Keywords
Arterial stroke - CVST - Seizure - Post stroke seizure
Abbreviations
CVST:
cerebral venous sinus thrombosis
CT:
computerized tomography
MRI:
magnetic resonance imaging
MRV:
MR venography
NIHSS:
National Institute of Health Stroke Scale
mRS:
modified rankin scale
CNS:
central nervous system
TOAST:
trial of ORG 1072 in acute stroke treatment
ICH:
intracerebral hemorrhage
SE:
status epilepticus
GCS:
Glasgow coma scale
EEG:
electroencephalography
ICP:
intracranial pressure
MV:
mechanical ventilation
1
Introduction
Stroke is the most important cause of mortality and disability after coronary artery
disease and cancer. The incidence of stroke is likely to increase with the increasing
age of the population. Cerebrovascular disease is the most important cause of epilepsy
in elderly population.[1]
[2] It has been estimated that 6.9–11.5% patients with stroke were at the risk of developing
post stroke seizure and the incidence of seizures increased with age.[3]
[4]
[5] Presence of structural brain lesion, EEG abnormalities and partial seizures may
have higher recurrence rate.[6] Male gender and cortical location of stroke independently predicted increased risk
of seizures. In India, the stroke occurs at younger age because of high incidence
of rheumatic heart disease, central nervous system (CNS) infection and high prevalence
of cerebral venous sinus thrombosis (CVST). In CVST, seizures are reported in 12–46.7%
patients whereas 44.3% of patients may have early seizures.[7]
[8]
[9] The patients with sensory motor deficit, cortical vein thrombosis, and supratentorial
lesion on CT or MRI are more likely to have seizures compared to those without these
features.[7]
[9]
[10] The pathophysiology of seizures in arterial and venous stroke is different. There
is no study comparing frequency, spectrum and predictors of seizure in arterial and
venous stroke. In the present study, we report the spectrum, predictors and prognosis
of patients with seizures in arterial and venous stroke.
2
Subjects and methods
In a hospital based observational study, the patients with stroke and CVST during
2010–2015 were included. The study was approved by the Institute Ethics Committee.
(PGI/BE/774)
2.1
Definitions
Stroke was defined as “rapidly developing clinical signs of focal (or global) disturbance
of cerebral function, lasting more than 24 h or leading to death, with no apparent
cause other than that of vascular origin”.[11] The diagnosis of stroke was confirmed on CT/MRI scan. To have a uniform classification,
Oxford classification was followed to define the arterial territory of ischemic stroke.[12] The etiological subtypes of ischemic stroke were classified according to the Trial
of ORG 1072 in Acute Stroke Treatment (TOAST) criteria. The ischemic stroke was categorized
into small vessel stroke, large vessel stroke, cardioembolic, stroke of undetermined
source and stroke of other determined source.[13] The hemorrhagic stroke was classified as primary ICH as lobar, ganglionic (caudate,
thalamic, putaminal), pontine and cerebellar. The size and intrventricular extension
of hematoma were noted.[14] The patients with CVST were defined on the basis of magnetic resonance venography
(MRV). The location and extent of thrombosis and number of sinuses involved were noted.
Presence of parenchymal lesions and its nature (infarction, hemorrhagic infarction
or hemorrhage) were noted.
2.2
Exclusion
The patients with trauma, vascular malformation, coagulopathy, malignancy and tumor
bleed were excluded. The patients with past history of seizures, febrile convulsion
and associated structural lesion other than arterial stroke or CVST were also excluded.
2.3
Seizure
The seizures were defined as early seizure if occurred within 14 days of stroke onset,
and late seizure if occurred after 14 day. The seizures occurring before the diagnosis
of stroke were defined as presenting seizure. The presenting seizure was included
in early seizure for comparison between early seizure and late seizures. Early seizure
witnessed by doctor or paramedical staff was documented and EEG was done. Late seizure
was documented in subsequent follow up of patients. Status epilepticus (SE) was categorized
into convulsive and nonconvulsive. Convulsive SE was defined as 5 min or more of continuous
clinical and/or electrographic seizure or recurrent seizure without recovery of consciousness
to baseline between the seizures. Nonconvulsive SE was defined as alteration of consciousness
lasting for 30 min or more with epileptiform discharges in EEG with or without suppression
of EEG activity by IV benzodiazepine.
