizures were induced by the patient's systemic factors, which may lower the threshold for cortical excitability. Patients with intraparenchymal lesions who undergo shunt placement should be carefully followed up for late seizures.
Keywords: aneurysmal subarachnoid hemorrhage, early seizure, late seizure, electroencephalogram
Introduction
A seizure occurring within 1 week after stroke onset is defined as an acute symptomatic seizure1) or an early seizure (ES). Contrarily, a seizure occurring beyond 1 week after stroke onset is defined as a late seizure (LS). Patients with ES are thought to have a greater likelihood of withdrawal of antiseizure medication (ASM), whereas those with LS are at a higher risk of seizure recurrence or post-stroke epilepsy.2,3) Given the differences in seizure prognosis, the etiology and the risk factors might be different between ES and LS. Both types could cause severe neurocognitive deficits if administration of appropriate ASM is delayed. Therefore, it is important to elucidate the risk factors for ES and LS, which could be shared with medical teams to promptly manage seizures.
Subarachnoid hemorrhage due to aneurysmal rupture (aSAH) accounts for 18%-24% of hemorrhagic cerebral strokes.4,5) The prevalence rate of epileptic seizure occurrence after aSAH is reported to be >10%.6-8) Notably, aSAH has a distinctive feature, as delayed vasospasm is possibly associated with seizures. Early infarction is another distinctive feature of aSAH, although the association between early infarction and seizures remains unclear.9) Numerous risk factors could be associated with aSAH-related seizures. As for ES, younger age,7,8) older age,5) poor grade,8) Fisher grades 3 and 4,7) acute hydrocephalus,8) and acute nonneurological infection5) have been reported to be associated with poor outcomes.5,7) As for LS, Fisher grades 3 and 4, cortical infarction,8) middle cerebral artery aneurysm, direct surgery,6) poor outcome,7) and hemosiderosis10,11) have been reported to be associated with poor outcomes.
Despite the background that treatment options have been shifting from clipping to coil embolization, the prevalence of aSAH-related seizures or epilepsy has not dramatically decreased. Some assessment scores have been proposed for LS, such as the SeLECT score for ischemic stroke3) and the CAVE score for hemorrhagic stroke,12) but there are currently no similar assessment scores for aSAH. The risk factors for aSAH-related seizures may be more complex as this type has both aspects of hemorrhage and ischemia. Furthermore, to the best of our knowledge, no previous studies have investigated neuroimaging and electroencephalogram (EEG) findings for aSAH-related seizures.
Therefore, this study examined the factors associated with ES and LS in patients with aSAH who were treated in our institute. We also investigated the characteristics of seizure symptoms, neuroimaging, and EEG findings.
Materials and Methods
Among 297 patients who had aSAH and underwent direct or endovascular surgery from May 1, 2011 to December 25, 2022, 25 had ES (mean age, 71.0 years; 95% confidence interval [CI], 64.4-77.6; 18 women) and 20 had LS (mean age, 64.1 years; 95% CI 56.9-71.3; 13 women). The factors associated with seizure occurrence were examined, such as age, sex, primary illness (hypertension, disease of the central nervous system), Hunt and Hess grade, Fisher's CT classification, aneurysmal location, intracerebral hematoma, Sylvian hematoma, treatment (craniotomy, ventricular drainage, shunt placement), electrolyte imbalance, meningitis, acute severe nonneurological infection (required antibiotic treatment for >3 days), symptomatic vasospasm, and volume and location of intraparenchymal lesion volume >10 mL after vasospasm. Patients without seizure within at least 2-years of follow-up were used as the control group (mean age, 59.3 years; 95% CI, 56.8-61.9; 64 women). Patients with impending rupture with no obvious subarachnoid hemorrhage (Fisher group 1) were excluded. Next, seizure semiology, EEG, neuroimaging, and the number of ASMs were investigated. Seizure symptoms were classified into motor or nonmotor types in accordance with the ILAE 2017 classification.13) Seizures manifesting as only motor impairment or eye deviation were classified as motor seizures whereas those with disturbance or impairment of consciousness as nonmotor seizures. Nonconvulsive status epilepticus (NCSE) was diagnosed in accordance with the Salzburg consensus criteria.14) Standard or portable scalp EEG was performed by a medical technologist on 32 patients, of whom 17 underwent EEG examination within 3 days after seizure occurred. The morphology and location of EEG abnormality were analyzed in accordance with the American Clinical Neurophysiological Society's Standard Critical Care EEG Terminology, 2021 version.15) Consistency of EEG and neuroimaging findings was assessed, specifically for hemispheric laterality and anterior-posterior distribution.
Prophylactic ASM was not administered in principle, except in three patients, and ASM was introduced to patients with motor or nonmotor seizure comfirmed by EEG.
A logistic regression model or Fisher's exact test was employed for statistical analysis using JMP pro16.0 (JMP Statistical Discovery LLC, Cary NC, USA). The study protocol was approved by the Wakayama Medical University Ethical Committee (No. 3536, approval date: Jun 8, 2022). The requirement for informed consent from each patient was waived due to the retrospective nature of the study.
