Predictors of Intractable Childhood Epilepsy

ORIGINAL ARTICLE Predictors of Intractable Childhood Epilepsy Muhammad Akbar Malik 1, Muhammad Haroon Hamid 2, Tahir Masood Ahmed 2 and Qurban Ali 3 ABSTRACT Objective: To determine the prognosis of seizures in epileptic children and identify early predictors of intractable childhood epilepsy. Study Design: Case-control study. Place and Duration of Study: The Epilepsy Centre of the Children s Hospital Lahore, from February 2005 to April 2007. Patients and Methods: All children (aged 1 month to 16 years) with idiopathic or cryptogenic epilepsy who were treated and followed at the centre during the study period were included. The patients who had marked seizures even after two years of adequate treatment were labeled as intractable epileptics (cases). Children who had no seizure for more than one year at last follow-up visit were the controls. Adequate treatment was described as using at least three anti-epileptic agents either alone or in combination with proper compliance and dosage. Records of these patients were reviewed to identify the variables that may be associated with seizure intractability. Results: Of 442 epileptic children, 325 (74%) intractable and 117 (26%) control epileptics were included in the study. Male gender (OR=3.92), seizures onset in infancy (OR=5.27), 10 seizures before starting treatment (OR=3.76), myoclonic seizures (OR=1.37), neonatal seizures (OR=3.69), abnormal EEG (OR=7.28) and cryptogenic epilepsy (OR=9.69) and head trauma (OR=4.07) were the factors associated with intractable epilepsy. Seizure onset between 5-7 years of age, idiopathic epilepsy, and absence seizures were associated with favourable prognosis in childhood epilepsy. Conclusion: Intractable childhood epilepsy is expected if certain risk factors such as type, age of onset, gender and cause of epilepsy are found. Early referral of such patients to the specialized centres is recommended for prompt and optimal management. Key words: Childhood epilepsy. Intractable epilepsy. Predictors. Antiepileptic drugs (AEDs). Myoclonic seizures. INTRODUCTION Assessment of the risk profile of a child with epilepsy may be used to tailor the treatment strategies. If known at onset, this risk profile can be used to identify children who might benefit from vigorous treatment. 1,2 Although the prognosis for the majority of epileptic children is good, upto 30-35% of such children do not have complete remission despite appropriate therapy with antiepileptic drugs (AEDs), a condition known as intractable epilepsy. 3-7 The potential of a Prognostic Group Specific Management Approach should be explored so that effective management plan could be used in a more rational and targeted fashion. However, determining the prognosis of patients when they first present with epilepsy is a difficult task. 8,9 The characteristics of patients are ill-defined and epileptic children with good and bad prognosis may be difficult to differentiate from each other at onset. 10 The potential 1 Department of Paediatric Neurosciences, The Children s Hospital, Lahore. 2 Department of Paediatric Medicine, The Children s Hospital, Lahore. 3 Department of EEG Lab, The Children s Hospital, Lahore. Correspondence: Dr. Akbar Malik, 218-D, Model Town, Lahore. E-mail: docmalikpk2000@yahoo.com Received July 13, 2007; accepted February 4, 2008. predictors of intractable epilepsy are age at onset, seizure etiology, EEG abnormalities at onset, number of seizures before treatment started and early response to antiepileptic drugs (AEDs). There is hardly any local data of epileptic children upon which the predictability of intractable epilepsy can be based. To resolve this issue, it remains essential to gather follow-up data of local patients and look for factors upon which predictability of prognosis can be based. Anticipation of the prognosis at onset is helpful in offering the best available evidence-based management. The objectives of this study were to determine the prognosis of seizures in epileptic children and identify early predictors of intractable childhood epilepsy. PATIENTS AND METHODS It was a case-control study conducted from February 2005 to April 2007. Children diagnosed to have epilepsy and being followed-up at the epilepsy centre of the Children s Hospital, Lahore, were included in the study. During the first visit, an extensive questionnaire was completed, containing pre-defined variables like gender, age of onset, number of seizures before starting the treatment, history of neonatal seizures, EEG findings, early response to antiepileptic drug(s), epilepsy type, seizure type and status epilepticus before commencing 158 Journal of the College of Physicians and Surgeons Pakistan 2008, Vol. 18 (3): 158-162

