 |
|
||||
| Home | Login | |||||
| About | Editorial board | Articles
|
NSI Publications
|
Search | Instructions | Online Submission | Subscribe | Videos | Etcetera | Contact |
|
||||||||||||||||||||
|
|
|
Seizures after stroke : a prospective clinical study.
Correspondence Address:
Stroke is one of the most common causes of epilepsy in elderly. However, there have been very few prospective studies to define the incidence, pattern and outcome of seizures in stroke. Most studies are based on retrospective analysis of hospital records. Hence, we planned this prospective study to see the clinical, radiological and electroencephalographic characteristics of seizures in stroke and their outcome, from a north Indian tertiary care centre. Over a span of approximately 6 years, 269 consecutive patients with stroke were studied and followed up. Thirty-five (13%) of these developed seizures, primarily related to stroke, during mean follow up period of 15.9 months. Twenty of these had infarctions while 15 had haemorrhages. Involvement of the cortical region was seen in most of the patients with seizures. In these patients, 86% of the lesions involved cortical areas exclusively or in addition to subcortical areas on CT scan of the brain. Twenty-seven (77%) developed early seizures, two third of them had immediate post-stroke seizures. None of the patients with early onset seizures developed recurrent seizures or epilepsy, while 50% of late onset seizures developed epilepsy. No specific EEG pattern was found in those who later developed epilepsy. In the present study, early onset seizures after stroke were rather common and did not affect outcome and did not recur even when not treated with anti-epileptics. Late onset seizures were less common but were associated with recurrent seizures.
The relation between seizures and stroke was recognized more than a century ago by John Hughling Jackson.1 As on today, stroke (cerebral infarction and haemorrhage) has been considered as one of the most important causes of epilepsy in the elderly.[2],[3] However, there have been very few prospective studies to define the incidence of seizures in stroke, their clinical pattern, response to therapy and outcome. Incidence of seizures has been reported to vary from 7.7% to 42.8%.[4],[5],[6],[7] Louis and McDowell studied non-embolic cerebral infarction and found an incidence of 7.7%.4 Other studies included both cerebral infarctions and haemorrhages and the reported incidences was 12.5% and 10% respectively.[5],[6] Meyer et al reported higher incidence of seizures (42.8%) in embolic cerebral infarctions.[7] The incidence of seizures in intracerebral haemorrhage has been reported to be 4.6%.[8] The present study was conducted prospectively to define the clinical features, CT findings, EEG correlation, response to therapy and prognosis of post-stroke seizures in consecutive patients of stroke.
Thirty five consecutive patients, admitted under department of Neurology at the Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow, India, with the diagnosis of post-stroke seizures between January 1990 to March 1996, were studied. Post-stroke seizures were defined as those either at the beginning of or after stroke in a patient without prior history of seizure disorder. All patients with history of previous seizures, previous brain surgery, head trauma, subarachnoid haemorrhage, aneurysmal / tumour, AVM related bleed, significant metabolic abnormality (hypoglycaemia, severe hyperglycaemia, hypo-natraemia, uraemia, alcohol intoxication etc.) and septicaemia were excluded from the study. Thus, out of 269 cases of stroke admitted during this period, 35 met these criteria. There were 20 male and 15 female patients. Mean age was 45.41 years (range-5 months to 76 years of age). Date of stroke, symptoms and signs of stroke and duration of follow-up were recorded in all patients and an attempt was made to establish the likely aetiology of stroke. Seizures were classified as generalised, localisation related (focal, with or without secondary generalisation) or unclassified.[9] Presence or absence of status epilepticus, the timing of occurrence after stroke, and frequency and recurrence were carefully noted. The seizures were considered 'early' if the initial seizure (the first seizure after stroke) occurred within two weeks of stroke, and 'late' if initial seizure occurred after two weeks of stroke. Early onset seizures were further classified as 'immediate' if seizure occurred within 24 hours of stroke. 'Recurrent' seizures were defined as those occurring at least two weeks after the onset of the initial seizure. We considered 2 or > 2 seizures as multiple. CT scan of head and EEG were performed in all the patients. These investigations were repeated as and when indicated. Other relevant investigations were performed to establish the likely aetiology (e.g. embolic infarction) and any provoking factor for seizures. CT findings were categorised as infarction and haematoma. Infarctions were further classified as cortical, subcortical and cortical-subcortical. The greatest diameter on that CT slice showing the largest area involved, was used to determine the size of the infarct as small (< 5 cms) or large (> 5 cms). Haematomas were noted for their location and volume. Haematomas were classified as small (0-29 ml) and large (30 ml or more). EEG findings were categorised as - (i) normal, (ii) diffuse slowing, (iii) focal slowing, with or without diffuse slowing and (iv) epileptiform discharges. An attempt was made to avoid medication for prevention of further seizure and to use only single medicine (phenytoin or carbamazepine or phenobarbitone) if required.
