Pediatric Epilepsy - An Indian Perspective
http://www.100md.com
《美国医学杂志》
Grant Medical College & JJ Group of Hospitals, PD Hinduja National Hospital and BJ Wadia Hospital for Children, Mumbai, India
Abstract
Prevalence studies from India suggest that epilepsy prevalence is similar to developed nations. Neurocysticercosis (NCC) predominates as an etiology. A large treatment gap is still a public health problem. Benign epilepsies and West syndrome appear to be underrepresented in studies on classification of seizures/syndromes. Febrile seizures prevalence in India is similar to other countries and appear to be as benign. Risk factors of intractable epilepsy (IE) in Indian studies include early age of onset, neurodevelopmental abnormalities and certain seizure types. Perinatal injuries underlie many IE. Many IE are not truly intractable and respond to simple therapeutic measures. The ketogenic diet and surgery are other methods now being used in Indian centers. Neurocysticercosis and neonatal hypoglycemic brain injury, two widely prevalent etiologies are reviewed in detail.
Keywords: Childhood epilepsy; India; Neurocysticercosis; Hypoglycemia
Epilepsy is probably the most common disorder seen by the child neurologist. Age specific prevalence rates are highest in the 1st decade of life especially below the age of 2 years. This short review focuses on some aspects of pediatric epilepsy as it pertains to the Indian subcontinent.
Epidemiology
Incidence studies are not available from the Indian subcontinent. There are several prevalence studies however and a recent meta-analysis[1] suggests that the prevalence rate is 5.59 per 1000 population with no gender or geographical differences. This rate is surprisingly similar to the rate in developed countries. The CRESS study, in progress in Andhra Pradesh[2] surveyed more than 74,000 people in 22 villages and showed a similar prevalence rate of 6.2 /1000 population. The prevalence in Kerala, a more developed state with a high awareness of health related issues is similar at 4.9 / 1000.[3]
Febrile seizures, head injury and a positive family history of epilepsy have been identified as risk factors.[4] Neurocysticercosis accounted for 47% of all identifiable etiologies in the ongoing CRESS study[2] with a significant percentage showing brain calcifications. We have found neonatal hypoglycemic brain injury accounting for 23% of all symptomatic epilepsies with onset below the age of 3 years in a tertiary center.[5] Hot water epilepsy, a peculiar type of reflex epilepsy is reported from South India with a prevalence rate of 1.14 -2.99/1000.[4]
The treatment gap is a reality in countries like India and is indicative of the number of epileptics who do not receive AEDs ever. It has been estimated to be between 6% in urban Parsi populations,[6] 38% in well developed states like Kerala[3] to 78% in rural Karnataka[7] suggesting a serious public health delivery failure. Knowledge, attitudes and practices are negative[3], [8] with several misconceptions widely prevalent e.g. epilepsy is contagious, a form of insanity, a hindrance to education, marriage and employment. These negative attitudes seem to be more of a problem for the patient than seizure control. One-third of all those surveyed in AP [8] and 2/3 in Kerala[3] believed that ayurveda and other traditional medical practices are useful in treating epilepsy and this probably worsens the treatment gap.
Issues in classification of epileptic seizures and syndromes
There are only two recent systematic Indian studies on classification and both are from metropolitan tertiary care institutes with an inherent referral bias.[9],[10] It is not surprising therefore that both report low numbers of idiopathic genetic syndromes like the benign partial epilepsies (BPEs). Our own experience suggests otherwise. More than 50% of children suffered from symptomatic or cryptogenic localization-related epilepsies. As imaging was carried out in only about a third, many symptomatic cases may have been labelled as 'cryptogenic'. Juvenile myoclonic epilepsy (JME) was frequently seen amongst patients from Hyderabad, contrary to the Mumbai cohort, which focused on only patients below 12 years.
Surprisingly both investigations reported a low incidence of West syndrome, which is contrary to our experience. Another notable finding was that the epilepsy of a large numbers of patients could not be classified at all; this may be because of incomplete investigations due to limited resources. Generalised epilepsies seem to be twice as common as partial epilepsies[3] in one community survey, possibly due to overestimation as generalized tonic clonic seizures are more dramatic and more likely to be noticed.
Community-based clinical surveys like the one in Yelundur[11] suggest that the majority of epilepsies are idiopathic or cryptogenic. This has its limitations as the conclusions were drawn from clinical accounts. What is needed is a community-based study of a large population where all identified cases are fully investigated. This is the approach in the CRESS study[2]
Febrile Seizures
Febrile convulsions (FS) are seizures precipitated by fever i.e. not due to an intracranial infection or other definable central nervous system cause and are not preceded by afebrile seizures. FS pose a common problem in paediatric neurology. The incidence in Indian children less than five years of age, is two to five %.[12] Most studies suggest either a polygenic mode of inheritance,[13] or rarely autosomal dominant.[14] Early Indian studies suggested a prevalence of 10% in the community or 3% of all hospital admissions. But in the Yelandur[11] survey prevalence of FS was estimated at 3.28-5.71/1000 population, similar to western figures. About half the patients had active epilepsy with a male preponderance (72%) and a peak age of onset around 1 year. 90% of the attacks occurred during the first three years but there were some as late as eight years. The average duration was 1.48 years, upto the age of nine years. The role of febrile status epilepticus in the genesis of subsequent mesial temporal sclerosis (MTS) is a question not yet resolved fully though there seems to be some evidence for it's occurrence.[15] A history of FS could be elicited in 65% of 26 children and adolescents with MTS.[16]
EEG is of limited value. 15% of all EEGs recorded in one laboratory in Mumbai were for febrile seizures (Udani; unpublished observations) and only 21 of the 433 records were 'abnormal', with background rather than epileptic abnormalities. There were more abnormalities noted in the records of patients with complex FS than simple FS. Our current practice is not to request for an EEG whether the FS is simple or complex.
A consensus report mentioned that patients with simple FS should not be prescribed a prophylactic anti-convulsant drug for an extended period.[17] In its place, intermittent prophylaxis with oral or rectal diazepam or oral clobazam is advised.[18] The effect of oral antipyretics alone is disappointing.[19]
The overall risk of recurrence is 25 to 30% and this happens within the first year of onset in children less than 1 year. The risk factors include young age of onset, seizure within first two hours of fever, seizures with low-grade fever, positive family history of FS or epilepsy, complex FS and those who have had already more than two episodes.