Predictors of intractable childhood epilepsy treatment etc. that may help predicting the intractability. Confirmation of the diagnosis of epilepsy by the paediatric neurologist was required for inclusion. The standard surface EEG records were performed by a qualified EEG technologist. Neuroimaging (CT/MRI brain) was performed when indicated to look for underlying structural abnormalities and such patients were excluded from the study. Information obtained from history, physical examination and EEG finding was used to classify epilepsy as idiopathic, symptomatic (excluded from the study) or cryptogenic (epilepsy with mental retardation without apparent etiology) according to ILAE. 3 All patients were prescribed the appropriate antiepileptic drugs (AEDs) according to the type of seizures; characteristics, efficiency, side-effects and interaction profile of the available drugs. 11,12 Patients were started on standard protocol of monotherapy (most commonly valproate, phenytoin, carbamazipine or phenobarbitone). Patients were subsequently evaluated at the epilepsy clinic after 4 weeks of commencement of treatment and then 3-4 monthly thereafter. A combination of drugs was used in patient, whose epilepsy remained uncontrolled despite treatment with two single drugs. 13,14 Dosage and compliance were monitored at the out-patient-clinics and adjusted as dictated by clinical circumstances, whereas patients having poor compliance or follow-up were excluded from study at the time of final analysis. Seizures were classified as generalized convulsive (e.g. tonic, tonic-clonic or myoclonic) or non-convulsive (e.g. absence) seizures and partial seizures, depending on the clinical presentation and the result of the studies described above. Outcome was defined as controlled or intractable epilepsy. Epilepsy was defined as intractable when despite taking adequate treatment for at least two years, patients continued to have significant seizures and were taken as cases. 16-19 Adequate treatment was described as using at least three antiepileptic drugs either alone or in combination with proper compliance and dosage. Control was defined as an epileptic having no seizures for a minimum of one year at last follow-up visit. Early response was defined as at least 75% reduction in seizure frequency and severity by 4 months of starting treatment. 15 Final evaluation of seizure control was done at minimum of 2 years of follow-up period. Demographic details, clinical characteristics and other relevant data was analyzed using statistical package SPSS (version 12.0). For categorical variables, Chisquare test or Fischer s exact test was applied and a p-value of < 0.05 was considered significantly. A univariate comparison between controls and cases was done for each possible predictor to calculate Odds Ratio (OR) and 95% Confidence Interval (CI). RESULTS Overall 506 epileptic children were enrolled at the clinic between February 2005 and April 2007. Sixty four patients were excluded from the study for failing to fulfill the inclusion criteria. Forty were lost to follow-up and 24 had poor compliance. The remaining 442 (61% male and 39% female) patients constituted the study population. Of those, 117 (26%) patients showed good control of epilepsy (controls), while 325 (74%) patients had intractable epilepsy according to the study criteria (Table I). Table I: Predictors of intractable epilepsy (n = 442). Variable Cases Controls Odds 95%CI p-value n=325 n=117 ratio (%) (%) Male gender 226 (70%) 43 (37%) 3.92 2.52-6.12 <0.001 Age at onset < 1 year 148 (46%) 16 (14%) 5.27 2.98-9.34 <0.001 Age at onset 5-7 years 35 (11%) 38 (32%) 0.25 0.14-0.42 <0.001 > 10 seizures before starting treatment 110 (34%) 14 (12%) 3.76 2.05-6.80 <0.001 History of neonatal seizures 46 (14%) 5 (4%) 3.69 1.43-9.53 0.004 Cryptogenic epilepsy 191 (59%) 15 (13%) 9.69 5.39-17.4 <0.001 Abnormal EEGs 204 (63%) 22 (19%) 7.28 4.34-12.18 <0.001 Early response to AED 79 (24%) 88 (75%) 0.10 0.06-0.17 <0.001 Absence seizures 2(1%) 29 (25%) 0.019 0.004-0.08 <0.001 Myoclonic seizures 13(4%) 0 1.37 1.29-1.45 0.02 Status epilepticus before presentation 32(9%) 5 (4%) 2.44 0.93-6.43 0.07 History of head trauma 41(13%) 4 (3.5%) 4.07 1.42-11.64 0.005 Sixty three percent (74/117) of the controlled epileptic children were female, compared to 70% (226/325) with intractable epilepsy being male [OR 3.92, 95% CI 2.52-6.12; p-value < 0.001]. Of the total, 164 patients had onset of