Results
Factors associated with aSAH-related ES and LS are listed in Table 1. Univariate analysis revealed that ES was associated with older age (>75 years), acute severe nonneurological infection, past history of hypertension requiring treatment, symptomatic vasospasm, shunt placement, and frontal lesion. However, it had weak association with cerebrospinal fluid drainage requirement. No significance was observed for sex, past illness related to disease of the central nervous system, intracerebral hematoma, Sylvian hematoma, craniotomy, ventricular tap, electrolyte imbalance, and intraparenchymal lesion volume >10 mL. Conversely, multivariate analysis revealed that acute severe nonneurological infection and older age (>75 years) were significant (Table 2). Patient demographics are presented in Table 3. Ischemia due to vasospasm progressed after ES in five patients (Table 3, Patients 11, 15, 18, 21, and 24).
Table 1.
Demographics and clinical differences between the control, early seizure, and late seizure groups
| Control n = 81 | Early (≤7 days) n = 25 | Late (>7 days) n = 20 | ||||||
|---|---|---|---|---|---|---|---|---|
| Factors | Frequency | Frequency | p value | OR (95% CI) | Frequency | p value | OR (95% CI) | |
| Higher age (>75) | 9 (11.1%) | 13 (52%) | <0.0001 | 16.4 (6.3-42.6) | 5 (25%) | 0.053 | 3.5 (0.99-12.6) | |
| Sex (M/F) | 17/64 | 7/18 | 0.466 | 1.5 (0.5-4.1) | 7/13 | 0.193 | 2.0 (0.7-5.9) | |
| Hypertension | 37 (45.7%) | 21 (84.0%) | 0.002 | 6.2 (2.0-19.8) | 9 (45.0%) | 0.957 | 0.97 (0.4-2.6) | |
| CNS disease | 8 (9.9%) | 4 (16.0%) | 0.403 | 1.74 (0.5-6.3) | 4 (20.0%) | 0.220 | 2.3 (0.6-8.5) | |
| Hunt and Hess grade | Higher HH (IV&V) | 14 (17.3%) | 9 (36.0%) | 0.052 | 2.7 (0.99-7.3) | 9 (45.0%) | 0.011 | 3.9 (1.4-11.2) |
| Fisher CT classification | Groups 3 & 4 | 63 (77.8%) | 24 (96%) | 0.068 | 6.9 (0.9-54.0) | 18 (90%) | 0.233 | 2.6 (0.5-12.1) |
| Intracerebral hematoma | 8 (9.9%) | 5 (20.0%) | 0.186 | 2.3 (0.7-7.7) | 8 (40.0%) | 0.002 | 6.1 (1.9-19.3) | |
| Sylvian hematoma | 5 (6.2%) | 2 (8.0%) | 0.748 | 1.3 (0.2-7.3) | 3 (15.0%) | 0.205 | 2.7 (0.6-12.3) | |
| Aneurysm location | ICA | 24 (29.6%) | 9 (36.0%) | 0.551 | 1.3 (0.5-3.4) | 8 (40.0%) | 0.379 | 1.6 (0.6-4.4) |
| MCA | 23 (28.4%) | 8 (32.0%) | 0.731 | 1.2 (0.5-3.1) | 7 (35.0%) | 0.567 | 1.4 (0.5-3.8) | |
| Acom | 21 (25.9%) | 5 (20.0%) | 0.541 | 0.7 (0.2-2.1) | 4 (20.0%) | 0.576 | 0.7 (0.2-2.4) | |
| Distal ACA | 4 (4.9%) | 0 | NA | 0 | NA | |||
| Posterior fossa | 5 (6.2%) | 2 (8.0%) | 0.748 | 1.3 (0.2-7.3) | 1 (5.0%) | NA | ||
| Multiple | 4 (4.9%) | 1 (4.0%) | NA | 0 | NA | |||
| Treatment | Craniotomy | 36 (44.4%) | 10 (40.0%) | 0.695 | 0.8 (0.3-2.1) | 14 (70.0%) | 0.046 | 12.9 (1.0-8.35) |
| Drainage | 19 (23.5%) | 12 (48.0%) | 0.021 | 3.0 (1.2-7.7) | 11 (55.0%) | 0.008 | 4.0 (1.4-11.1) | |
| Ventricular tap | 16 (19.8%) | 9 (36.0%) | 0.099 | 2.3 (0.9-6.1) | 10 (50.0%) | 0.003 | 5.0 (1.8-14.0) | |
| Shunt placement | 16 (19.8%) | NA | 13 (65.0%) | <0.001 | 7.5 (2.6-22.0) | |||
| Electrolyte imbalance | 7 (8.6%) | 4 (16.0%) | 0.299 | 2.0 (0.5-7.5) | 5 (25.0%) | 0.0528 | 3.5 (0.99-12.6) | |
| Meningitis | 5 (6.1%) | 1 (4.0%) | NA | 5 (25.0%) | 0.0192 | 5.1 (1.3-19.7) | ||
| Acute severe nonneurological infection | 7 (8.6%) | 12 (48.0%) | <0.0001 | 9.8 (3.2-29.4) | 5 (25.0%) | 0.156 | 2.6 (0.7-10.1) | |
| Symptomatic vasospasm | 6 (7.4%) | 9 (36.0%) | 0.001 | 7.0 (2.2-22.6) | 9 (45.0%) | <0.001 | 10.2 (3.1- 34.3) | |
| Neuroimaging | Intraparenchymal lesion volume > 10 mL | 15 (18.5%) | 7 (28.0%) | 0.310 | 1.7 (0.6-4.8) | 14 (70.0%) | <0.0001 | 10.3 (3.4-31.1) |
| Frontal | 7 (8.6%) | 8 (32.0%) | 0.006 | 5.0 (1.6-15.6) | 14 (70.0%) | <0.0001 | 24.7 (7.2-84.5) | |