The possibility of suffering from epilepsy in the future is 2 to 5%. This is increased in the presence of developmental or neurological abnormality, family history of epilepsy and in those with complex FS. It is as high as 50% when all unfavorable factors exist.[20] Only 2% of the Yelundur patients were neurologically and developmentally handicapped, and only 1.2% had later epilepsy. Most of these seizures were a part of an idiopathic partial or generalized epilepsy syndrome like BPE, JME and the newly described GEFS + (generalized epilepsy febrile seizures plus).[21]
Neuro-developmental outcome is not compromised in children with FS[22] and is linked to pre-existing developmental status and appropriate management of prolonged seizures.
Intractable Epilepsy and Epileptic Encephalopathies
These severe epilepsies constitute about 10-20% of all the epilepsies and up to 50% in specialty epilepsy clinics. Epileptic encephalopathies (EE) are syndromes in which the epilepsy itself is responsible for cognitive deterioration and amelioration of the seizures and/or regression of abnormal rhythms on EEG results in improvement. Syndromes like the West (WS), Lennox - Gastaut (LGS), Landau-Kleffner (LKS) / electrical status in slow wave sleep (ESES) and severe myoclonic epilepsy of infancy (SMEI) are some examples.
There is a paucity of information on this subject from India. In a study of 123 'difficult to control' epileptic children from Mumbai,[23] risk factors for intractability were, onset below two years of age, male sex, neurological and developmental handicaps and certain seizure types (complex partial, GTC, mixed seizures). Static encephalopathies were noted in about two-third, suggesting that epilepsy results from an abnormal substrate. Only 50% had an identifiable syndrome. An investigation carried out at AIIMS[24] found that 3.5% of all children attending the neurology clinic were suffering from epileptic encephalopathies. Of 94 patients, WS affected more than 50%. They noted an evolution of syndromes in four children from early myoclonic encephalopathy to WS to finally LGS. Prenatal aetiology predominated. We have found perinatal brain - damaging events to be the most common risk factor.[5] Perinatal encephaloclastic aetiologies were noted in 50%, with about a fourth of all cases due to neonatal hypoglycemia. Tuberous sclerosis, cortical dysplasias and neuronal migration disorders constituted another 20%. Infantile spasms and partial seizures were the most common seizure type and were often refractory to treatment. Seth et al examined 37 children with EE[25] and noted that more than 50% were getting spasms and in 70% there was history suggesting perinatal brain injury. The frequency of seizures lessened in only a third of their patients and only 6% were seizure- free.
Differing results are reported with the use of ACTH or steroids. In our randomised controlled trial of ACTH vs oral prednisolone given to children with WS, only 33% in the ACTH group were rendered free of spasm. Prednisolone was even less effective, only 14% were fully relieved.[26] Additional vigabatrin was effective in 50% of non-responders. These disappointing results are probably due to a high number of patients with symptomatic epilepsies and the long lag before the start of appropriate treatment.
In many older children and adults 'intractable' epilepsy is due to inappropriate polytherapy or low dosages. Simple measures like switching from polytherapy to monotherapy at correct dosage resulted in adequate control.[23] This has the added advantage of less adverse effects and a reduction in the cost per patient. These conclusions were reinforced when only 8% of samples referred for estimating drug levels were in the therapeutic range.[27] Compliance may be affected by cost of even standard anti-convulsant drugs (AEDs) like sodium valproate and carbamazepine. Hence earlier-marketed AEDs like phenytoin and phenobarbitone are still preferred for rural and low-income groups. Indeed phenobarbitone has been recommended in preference to phenytoin[28] because, besides being cheaper it has less side effects and its effect on behaviour is no worse.[29] New AEDs have not yet made a significant impact on the control of IE in countries like India except in small numbers in urban/semi-urban areas where costs and availability are less of an issue than in the rural hinterland.
The ketogenic diet is a high fat, low carbohydrate diet coming into its own in several centers. It seems to be useful at all ages and for all types of seizures. It is however fairly demanding and needs a lot of discipline. It is mainly used in young children where feeding behaviour is completely under the caretaker's control and appears to be a little more difficult and restrictive in vegetarian patients. In our own analysis of 49 patients between the ages of 18 m-16 years it was very successful in about 38% while about another 40% dropped out because of various reasons (Prabhu unpublished observations). Improved behaviour and decreased hyperactivity are other benefits.
Epilepsy surgery is a viable option in intractable seizures even in the developing world. Surgically remediable syndromes are mainly the lesional epilepsies, mesial temporal lobe epilepsies caused by hippocampal sclerosis, the hemispheric epilepsies like those caused by Rasmussen;s encephalitis or an unilateral destructive lesion and finally generalized epilepsies where corpus callosotomy would be very useful in reducing drop attacks. The availability of surgery is increasing with several tertiary centers offering comprehensive epilepsy services. Cost is a limiting factor though a recent investigation comparing surgical and medical management of adult TLE[30] concluded that surgery was much more cost-effective and was far cheaper in the long run. Our own experience in surgical treatment of pediatric epilepsy is similar. However pediatric epilepsy often has extra-temporal, multiple foci which make surgery more difficult and less successful than in adults.
Two etiologies are commonly seen in India and deserve separate mention. These are neurocysticercosis commonly seen in older children and adolescents and neonatal hypoglycemic brain injury commonly presenting in infancy.