epilepsy under one year of age and these infants showed more chance of intractable epilepsy [46% (148/325) cases vs. 14% (16/117) control; OR 5.27, 95% CI 2.98-9.34; p-value < 0.001]. Onset of epilepsy beyond infancy showed less chances of intractable epilepsy, more so when onset was between 5-7 years of age, Table II (OR 0.25, 95% CI 0.14-0.42; p-value < 0.001). Table II: Poor predictors of intractable epilepsy (n = 442). Variable Cases Controls Odds 95%CI p-value n =325 (%) n=117 (%) ratio Age of onset > 7 years 72 (22%) 29 (25%) 0.86 0.52-1.4 0.56 Generalized seizures at onset 160 (49%) 50 (43%) 1.29 0.85-1.98 0.22 Family history of epilepsy 30 (9%) 14 (12%) 0.74 0.38-1.46 0.39 Journal of the College of Physicians and Surgeons Pakistan 2008, Vol. 18 (3): 158-162 159

Muhammad Akbar Malik, Muhammad Haroon Hamid, Tahir Masood Ahmed and Qurban Ali Of 325 children with intractable epilepsy (cases), 110 (34%) had more than 10 seizure episodes before treatment was started compared to 14 (12%) children in control group (OR 3.76, 95% CI 2.05-6.88; p-value < 0.001). History of neonatal seizures was a risk factor [14% cases vs. 4% controls; OR 3.69, 95% CI 1.43-9.53; p-value 0.004]. Forty five (10%) patients had history of head injury and it was a poor prognostic variable (13% cases vs. 3% controls; OR 4.07, 95% CI 1.42-11.64; p-value 0.005). Statistically, family history of epilepsy did not show any impact on intractability of seizures (OR 0.74, 95%, CI 0.38-1.46; p-value 0.39). Significant abnormalities of EEG were documented among 226 (51%) of the 442 patients. A higher proportion of patients with abnormal EEG continued to have seizures during the study period compared to patients with normal EEG (63% cases vs. 19% controls; OR 7.28, 95%, CI 4.34-12.18; p-value < 0.001). Among the study population, 236 (53%) patients were classified to have idiopathic epilepsy and 206 (47%) patients had cryptogenic epilepsy. A higher proportion of patients with cryptogenic epilepsy continued to have seizures compared to patients with idiopathic epilepsy (59% case vs. 13% controls with cryptogenic epilepsy; OR 9.69, 95%, CI 5.39-17.4; p-value < 0.001). Early response to single antiepileptic drug was documented in 167 (38%) epileptic children and there was a significant trend of higher proportion of controls among patients who had early response to a single antiepileptic drugs (24% cases vs. 75% controls; OR 0.10, 95%, CI 0.06-0.17; p-value < 0.001). At the time of presentation, 238 (54%) patients had generalized seizures [194 (44%) tonic-clonic, 31 (7%) absence, 13 (3%) myoclonic] and 204 (46%) epileptic children presented with partial seizures. Overall, there was no significant difference between the proportion of patients with generalized onset seizures and partial onset seizures concerning intractability (49% cases vs. 43% controls with generalized epilepsy and 46% cases vs. 32% controls with partial seizures; OR 0.81, 95% CI 0.49-1.34; p-value 0.38). Thirty one (7%) patients had absence seizures at onset and higher proportion of these patients were seizures-free at the time of final analysis (25% controls vs. 1% cases; OR 0.02, 95% CI 0.004-0.008; p-value < 0.001). Myoclonic seizures were present among 13 (3%) patients and all of these had intractable epilepsy (OR 1.37, 95% CI 1.29-1.45; p-value 0.02). Though absence and myoclonic were generalized seizures; they were separated because of their excellent and poor prognosis, respectively. Status epilepticus before commencing antiepileptic drug(s) was (statistically non-significant) a risk factor concerning intractability (9% cases vs. 4% cases; OR 2.44, 95% CI 0.93-6.43, p-value 0.07). DISCUSSION The long-term prognosis of childhood seizures is uncertain. For the purpose of this study, it was hypothesized that it might be possible to predict the outcome of childhood epilepsy based on variables that could be identified at the time of diagnosis of epilepsy or shortly thereafter. Such a strategy might lead to individualized treatment regimens depending on the presumptive prognosis of the child. There seems to be a good general consensus that both adults and children with newly diagnosed epilepsy have a 63-65% chance of entering long-term remission. 20 In this cohort large proportion (74%) was found with an uncontrolled epilepsy over study time, whereas the proportion with controlled epilepsy was relatively small (26%). These findings are in contrast to those of Arts et al. who have reported 76% terminal remission after two years of starting treatment in childhood epilepsy. 