Thirty five patients (20 males and 15 females) in age range of 5 months to 76 years (Mean 45.41 years) from 269 consecutive patients of stroke, fulfilled the selection criteria. Twenty patients had infarction and 15 had haemorrhage. Five of 20 patients with infarction had an embolic stroke from cardiac source (valvular disease, atrial fibrillation and myocardial infarctions). Seventeen out of 20 patients of infarction involved cortical areas, with or without subcortical regions. Only 3 patients had pure subcortical lesions on cranial CT. Five haematomas were in the cerebral cortex while 10 were primarily located to ganglionic region (Table I). However, 8 of these ganglionic haematomas were large, with extension of haematoma and/or surroundings oedema to the cortical regions. Twenty-seven (77%) patients had early seizures including 17 who had immediate seizures (i.e. within 24 hrs of stroke) and 8 (23%) had late onset seizures. Fourteen (40%) initial seizures were single while 21 (60%) had multiple seizures. Eleven (78.6%) of single initial seizures were focal while 3 (21.4%) were generalised. Fifteen (71.4%) of 21 initial multiple seizures were focal including one focal status, while 6 (28.6%) were generalised. Recurrent seizures developed in 4 patients (focal-3, generalised-1) all of whom had late onset seizures (onset ranged from 20 days to 12 months). None of the early onset seizures developed epilepsy or recurrent seizure, during mean follow-up period of 15.9 months (range 3 to 60 months) (Table II). EEG recordings were normal in 12 (34.28%), showed diffuse slowing in 9 (25.71%), focal slowing or significant asymmetry in 8 (22.86%) and epileptiform discharges in 6 (17.14%). One of the 4 patients with recurrent seizures had focal slowing, one had focal spikes, while other two (50%) had normal EEG. Thus, no specific EEG pattern was seen with early versus late seizures and recurrent versus non-recurrent seizures. Presence of epileptiform activity on EEG was not correlated with multiple or recurrent seizures. Nine patients out of 14 with early single seizures did not receive antiepileptics, while 5 received antiepileptics. None in the either group developed recurrence. 10 out of 13 patients with early multiple seizures were given antiepileptics while 3 did not receive the same. None developed chronic epilepsy/recurrent seizures. All 8 patients with late onset seizures, were given antiepileptics including 4 cases, who developed recurrent seizures. In all cases, monotherapy with carbamazepine, phenytoin or phenobarbitone was sufficient to control seizures. There were only 4 deaths in the follow up period. 17 patients had good recovery while 14 had poor recovery. Of the 4 deaths, 2 patients had ganglionic haemorrhages while the other two had MCA territory infarction. Two of them had developed seizures at onset of stroke while remaining two within next two weeks.
35 out of 269 patients (13%) had seizures following stroke. 27 cases (77%) had early onset seizures of which 2/3rd had immediate post-stroke seizures. Out of 97 patients in five series, who had experienced seizures in the post-stroke period, seizures occurred in 55 at the onset of stroke.[4],[10],[11],[12],[13] We had similar findings. However, in a recent series of 72 patients, where only post-ischaemic strokes were considered, only 24% initial seizures were of early onset.[14] In intracerebral haemorrhage, incidence of early seizures tends to be higher than the late onset seizures.[8] Focal motor seizures, with or without secondary generalisation, were the predominant types of seizures both in early onset (74%) and late onset (75%) group. Rest of the patients had generalised seizures. Complex partial seizures were not observed. Similar findings have been reported by other authors as well.[12],[13],[15] However, in series of Susanna et al, early onset seizures were more likely to be generalised.[14] Complex partial or unclassified seizures were rare. We found recurrent seizures only in cases of late onset seizures. Susanna et al[14] had also reported recurrent seizures mainly in late onset seizures group though few previous studies disagree.[16],[17] All our patients had undergone EEG. Periodic lateralised epileptiform discharges were not recorded while diffuse slowing (25.71%) and focal slowing with or without diffuse slowing (22.86%) were seen in approximately equal number of cases. There was no specific pattern of EEG correlating to recurrence. However, the total number of cases of recurrent seizures (n=4) is too small for any conclusion. Recurrent seizures in 100% of the four patients with periodic lateralised epileptiform discharges and in 75% of the eight patients with diffuse slowing, have been reported.[18] In our patient, late onset of initial seizure was related to recurrence rather than pattern of EEG abnormality. The involvement of cerebral cortex has been emphasised in the pathogenesis of epilepsy caused by stroke.[12],[14] Some authors have given more stress on the size of cerebral infarction.[18] Incidence of seizures in lobar haemorrhage has been reported to be 32%, while it is 2% in putaminal, thalamic and pontine haemorrhages.[8] In present study, 85% of infarctions leading to seizure involved cerebral cortex with or without involvement of subcortical region. 33% of haematomas were localised exclusively to the cortical region while 80% of ganglionic haematomas were large with extension of haematoma/oedema to the cortical area. Thus, collectively only 14.28% of lesions (infarcts or haematoma) were localised to subcortical regions. However, MRI studies were not done in present series, hence possibility of cortical involvement still cannot be ruled out, in these cases of 'pure subcortical' involvement as seen on CT scan. Early seizures, whether treated with antiepileptics or not, did not lead to chronic epilepsy. Even recurrent seizures were easily controlled with one anti epileptic drug. Thus, inspite of presence of structural lesion, post-stroke seizures did not present as therapeutic challenge. Similar conclusion has been reached by earlier authors as well.[15],[18] All four patients who died, had early onset seizures. Late onset seizures probably correlated with recurrent seizures (chronic epilepsy), thus enhancing morbidity. They are however, not significantly correlated with mortality.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||
