Poststroke Late Seizures and Their Role in Rehabilitation of Inpatients

Size: px

Start display at page:

Download "Poststroke Late Seizures and Their Role in Rehabilitation of Inpatients"

Bruce Little
6 years ago
Views:

Epilepsia, 38(3):266-270, 1997 Lippincott-Raven Publishers, Philadelphia 0 International League Against Epilepsy Poststroke Late Seizures and Their Role in Rehabilitation of Inpatients Stefan0

1 Epilepsia, 38(3): , 1997 Lippincott-Raven Publishers, Philadelphia 0 International League Against Epilepsy Poststroke Late Seizures and Their Role in Rehabilitation of Inpatients Stefan0 Paolucci, *Giulia Silvestri, Sergio Lubich, Luca Pratesi, Marco Traballesi, and *Gian Luigi Gigli IRCCS S. Lucia, and *Department of Neurology, Tor Vergata University, Rome, Italy Summary: Purpose: This study was designed to (a) identify the prevalence of poststroke late seizures in a population of patients admitted to rehabilitation of neurologic sequelae of their first stroke, (b) recognize reliable prognostic factors associated with the occurrence of poststroke late seizures, and (c) evaluate the impact of seizures on the results of rehabilitation treatment. Methods: In a prospective study of 306 consecutive patients admitted to a rehabilitation hospital for sequelae of their first stroke, we assessed the relation among 15 independent variables and the development of seizures by using multiple regression analysis (forward stepwise). In addition, we evaluated the impact of occurrence of poststroke seizures on both efficiency and effectiveness of rehabilitation and length of stay. Results: Poststroke late seizures occurred in 46 (15.03%) patients, with a mean interval from stroke of f days. In multiple regression analysis, putaminal and lobar hemorrhages showed a significant positive association with the development of seizures (p < 0.005), whereas high scores on the Canadian Neurological Scale (CNS) (indicating less severe strokes) and increasing age were negatively associated (p < 0.01 and p < 0.05, respectively). Patients with putaminal and lobar hemorrhages and patients with severe stroke (CNS score at admission, <7) were at significantly greater relative risk of seizures [relative risk (RR) = 1.99, 95% confidence interval (CI), ; RR = 3.00, CI, ; and RR = 2.41, CI, , respectively). No significant association was found between poststroke seizures and results of rehabilitation. Conclusions: Poststroke late seizures occurred mainly in patients with putaminal and lobar hemorrhagic strokes but, if treated, did not affect rehabilitation therapy. Key Words: Stroke-Epilepsy-Rehabilitation. Stroke incidence is very high (200/ ) in western societies (l), and stroke is one of the most common causes of death or disability. Despite the relatively low incidence of epilepsy (69%) after cerebral stroke, because of its high incidence, stroke is one of the most common causes of epilepsy (2-7). Epilepsy was defined as a condition of recurring seizures, according to the International League Against Epilepsy (8). This applies particularly to the elderly, the age group most at risk for stroke. Patients with hemorrhagic stroke, cortical lesions, lesions involving more than one cerebral lobe, and persisting paresis are at higher risk of developing seizures (3,5,7,9-1 l).the higher incidence of epilepsy in patients needing rehabilitation was confirmed by Kotila and Waltimo (12). Computed tomography (CT) scan is valuable in the identification of patients at risk for developing epilepsy, whereas EEG is not helpful in predicting the Accepted September 13, Address correspondence and reprint requests to D ~, G, L, ~ i ~ at l i IRCCS S. Lucia, via Ardeatina, 306, Roma, Italy. occurrence of seizures (4-7,13). The risk of recurring seizures is higher if the first seizures occur in the chronic stage, compared with seizures occurring in the acute phase after stroke (14-16). Despite the higher risk of developing epilepsy among patients in need of rehabilitation for neurologic sequelae after stroke, the problem of the consequences of epilepsy occurrence on rehabilitation programs and functional outcome has not been addressed. The aims of our study were (a) to identify the prevalence of poststroke late seizures among inpatients admitted to rehabilitation for neurologic deficits, (b) to recognize reliable prognostic factors associated with the occurrence of poststroke late seizures, and (c) to assess their influence on the results of rehabilitative treatment. METHODS Subject selection The study included 306 consecutive patients admitted to our rehabilitation unit for neurologic sequelae of their first stroke. The sample was not community based (pa- 266

POSTSTROKE LATE SEIZURES AND REHABILITATION 267 tients were referred by general hospitals), and no selection was made before admission, except for the clinical indication for physical rehabilitation.