| Temporal | 8 (9.9%) | 5 (20.0%) | 0.186 | 2.3 (0.7-7.7) | 10 (50.0%) | <0.001 | 9.1 (2.9-28.6) | |
| Parietal | 4 (4.9%) | 1 (4.0%) | NA | 7 (35.0%) | <0.001 | 10.4 (2.7-40.5) | ||
| Polypharmacy | NA | 3 (12.0%) | NA | 9 (45.0%) | NA | |||
| Poor outcome at discharge | mRS 3-6 | 12 (14.8%) | 20 (80.0%) | NA | 13 (65.0%) | <0.0001 | 10.7 (3.5-32.2) | |
Abbreviations used:
M = male, F = female, CI = confidence interval, CNS = central nervous system, HH = Hunt and Hess, ICA = internal carotid artery, MCA = middle cerebral artery, Acom = anterior communicating artery, ACA = anterior cerebral artery, mRS = modified Rankin scale, OR = odds ratio, NA = not applicable
Table 2.
Results of multivariate analysis
| Early (≤7 days) n = 25 | Late (>7 days) n = 20 | ||||
|---|---|---|---|---|---|
| Variable | p value | OR (95% CI) | Variable | p value | OR (95% CI) |
| Acute nonneurological systemic infection | 0.0007 | 9.1 (2.5-32.5) | Intraparenchymal lesion volume> 10 mL | 0.0002 | 10.4 (3.0-35.4) |
| Higher age | 0.0037 | 6.1 (1.8-21.0) | Shunt placement | 0.0012 | 7.6 (2.2-26.0) |
| Hypertension | 0.0604 | 3.4 (0.9-12.5) | |||
Abbreviations used:
OR = odds ratio
Table 3.
Demographics of patients with early seizures and patients with late seizures
| Patient | Age | Sex | Past history | HH | Aneurysm | ICH | Sylvian hematoma |
Treatment | Symptomatic spasm |
Drainage/ shunt | Intraparenchymal lesion | Complication | Seizure | mRS | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HT | CNS | Side | Location | Volume > 10 mL | Side | Fr | T | P | Procedual | EI | Meningitis | Systemic | Phase (day) | Symptom | ASD | |||||||||||
| 1 | 50 | F | + | - | IV | R | MCA | - | - | Coil | - | -/- | - | - | - | - | - | - | - | - | Early | 0 | Motor | 0 | 0 | |
| 2 | 46 | M | + | - | II | R | MCA | - | - | Coil | - | -/- | - | - | - | - | - | - | - | - | Early | 0 | Motor | 1 | 1 | |
| 3 | 48 | F | - | - | II | L | ICA | - | - | Coil | - | -/- | - | - | - | - | - | - | - | - | Early | 0 | Motor | 0 | 1 | |
| 4 | 53 | F | + | - | II | R | MCA | - | - | Clip | + | +/+ | - | - | - | - | - | - | - | - | Early | 0 | Motor | 1 | 5 | |
| 5 | 48 | F | - | - | III | L | ICA | - | - | Coil | - | -/- | - | - | - | - | Small embolism | - | - | - | Early | 0 | Motor | 0 | 0 | |
| 6 | 57 | M | + | - | V | L | VA dissection | - | - | Coil | - | -/- | - | - | - | - | - | + | - | Cellulitis | Early | 0 | Motor | 1 | 3 | |
| 7 | 87 | F | + | - | III | R | Pcom | - | - | Coil | - | -/- | - | - | - | - | - | + | - | - | Early | 0 | Motor | 0 | 2 | |
| 8 | 58 | M | + | - | II | Acom | - | - | Clip | - | +/+ | + | L | + | - | - | - | - | - | - | Early | 0 | Nonmotor | 1 | 3 | |
| 9 | 69 | F | + | Depression | III | R | MCA | + | - | Clip | + | +/- | - | - | - | - | - | - | - | Pneumonia | Early | 1 | Motor | 0 | 4 | |
| 10 | 81 | F | + | - | III | R | MCA | - | - | Clip | + | -/+ | - | - | - | - | - | - | - | - | Early | 1 | Motor | 0 | 4 | |
| 11 | 90 | F | + | Dementia | IV | AcomA | + | - | Coil | + | +/+ | + | R = L | + | + | - | - | - | - | - | Early | 1 | Motor | 1 | 5 | |
| 12 | 80 | M | + | - | III | L | ICA | - | - | Coil | - | +/+ | - | - | - | - | Small embolism | + | - | Pneumonia | Early | 1 | Motor | 0 | 4 | |
| 13 | 93 | F | + | - | II | Acom | - | - | Coil | + | +/+ | - | - | - | - | - | - | - | Pneumonia | Early | 1 | NCSE | 0 | 5 | ||
| 14 | 77 | F | - | - | III | L | ICPC | - | - | Clip | - | -/+ | - | - | - | - | - | - | - | - | Early | 1 | NCSE | *1 | 3 | |
| 15 | 73 | F | + | - | IV | R | ICPC | - | - | Clip | + | +/+ | + | R > L | + | + | - | Edema | - | - | - | Early | 2 | Motor | 3 | 4 |