Neurocysticercosis (NCC)
NCC needs special mention as it accounted for 47% of the identifiable etiology in the CRESS study.[2] Crude prevalence rate for NCC-related seizure disorder was 0.98 per 1000 population and for NCC-related acute symptomatic seizures it was 0.15 per 1000
NCC is caused by the larval stage of Taenia solium and is a major public health problem in non-Islamic developing countries including India.[31] Autopsy studies have shown a 1% rate of NCC, while clinical surveys implicated it in 0.4% of all children with neurological complaints.[32] There are state-wise differences with high rates in Delhi and surrounding areas and a much lower incidence in Kerala. Several types of NCC are described - parenchymal, subarachnoid, intraventricular and disseminated[31],[32] and this probably represents an interplay between infection load and host immune factors. Live cysts and granulomas are symptomatically 'active' while residual calcifications and gliosis are considered 'inactive'. In children 60-76% are single parenchymal NCC granulomas.[33] Calcifying lesions are much less common in children (15%) as compared to adults (55%).[33]
School-age children are most affected with a slight male preponderance.[31],[32],[33] The youngest child described was 9 months of age.[32] Recent onset focal seizures are the most common presenting symptom (70-95%). In the granuloma stage, seizures are acutely symptomatic, but seizures continue to occur in some children even after the lesion becomes calcified. In a series of 558 children, 51% of symptomatic partial epilepsies were due to either active or inactive NCC.[34] In most children neurological examination is normal. Neuroimaging is mandatory for diagnosis. Though the MRI is more sensitive, a contrast enhanced CT scan is probably good enough for transitional forms and is in fact superior for detection of calcifications.[35] Lesions are mainly at grey - white junctions in parietal, frontal and occipital lobes[32] and are rare in temporal lobes and the posterior fossa. Tuberculoma, a common differential diagnosis, on the other hand is seen more in basal regions.
Serology, CSF studies, stool examinations, radiology of leg muscles and fundoscopy are usually not cost effective[31],[32],[33] in childhood as the parasite load is not high. EEG must be recorded, as focal background slowing often indicates acute symptomatic seizures due to oedema, while focal spikes are seen in chronic remote symptomatic epilepsy.
The natural history of the single live cysticercus is variable and as seizures tend to be chronic and unremitting in this group, it seems reasonable to prescribe cysticidal drugs (albendazole preferred over praziquantel) along with a short course of steroids.[36] Caution should be exercised when there are multiple lesions as simultaneous death/degeneration of multiple cysts may cause severe oedema and even death.[37] The natural history of the cysticercus granuloma appears more uniform. 85-90% disappear decrease in size or become calcified over a few months. Studies on efficacy of cysticidal treatment in this granuloma group are conflicting.[33] Padma et al found no benefit,[37] but the group from the Post-Graduate Institute, Chandigarh (PGI) report short-term clinical benefit and regression of the CT/MRI lesion.[33] Similarly there are conflicting reports about control of seizures: Retrospective studies recommend treatment[38] but a long term, prospective, randomised trial argued against it.[39] What is clear is that treatment in children will not change the seizure remission rate, which is high anyway. What is questioned is whether treatment decreases residual calcifications and perilesional gliosis, in which the remission rate is lower. Some evidence is available supporting this notion.[40] Steroids given without albandazole along with AEDs may augment complete resolution of the granuloma[41] and reduce the risk of seizure recurrence. However another study has shown conflicting results.[42] Our policy is to treat all single and multiple active lesions with albandazole and steroids. The only time we do not prescribe cysticidal drugs is when there is disseminated disease or cysticercal encephalitis where there is danger of severe, often fatal increases in intracranial pressure. Anti-epileptic drugs are given till lesions resolve or become inactive. The problem is in managing calcified lesions. Some "re-activate" periodically with surrounding oedema and acute symptomatic seizures occur over several years necessitating prolonged AED therapy.[43] Gradient -echo MR imaging suggests that remnants of the scolex persist in these lesions and periodiacally incite surrounding oedema.[44] Albendazole would probably not be helpful at this stage as the parasite does not seem to be viable. The more worrying calcified lesion is one with associated gliosis where seizures may become refractory and where there is a true transition to an "epileptic" focus with presumably change in electrical circuitry and consequent persistent seizures. Surgical resection should be a consideration if AED therapy fails.
In conclusion there are several unanswered questions in NCC and epilepsy that would need a long-term multi-center trial to decide optimal management.
Neonatal Hypoglycaemic Brain Injury and Infantile Epilepsy
It is becoming clear that this very preventable condition is widely prevalent in India and accounts for a lot of long-term neurologic morbidity including epilepsy, mental retardation and visual impairment. We studied 100 consecutive cases of remote symptomatic epilepsy with onset in the first 3 years of life using imaging as a primary diagnostic tool for the etiology and found that 50% of children had perinatal brain damage as their epilepsy substrate.[5] As many as 23% had MRI evidence of neonatal hypoglycaemic brain injury. Birth histories were compatible in all while low plasma glucose had been documented in 14 /23 patients in the neonatal period. Low birth weight was a risk factor but interestingly 6/23 had birth weights of >2.5 kg. LSCS and poor feeding were other important risk factors suggesting that establishment of early feeding is mandatory after LSCS and if neglected can have disastrous consequences even in AGA babies. Infantile spasms was the most common seizure type. Associated disabilities were noted in > 50% and consisted of severe MR, autism, visual impairment and poor hand use. Interestingly, motor disabilities were much less common.
References
1. Sridharan R, Murthy BN. Prevalence and pattern of epilepsy in India. Epilepsia 1999; 40 : 631-636.
2. Chowdary GVS, Murthy JMK, Vijay S et al. Prevalence of Seizure Disorders Associated with Neurocysticercosis: A Community-based Study - Comprehensive Rural Epilepsy Study - South India (CRESSI). Presented at the Asian and Oceanic Congress of Neurology, Singapore 2004.
3. Radhakrishna J, Pandian D, Santoshkumar T et al. Prevalence, knowledge, attitude and practice of epilepsy in Kerala, South India. Epilepsia 2000; 41(8) : 1027-1035.
4. Bharucha NE. Epidemiology of epilepsy in India. Epilepsia 2003. 44(Suppl.1);9-11.
5. Munot P. Etiology of symptomatic epilepsy with onset in the first three years of life. Presented at the National conference of the IAP (Pedicon) 7th Jan 2005, Kolkata
6. Bharucha NE, Bharucha EP, Bharucha AE et al. Prevalence of epilepsy in the Parsi community of Bombay. Epilepsia 1988;29 : 111-115.
7. Mani KS. Epidemiology of epilepsy in Karnataka, India. Neurosci Today 1997; 1 : 167-174.
8. Thomas J, Seshadri V, Murthy JMK et al. Knowledge, attitude and practice of epilepsy in rural Andhra Pradesh, South India. Presented at the joint annual conference of the IAE & IES 2003, Vishakapatnam.