2.4
Evaluation
Stroke risk factors such as diabetes, hypertension, hyperlipidemia, heart disease,
obesity and hyperhomocysteinemia were noted. The patient was considered hypertensive
if there was documented history of hypertension, received antihypertensive treatment
or blood pressure was above 140/90 mm of Hg after 2 weeks of stroke or on follow up.[15] The patients were considered diabetic if they were on antidiabetic treatment or
fasting blood sugar was ≥126 mg/dl or 2 h post prandial blood sugar ≥200 mg/dl. The
underlying etiology of CVST such as pregnancy, oral contraceptive pills, factor V
Leiden mutation, prothombin gene mutation, hyperhomocysteinemia, antinuclear antibody,
antiphospholipid antibody and hematological causes (anemia, paroxysmal nocturnal hematuria,
polycythemia, and thrombocytopenia) were noted.
Consciousness was assessed by Glasgow Coma Scale (GCS). Presence of cranial nerve
palsy and fundus abnormalities was noted. Focal weakness was categorized as hemiplegia,
monoplegia or quadriplegia (as partial or complete weakness). Muscle tone and tendon
reflex were graded as increased, normal or decreased. Coordination and sensations
were also tested in the patients who could co-operate. Severity of stroke was graded
using NIHSS score. Outcome on discharge was defined by modified Rankin Scale (mRS)
as good (≤2) or poor (>2).
2.5
Treatment
The patients were given symptomatic and supportive treatment including management
of underlying cause. Seizures were treated with antiepileptic drugs and SE by 0.1
mg/kg lorazepam IV followed by sodium valproate 20 mg/kg, levetiracetam 30 mg/kg or
phenytoin 10 mg/kg IV. Raised intracranial pressure (ICP) was treated by mannitol,
respiratory failure by mechanical ventilation, fever by cold sponging and paracetamol.
Fluid, electrolyte and calories were administered by nasogastric and/or intravenous
route. The patients with arterial stroke received management of the risk factors.
The patients with CVST were treated with low molecular weight heparin (LMWH) or unfractionated
heparin (UFH) for 14 days followed by oral anticoagulant to maintain INR of 2–2.5.
3
Results
Our results are based on 128 patients with CVST and 870 patients with arterial stroke
which were admitted during 2010–2015. Seizures were present in 74 (57.8%) patients
with CVST and 119 (13.7%) with arterial stroke. In the patients with arterial stroke,
29 patients had ICH; 18 (62.1%) of them had early and 11(37.9%) had late seizures.
The hematomas were lobar in 18 and ganglionic/thalamic in 19 patients. In ischemic
arterial stroke, 51 (56.7%) patients had late seizure and 39 (43.3%) had early seizure,
and status epilepticus was present in 24 (26.7%). Status epilepticus was present in
both CVST (24.3%) and arterial (25.2%) stroke. Secondary generalized seizures were
more common in arterial stroke (90.8%) than CVST (81.1%) (p = 0.05). Early seizures
were also more common in CVST (98.6%) compared to arterial (47.9% p = 0.001), whereas
late seizures were more common in arterial (52.1%) compared to CVST (1.4%; p = 0.001).
Presenting seizures resulted in SE more commonly in arterial (19.3%) than CVST (16.2%).
Nonconvulsive SE was insignificantly more common in arterial stroke (p = 0.23) ([Table 1]). The mean follow up of venous stroke patients with seizure was 26.41 ± 12.99 month
and 4 patients lost from follow up, while in arterial stroke patients, the mean follow
up was 19.40 ± 10.46 month and 28 (23.5%) patients lost from follow up.
Table 1
Comparison of demographic and clinical parameters in arterial stroke and cerebral
venous sinus thrombosis (CVST) with seizure.
|
Parameters
|
CVST, n = 74 (%)
|
Arterial, n = 119 (%)
|
P value
|
|
GCS = Glasgow Coma Scale, mRS = Modified Rankin Scale.