5 This higher prevalence of intractability in referral centres like ours is quite expected as observed by Semah. 21 Kwan 1 reported remission of seizures in 63% of epileptic children in a hospital-based study, confirming the variable occurrence of intractable childhood epilepsy in different settings. Community-based studies would give the true prevalence of intractable epilepsies and its associated factors, as it would eliminate referral bias. This was a limitation of this study, as the settings were the tertiary care paediatric neurology centre with higher percentage of intractable epileptic referrals. Statistically, male gender was a risk factor for intractable seizures, in agreement with Akhondian 22 et al. but in contrast to the findings of Kwan and Brodie. 12 Another risk factor for developing intractable seizures was the onset of seizures before the age of one year, which was similar to other studies. 12,23-24 Berg et al. 25 demonstrated this association throughout the childhood but in this study, this was not significant after infancy except when onset was between the age of 5-7 years, where chance of intractable seizures was statistically less. In the present data, > 10 seizures before starting the treatment was a risk factor for intractable childhood epilepsy in agreement with findings of Kwong et al. 7 whereas Sillanpaa et al. 6 demonstrated that the number of seizures recorded before drug therapy was not related to the probability of remission. In another study of children with epilepsy, the initiation of treatment after 10 or fewer seizures did not influence the remission rate. 13 It may be that a large number of seizures before treatment be the result of the pathophysiological changes of epilepsy rather than the cause of refractory epilepsy. 8-9,16 We have documented that the history of neonatal seizures was a risk variable for drug resistant seizures as has been reported in other studies. 6,23,25 Neurological findings (cryptogenic epilepsy) was a risk factor for 160 Journal of the College of Physicians and Surgeons Pakistan 2008, Vol. 18 (3): 158-162

Predictors of intractable childhood epilepsy intractable childhood epilepsy in this study, which is in agreement with findings by Berg et al. 25 who reported intractable seizures in 34.6% of children with symptomatic/cryptogenic and 2.7% with idiopathic generalized childhood epilepsy. An early response to drug therapy confers a favourable prognosis and the present findings found that the response to the first antiepileptic drug was also a powerful prognostic factor (75% vs. 24%). This is also supported by other authors like Schmidth, Kawan and Dlugos. 26-28 On the whole, generalized seizures were not strongly associated with intractability, in agreement with Stephen. 29 In contrast, absence onset seizures were a strong favourable prognostic variable, which is in agreement with other studies. 5,16-18 Similarly, all 13 children with myoclonic seizures had intractable epilepsy and it appears to be a strong predictor as has also been observed by others. 6,25,30 A history of status epilepticus was an important factor for predicting development of intractable seizures, was in agreement with Ohtsuka, 16 Ko and Holmes, 31 but in contrast to the observations of Akhondian et al. 22 A strong association was noted between intractability and initial abnormal EEG, which is in agreement with other studies. 2,31-32 CONCLUSION A large number of patients with childhood refractory epilepsy may be identified at the onset or in the early course of disease based on gender, age, onset of seizures. This type of seizure response to first drug etc. conceptual framework has important implications for formulation of a Prognostic Group Base Approach to the management of newly diagnosed childhood epilepsy. Early identification of anticipated intractable epilepsy would allow prompt referral to the specialist services, where the diagnosis can be confirmed, seizures and syndromes classified, AED therapy optimized and suitability for additional combination therapies assessed. REFERENCES 1. Kwan P, Brodie MJ. Early identification of refractory epilepsy. N Engl J Med 2000; 342:314-9. 2. Berg AT, Levy SR, Testa FM, Shinnar S. Treatment of newly diagnosed paediatric epilepsy: a community-based study. Arch Pediatr Adolesc Med 1999; 153:1267-71. 3. ILAE Commission Report. The epidemiology of the epilepsies: future directions. International League Against Epilepsy. Epilepsia 1997; 38:614-8. 4. Kwan P, Sander JW. The natural history of epilepsy: an epidemiology view. J Neurol Neurosurg Psychiatry 2004; 75:1376-81. 