2 POSTSTROKE LATE SEIZURES AND REHABILITATION 267 tients were referred by general hospitals), and no selection was made before admission, except for the clinical indication for physical rehabilitation. Stroke was defined, according to the Stroke Data Bank (17), as a sudden, nonconvulsive, focal neurologic deficit persisting for >24 h. The diagnosis of stroke was based on history, clinical examination, and neuroradiologic findings [CT scans or magnetic resonance imaging (MRI)]. Each patient underwent either a CT or a MRI examination, at least. Exclusion criteria included patients older than 79 years, previous cerebrovascular accidents, subarachnoid hemorrhage, presence of other chronic neurologic (i.e., multiple sclerosis, amyotrophic lateral sclerosis, and polyneuropathy) or disabling diseases (severe cardiac, liver, or renal failure; cancer, limb amputation, etc.). Patients with a history of seizures before the stroke, at stroke onset, or after stroke but before transfer to the rehabilitation ward were excluded from the cohort. Neurologic and functional assessment To measure severity of stroke, we used the revised and validated version (18) of the Canadian Neurological Scale (CNS), proposed by Cot6 et al. in 1986 as impairment scale (19), with a cut-off score of 11.5 for normal patients. Activities of daily living (ADL) status was measured 6y means of Barthel Index (20) (BI), the validity and reliability of which has been well established (21). The scale gives scores ranging from 0 to 100, with the top score implying full functional independence, even if not necessarily a normal status. Specific mobility was monitored by means of Rivermead Mobility Index (22) (RMI; derived from the Rivermead Motor Assessment), which assesses the ability of the patient through 15 common daily movements. The scale, whose score ranges from 0 (totally unable) to 15, proved to be valid and reliable for evaluating mobility after stroke and head injury (22). It has recently been used (in an earlier version, with a maximum score of 20) in a randomized crossover trial in stroke rehabilitation (23). Individual physiotherapy, essentially based on Bobath s therapeutic exercises and adapted to the needs of each patient, was performed for 60 min twice daily, 6 days a week. The stay in the rehabilitation hospital usually continued until clinical recovery was judged to be stabilized, but in a few cases, discharge had to be delayed because of nonmedical factors. At admission all patients were submitted to CNS, BI, and RMI, whereas at discharge, CNS was not used because the target of the rehabilitation was the functional and not the neurologic recovery. All evaluations were made by the same neurologic staff. We calculated rehabilitation results by using efficiency and effectiveness of treatment (24). Efficiency is the amount of improvement in the rating score of each scale divided by duration of rehabilitation stay; it represents the average increase per day obtained by therapy. Effectiveness reflects the proportion of potential improvement achieved during rehabilitation, calculated by this formula: (Discharge score - initial score)/(maximum score - initial score) x 100. Therefore if a patient achieves the top score after the rehabilitation, effectiveness is 100%. Data analysis and statistics The incidence of seizures in each type of stroke and the timing relative to stroke onset were recorded. A 1 -year follow-up was performed in patients who had seizures. In the other patients, the observation was stopped after discharge, but patients were instructed to inform the medical staff if seizure or any other relevant clinical event occurred. Seizures have been classified according to their respective clinical signs and electroencephalographic characteristics, as developed by the Commission on Classification and Terminology of the International League Against Epilepsy (8). To evaluate reliable factors predicting the occurrence of poststroke seizures, a multiple regression analysis (forward stepwise) was performed by using as dependent variable the development of seizures (coded as 1 = present and 0 = absent) and 15 different independent variables. Independent variables were age, sex (coded as 1 = male and 2 = female), days elapsed from stroke to admission, side of motor deficit (coded as 1 = right hemiparesis/hemiplegia and 2 = left hemiparesis/ hemiplegia), stroke severity (CNS score at admission), hypertension, diabetes, heart disease, middle cerebral artery infarction, deep lacunar lesions, vertebrobasilar ischemia, infarctions in uncertain areas, lobar hemorrhages, putaminal hemorrhages, or thalamic hemorrhages. The last eleven variables were coded as 1 or 0, according to the presence or absence of the event. Relative risks of occurrence of seizures were calculated according to results of multivariate analysis. To evaluate the relations between seizure occurrence, clinical variables, and rehabilitation results, we performed five other multiple regressions (forward stepwise), by using as dependent variable, length of stay, efficiency, and effectiveness on both BI and RMI, respectively. Independent variables were the same as the first analysis plus poststroke seizures, Broca s aphasia, global aphasia, and hemineglect. These four latter variables were coded as 1 or 0, according to the presence or absence of the disorder. Data analyses were performed by using the CSS/3 (Statsoft, Inc.) Statistical package. RESULTS All patients but one were white. Sex distribution was equal between men and women. Mean age was f

