Brief Reports

Indian Pediatrics 1999;36:1032-1038

Cognitive Function and Behavior in Epileptic Children of School Going Age

K. Jayashree, B. Talukdar, P.K. Srivastava, Usha Sharma

From the Department of Pediatrics, Maulana Azad Medical College, New Delhi 110 002, India and Department of Psychiatry, G.B. Pant Hospital, New Delhi 110 002, India.

Reprint requests: Dr. B. Talukdar, Professor, Department of Pediatrics, Maulana Azad Medical College, New Delhi 110 002, India.
Manuscript received: January 25, 1999;
Initial review completed: February 26, 1999;
Revision accepted: May 26, 1999

Cognitive function and behavior have been areas of special interest in pediatric epileptology as these are closely related to academic achievement and development of a healthy and useful personality. It has been observed that even with normal intelligence epilepsy increases the chances of academic under-achievement and school failure(1) and this is often attributed to specific cognitive dysfunc-tions and behavioral and psychosocial distur-bances(2). Besides primary brain pathology, certain factors like age at onset of seizure, seizure type and antiepileptic drugs (AEDs) have been observed to be closely associated with cognitive dysfunctions and behavioral disturbances although it is often difficult to determine the relative contribution of any given factor(3). There is paucity of data in this important area in our country, which should be useful in management of the educational aspect of epileptic children. This communication pertains to the study of cognitive function and behavior in epileptic children of school going age with special reference to certain seizure characteristics and medication.

Subjects and Methods

The study subjects were 90 school going epileptic children selected randomly from those attending our Pediatric Neurology Clinic for follow up and medication who were fully investigated. Thirty controls were selected from patients attending the Outpatient Department for minor ailments or sibs of patients admitted into the hospital. The controls were healthy children without personal and family history of seizure and belonged to similar socio-economic class as the epileptic subjects. Any child with clinically obvious or diagnosed mental retarda-tion and/or sensorimotor handicap was exclu-ded from the study.

After initial evaluation, the epileptics and the controls were subjected to tests of attention, concentration, memory and behavior. Attention and concentration were tested by Digit Forward, Digit Backward and Knox Cube Imitation Test(4). These standard psychometric tests are age independent and have inbuilt scoring pattern. Impairment of attention and concen- tration is indicated by lowering of test scores. Memory was tested by a questionnaire based on Wechsler's memory scale having three components of immediate memory, recent memory and remote memory. The questio-nnaire was pre-tested on 30 normal children for standardization. Impairment of memory is indicated by lowering of test scores. Behavior was tested using the Conner's Parent's Rating Scale(5). Disturbance in behavior in this scale is indicated by increase in test scores. The results of these tests were studied in the epileptic children as a whole and also corelated to seizure characteristics, namely, (a) age at onset of seizure, (b) seizure type, and (c) seizure control and medication characteristics, namely, (a) type of anti-epileptic drugs (AEDs) used, (b) doses of these AEDs, and (c) duration of AED therapy. Seizure type was based on ILAE classification. Seizure control was divided into two groups _`good' and `poor'. The control was arbitrarily considered to be good if seizures were controlled by one year of initiation of AED therapy and no subsequent recurrence, otherwise the control was arbitrarily considered to be poor.

The data was analyzed in computer using univariate and multivariate methods. Univariate analyses were done using ANOVA and Student `t' test. Multivariate analysis, using multiple regression analysis was done to find the best predictor for each cognitive function test taking those variables into account which showed statistically significant values in univariate analysis in several test parameters; these variables were age at onset of seizure, seizure control and number of drugs (polytherapy).

Results

The age of the epileptic children ranged from 5-14 years (median 10 yr) and that of the controls from 5-12 years (median 9 years). There were 56 males and 34 females (M:F = 1.6:1) in the former group and 20 males and 10 females, in the controls (M:F = 2:1). The distribution of cases according to the seizure characteristics studied namely age at onset, seizure type and seizure control and the results of cognitive function tests in these categories and also the controls are shown in Table I.

Table I^__ Results of the Tests of Cognitive Functions and Behavior in Epileptic Subjects and Their Relation to Seizure Characteristics.