9. Shah KN, Rajadhyaksha SB, Shah VS et al. Experience with the International league against epilepsy classifications of epileptic seizures (1981) and epilepsies and epileptic syndrome (1989) in epileptic children in a developing country. Epilepsia 1992; 33: 1072-1077.
10. Murthy JM, Yangala R, Srinivas M. The syndromic classification of the International League against epilepsy: A hospital based study from South India. Epilepsia 1998; 39 : 48 - 54.
11. Mani KS, Rangan G. The Yelandur Model for Rural Epilepsy Control in India in Epilepsy. In Singhal BS, Nag D, eds. Indian Epilepsy Association Publication 2000; 16-32.
12. Febrile seizures: Long-term management of children with fever associated seizures. Summary of an NIH consensus statement. Brit Med J 1980; 281 : 277-279.
13. Anderson VE, WilcoxKJ, Hauser WA et al. A test of autosomal dominant inheritance in febrile convulsions. Epilepsia 1988;29 : 705-706.
14. Rich SS, Annegers JF, Hauser WA et al. Complex segregation analysis of febrile convulsions. Am J Human Genet 1987; 41 : 249-257.
15. Van Landingham KE, Heinz ER, Cavazos JE et al. Magnetic resonance imaging evidence of hippocampal injury after prolonged focal febrile convulsions. Ann Neurol 1998; 43: 413-426.
16. Shah KN, Rajadhyaksha. In Singhal BS, Nag, eds. Febrile Convulsions in Epilepsy in India. Indian Epilepsy Association publication 2000: 92-108.
17. Joint working group of the research unit of the Royal College of Physicians and the British Pediatric Association. Guidelines for the management of convulsions with fever. Brit Med J 1991; 303 : 634-636.
18. Knudsen FU. Febrile seizures : treatment and prognosis. Epilepsia 2000; 41 : 2-9.
19. Van Stuijvenberg M, Derksen-Lubsen G, Steyerberg EW et al. Randomised, controlled trial of ibuprofen syrup administered during febrile illnesses to prevent febrile seizure recurrences. Pediatrics 1998; 102: E51.
20. Annegers JF, Shirts SB, Hauser WA et al. Risk of recurrence with an initial unprovoked seizure. Epilepsia 1986; 27 : 43-50.
21. Singh R, Scheffer IE, Crossland K et al. Generalised epilepsy with febrile seizures plus: a common childhood - onset genetic epilepsy syndrome. Ann Neurol 1999; 45 : 75-81.
22. Nelson KB, Ellenberg JH. Prognosis in children with febrile seizures. Pediatrics 1978; 61 : 720-727.
23. Udani VP, Dharnidharka V, Nair A et al. Difficult to control epilepsy in childhood - a long term study of 123 cases. Indian Pediatr 1993; 30 : 1199-1206.
24. Kalra V, Gulati S, Pandey RM et al. West syndrome and other infantile epileptic encephalopathies - Indian hospital experience. Brain Dev 2002; 24: 130-139.
25. Seth A, Aneja S, Taluja V. Epileptic encephalopathies of early childhood. Indian Pediatr 2001; 38: 390 - 396.
26. Moharir M, Udani VP. Randomised controlled trial of high dose natural ACTH vs prednisolone in treatment of infantile spasms. Paper presented at Golden jubilee annual conference, Neurological Society of India;
Abstract: 217.
27. Karande SC, Dalvi SS, Kshirsagar NA. Shortcomings in the pharmacotherapy of epileptic children in Bombay. India J Trop Pediatr 1995; 41: 247-249.
28. Mani KS, Rangan G, Srinivas HV et al. Epilepsy control with phenobarbital or phenytoin in rural south India: the Yelandur study. Lancet 2001; 357 : 1316-1320.
29. Pal DK, Das T, Chaudhary G et al. Randomised controlled trial to assess acceptability of phenobarbital for childhood epilepsy in rural India. Lancet 1998; 351:19-23.
30. Rao MB, Radhakrishna K. Is epilepsy surgery possible in countries with limited resources. Epilepsia 2000; 41 (Suppl.4) : S31-S34.
31. Garg RK. Childhood Neurocysticercosis : Issues in diagnosis and management. Indian Pediatr 1995; 32 : 1023-1029.
32. Kalra V, Suri M, Jailkhani BL. A profile of childhood neurocysticercosis. Indian J Pediatr 1994; 61: 33-42.
33. Singhi P, Ray M, Singhi S et al. Clinical spectrum of 500 children with neurocysticercosis and response to albendazole therapy. J Child Neurol 2000; 15: 207-213.
34. Murthy JM, Yangala R. Etiological spectrum of localization-related epilepsies in childhood and the need for CT scan in children with partial seizures with no obvious causation - a study from south India. J Tropical Pediatr 2000; 46 : 202-206.
35. Thakur LC, Anand KS. Childhood neurocysticercosis in South India. Indian J Pediatr 1991; 58 : 815-819.
36. Garg RK. Medical management of neurocysticercosis. Neurol India 2001; 49 : 329-337.
37. Padma MV, Behari M, Misra NK et al. Albendazole in neurocysticercosis. Natl Med J India 1995; 8 : 255-258.
38. Del Brrutto OH, Santibanez R, Noboa CA et al. Epilepsy due to neurocysticercosis: analysis of 203 patients. Neurology 1992;42 : 389-392.
39. Carpio A, Santillan F, Leon P et al. Is the course of neurocysticercosis modified by treatment with antihelminthic agents Arch Intern Med 1995; 155 : 1982-1988.
40. Del Brutto OH. Prognostic factors for seizure recurrence after withdrawal of antiepileptic drugs in patients with neurocysticercosis. Neurology 1994; 44 : 1706-1709.
41. Mall RK, Agarwal A, Garg RK et al. Short course of prednisone in Indian patients with solitary cysticercus granuloma and new-onset seizures. Epilepsia 2003; 44(11) : 1397-1401.
42. Singhi P, Jain V, Khandelwal N. Corticosteroids versus albendazole for treatment of single small enhancing computed tomographic lesions in children with neurocysticercosis. J Child Neurol 2004; 19 : 323-327.