3 268 S. PAOLUCCI ET AL years (median, 67 years), and mean interval between stroke and admission was f days (median, 47 days). The length of this interval is because in Italy, the number of beds available in rehabilitation is considerably lower than the potential demand. Before stroke, none of the patients was institutionalized. Clinical and neuroradiologic diagnosis was of ischemic stroke in 247 (80.72%) of 306 patients, and of cerebral hemorrhage in 59 (19.28%); 161 (52.61%) patients had a right hemiparesis or hemiplegia, whereas 145 (47.39%) were affected by a left-sided motor deficit. Hemorrhagic patients were younger (F = 5.51; p = 0.02) than ischemic patients (60.45 k years vs f years). Among ischemic strokes, 144 (58.3%) of 247 were infarcts in the territory of the middle cerebral artery (MCA), 57 (23.08%) were deep lacunar infarcts, 17 (6.88%) were infarcts in vertebrobasilar territory, and the remaining 29 (1 1.74%) were of uncertain localization. Among hemorrhagic lesions, 29 (49.15%) were lobar, 14 (23.73%) putaminal, and the remaining 16 (27.12%) were thalamic. In all patients, the presence of hypertension, diabetes, heart disease (i.e., previous myocardial infarction, atrial fibrillation) or of other risk factors was recorded. Mean length of stay in rehabilitation hospital was f days. Seizures occurred in 46 (15.03%) patients, with a mean interval from stroke of k days. Seven ( 15.22%) patients had generalized tonic-clonic seizures, probably of focal origin; the remaining 38 had partial seizures (simple partial or complex partial). All patients received anticonvulsant drugs, 38 with carbamazepine monotherapy and seven (15.22%) with polytherapy. In consideration of the lesional substrate of epileptogenicity in this group of patients, antiepileptic treatment was started immediately after the first seizure. Despite this, 12 (26.09%) patients had a second seizure during their stay in rehabilitation hospital, and this percentage increased to 89.13% at 1 year after discharge from the hospital. Table 1 shows medical, demographic, and neuroradiologic findings of the sample, divided according to the occurrence of seizures. Patients with seizures showed at admission a greater severity of stroke, with an impairment score (CNS) significantly lower than that of patients without seizures (5.04 k 1.98 vs f 2.14; z = ; p < 0.005). In addition, the prevalence of hemorrhagic lesions was significantly higher (x2 = 8.36; p < 0.01) in patients with seizures than in patients without seizures. In multiple regression analysis, the first significant predictive variable (p < 0.005) to enter the equation was putaminal hemorrhages, followed by lobar hemorrhages, with p positive regression coefficients (p = and p = 0.145, respectively). The other significant variables were CNS score at admission (severity of stroke) (p = ; p < 0.01) and age (p = ; p < 0.05), both with a p negative regression coefficient. In this model, days of interval between stroke and admission, hypertension, and deep lacunar lesions were not significant, whereas sex, side of motor deficits, diabetes, heart disease, MCA infarctions, infarctions with uncertain localization, vertebrobasilar ischemia, and thalamic hemorrhages did not enter the equation. As shown in Table 2, patients with putaminal and lobar hemorrhages and patients with severe stroke (CNS score at admission <7) had a significantly higher RR of occurrence of seizures (RR = 1.99,95% CI, ; RR = 3.00, CI, ; and RR = 2.41, CI, , respectively). In the other five multiple regressions analyses with functional measures as dependent variables (length of TABLE 1. Demographic and neuroradiologic characteristics of the sample Gender Female Male Age & Ischemic strokes MCA territory Lacunar deep infarcts Border or uncertain areas VB territory Hemorrhagic strokes htaminal Thalamic Lobar MCA, middle cerebral artery; VB, vertebrobasilar. n With seizures (% of no.) 22 (14.38) 24 (15.69) 8 (25.81) 7 (22.58) 12 (15.19) 18 (13.33) l(3.33) 21 (14.58) 7 (12.29) l(3.57) l(6.25) 4 (28.57) l(6.25) 11 (37.93)

POSTSTROKE LATE SEIZURES AND REHABILITATION 269 TABLE 2. Relative risks of occurrence of seizures Putaminal hemorrhages Lobar hemorrhages Seventy of stroke (CNS score <7 vs. 27) Age (<46 vs.