| 16 | 77 | F | + | - | III | R | MCA, Acom | - | - | Clip | - | +/+ | + | R | + | - | - | Surgical contusion | - | - | Pneumonia, ileus | Early | 2 | Motor | 0 | 5 |
| 17 | 81 | F | + | Cerebral infarction | III | R | MCA | + | + | Clip | - | -/- | - | - | - | - | - | + | - | Pneumonia, cardiac failure | Early | 3 | Motor | 1 | 6 | |
| 18 | 83 | M | + | - | V | R | IC | - | - | Coil | + | +/+ | + | R | + | - | - | - | - | - | Pneumonia, sepsis | Early | 3 | Motor | 2 | 5 |
| 19 | 49 | M | - | - | I | Acom | + | - | Coil | + | -/- | - | + | - | - | - | - | - | - | Early | 3 | Motor | 1 | 4 | ||
| 20 | 86 | F | + | Dementia | II | R | ICA | - | - | Coil | - | +/+ | - | - | - | - | - | - | - | UTI | Early | 3 | NCSE | 1 | 5 | |
| 21 | 72 | M | + | - | I | R | ICPC | - | - | Clip | + | +/- | + | R > L | + | + | - | - | - | + | Pneumonia | Early | 4 | Motor | 0 | 6 |
| 22 | 89 | F | + | - | IV | L | MCA | + | + | Clip | - | -/- | + | L | - | + | - | - | - | - | UTI | Early | 4 | Motor | 1 | 4 |
| 23 | 51 | M | + | - | V | R | VA dissection | - | - | Coil | - | -/+ | - | - | - | - | - | - | - | Pulmonary edema | Early | 4 | Motor | *1 | 5 | |
| 24 | 88 | F | + | - | IV | L | MCA | - | - | Coil | + | -/- | + | L > R | + | + | + | - | - | - | - | Early | 5 | Motor | 1 | 6 |
| 25 | 85 | F | + | - | IV | Acom | - | - | Coil | - | +/+ | - | - | - | - | - | - | - | UTI | Early | 5 | NCSE | 2 | 5 | ||
| 26 | 81 | F | - | - | IV | Acom | - | - | Clip | + | +/+ | + | R = L | + | - | - | - | - | - | Pneumonia | Late | 10 | Motor | 1 | 5 | |
| 27 | 68 | M | + | - | III | R | ICA | - | - | Coil | - | -/- | - | - | - | - | - | + | - | - | Late | 11 | Motor | 2 | 2 | |
| 28 | 80 | F | - | - | II | R | MCA | - | - | Clip | - | +/+ | + | R | + | - | - | Surgical contusion | - | - | Pneumonia | Late | 16 | Nonmotor | 2 | 4 |
| 29 | 48 | F | - | Migraine | II | L | ICA | - | - | Coil | - | -/- | - | L | - | + | - | - | - | - | - | Late | 17 | Nonmotor | 1 | 1 |
| 30 | 87 | F | + | - | II | L | ICPC | - | - | Coil | - | +/+ | - | - | - | - | - | - | - | - | Late | 19 | Motor | 1 | 4 | |
| 31 | 65 | F | + | - | IV | L | MCA | + | - | Coil+ craniotomy | - | -/+ | - | + | + | - | - | - | - | Pneumonia | Late | 21 | Motor | 1 | 4 | |
| 32 | 57 | M | - | - | II | Acom | + | - | Clip | + | +/+ | + | L > R | + | - | - | - | + | + | Pneumonia | Late | 31 | Nonmotor | 1 | 4 | |
| 33 | 72 | F | - | - | III | L | MCA | - | - | Clip+coil | + | +/+ | + | L | + | - | + | Hemorrhage | - | - | - | Late | 36 | Motor | 1 | 2 |
| 34 | 57 | F | + | - | II | R | MCA | - | - | Clip | + | -/- | + | R > L | + | - | - | - | - | - | - | Late | 37 | Motor | 0 | 0 |
| 35 | 94 | F | + | Cerebral infarction | II | L | ICA | - | - | Coil | - | -/- | + | R | - | + | + | - | - | - | Pneumonia | Late | 42 | NCSE | 3 | 5 |
| 36 | 81 | F | + | Cerebral infarction, dementia | IV | R | VA-PICA | - | - | Coil | - | +/+ | - | - | - | - | - | - | + | Pneumonia | Late | 67 | Motor | 1 | 5 | |
| 37 | 67 | F | + | - | IV | R | IC-AchoA | + | - | Trap | + | +/+ | + | R > L | + | + | + | - | - | + | - | Late | 105 | Motor | 3 | 5 |
| 38 | 46 | M | - | - | IV | R | MCA | + | - | Clip | + | +/+ | + | R | + | + | + | - | - | + | - | Late | 6mos | Motor | 2 | 3 |
| 39 | 65 | M | - | - | IV | Acom | - | - | Clip | + | +/+ | + | R | + | - | + | - | - | - | Diabetes insipidus | Late | 7mos | Motor | 2 | 3 | |
| 40 | 53 | M | - | - | III | R | MCA | + | + | Clip | - | -/- | + | R | + | + | - | - | - | - | - | Late | 8mos | Motor | 1 | 2 |
| 41 | 72 | F | + | - | III | L | ICA | + | - | Clip | - | -/+ | - | - | + | - | - | - | - | - | Late | 8mos | Motor | 2 | 4 | |
