Profile of intractable epilepsy in a tertiary referral center.
This study was undertaken to find out the profile of intractable epilepsy (IE) in a tertiary referral centre. 100 patients (males 67; females 33) with IE attending the epilepsy clinic were evaluated. Detailed history, examination, investigations like EEG and CT scan and details regarding pharmacotherapy were analysed. The age of the patients ranged from 5 to 70 yrs (mean=23.2 yrs). Mean duration of seizures was 11.44 years. Commonest seizure type was partial seizures (74%). Amongst patients with generalised seizures (26%), 14% had multiple seizure types. The seizure frequency was 12.39 +/- 21.57 (mean +/- SD) per month. Fifty seven patients were in the symptomatic group with CNS infections being the leading cause (19%) of epilepsy. Fifty patients had one or more abnormal predictors of IE. There was no difference in the severity of epilepsy in patients with no abnormal feature when compared with patients having abnormal features. EEG was abnormal in 69% cases with background abnormality in 20% and focal abnormality in 36% cases. CT scan was abnormal in 41% cases with commonest abnormality being neurocysticercosis (11%) followed by gliosis (9%) and chronic infarct (9%). Sixty patients were receiving a combination of two drugs, 32 patients 3 drugs and 8 patients were on 4 drugs. There was no difference in seizure control in patients who were on 2 drugs or more than 2 drugs. Partial seizures were the commonest seizure type leading to IE; CNS infection being the leading aetiological factor. The presence or absence of predictors of intractability does not predict severity of epilepsy. Addition of third primary drug to existing combination only increases adverse effects without better control of seizures.
Epilepsy is usually a chronic disorder with a varying outcome. There is now general agreement that 7080% of patients with newly diagnosed epilepsy can expect one year remission on treatment,, but 15-35% of patients develop intractable epilepsy (IE).,, There are a number of factors which are predictors of poor prognosis., These include organic brain lesion, partial seizures, multiple seizure types, high seizure frequency and abnormal background EEG activity. Though IE has been extensively studied in the developed countries, there are not many reports of IE from developing countries. The present study has been undertaken in a tertiary care hospital to find out the profile of IE in North West India.
A total of 100 patients of IE attending the intractable epilepsy clinic of Postgraduate Institute of Medical Education and Research, Chandigarh between May 1997-June 1998 were enroled in the present study. The study included both old patients of uncontrolled epilepsy and new cases registered during the period of study. Only those patients who had uncontrolled seizures despite two appropriate first line drugs for two years, were included.
Age, sex, details of seizures, any history of perinatal brain injury in the form of birth hypoxia, forceps delivery, history suggestive of CNS infections i.e. meningitis/meningoencephalitis; history of head trauma with unconsciousness; behavioural disorders such as psychosis and febrile seizures were recorded in addition to family history of epilepsy. Seizures were classified according to International League Against Epilepsy Classificatons. All patients were examined by one of the consultant neurologists with special interest in epilepsy. Patients were examined in detail for any neurocutaneous markers, subcutaneous nodules, behavioural disorders, mental retardation and neurological deficits. All the patients were subjected to EEG examination. CT scans were carried out in 87 patients. Additional MRI was done in 14 patients. Details regarding antiepileptic drugs (AED) were recorded i.e. number of drugs, duration of therapy, dosage, and compliance. It was noted whether the choice of drugs were correct or incorrect in relation to the seizure type and the dosage schedule was proper according to the body weight. For statistical analysis, mean+standard deviation were calculated for various characteristics. Further characteristics were compared using students 't' test.
Clinical profile : One hundred patients (males 67, females 33) of IE were registered over 14 months period. The age of the patients ranged from 5-70 (mean 23.2) years. Maximum patients (30%) were in the age group of 21-30 years (30%). The duration of seizures varied from 2-60 years (mean+SD = 11.44+8.63 years).
Seizure Characteristics : Seventy four patients had partial seizures and 26 patients had generalised seizures. Amongst patients with partial seizures, 8 had simple partial seizures, 13 had complex partial seizures and 53 had partial seizures with secondary generalisation. In the generalised seizure group, 9 patients had generalised tonic-clonic type, 2 had atypical absences and one had atonic seizures. Fourteen out of 26 patients in the generalised group had >1 type of generalised seizures [Table I]. The seizure frequency varied from <1 per month to 99 per month (mean+SD = 12.39+21.57).
Aetiology : Fifty seven patients had symptomatic epilepsy. Central nervous system infections were the leading cause of epilepsy in 19 cases [Table I]. No cause could be ascertained in 43 cases. Sixteen patients had positive family history of epilepsy and 7 patients has past history of febrile seizures.
Neurological deficit : Five patients had neurological deficit (hemiparesis 2, rigidity and dystonia 1, optic atrophy 2). Seventeen patients were mentally retarted and 19 patients had behavioural disorders. Fifty patients had either significant history of neurological insult and/or abnormal findings [Table II]. Twenty two patients had single poor prognostic factor (PPF) and the rest had multiple factors [Figure. 1].
Severity of epilepsy in relation to PPF : The duration of epilepsy in patients without PPF was 12.1+10.6 years (mean+SD) and in those with PPF was 0.2+5.9 years. The frequency of seizures in the two groups was 12.3+21.3 per month and, 12.4+21.9 per month respectively. Further, patients with one PPF were compared with patients having >1 PPF. There was no statistical difference in the severity of epilepsy between the group with or without PPF, as well as amongst the patients with one or more abnormal features [Figure. 2] (p=0.5, NS).
Investigations : The details of EEG findings are shown in [Table III]. The EEG in patients with seizures of temporal lobe origin were compared with those in seizures of frontal lobe origin. There was no significant difference in severity of epilepsy in terms of duration and seizure frequency (p 0.2, 0.4).