4 POSTSTROKE LATE SEIZURES AND REHABILITATION 269 TABLE 2. Relative risks of occurrence of seizures Putaminal hemorrhages Lobar hemorrhages Seventy of stroke (CNS score <7 vs. 27) Age (<46 vs. 246) Any group was compared with all other patients. CNS, Canadian Neurological Scale; CI, confidence interval. Relative risks % CI stay, efficiency, and effectiveness on both BI and RMI), development of epileptic seizures entered into all but one equation but was never significant. Seizures did not enter into analysis with efficiency on RMI as dependent variable. Multiple R2 were (length of stay); and (both efficiency); and and (both effectiveness). In all analyses, CNS score at admission (severity of stroke) was the most powerful prognostic factor, positively correlated with rehabilitation results, and negatively, with length of stay. Relevant prognostic factors were also hemineglect, days of interval before admission, and age. DISCUSSION In our study, the percentage of patients developing late seizures after a first stroke (15.03%) is higher than those usually reported in similar studies conducted on unselicted patient populations (2-7). This is probably because our case series was selected for the need of rehabilitation, usually for permanent language or motor deficits, indicating a higher severity of stroke. Higher prevalences of late seizures were found by other authors in patients with permanent neurologic deficits (7) or needing rehabilitation (12). From our data, patients at risk for developing late seizures are younger patients affected by severe strokes with hemorrhagic lesions. In particular, putaminal and lobar hemorrhages are significantly associated with seizures. However, in our series, younger age is a weak predicting factor, showing only a slight association (p < 0.05) with occurrence of seizures at multivariate analysis. In fact, relative risk of seizures in younger patients (younger than 46 years) compared with the older patients did not reach statistical significance. We have been unable to find this association in other studies on poststroke late seizures. A tentative explanation of the slight association of seizures with younger age could be found in the generally weaker epileptogenicity of older brains. Altogether our findings indicate that younger patients with putaminal and lobar hemorrhagic lesions in need of rehabilitation constitute a group that should be carefully monitored to examine a possible development of epilepsy. At 1-year follow-up, the percentage of patients with recurrence of seizures, for whom it is possible to speak of a true epileptic condition, is very high (89.13% of the group with seizures). This percentage, which could have been even higher if a treatment had not been started immediately, is in accordance with data from literature, indicating that epilepsy developed in 90% of patients with late poststroke seizures, compared with 35% of patients with early seizures (16). Similar to other previous studies, a higher prevalence (27.12%) of late seizures was found among patients with hemorrhagic lesions (3,5,7,9,10). In ischemic strokes, this was only 12.15%, with the higher frequency in MCA lesions. All but one of MCA lesions associated with seizures were observed in cortical lesions. The most important results in terms of risk factors predicting seizure development are putaminal and lobar hemorrhages, followed by severity of stroke and younger age. Therefore the relevant role of cortical hemorrhages is independent of the severity of stroke. In our series, the effect of younger age is partially dependent of the other variables, because hemorrhagic patients were significantly younger than ischemic patients. On the contrary, the percentage of thalamic hemorrhages was higher among patients without seizures, indicating that purely deep lesions are less at risk for the development of seizures. The most interesting findings of our study concern the poor impact of the late seizures on rehabilitation and functional outcome. In fact, no significantly association was found in multiple regression analyses with length of stay and efficiency and effectiveness on both BI and RMI. The assessment of these three parameters permits a global evaluation of rehabilitation results. Relevant prognostic factors were severity of stroke (CNS score at admission), hemineglect, days of interval before admission, and age. In our series, the development of seizures is not associated with a worsening of functional outcome. Because worsening of neurologic status is not uncommon in patients with poststroke seizures, especially after seizures of longer duration (25), we wonder if the lack of negative consequences on functional outcome in our series could be interpreted as an effect of the immediate phmacologic treatment. In this case, treatment after late poststroke seizures should be started as soon as possible. In addition, the presence of groups more at risk for developing seizures suggests the interest