| 42 | 44 | M | - | Alcoholism | III | R | MCA | - | + | Clip | - | -/- | + | + | + | - | - | - | - | Pulmonary edema | Late | 10mos | Motor | 1 | 1 | |
| 43 | 51 | F | - | - | IV | L | ICPC | + | - | Coil | + | -/- | + | L | + | + | + | - | - | - | - | Late | 18mos | Motor | *2 | 3 |
| 44 | 53 | M | + | - | II | Acom | - | - | Clip | + | +/+ | + | R > L | + | - | + | - | - | + | - | Late | 19mos | Motor | 1 | 2 | |
| 45 | 41 | F | - | - | IV | R | ICA | + | + | Clip | - | +/+ | + | R | + | + | - | - | + | - | - | Late | 2Y | Motor | 2 | 3 |
Abbreviations used:
F = female, M = male, HT = hypertension, CNS = central nervous system, HH = Hunt and Hess, R = right, L = left, MCA = middle cerebral artery, ICA = internal carotid artery, ICPC = internal carotid artery posterior communicating artery, Acom = anterior communicating artery, Pcom = posterior communicating artery, VA = vertebral artery, VA-PICA = vertebral artery-posterior inferior cerebellar artery, IC-AchoA = internal carotid artery-anterior choroidal artery, ICH = intracerebral hematoma, Fr= frontal, T = temporal, P = parietal, EI = electrolyte imbalance, UTI = urinary tract infection, mos = month, Y = year, NCSE = nonconvulsive status epilepticus, ASD = antiseizure drug, mRS = modified Rankin scale, * received prophylactic antiseizure medication
Univariate analysis revealed that LS was associated with intraparenchymal lesion volume >10 mL, frontal and temporal lesions, shunt placement, symptomatic vasospasm, parietal lesion, intracerebral hematoma, ventricular tap, and drainage (in the descending order of odds ratio). In Patient 35 (Table 3), previous cerebral lesions were considered a critical factor for LS. Hunt and Hess grade, meningitis and craniotomy had weak associations with LS. Unlike in ES, older age, acute severe nonneurological infection, and past history of hypertension were insignificant. According to multivariate analysis, intraparenchymal lesion volume >10 mL and shunt placement were significant (Table 2).
As for clinical seizures, the proportion of motor and nonmotor seizures was not different between ES and LS. Five patients were diagnosed with NCSE.
Poly-ASM was more frequently observed in the LS than in the ES group (Fisher's exact test, p = 0.0188). Modified Rankin scale at discharge was poorer in the LS than in the control group. Among patients in the LS group with poor mRS, five had worsened, mainly due to LSs (Table 3, Patients 4, 16, 20, 30, and 35). Three patients had a seizure despite the administration of prophylactic ASM.
The distribution of interictal epileptiform discharge (IED) tended to be different between the groups. The ES group exhibited nonlocalizing IEDs without intraparenchymal lesions, whereas the LS group exhibited localizing IEDs consistent with coarse lesion. Notably, the frontopolar electrode was shown to have high-voltage EEG abnormality in all the examined patients with ES (Table 4). Representative cases are presented in Fig. 1. A 93-year-old woman with ES (Patient 13, Hunt and Hess grade II, ruptured aneurysm treated with coil embolization, NCSE observed 1 day after admission) exhibited diffuse SAH on CT and rhythmic delta activity with sharp waves in the bilateral fronto-centro-temporal area (Fig. 1, left). Conversely, a 94-year-old woman with LS (Patient 35, Hunt and Hess grade II, left IC-PC ruptured aneurysm treated with coil embolization, NCSE observed 43 days after admission) had left Sylvian SAH on CT and a previous, old infarction with perifocal high intensity on diffusion weighted imaging in the right temporoparietal area. EEG revealed a spike or polyspike predominantly on P4, consistent with the previously existing lesion (Fig. 1, right).
Table 4.