|
|
Age (in years)
|
31.9 ± 12.5
|
54.6 ± 16.3
|
<0.001
|
|
|
Female
|
34 (45.9)
|
39 (32.8)
|
0.07
|
|
|
Seizures-
|
|
|
|
|
Early
|
73 (98.6)
|
57 (47.9)
|
0.001
|
|
Late
|
1 (1.4)
|
62 (52.1)
|
|
|
|
Status epilepticus
|
18(24.3)
|
30 (25.2)
|
1.00
|
|
|
Focal deficit
|
48 (64.9)
|
106 (89.1)
|
0.001
|
|
|
Admission GCS (mean ± SD)
|
12.11 ± 3.50
|
12.28 ± 3.044
|
0.72
|
|
|
Mechanical ventilation
|
9 (12.2)
|
6 (5)
|
0.09
|
|
|
Types of lesion
|
|
|
|
|
Ischemic
|
13 (17.6)
|
90 (75.6)
|
0.001
|
|
Hemorrhage
|
49 (66.2)
|
29 (24.4)
|
|
|
|
3 month mRS
|
|
|
|
|
Good (≤2)
|
58 (78.4)
|
53 (44.5)
|
0.001
|
|
Poor (>2)
|
16 (21.6)
|
66 (55.5)
|
|
|
|
Death
|
10 (13.5)
|
3 (2.5)
|
0.006
|
3.1
Predictors of seizures
Comparing the arterial stroke patients with and without seizure, the type of stroke
(hemorrhagic vs. ischemic; p = 0.001), arterial territory of ischemic stroke (anterior
vs. posterior circulation; p = 0.008), admission GCS score (p = 0.004), admission
NIHSS score (p = 0.002), outcome on discharge (p = 0.005) and age (p = 0.01) were
the predictors of seizure on univariate analysis. On multivariate analysis, anterior
territory of ischemic stroke was independent predictor of seizure (OR 3.95, 95% CI
1.51–10.32, p = 0.005).
Comparing the CVST patients with and without seizure, focal deficit (p = 0.004), parenchymal
lesion (p = 0.001), frontal lobe involvement (p = 0.003) and papilledema (p = 0.01)
were the predictors of seizure on univariate analysis. On multivariate analysis presence
of parenchymal lesion was independent predictor of seizure (OR 2.61, 95% CI 1.04–6.55,
p = 0.04) ([Table 2]).
Table 2
Comparison of CVST (cerebral venous sinus thrombosis) patients with and without seizure.
|
Parameters
|
CVST with seizure, n = 74(%)
|
CVST without seizure, n = 54(%)
|
P value
|
|
GCS = Glasgow Coma Scale, mRS = Modified Rankin Scale.
|
|
Age (mean ± SD)
|
31.86 ± 12.50
|
33.76 ± 16.12
|
0.47
|
|
|
Sex (Female)
|
34 (45.9)
|
32 (59.3)
|
0.16
|
|
|
GCS score at admission (mean ± SD)
|
12.11 ± 3.50
|
13.09 ± 3.03
|
0.09
|
|
|
Ventilation
|
9 (12.2)
|
4 (7.4)
|
0.56
|
|
|
Duration of hospitalization (mean ± SD)
|
17.89 ± 11.77
|
17.41 ± 8.95
|
0.80
|
|
|
Focal deficit
|
48 (64.9)
|
21 (38.9)
|
0.004
|
|
|
Papilloedema
|
15 (20.3)
|
23 (42.6)
|
0.01
|
|
|
Parenchymal lesion
|
63 (85.1)
|
31 (57.4)
|
0.001
|
|
|
Frontal lobe
|
29 (39.2)
|
8 (14.8)
|
0.003
|
|
|
Type of lesion
|
|
|
|
|
Ischemic
|
13 (17.6)
|
11 (20.4)
|
0.14
|
|
Hemorrhagic
|
49 (66.2)
|
20 (37)
|
|
|
3 month mRS
|
|
|
|
|
Good (≤2)
|
58 (78.4)
|
45 (83.3)
|
0.51
|
|
Poor (>2)
|
16 (21.6)
|
9 (16.7)
|
|
|
Death
|
10 (13.5)
|
6 (11.1)
|
0.80
|
3.2
Outcome
There were more deaths in CVST compared to arterial stroke (13.5% vs. 2.5%; p = 0.006).
The functional outcome however was better in the CVST patients with seizure compared
to arterial stroke (73% vs. 35.3%, p = 0.001) ([Fig. 1]).
Fig. 1– Bar diagram showing outcome (mRS grade–Good ≤ 2, Poor > 2) at discharge and
three month in patients of CVST and arterial stroke with seizure.
On comparing the CVST patients with seizure, with good (mRS up to 2) and poor (mRS
3–6) outcome at 3 month follow up; GCS score on admission (p = 0.001), MV (p = 0.001),
focal weakness (p = 0.03), bilateral lesions (p = 0.03), and involvement of both superficial
and deep sinus (p = 0.03) were the predictors of poor outcome on univariate analysis.
Using binary logistic regression, involvement of both superficial and deep sinuses
(OR 63.47, CI 1.36–2946.69, p = 0.03), lower admission GCS (OR 1.92, 95% CI 1.25–2.96,
p = 0.003) and use of MV (OR 39.85, 95% CI 2.95–537.33, p = 0.005) were significant
predictors of poor outcome ([Table 3]).