43. Nash TE, Patronas NJ. Edema associated with calcified lesions in neurocysticercosis. Neurology 1999; 53: 777-781.
44. Gupta RK, Kumar R, Chawla S et al. Demonstration of scolex within calcified cysticercus cyst: Its possible role in the pathogenesis of perilesional edema. Epilepsia 2002; 43: 1502-1508.(Udani Vrajesh)
Abstract
Prevalence studies from India suggest that epilepsy prevalence is similar to developed nations. Neurocysticercosis (NCC) predominates as an etiology. A large treatment gap is still a public health problem. Benign epilepsies and West syndrome appear to be underrepresented in studies on classification of seizures/syndromes. Febrile seizures prevalence in India is similar to other countries and appear to be as benign. Risk factors of intractable epilepsy (IE) in Indian studies include early age of onset, neurodevelopmental abnormalities and certain seizure types. Perinatal injuries underlie many IE. Many IE are not truly intractable and respond to simple therapeutic measures. The ketogenic diet and surgery are other methods now being used in Indian centers. Neurocysticercosis and neonatal hypoglycemic brain injury, two widely prevalent etiologies are reviewed in detail.
Keywords: Childhood epilepsy; India; Neurocysticercosis; Hypoglycemia
Epilepsy is probably the most common disorder seen by the child neurologist. Age specific prevalence rates are highest in the 1st decade of life especially below the age of 2 years. This short review focuses on some aspects of pediatric epilepsy as it pertains to the Indian subcontinent.
Epidemiology
Incidence studies are not available from the Indian subcontinent. There are several prevalence studies however and a recent meta-analysis[1] suggests that the prevalence rate is 5.59 per 1000 population with no gender or geographical differences. This rate is surprisingly similar to the rate in developed countries. The CRESS study, in progress in Andhra Pradesh[2] surveyed more than 74,000 people in 22 villages and showed a similar prevalence rate of 6.2 /1000 population. The prevalence in Kerala, a more developed state with a high awareness of health related issues is similar at 4.9 / 1000.[3]
Febrile seizures, head injury and a positive family history of epilepsy have been identified as risk factors.[4] Neurocysticercosis accounted for 47% of all identifiable etiologies in the ongoing CRESS study[2] with a significant percentage showing brain calcifications. We have found neonatal hypoglycemic brain injury accounting for 23% of all symptomatic epilepsies with onset below the age of 3 years in a tertiary center.[5] Hot water epilepsy, a peculiar type of reflex epilepsy is reported from South India with a prevalence rate of 1.14 -2.99/1000.[4]
The treatment gap is a reality in countries like India and is indicative of the number of epileptics who do not receive AEDs ever. It has been estimated to be between 6% in urban Parsi populations,[6] 38% in well developed states like Kerala[3] to 78% in rural Karnataka[7] suggesting a serious public health delivery failure. Knowledge, attitudes and practices are negative[3], [8] with several misconceptions widely prevalent e.g. epilepsy is contagious, a form of insanity, a hindrance to education, marriage and employment. These negative attitudes seem to be more of a problem for the patient than seizure control. One-third of all those surveyed in AP [8] and 2/3 in Kerala[3] believed that ayurveda and other traditional medical practices are useful in treating epilepsy and this probably worsens the treatment gap.
Issues in classification of epileptic seizures and syndromes
There are only two recent systematic Indian studies on classification and both are from metropolitan tertiary care institutes with an inherent referral bias.[9],[10] It is not surprising therefore that both report low numbers of idiopathic genetic syndromes like the benign partial epilepsies (BPEs). Our own experience suggests otherwise. More than 50% of children suffered from symptomatic or cryptogenic localization-related epilepsies. As imaging was carried out in only about a third, many symptomatic cases may have been labelled as 'cryptogenic'. Juvenile myoclonic epilepsy (JME) was frequently seen amongst patients from Hyderabad, contrary to the Mumbai cohort, which focused on only patients below 12 years.
Surprisingly both investigations reported a low incidence of West syndrome, which is contrary to our experience. Another notable finding was that the epilepsy of a large numbers of patients could not be classified at all; this may be because of incomplete investigations due to limited resources. Generalised epilepsies seem to be twice as common as partial epilepsies[3] in one community survey, possibly due to overestimation as generalized tonic clonic seizures are more dramatic and more likely to be noticed.
Community-based clinical surveys like the one in Yelundur[11] suggest that the majority of epilepsies are idiopathic or cryptogenic. This has its limitations as the conclusions were drawn from clinical accounts. What is needed is a community-based study of a large population where all identified cases are fully investigated. This is the approach in the CRESS study[2]
Febrile Seizures
Febrile convulsions (FS) are seizures precipitated by fever i.e. not due to an intracranial infection or other definable central nervous system cause and are not preceded by afebrile seizures. FS pose a common problem in paediatric neurology. The incidence in Indian children less than five years of age, is two to five %.[12] Most studies suggest either a polygenic mode of inheritance,[13] or rarely autosomal dominant.[14] Early Indian studies suggested a prevalence of 10% in the community or 3% of all hospital admissions. But in the Yelandur[11] survey prevalence of FS was estimated at 3.28-5.71/1000 population, similar to western figures. About half the patients had active epilepsy with a male preponderance (72%) and a peak age of onset around 1 year. 90% of the attacks occurred during the first three years but there were some as late as eight years. The average duration was 1.48 years, upto the age of nine years. The role of febrile status epilepticus in the genesis of subsequent mesial temporal sclerosis (MTS) is a question not yet resolved fully though there seems to be some evidence for it's occurrence.[15] A history of FS could be elicited in 65% of 26 children and adolescents with MTS.[16]
EEG is of limited value. 15% of all EEGs recorded in one laboratory in Mumbai were for febrile seizures (Udani; unpublished observations) and only 21 of the 433 records were 'abnormal', with background rather than epileptic abnormalities. There were more abnormalities noted in the records of patients with complex FS than simple FS. Our current practice is not to request for an EEG whether the FS is simple or complex.
A consensus report mentioned that patients with simple FS should not be prescribed a prophylactic anti-convulsant drug for an extended period.[17] In its place, intermittent prophylaxis with oral or rectal diazepam or oral clobazam is advised.[18] The effect of oral antipyretics alone is disappointing.[19]
The overall risk of recurrence is 25 to 30% and this happens within the first year of onset in children less than 1 year. The risk factors include young age of onset, seizure within first two hours of fever, seizures with low-grade fever, positive family history of FS or epilepsy, complex FS and those who have had already more than two episodes.