Consistency between neuroimaging and EEG findings
| Neuroimaging | EEG | Consistency | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Patient | Phase | Intraparenchymal lesion volume >10 mL (etiology) | Side | Morphology | Location | High voltage | Side | A-P | |
| 2 | Early | - | None (atrophy in R-temporal tip) | Sharp wave | Bilateral Fp, F, C | Fp2 | NA | NA | |
| 8 | Early | + (Early brain injury) | L | Infarction (L-supplementary motor area) | RDA + S | Fp, F | Fp1 | C | C |
| 10 | Early | - | None | RDA + S | Bilateral Fp, C, T | Fp1 | NA | NA | |
| 12 | Early | - | Hematoma (L-uncus), embolism (L-corona radiata) | PDs + S | Bilateral anterior | bilateral Fp, C | NA | NA | |
| 13 | Early | - | None | RDA + S | Bilateral Fp, C, T | Fp, T | NA | NA | |
| 19 | Early | - | R | Small ICH (R-mediobasal frontal) | RDA | Fp1, F3, F7 | Fp1 | N | C |
| 20 | Early | - | Chronic ischemic change, ventricular dilatation | PDs + S | Generalized | Fp, C3, O1, Cz | NA | NA | |
| 27 | Late | - | Brain edema (R-putamen) | Sharp wave | Fp2, F8 | Fp2 | NA | NA | |
| 28 | Late | + (Surgical contusion) | R | Cortico-subcortical edema (R-frontal) | RDA | Bilateral Fp, F | bilateral Fp | C | C |
| 29 | Late | - | L | Small infarction (L-uncus) | Not remarkable | NA | NA | NA | NA |
| 32 | Late | + (Vasospasm) | L > R | Infarction (L>R-medial frontal) | Sharp wave | C3, P3 | C3, P3 | C | C |
| 33 | Late | + (Vasospasm) | L | Infarction (L-fronto-temporo-parietal) | Sharply contoured wave | L-hemisphere | Fp1, T3 | C | C |
| 35 | Late | + (Old infarction) | R | Old infarction (R-parietotemporal) | Spike, sharp wave | R-hemisphere | P4 | C | C |
| 40 | Late | + (ICH, Sylvian hematoma) | R | Hematoma (R-lateral, opercular frontal>L-temporal tip) | Sharply contoured wave | Fp2 | Fp2 | C | C |
| 41 | Late | - | L | Small hematoma (L-temporal, R-frontal (ventricular tap) ) | RDA | L-hemisphere | Fp1, F3, F7 | NA | NA |
| 44 | Late | + (Vasospasm) | R > L | Infarction (R-frontobasal, dorsolateral, temporoparietal, L-parietal) | PDs + S | Fp2, C4, T4 | Fp2 | C | C |
| 45 | Late | + (ICH, Sylvian hematoma) | R | Hematoma (R-lateral, basal frontal) | PDs + S | Fp2, F4, T4 | Fp2, F4, T4 | C | C |
Abbreviations used:
ICH = intracerebral hemorrhage, R = right, L = left, RDA = rhythmic delta activity, PDs = periodic discharges, S = sharp waves or spikes, NA = not applicable, N = not consistent, C = consistent, A-P = anterior-posterior
Fig. 1.
Neuroimaging and electroencephalogram of representative cases of early seizure (A) and late seizure (B). Left: Diffuse subarachnoid hemorrhage on CT (upper) and rhythmic delta activity + sharp on bilateral frontopolar, C, T (lower). (Patient 13, 93-year-old woman, Hunt and Hess grade II; Acom ruptured aneurysm treated with coil embolization; NCSE was observed 1 day after admission). Right: Left Sylvian subarachnoid hemorrhage on CT and the previous old infarction and perifocal high intensity in the right temporoparietal area on diffusion weighted imaging. EEG shows spike or polyspike predominantly on P4, consistent with the previously existing lesion (Patient 35, 94-year-old woman, Hunt and Hess grade II, left internal carotid artery posterior communicating artery ruptured aneurysm treated with coil embolization, NCSE was observed 43 days after admission).
Discussion
The main finding of this study is the difference in factors associated with aSAH-related seizure between ES and LS.
Regarding ES, patient's systemic factors, such as older age and acute severe nonneurological infection, seem important. In a recent report of a multicenter registry study, older patients had a greater risk for epilepsy in aSAH.5) In general, aSAH occurs at a younger age than intracerebral hemorrhage or cerebral infarction, although its onset in elderly people has been increasing, even in aSAH, along with the aging of society.16) Elderly people are considered vulnerable to post-stroke epilepsy,17) which might reduce the quality of life. Hypertension is considered to be associated with post-stroke epilepsy after aSAH.18) Middle-aged hypertension might cause various types of brain damage, such as gliosis, blood-brain barrier disruption, and amyloid deposition.19) Patients with antiplatelet therapy more frequently experienced ES in SAH.5) The risk factors for cerebral infarction might lower the threshold of epileptic seizure. Furthermore, acute severe nonneurological infection may lower the threshold of seizure occurrence. Vasospasm is a distinctive condition associated with aSAH, which reaches its peak 7 to 10 days after the stroke onset. In many of our patients, ES seemed to occur as a prodromal symptom followed by severe symptomatic vasospasm. The causal association between ES, vasospasm, and acute systemic infection is unclear, but all three influence each other and need to be comprehensively treated in the acute phase of aSAH.