Table 3
Comparison of outcome of CVST (cerebral venous sinus thrombosis) patients with seizure
at 3 month.
|
Parameters
|
CVST, (mRS > 2)
N = 16 (%)
|
CVST (mRS ≤ 2)
N = 58 (%)
|
P value
|
|
Age (mean ± SD)
|
35.31 ± 15.37
|
30.91 ± 11.56
|
0.30
|
|
Sex (Female)
|
7 (43.8)
|
27(46.6)
|
0.84
|
|
Altered sensorium
|
13 (81.3%)
|
33 (56.9)
|
0.075
|
|
Admission GCS (mean ± SD)
|
8.75 ± 4.10
|
13.03 ± 2.68
|
0.001
|
|
Ventilation
|
8 (50)
|
1 (1.7)
|
0.001
|
|
Duration of hospitalization (mean ± SD)
|
16.18 ± 17.75
|
18.41 ± 9.44
|
0.63
|
|
Focal deficit
|
14 (87.5)
|
34 (58.6)
|
0.03
|
|
Papilloedema
|
4 (25)
|
11 (19)
|
0.60
|
|
Parenchymal lesion
|
15 (93.8)
|
48 (82.8)
|
0.27
|
|
Type of lesion
|
|
|
|
|
Ischemic
|
4 (25)
|
9 (15.5)
|
0.53
|
|
Hemorrhagic
|
11(68.8)
|
38 (65.5)
|
|
|
Bilateral lesions
|
8 (50)
|
13 (22.4)
|
0.03
|
|
Total number of sinus involved (mean ± SD)
|
2.06 ± 1.28
|
1.98 ± 1.23
|
0.82
|
|
Superficial sinus
|
12 (75)
|
51 (87.9)
|
0.19
|
|
Deep sinus
|
0
|
3 (5.2)
|
0.35
|
|
Both superficial and deep sinus
|
4 (25.0)
|
4 (6.9)
|
0.03
|
On comparing the arterial patients with good (mRS up to 2) and poor (mRS 3–6) outcome
on discharge; admission GCS score (p = 0.002), NIHSS score on admission (p = 0.001)
and duration of hospitalization (p = 0.002) were the predictors of outcome on univariate
analysis. Using binary logistic regression, GCS score (OR 1.37, CI 1.02–1.84, p =
0.03) and NIHSS on admission (OR 1.42, 95% CI 1.19–1.70, p = 0.001) and duration of
hospitalization (OR 1.15, 95% CI 1.02–1.29, p = 0.023) were significant predictors
of outcome ([Table 4]).
Table 4
Comparison of outcome in the patients of arterial stroke with seizure at discharge.
|
Parameters
|
Arterial stroke (mRS ≤ 2)
N = 42 (%)
|
Arterial stroke, (mRS > 2)
N = 77 (%)
|
P value
|
|
Age (mean ± SD)
|
51.55 ± 15.86
|
56.29 ± 16.33
|
0.12
|
|
|
Sex (Female)
|
18 (42.9)
|
21 (27.3)
|
0.08
|
|
|
Hospitalization duration (mean ± SD)
|
7.90 ± 4.23
|
18.29 ± 17.93
|
0.002
|
|
|
NIHSS score admission (mean ± SD)
|
6.95 ± 4.42
|
14.41 ± 6.92
|
0.001
|
|
|
GCS score at admission (mean ± SD)
|
13.37 ± 2.25
|
11.70 ± 3.25
|
0.002
|
|
|
Ventilation
|
2 (4.8)
|
4 (5.2)
|
0.91
|
|
|
Focal deficit
|
36 (85.7)
|
70 (90.9)
|
0.38
|
|
|
Death
|
0
|
3 (3.9)
|
0.19
|
|
|
Type of stroke
|
|
|
|
|
Ischemic
|
33 (78.6)
|
57 (74)
|
0.66
|
|
Hemorrhagic
|
9 (21.4)
|
20 (26)
|
|
|
|
Type of ischemic stroke
|
|
|
|
|
Anterior circulation
|
26 (61.9)
|
49 (63.6)
|
0.25
|
|
Posterior circulation
|
8 (19)
|
8 (10.4)
|
|
4
Discussion
In the present study, the seizures were more common in CVST (57.8%) than in arterial
stroke (13.7%). CVST related seizures were more intractable and resulted in higher
mortality but the survivors had better outcome than arterial stroke. The arterial
strokes resulted in more frequent late seizures. The predictor of seizure in CVST
was parenchymal involvement, and in arterial stroke was carotid territory infarction.