The possibility of suffering from epilepsy in the future is 2 to 5%. This is increased in the presence of developmental or neurological abnormality, family history of epilepsy and in those with complex FS. It is as high as 50% when all unfavorable factors exist.[20] Only 2% of the Yelundur patients were neurologically and developmentally handicapped, and only 1.2% had later epilepsy. Most of these seizures were a part of an idiopathic partial or generalized epilepsy syndrome like BPE, JME and the newly described GEFS + (generalized epilepsy febrile seizures plus).[21]
Neuro-developmental outcome is not compromised in children with FS[22] and is linked to pre-existing developmental status and appropriate management of prolonged seizures.
Intractable Epilepsy and Epileptic Encephalopathies
These severe epilepsies constitute about 10-20% of all the epilepsies and up to 50% in specialty epilepsy clinics. Epileptic encephalopathies (EE) are syndromes in which the epilepsy itself is responsible for cognitive deterioration and amelioration of the seizures and/or regression of abnormal rhythms on EEG results in improvement. Syndromes like the West (WS), Lennox - Gastaut (LGS), Landau-Kleffner (LKS) / electrical status in slow wave sleep (ESES) and severe myoclonic epilepsy of infancy (SMEI) are some examples.
There is a paucity of information on this subject from India. In a study of 123 'difficult to control' epileptic children from Mumbai,[23] risk factors for intractability were, onset below two years of age, male sex, neurological and developmental handicaps and certain seizure types (complex partial, GTC, mixed seizures). Static encephalopathies were noted in about two-third, suggesting that epilepsy results from an abnormal substrate. Only 50% had an identifiable syndrome. An investigation carried out at AIIMS[24] found that 3.5% of all children attending the neurology clinic were suffering from epileptic encephalopathies. Of 94 patients, WS affected more than 50%. They noted an evolution of syndromes in four children from early myoclonic encephalopathy to WS to finally LGS. Prenatal aetiology predominated. We have found perinatal brain - damaging events to be the most common risk factor.[5] Perinatal encephaloclastic aetiologies were noted in 50%, with about a fourth of all cases due to neonatal hypoglycemia. Tuberous sclerosis, cortical dysplasias and neuronal migration disorders constituted another 20%. Infantile spasms and partial seizures were the most common seizure type and were often refractory to treatment. Seth et al examined 37 children with EE[25] and noted that more than 50% were getting spasms and in 70% there was history suggesting perinatal brain injury. The frequency of seizures lessened in only a third of their patients and only 6% were seizure- free.
Differing results are reported with the use of ACTH or steroids. In our randomised controlled trial of ACTH vs oral prednisolone given to children with WS, only 33% in the ACTH group were rendered free of spasm. Prednisolone was even less effective, only 14% were fully relieved.[26] Additional vigabatrin was effective in 50% of non-responders. These disappointing results are probably due to a high number of patients with symptomatic epilepsies and the long lag before the start of appropriate treatment.
In many older children and adults 'intractable' epilepsy is due to inappropriate polytherapy or low dosages. Simple measures like switching from polytherapy to monotherapy at correct dosage resulted in adequate control.[23] This has the added advantage of less adverse effects and a reduction in the cost per patient. These conclusions were reinforced when only 8% of samples referred for estimating drug levels were in the therapeutic range.[27] Compliance may be affected by cost of even standard anti-convulsant drugs (AEDs) like sodium valproate and carbamazepine. Hence earlier-marketed AEDs like phenytoin and phenobarbitone are still preferred for rural and low-income groups. Indeed phenobarbitone has been recommended in preference to phenytoin[28] because, besides being cheaper it has less side effects and its effect on behaviour is no worse.[29] New AEDs have not yet made a significant impact on the control of IE in countries like India except in small numbers in urban/semi-urban areas where costs and availability are less of an issue than in the rural hinterland.
The ketogenic diet is a high fat, low carbohydrate diet coming into its own in several centers. It seems to be useful at all ages and for all types of seizures. It is however fairly demanding and needs a lot of discipline. It is mainly used in young children where feeding behaviour is completely under the caretaker's control and appears to be a little more difficult and restrictive in vegetarian patients. In our own analysis of 49 patients between the ages of 18 m-16 years it was very successful in about 38% while about another 40% dropped out because of various reasons (Prabhu unpublished observations). Improved behaviour and decreased hyperactivity are other benefits.
Epilepsy surgery is a viable option in intractable seizures even in the developing world. Surgically remediable syndromes are mainly the lesional epilepsies, mesial temporal lobe epilepsies caused by hippocampal sclerosis, the hemispheric epilepsies like those caused by Rasmussen;s encephalitis or an unilateral destructive lesion and finally generalized epilepsies where corpus callosotomy would be very useful in reducing drop attacks. The availability of surgery is increasing with several tertiary centers offering comprehensive epilepsy services. Cost is a limiting factor though a recent investigation comparing surgical and medical management of adult TLE[30] concluded that surgery was much more cost-effective and was far cheaper in the long run. Our own experience in surgical treatment of pediatric epilepsy is similar. However pediatric epilepsy often has extra-temporal, multiple foci which make surgery more difficult and less successful than in adults.
Two etiologies are commonly seen in India and deserve separate mention. These are neurocysticercosis commonly seen in older children and adolescents and neonatal hypoglycemic brain injury commonly presenting in infancy.
Neurocysticercosis (NCC)
NCC needs special mention as it accounted for 47% of the identifiable etiology in the CRESS study.[2] Crude prevalence rate for NCC-related seizure disorder was 0.98 per 1000 population and for NCC-related acute symptomatic seizures it was 0.15 per 1000
NCC is caused by the larval stage of Taenia solium and is a major public health problem in non-Islamic developing countries including India.[31] Autopsy studies have shown a 1% rate of NCC, while clinical surveys implicated it in 0.4% of all children with neurological complaints.[32] There are state-wise differences with high rates in Delhi and surrounding areas and a much lower incidence in Kerala. Several types of NCC are described - parenchymal, subarachnoid, intraventricular and disseminated[31],[32] and this probably represents an interplay between infection load and host immune factors. Live cysts and granulomas are symptomatically 'active' while residual calcifications and gliosis are considered 'inactive'. In children 60-76% are single parenchymal NCC granulomas.[33] Calcifying lesions are much less common in children (15%) as compared to adults (55%).[33]
School-age children are most affected with a slight male preponderance.[31],[32],[33] The youngest child described was 9 months of age.[32] Recent onset focal seizures are the most common presenting symptom (70-95%). In the granuloma stage, seizures are acutely symptomatic, but seizures continue to occur in some children even after the lesion becomes calcified. In a series of 558 children, 51% of symptomatic partial epilepsies were due to either active or inactive NCC.[34] In most children neurological examination is normal. Neuroimaging is mandatory for diagnosis. Though the MRI is more sensitive, a contrast enhanced CT scan is probably good enough for transitional forms and is in fact superior for detection of calcifications.[35] Lesions are mainly at grey - white junctions in parietal, frontal and occipital lobes[32] and are rare in temporal lobes and the posterior fossa. Tuberculoma, a common differential diagnosis, on the other hand is seen more in basal regions.