Intraparenchymal lesion volume >10 mL caused by intracerebral hemorrhage or vasospasm seems to be responsible in part for LS. Previously existing cerebral lesions should also be a reason for caution regarding aSAH-related seizures.5) Aneurysm location was not significant in our study, but a literature review reported that LS was more likely to occur in middle cerebral artery aneurysm.6) In cases with internal carotid or middle cerebral artery aneurysm, regions with low seizure threshold, such as the face motor area, insula, or hippocampus, could be exposed by aSAH or intraparenchymal hematoma. In addition, shunt placement in association with LS should not be ignored. In general, hydrocephalus occurs in cases with thick SAH. In such cases, a considerable amount of hemosiderin, which enhances neuronal hyperexcitability,10,11) would accumulate on the cortical surface due to insufficient cerebrospinal fluid circulation. Similar to International Subarachnoid Aneurysm Trial reports,20) LS was frequently observed in our patients who underwent craniotomy. Endovascular surgery requiring heparinization may have been avoided in patients with intraparenchymal hemorrhage owing to the risk of enlargement of hematoma. 4Within the LS group with poor modified Rankin scale at discharge, some patients exhibited neurological deterioration mainly due to LS, not the severity of aSAH. It is noteworthy that LS could occur over a month after aSAH in a well-operated patient, and that diagnosis might be difficult.
Neuroimaging and EEG findings have not been previously investigated in detail for aSAH-related seizures. In general, aSAH tends to distribute in the basal cistern and then spreads to the remote subarachnoid space along the main trunk. It occasionally causes intraparenchymal hemorrhage or symptomatic vasospasm along the parent artery. Therefore, abnormal EEG findings may show characteristic distribution in aSAH-related hemorrhage. In the present study, we observed a tendency of difference between ES and LS; ES had fewer intraparenchymal lesions and exhibited diffuse EEG abnormality, whereas LS frequently had intraparenchymal lesions consistent with EEG abnormality. These EEG findings suggest that the main epileptogenesis of aSAH-related LS is intraparenchymal lesions associated with aneurysmal rupture. Consequently, EEG abnormality could be easily recorded at the frontopolar or temporal region. EEG monitoring has been reported to be effective for cerebral strokes,21,22) which may also be true for aSAH-related seizures. As the frontopolar electrode was highly sensitive to ES, simple EEG monitoring using the frontopolar electrode may be beneficial for the administration of ASM for ES.
As regards ASM, the rate of polypharmacy was significantly higher in patients with LS. The American Heart Association/American Stroke Association guidelines recommended considering prophylactic administration of ASM and fast withdrawal for aSAH.23) Alternatively, one study reported no correlation between prophylactic ASM under continuous EEG monitoring and functional outcome.24) No standardized methods for prophylactic ASM have been developed, and in our limited cases, seizures actually occurred under prophylactic ASM.
Regarding limitations, this study was retrospective in nature with a small sample size. There were missing cases over the 2-year follow-up, and nonmotor cases and NCSE were potentially overlooked. The proportion of patients with ES and LS to all aSAH patients in this study was similar to those in previous studies,7,8) but lower than that in a recent study.25) EEG could not be promptly performed in all patients following seizure occurrence. The transition from ES to LS could not be investigated as there was insufficient follow-up of a considerable number of ES cases. The frequency of the transition from ES to LS remains unclear, although seizures at presentation or onset seizures are not considered to be associated with LS according to previous reports.6,8) The decision to discontinue ASM is critical as seizure recurrence could cause functional decline in stroke survivors.26) This requires clarification by longitudinal studies. Considering the prevalence of aSAH-related seizures, multicenter prospective studies are warranted to elucidate the risk factors and other issues for aSAH-related seizures.
Conclusion
Factors associated with aSAH-related seizures seem to differ between ES and LS. ES is perhaps induced by the patient's systemic factors such as older age and general condition, and this may lower the threshold for cortical excitability. Patients with intraparenchymal lesions undergoing shunt placement should be carefully monitored for potential LS. EEG findings on the frontopolar electrode may be useful for administering ASM to patients with aSAH-related seizures, particularly those with ES.
Conflicts of Interest Disclosure
All the authors report that there are no conflicts of interests to declare in association with this manuscript.
Acknowledgments
We thank EEG technologists at Wakayama Medical University for their contributions.
We acknowledge proofreading and editing by Benjamin Phillis at the Clinical Study Support Center at Wakayama Medical University.