Radiology : CT findings are shown in [Table IV]. Additional MRI done in 14 patients showed medial temporal sclerosis and porencephalic cyst in 2 patients each. One patient each showed arachnoid cyst and cerebellar atrophy with ex-vacuo prominency of cisterna magna and left periventricular gliosis. It was normal in 7 patients.
Pharmacotherapy : Sixty patients were receiving a combination of 2 drugs, 32 patients were on 3 drugs and 8 patients were on 4 drugs [Table V].
Phenobarbitone was used for the longest duration. The seizure control was compared between patients on 2 drugs with those on >2 drugs. There was no difference in seizure control in the two groups [Table V]. The newer anticonvulsants were only available after the study was over, hence they were not tried in these cases. Seizure control amongst various two drug combinations was also compared. No combination was found better than the other [Table VI]. Ninety [eight] out of 100 patients had good compliance for drug therapy.
In general, the prognosis of a newly diagnosed case of epilepsy is good. However, 15-35% patients develop IE i.e. seizure persisting despite treatment.,, The magnitude of the problem of IE in India is unknown. Radhakrishnan et al estimated 20,000-40,000 patients with IE in India. There are a number of factors which predict poor prognosis., These include organic brain lesions, partial seizures, multiple seizure types, high seizure frequency, seizure onset in infancy and abnormal EEG.
In our study, 73% patients were between 11 and 40 years of age. Earlier reports suggested that onset of seizure before 10 years of age had better remission rate than seizure after 10 years. Intractability occurs with equal frequency in both sexes. The male predominance (67%) in this study is more likely to be because of a selection bias. Majority of the patients (74%) were having partial seizures, and further 14 out of 26 patients with generalised seizures had multiple types of seizures. Earlier reports suggested that partial seizures and multiple or mixed types of seizures are important adverse prognostic factors for the development of intractability. There are controversial reports regarding outcome of seizure in relation to seizure frequency.,, In the present study, the mean seizure frequency was 12.39 per month, substantiating earlier reports that high seizure frequency is a PPF., However, Sillanpaa showed no correlation between seizure frequency and outcome of seizures in a long term follow up.
The incidence of mental retardation (MR) and/or cerebral palsy in epilepsy has been reported to be 8%. Our finding of high frequency of birth hypoxia (13%) and MR (11%) confirms various studies showing similar results., The remission rate for seizures secondary to perinatal disturbance is 55%. Infection is one of the common causes of symptomatic epilepsy in India., The chances of remission of epilepsy, as a sequelae to CNS infection has been reported to be 46%. In this series, CNS infection was the most common aetiological agent seen in 19 cases, neurocysticercosis being the leading cause (11 cases). Fourteen patients (14%) had history of head injury which is higher than the overall incidence of head injury in epilepsy (5.5%). All the patients with history of febrile seizures (7%) had additional PPF for intractability. This factor has not been considered predictor of intractability in previous reports. Presence of neurological deficit carries poor prognosis in epilepsy., In this study, 5% patients had neurological deficit which is not higher when compared to incidence of deficits in epilepsy as a whole.
The chances of remission depend on presence of PPF for epilepsy. The chances of remission with no PPF may be as high as 96%, as compared to patients with 4 PPF, where remission rate is almost nil. In the present study, 50% patients had one or more PPF. The severity of epilepsy in patients with and without PPF was the same [Figure. 2]. Thus, in contradiction to earlier report, we found that presence or absence of PPF had no positive correlation with the severity of epilepsy.
EEG was abnormal in 69% cases, with background abnormality being present in 20% of the cases. The outcome of seizure is poorer if EEG shows background abnormality and focal epileptiform activity. But Shafer et al noted no difference in 5 years seizure free period with or without EEG abnormality. In general, patients with anterior temporal and frontal focus tend to have lower remission rate than those with mid temporal focus. However, we did not notice any significant difference in severity of epilepsy with frontal or temporal lobe focus. Forty one out of 87 CT scans done in 100 patients showed abnormalities [Table IV]. The seizure remission is less likely in symptomatic epilepsy, caused by either defined acquired brain damage or less well defined perinatal causes, manifested by gross neurological deficits or MR.,,,, Our results substantiate previous reports.
Even though the combination of drugs used and their doses were proper, 60% patients on 2 drugs, 32% on 3 drugs and 8% patients on 4 drugs had uncontrolled seizures. Many studies,, have shown that addition of second drug in patients with IE controlled seizures only in additional 10-15% of cases with increased toxicity. In the present study it was interesting to note that seizure control in patients with >2 primary drugs was no better than with 2 primary drugs. Thus if seizures are uncontrolled with 2 primary drugs, further addition of primary drugs has no clinical benefit and leads to more adverse effects. Though combination of carbamazepine and phenobarbitone appeared to have better control of seizures than others, the results were not statistically significant (p = 0.07-0.9, NS) [Table VI]. This points towards the need of newer AED in managing patients with IE. A large number of refractory cases in a developing country like India may be unable to afford newer AED because of high cost. Further, these drugs reduces the frequency of seizures by >50% in 50-70% cases, thus epilepsy surgery is a realistic approach for these cases.
In the present study, 50% patients had one or more abnormal predictors of intractability. It is possible that, there other unknown factors, possibly of genetic origin which may lead to IE. This opens up a new area in epilepsy research. Since presence or absence of known risk factors results in equal chance of developing IE, there has to be some other risk factors responsible for intractability. The identification of the factors will ultimately lead to better management as well as possible prevention of IE. However, the limitation of this study was small number of patients with MRI study because of economical reasons.