5. Arts WF, Geerts AT, Brouwer OF, Boudewyn Peters AC, Stroink H, van Donselaar CA. The early prognosis of epilepsy in childhood: the prediction of a poor outcome. The Dutch study of epilepsy in childhood. Epilepsia 1999; 40:726-34. 6. Sillanpaa M, Schmidt D. Natural history of treated childhoodonset epilepsy: prospective, long-term population-based study. Brain 2006; 129 (Pt 3):617-24. 7. Kwong KL, Sung WY, Wong SN, So KT. Early predictors of medical intractability in childhood epilepsy. Pediatr Neurol 2003; 29:46-52. 8. Berg AT, Shinnar S, Testa FM, Levy SR, Smith SN, Beckerman B. Mortality in childhood-onset epilepsy. Arch Pediatr Adolesc Med 2004; 158:1147-52. 9. Shinnar S, Berg AT, O Dell C, Newstein D, Moshe SL, Hauser WA. Predictors of multiple seizures in a cohort of children prospectively followed from the time of their first unprovoked seizure. Ann Neurol 2000; 48:140-7. 10. French JA. Response to early AED therapy and its prognostic implications. Epilepsy Curr 2002; 2:69-71. 11. Shinnar S, Berg AT. Does anti-epileptic drug therapy prevent the development of "chronic" epilepsy? Epilepsia 1996; 37:701-8. 12. Kwan P, Brodie MJ. Clinical trials of anti-epileptic medications in newly diagnosed patients with epilepsy. Neurology 2003; 60 (11 Suppl 4):S2-12. 13. Camfield PR, Camfield CS, Gordon K, Dooley JM. If a first antiepileptic drug fails to control a child's epilepsy, what are the chances of success with the next drug? J Pediatr 1997; 131: 821-4. 14. Brodie MJ, Kwan P. Staged approach to epilepsy management. Neurology 2002; 58 (8 Suppl 5):S2-8. 15. Percucca E. Pharmacoresistance in epilepsy: how should it be defined? CNS Drugs 1998; 10:171-9. 16. Ohtsuka Y, Yoshinaga H, Kobayashi K, Murakami N, Yamatogi Y, Oka E, et al. Predictors and underlying causes of medically intractable localization-related epilepsy in childhood. Pediatr Neurol 2001; 24:209-13. 17. Sillanpaa M, Jalava M, Kaleva O, Shinnar S. Long-term prognosis of seizures with onset in childhood. N Engl J Med 1998; 338:1715-22. 18. Arts WF, Brouwer OF, Peters AC, Stroink H, Peeters EA, Schmitz PI, et al. Course and prognosis of childhood epilepsy: 5-year follow-up of the Dutch study of epilepsy in childhood. Brain 2004; 127(Pt 8):1774-84. 19. Berg AT, Shinnar S, Levy SR, Testa FM, Smith-Rapaport S, Beckerman B. Early development of intractable epilepsy in children: a prospective study. Neurology 2001; 56:1445-52. 20. Cockerell OC, Johnson AL, Sander JW, Shorvon SD. Prognosis of epilepsy: a review and further analysis of the first nine years of the British National General Practice Study of Epilepsy: a prospective population-based study. Epilepsia 1997; 38:31-46. 21. Semah F, Picot MC, Adam C, Broglin D, Arzimanoglou A, Bazin B, et al. Is the underlying cause of epilepsy a major prognostic factor for recurrence? Neurology 1998; 51:1256-62. 22. Akhondian J, Heydarian F, Jafari SA. Predictive factors of pediatric intractable seizures. Arch Iran Med 2006; 9: 236-9. 23. Udani V. Evaluation and management of intractable epilepsy. Indian J Pediatr 2000; 67 (1 Suppl): S61-70. 24. Berg AT, Langfitt J, Shinnar S, Vickrey BG, Sperling MR, Walczak T, et al. How long does it take for partial epilepsy to become intractable? Neurology 2003; 60:186-90. 25. Berg AT, Levy SR, Novotny EJ, Shinnar S. Predictors of intractable epilepsy in childhood: a case-control study. Epilepsia Journal of the College of Physicians and Surgeons Pakistan 2008, Vol. 18 (3): 158-162 161

Muhammad Akbar Malik, Muhammad Haroon Hamid, Tahir Masood Ahmed and Qurban Ali 1996; 37:24-30. 26. Schmidt D, Gram L. Monotherapy versus polytherapy in epilepsy: a reappraisal. CNS Drugs 1995; 3:194-208. 27. Kawan P, Brodie MJ. Effectiveness of first anti-epileptic drug. Epilepsia 2001; 42:1255-60. 28. Dlugos DJ, Sammel MD, Strom BL, Farrar JT. Response to first drug trial predicts outcome in childhood temporal lobe epilepsy. Neurology 2001; 57: 2259-64. 29. Stephen LJ, Kwan P, Brodie MJ. Does the cause of localizationrelated epilepsy influence the response to anti-epileptic drug treatment? Epilepsia 2001; 42: 357-62. 30. Chawla S, Aneja S, Kashyap R, Mallika V. Etiology and clinical predictors of intractable epilepsy. Pediatr Neurol 2002; 27:186-91. 31. Ko TS, Holmes GL. EEG and clinical predictors of medically intractable childhood epilepsy. Clin Neurophysiol 1999; 110: 1245-51. 32. Casetta I, Granieri E, Monetti VC, Gilli G, Tola MR, Paolino E, et al. Early pedictors of intractability in childhood epilepsy: a community-based case-control study in Copparo, Italy. Acta Neurol Scand 1999; 6:329-33. 162 Journal of the College of Physicians and Surgeons Pakistan 2008, Vol. 18 (3): 158-162