References
- 1). Beghi E, Carpio A, Forsgren L, et al. : Recommendation for a definition of acute symptomatic seizure. Epilepsia 51: 671-675, 2010 [DOI] [PubMed] [Google Scholar]
- 2). Tanaka T, Ihara M: Post-stroke epilepsy. Neurochem Int 107: 219-228, 2017 [DOI] [PubMed] [Google Scholar]
- 3). Galovic M, Dohler N, Erdelyi-Canavese B, 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 Neruol 17: 143-152, 2018 [DOI] [PubMed] [Google Scholar]
- 4). Szaflarski JP, Rackley AY, Kleindofer DO, et al. : Incidence of seizures in the acute phase of stroke: A population-based study. Epilepsia 49: 974-981, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5). Zöllner JP, Konczalla J, Stein M, et al. : Acute symptomatic seizures in intracerebral and subarachnoid hemorrhage: A population study of 19,331 patients. Epilepsy Res 161: 106286, 2020 [DOI] [PubMed] [Google Scholar]
- 6). Raper DM, Starke RM, Komotar RJ, Allan R, Connoly ES: Seizures after aneurysmal subarachnoid hemorrhage: A systemic review of outcomes. World Neurosurg 79: 682690, 2013 [DOI] [PubMed] [Google Scholar]
- 7). Lin CL, Dumont AS, Lieu AS, et al. : Characterization of perioperative seizures and epilepsy following aneurysmal subarachnoid hemorrhage. J Neurosurg 99: 978-985, 2003 [DOI] [PubMed] [Google Scholar]
- 8). Choi KS, Chun HJ, Yi HJ, Ko Y, Kim YS, Kim JM: Seizures and epilepsy following aneurysmal subarachnoid hemorrhage: incidence and risk factors. J Korean Neurosurg Soc 46: 93-98, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9). Shimoda M, Hoshikawa K, Shiramizu H, et al. : Early infarction detected by diffusion-weighted imaging in patients with subarachnoid hemorrhage. Acta Neurochir 152: 1197-1205, 2010 [DOI] [PubMed] [Google Scholar]
- 10). Tanaka T, Fukuma K, Abe S, et al. : Association of cortical superficial siderosis with post-stroke epilepsy. Ann Neurol 93: 357-370, 2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11). Hirano T, Enatsu R, Iihoshi S, et al. : Effects of hemosiderosis on epilepsy following subarachnoid hemorrhage. Neurol Med Chir (Tokyo) 59: 27-32, 2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12). Haapaniemi E, Strbian D, Rossi C, et al. : The CAVE score for predicting late seizures after intracerebral hemorrhage. Stroke 45: 1971-1976, 2014 [DOI] [PubMed] [Google Scholar]
- 13). Fisher RS, Cross JH, French JA, et al. : Operational classification of seizure types by the International League Against Epilepsy: position paper of the ILAE Commission for Classification and terminology. Epilepsia 58: 522-530, 2017 [DOI] [PubMed] [Google Scholar]
- 14). Beniczky S, Hirsch LJ, Kaplan PW, et al. : Unified EEG terminology and criteria for nonconvulsive status epilepticus. Epilepsia 54: 28-29, 2013 [DOI] [PubMed] [Google Scholar]
- 15). Hirsch LJ, Fong MWK, Leitinger M, et al. : American Clinical Neurophysiology Society's standardized critical care EEG terminology: 2021 version. J Clin Neurophysiol 38: 1-29, 2021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16). Kaminogo M, Yonekura M: Trends in subarachnoid haemorrhage in elderly persons from Nagasaki, Japan: Analysis of the Nagasaki SAH Data Bank for cerebral aneurysm, 1989-1998. Acta Neurochir 144: 1133-1139, 2002 [DOI] [PubMed] [Google Scholar]
- 17). Sen A, Jette N, Husain M, Sander JW: Epilepsy in older people. Lancet 395: 735-748, 2020 [DOI] [PubMed] [Google Scholar]
- 18). Ohman J: Hypertension as a risk factor for epilepsy after aneurysmal subarachnoid hemorrhage and surgery. Neurosurgery 27: 578-581, 1990 [DOI] [PubMed] [Google Scholar]
- 19). Johnson EL, Krauss GL, Lee AK, et al. : Association between midlife risk factors and late-onset epilepsy: results from the atherosclerosis risk in communities study. JAMA Neurol 75: 1375-1382, 2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20). Hoh BL, Nathoo S, Chi YY, Mocco J, Baker FG 2nd: Incidence of seizures or epilepsy after clipping or coiling of ruptured and unruptured cerebral aneurysms in the nationwide inpatient sample database: 2002-2007. Neurosurgery 69: 644-650, 2011 [DOI] [PubMed] [Google Scholar]
- 21). Bentes C, Peralta AR, Martins H, et al. : Seizures, electroencephalographic abnormalities, and outcome of ischemic stroke patients. Epilepsia Open 2: 441-452, 2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22). Bentes C, Martins H, Peralta AR, et al. : Early EEG predicts poststroke epilepsy. Epilepsia Open 3: 203-212, 2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23). Connolly ES, Rabinstein AA, Carhuapoma JR, et al. : Guidelines for the management of aneurysmal subarachnoid hemorrhage. a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 43: 1711-1737, 2012 [DOI] [PubMed] [Google Scholar]
- 24). Zafar SF, Rosenthal ES, Postma EN, et al. : Antiseizure medication treatment and outcomes in patients with subarachnoid hemorrhage undergoing continuous EEG monitoring. Neurocrit Care 36: 857-867, 2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25). Vychopen M, Lampmann T, Asoglu H, et al. : Non-convulsive status epilepticus in aneurysmal subarachnoid hemorrhage: a prognostic parameter. Brain Sci 13: 184, 2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26). Yoshimura H, Tanaka T, Fukuma K, et al. : Impact of seizure recurrence on 1-year functional outcome and mortality in patients with poststroke epilepsy. Neurology 99: e376-e384, 2022 [DOI] [PubMed] [Google Scholar]