5 2 70 S. PAOLUCCI ET AL. in prospective studies with prophylactic treatment of patients. Acknowledgment: This study was supported by a fund of the Italian Ministry of Health. REFERENCES 1. Hansen BS, Marquardsen J. Incidence of stroke in Frederiksberg, Denmark. Stroke 1977;8: Aring CC, Merritt HH. Differential diagnosis between cerebral hemorrhage and cerebral thrombosis: a clinical and pathologic study of 245 cases. Arch Intern Med 1935;56: Dodge PR, Richardson EP, Victor M. Recurrent convulsive seizures as a sequel to cerebral infarction: a clinical and pathological study. Brain 19S4;77: , 4. Louis S, McDowell F. Epileptic seizures in nonembolic cerebral infarction. Arch Neurol 1967;17: Marquardsen J. The natural history of acute cerebrovascular disease. Copenhagen: Munksgaard, 1969:l Fentz V. Epileptiske anfaldsfaenomener hos patienter med apoplexia cerebri. Nord Med 1971;86: Olsen TS, Hdgenhaven H, Thage 0. Epilepsy after stroke. Neurology 1987;37: Commission on Classification and Terminology of the Intemational League Against Epilepsy. Proposal for revised clinical and electroencephalographic classifications of epileptic seizures. Epilepsia 198 1;22: Kilpatrick CJ, Davis SM, Tress BM, Rossiter SC, Hopper JL, Vandendriesen ML. Epileptic seizures in acute stroke. Arch Neurol 1990;47: Davalos A, De Cendra E, Molins A, Ferrandiz M, Lopez-Pousa S, Genis D. Epileptic seizures at the onset of stroke. Cerebrovasc Dis 1 992; 2: Lancman ME, Golimstok A, Norscini J, Granillo R. Risk factors for developing seizures after a stroke. Epilepsia 1993;34: Kotila M, Waltimo 0. Epilepsy after stroke. Epilepsia 1992;33: Liidorf K, Jensen LK, Plesner AM. The value of EEG in the investigation of postapoplectic epilepsy. Acta Neurol Scand 1986; 74: Fish DR, Miller DH, Roberts RC, Blackie JD, Gilliatt RW. The natural history of late-onset epilepsy secondary to vascular disease. Acta Neurol Scand 1989;80: Homig CR, Biittner T, Hufnagel A, Schroder-Rosenstock K, Domdorf W. Epileptic seizures following ischaemic cerebral infarction. Eur Arch Psychiatry Neurol Sci : Sung CY, Chu NS. Epileptic seizures in thrombotic stroke. J Neurol 1990;237:16& Foulkes MA, Wolf PA, Price TR, Mohr JP, Hier DB. The Stroke Data Bank: design, methods, and baseline characteristics. Stroke 1988; Cot6 R, Battista RN, Wolfson C, Boucher J, Adam J, Hachinski V. The Canadian Neurological Scale: validation and reliability assessment. Neurology 1989;39: Cote R, Hachinski VC, Shurvell BL, Nonis JW, Wolfson C. The Canadian Neurological Scale: a preliminary study in acute stroke. Stroke 1986;17: Mahoney F, Barthel D. Functional evaluation: the Barthel Index. Md Med J 1965;2: Wade DT, Langton Hewer R. Functional abilities after stroke: measurement, natural history and prognosis. J Neurol Neurosurg Psychiatry 1987;50: Collen FM, Wade DT, Robb DT, Bradshaw CM. The Rivermead Mobility Index: a further development of the Rivermead Motor Assessment. Int Disabil Stud 1991; 13: Wade DT, Collen FM, Robb GF, Warlow CP. Physiotherapy intervention late after stroke and mobility. BMJ 1992;304: Shah S, Vanclay F, Cooper B. Efficiency, effectiveness and duration of stroke rehabilitation. Stroke : Bogousslavsky J, Martin R, Regli F, Despland PA, Bolyn S. Persistent worsening of stroke sequelae after delayed seizures. Arch Neurol 1992;49:385-8.

SEVERAL STUDIES on stroke recovery have been conducted

SEVERAL STUDIES on stroke recovery have been conducted 695 Early Versus Delayed Inpatient Stroke Rehabilitation: A Matched Comparison Conducted in Italy Stefano Paolucci, MD, Gabriella Antonucci, PhD, Maria Grazia Grasso, MD, PhD, Daniela Morelli, MD, Elio