This is the first study comparing the frequency and spectrum of seizure in arterial
and venous stroke and its effect on outcome. The frequency of seizure in the present
study in arterial stroke is in agreement with the reported frequency of post stroke
seizures which is up to 11.5%[3]
[4] and in CVST up to 43.5%.[7] Slightly higher frequency of seizures in CVST in the present study may be due to
difference in patient population. In ischemic stroke, severity of neurological deficit,
severity of persistent disability after stroke, involvement of multiple sites, large
lesion, cortical damage or hippocampal involvement predicted development of post stroke
seizure.[6] In our study, 29 patients (24.4%) with ICH had seizures; early seizure in 62.1%
and late seizure in 37.9% patients. Similar results have been found in other studies
as well. In a study on 123 patients with ICH followed up for a period of 4.6 years,
seizure occurred in 25% of patients. In the more recent studies, the seizure frequency
is higher in ICH compared to the earlier, because small lobar hemorrhages are more
likely to produce seizure which can be detected by CT scan.[16] The products of blood metabolites such as hemosiderin and iron may cause seizure
by local cerebral irritation.[17] The epileptogenic lobar ICH may involve the gray-white matter interface, creating
a condition similar to the surgical isolation of cortex resulting in its sustained
paroxysmal activity. Direct irritation of the cortex is also a possible mechanism
of post seizures in ICH.[18] The higher frequency of seizures in hemorrhagic infarction both in CVST and arterial
strokes have been reported.[9]
In prospective studies, seizure was not related to severity of stroke.[19]
[20] In a Chinese study, cortical location of stroke was associated with 3.8 times higher
risk of seizure.[21]
In our study, SE is the predictor of poor outcome on discharge in arterial stroke
patients but not CVST. Development of epilepsy and the independent effect of SE on
mortality is controversial.[22]
[23]
[24] In our study, CVST had higher mortality (13.5%) than arterial stroke (2.5%). Severity
of stroke is a predictor of outcome but the effect of seizure on outcome of stroke
is uncertain. In a prospective cohort study, 48 h mortality was found to be higher
in the patients with early seizures (38.4%) compared to those without early seizure
(7.4%; p < 0.01).[25] In this study the effect of location and stroke severity were not adjusted. Moreover
the effect on mortality rates was not significant at the time of discharge or follow
up at 27 months. A large hospital based study reported that early seizures were associated
with increase in hospital mortality in atherothrombotic infarction.[26] However after accounting for stroke severity, population based studies have not
found association between early post stroke seizure and mortality.[20]
[27] Higher death in CVST and worse outcome in the survivors of ischemic stroke may be
related to underlying pathophysiology of arterial and venous stroke.
The mechanism of seizures in stoke depends on the timing. The early seizures are attributable
to biochemical alterations leading to electrical instability.[28] Acute ischemia results in increased extracellular concentration of glutamate which
results in secondary neuronal injury. Epileptiform neuronal discharges can occur in
surviving neurons exposed to high levels of glutamate.[29] Late seizures are attributable to gliosis and meningocerebral cicatrix.[30] Changes in membrane properties, selective neuronal loss and collateral sprouting
may result in hyperexcitabilty and seizures.[31] The pathophysiological mechanisms in venous and arterial stroke also differ. CVST
is less acute, Occlusion of venous sinuses results in stagnation of blood, hypoxia,
diapediasis of RBC, edema, raised intracranial pressure and venous infarction. Oxidative
stress, apoptosis, inflammation, necrosis and endoplasmic reticulum stress have been
reported in CVST which manifest at a slower rate than arterial stroke.[32] Following therapy in CVST or even spontaneously, there may be thrombolysis or recanalization
of venous sinuses which may account for the lower incidence of late seizures in CVST
and fewer sequelae compared to arterial stroke. The arterial stroke on the other hand
is more acute and results in persistent changes, including necrosis, apoptosis and
infarction especially involving epileptogenic area of cortex account for higher frequency
of late post stroke seizure in arterial stroke.
Ours is the first study comparing the frequency, spectrum and outcome of seizures
in arterial and venous stroke. This study is however limited by retrospective design.
This hospital based study has a referral bias of a tertiary care hospital. In which
the patients included are more sick and do not represent the patients in the community.
It can be concluded that CVST results in more frequent and early seizures whereas
in arterial stroke, late seizures are common. Though CVST had higher mortality but
their survivors have better outcome.
Conflict of interest
The authors have none to declare.
Ethical approval
This study was approved by Institutional Ethics Committee, SGPGIMS, Lucknow INDIA.