Serology, CSF studies, stool examinations, radiology of leg muscles and fundoscopy are usually not cost effective[31],[32],[33] in childhood as the parasite load is not high. EEG must be recorded, as focal background slowing often indicates acute symptomatic seizures due to oedema, while focal spikes are seen in chronic remote symptomatic epilepsy.
The natural history of the single live cysticercus is variable and as seizures tend to be chronic and unremitting in this group, it seems reasonable to prescribe cysticidal drugs (albendazole preferred over praziquantel) along with a short course of steroids.[36] Caution should be exercised when there are multiple lesions as simultaneous death/degeneration of multiple cysts may cause severe oedema and even death.[37] The natural history of the cysticercus granuloma appears more uniform. 85-90% disappear decrease in size or become calcified over a few months. Studies on efficacy of cysticidal treatment in this granuloma group are conflicting.[33] Padma et al found no benefit,[37] but the group from the Post-Graduate Institute, Chandigarh (PGI) report short-term clinical benefit and regression of the CT/MRI lesion.[33] Similarly there are conflicting reports about control of seizures: Retrospective studies recommend treatment[38] but a long term, prospective, randomised trial argued against it.[39] What is clear is that treatment in children will not change the seizure remission rate, which is high anyway. What is questioned is whether treatment decreases residual calcifications and perilesional gliosis, in which the remission rate is lower. Some evidence is available supporting this notion.[40] Steroids given without albandazole along with AEDs may augment complete resolution of the granuloma[41] and reduce the risk of seizure recurrence. However another study has shown conflicting results.[42] Our policy is to treat all single and multiple active lesions with albandazole and steroids. The only time we do not prescribe cysticidal drugs is when there is disseminated disease or cysticercal encephalitis where there is danger of severe, often fatal increases in intracranial pressure. Anti-epileptic drugs are given till lesions resolve or become inactive. The problem is in managing calcified lesions. Some "re-activate" periodically with surrounding oedema and acute symptomatic seizures occur over several years necessitating prolonged AED therapy.[43] Gradient -echo MR imaging suggests that remnants of the scolex persist in these lesions and periodiacally incite surrounding oedema.[44] Albendazole would probably not be helpful at this stage as the parasite does not seem to be viable. The more worrying calcified lesion is one with associated gliosis where seizures may become refractory and where there is a true transition to an "epileptic" focus with presumably change in electrical circuitry and consequent persistent seizures. Surgical resection should be a consideration if AED therapy fails.
In conclusion there are several unanswered questions in NCC and epilepsy that would need a long-term multi-center trial to decide optimal management.
Neonatal Hypoglycaemic Brain Injury and Infantile Epilepsy
It is becoming clear that this very preventable condition is widely prevalent in India and accounts for a lot of long-term neurologic morbidity including epilepsy, mental retardation and visual impairment. We studied 100 consecutive cases of remote symptomatic epilepsy with onset in the first 3 years of life using imaging as a primary diagnostic tool for the etiology and found that 50% of children had perinatal brain damage as their epilepsy substrate.[5] As many as 23% had MRI evidence of neonatal hypoglycaemic brain injury. Birth histories were compatible in all while low plasma glucose had been documented in 14 /23 patients in the neonatal period. Low birth weight was a risk factor but interestingly 6/23 had birth weights of >2.5 kg. LSCS and poor feeding were other important risk factors suggesting that establishment of early feeding is mandatory after LSCS and if neglected can have disastrous consequences even in AGA babies. Infantile spasms was the most common seizure type. Associated disabilities were noted in > 50% and consisted of severe MR, autism, visual impairment and poor hand use. Interestingly, motor disabilities were much less common.
References
1. Sridharan R, Murthy BN. Prevalence and pattern of epilepsy in India. Epilepsia 1999; 40 : 631-636.
2. Chowdary GVS, Murthy JMK, Vijay S et al. Prevalence of Seizure Disorders Associated with Neurocysticercosis: A Community-based Study - Comprehensive Rural Epilepsy Study - South India (CRESSI). Presented at the Asian and Oceanic Congress of Neurology, Singapore 2004.
3. Radhakrishna J, Pandian D, Santoshkumar T et al. Prevalence, knowledge, attitude and practice of epilepsy in Kerala, South India. Epilepsia 2000; 41(8) : 1027-1035.
4. Bharucha NE. Epidemiology of epilepsy in India. Epilepsia 2003. 44(Suppl.1);9-11.
5. Munot P. Etiology of symptomatic epilepsy with onset in the first three years of life. Presented at the National conference of the IAP (Pedicon) 7th Jan 2005, Kolkata
6. Bharucha NE, Bharucha EP, Bharucha AE et al. Prevalence of epilepsy in the Parsi community of Bombay. Epilepsia 1988;29 : 111-115.
7. Mani KS. Epidemiology of epilepsy in Karnataka, India. Neurosci Today 1997; 1 : 167-174.
8. Thomas J, Seshadri V, Murthy JMK et al. Knowledge, attitude and practice of epilepsy in rural Andhra Pradesh, South India. Presented at the joint annual conference of the IAE & IES 2003, Vishakapatnam.
9. Shah KN, Rajadhyaksha SB, Shah VS et al. Experience with the International league against epilepsy classifications of epileptic seizures (1981) and epilepsies and epileptic syndrome (1989) in epileptic children in a developing country. Epilepsia 1992; 33: 1072-1077.
10. Murthy JM, Yangala R, Srinivas M. The syndromic classification of the International League against epilepsy: A hospital based study from South India. Epilepsia 1998; 39 : 48 - 54.
11. Mani KS, Rangan G. The Yelandur Model for Rural Epilepsy Control in India in Epilepsy. In Singhal BS, Nag D, eds. Indian Epilepsy Association Publication 2000; 16-32.
12. Febrile seizures: Long-term management of children with fever associated seizures. Summary of an NIH consensus statement. Brit Med J 1980; 281 : 277-279.
13. Anderson VE, WilcoxKJ, Hauser WA et al. A test of autosomal dominant inheritance in febrile convulsions. Epilepsia 1988;29 : 705-706.
14. Rich SS, Annegers JF, Hauser WA et al. Complex segregation analysis of febrile convulsions. Am J Human Genet 1987; 41 : 249-257.
15. Van Landingham KE, Heinz ER, Cavazos JE et al. Magnetic resonance imaging evidence of hippocampal injury after prolonged focal febrile convulsions. Ann Neurol 1998; 43: 413-426.
16. Shah KN, Rajadhyaksha. In Singhal BS, Nag, eds. Febrile Convulsions in Epilepsy in India. Indian Epilepsy Association publication 2000: 92-108.
17. Joint working group of the research unit of the Royal College of Physicians and the British Pediatric Association. Guidelines for the management of convulsions with fever. Brit Med J 1991; 303 : 634-636.
18. Knudsen FU. Febrile seizures : treatment and prognosis. Epilepsia 2000; 41 : 2-9.
19. Van Stuijvenberg M, Derksen-Lubsen G, Steyerberg EW et al. Randomised, controlled trial of ibuprofen syrup administered during febrile illnesses to prevent febrile seizure recurrences. Pediatrics 1998; 102: E51.
20. Annegers JF, Shirts SB, Hauser WA et al. Risk of recurrence with an initial unprovoked seizure. Epilepsia 1986; 27 : 43-50.
21. Singh R, Scheffer IE, Crossland K et al. Generalised epilepsy with febrile seizures plus: a common childhood - onset genetic epilepsy syndrome. Ann Neurol 1999; 45 : 75-81.
22. Nelson KB, Ellenberg JH. Prognosis in children with febrile seizures. Pediatrics 1978; 61 : 720-727.
23. Udani VP, Dharnidharka V, Nair A et al. Difficult to control epilepsy in childhood - a long term study of 123 cases. Indian Pediatr 1993; 30 : 1199-1206.
24. Kalra V, Gulati S, Pandey RM et al. West syndrome and other infantile epileptic encephalopathies - Indian hospital experience. Brain Dev 2002; 24: 130-139.
25. Seth A, Aneja S, Taluja V. Epileptic encephalopathies of early childhood. Indian Pediatr 2001; 38: 390 - 396.
26. Moharir M, Udani VP. Randomised controlled trial of high dose natural ACTH vs prednisolone in treatment of infantile spasms. Paper presented at Golden jubilee annual conference, Neurological Society of India;
Abstract: 217.
27. Karande SC, Dalvi SS, Kshirsagar NA. Shortcomings in the pharmacotherapy of epileptic children in Bombay. India J Trop Pediatr 1995; 41: 247-249.
28. Mani KS, Rangan G, Srinivas HV et al. Epilepsy control with phenobarbital or phenytoin in rural south India: the Yelandur study. Lancet 2001; 357 : 1316-1320.
29. Pal DK, Das T, Chaudhary G et al. Randomised controlled trial to assess acceptability of phenobarbital for childhood epilepsy in rural India. Lancet 1998; 351:19-23.
30. Rao MB, Radhakrishna K. Is epilepsy surgery possible in countries with limited resources. Epilepsia 2000; 41 (Suppl.4) : S31-S34.
31. Garg RK. Childhood Neurocysticercosis : Issues in diagnosis and management. Indian Pediatr 1995; 32 : 1023-1029.
32. Kalra V, Suri M, Jailkhani BL. A profile of childhood neurocysticercosis. Indian J Pediatr 1994; 61: 33-42.
33. Singhi P, Ray M, Singhi S et al. Clinical spectrum of 500 children with neurocysticercosis and response to albendazole therapy. J Child Neurol 2000; 15: 207-213.
34. Murthy JM, Yangala R. Etiological spectrum of localization-related epilepsies in childhood and the need for CT scan in children with partial seizures with no obvious causation - a study from south India. J Tropical Pediatr 2000; 46 : 202-206.
35. Thakur LC, Anand KS. Childhood neurocysticercosis in South India. Indian J Pediatr 1991; 58 : 815-819.
36. Garg RK. Medical management of neurocysticercosis. Neurol India 2001; 49 : 329-337.
37. Padma MV, Behari M, Misra NK et al. Albendazole in neurocysticercosis. Natl Med J India 1995; 8 : 255-258.
38. Del Brrutto OH, Santibanez R, Noboa CA et al. Epilepsy due to neurocysticercosis: analysis of 203 patients. Neurology 1992;42 : 389-392.
39. Carpio A, Santillan F, Leon P et al. Is the course of neurocysticercosis modified by treatment with antihelminthic agents Arch Intern Med 1995; 155 : 1982-1988.
40. Del Brutto OH. Prognostic factors for seizure recurrence after withdrawal of antiepileptic drugs in patients with neurocysticercosis. Neurology 1994; 44 : 1706-1709.
41. Mall RK, Agarwal A, Garg RK et al. Short course of prednisone in Indian patients with solitary cysticercus granuloma and new-onset seizures. Epilepsia 2003; 44(11) : 1397-1401.
42. Singhi P, Jain V, Khandelwal N. Corticosteroids versus albendazole for treatment of single small enhancing computed tomographic lesions in children with neurocysticercosis. J Child Neurol 2004; 19 : 323-327.
43. Nash TE, Patronas NJ. Edema associated with calcified lesions in neurocysticercosis. Neurology 1999; 53: 777-781.
44. Gupta RK, Kumar R, Chawla S et al. Demonstration of scolex within calcified cysticercus cyst: Its possible role in the pathogenesis of perilesional edema. Epilepsia 2002; 43: 1502-1508.(Udani Vrajesh)