Parameters	DF	DB	KCIT	IMM	RCM	REM	CPRS
Epileptics (n=90)	4.9 (1.4)	2.4 (1.6)	6.7 (1.9)	4.8 (1.6)	6.7 (2.1)	9.8 (3.1)	8.2 (5.5)
Controls (n=30)	5.96 (0.9)	3.1 (1.0)	7.2 (1.2)	5.4 (0.7)	7.4 (0.8)	11.8 (1.1)	4.7 (2.8)
	p = 0.0002*	p=0.03*	p=0.002*	p=0.04*	p=0.06	p=0.03*	p=0.003*

Age at onset
<5 yrs (n=32)	4 (1.3)	1.7 (1.7)	5.5 (2.2)	4 (2.1)	5.7 (2.8)	8.3 (4.0)	10.1 (5.6)
>5 yrs (n=58)	5.4 (1.1)	2.7 (1.5)	7.4 (1.4)	5.3 (1.1)	7.2 (1.3)	10.7 (2.1)	7.2 (5.2)
	p = 0.00003*	p=0.006*	p=0.004*	p=0.0007*	p=0.008*	p=0.007*	p=0.01*

Seizure Type
GTCS (n=56)	4.98 (1.3)	2.4 (1.7)	6.8 (1.9)	4.9 (1.6)	6.7 (2.1)	9.9 (3.3)	8.6 (5.8)
CPS (n=22)	4.9 (1.3)	2.4 (1.3)	6.3 (1.9)	4.7 (1.6)	7.0 (1.5)	10.1 (2.4)	7.8 (4.8)
SPS (n=12)	4.5 (1.2)	2.1 (1.8)	7.1 (2.1)	4.2 (1.9)	5.6 (2.1)	8.5 (3.0)	7.6 (5.0)
	p =0.25*	p=0.071	p=0.20	p=0.11	p=0.04*	p=0.10	p=0.26

Seizure Control
Good (n=72)	5.1 (1.2)	2.6 (1.6)	7.0 (1.7)	5.0 (1.5)	6.9 (1.9)	10.1 (2.8)	7.8 (5.4)
Poor (n=18)	4.1 (1.4)	1.5 (1.5)	5.6 (2.5)	4.0 (2.0)	5.8 (2.6)	8.6 (3.9)	9.9 (5.8)
	p = 0.003*	p=0.01*	p=0.02*	p=0.04*	p=0.06	p=0.05*	p=0.10

Values are depicted as Mean (SD).

Abbreviation: DF = Digit Forward, DB= Digit Backward, KCIT = Knox Cube Imitation Test,
                     IMM = Immediate Memory, RCM = Recent Memory, REM = Remote Memory,
                     CPRS = Conner's Parent's Rating Scale, GTCS = Generalized Tonic Clonic Seizure,
                     CPS = Complex Partial Seizure, SPS = Simple Partial Seizure.

* p value is statistically significant. (<0.05).

AEDs namely phenytoin, pheno-barbitone and carbamazepine did not show any significant variation compared to the control. However, the mean test scores for attention, concentration and all forms of memory were significantly lower in cases on polytherapy than the cases on monotherapy (p <0.05), despite the doses of AEDs in almost all the cases being within the therapeutic range (Table II). With respect to the variation in doses of the same drug, with phenytoin, the mean test scores for attention and the doses of AEDs in all others were within therapeutic range. The duration of AED therapy was as follows: 1 year = 29 (32.2%), 1-2 years = 31 (34.4%), 2-3 years = 16 (17.8%) and >3 years = 14 (15.6%) cases.

The mean test scores for attention, concentration, immediate memory and remote memory in the epileptic children as a whole were significantly lower than the controls (p <0.05) (Table I). The mean test score for attention, concentration and all forms of memory in the cases with age at onset of seizure <5 years were significantly lower than the cases

The distribution of epileptic subjects and the results of cognitive function tests related to AEDs are shown in Table II.

Table II^__ Results of the Tests of Cognitive Function and Behavior Corelated to Number and Doses of AEDs.

Parameters	DF	DB	KCIT	IMM	RCM	REM	CPRS
One Drug (n=74)	5.2 (1.2)	2.6 (1.6)	7.1 (1.7)	5.1 (1.3)	7.0 (1.7)	10.4 (2.5)	7.6 (5.2)
Two Drug (n=13)	3.8 (1.7)	1.3 (1.6)	5.3 (2.5)	3.3 (2.1)	5.1 (3.0)	7.4 (4.4)	10.0 (6.5)
Three Drug (n=3)	4.0 (1.0)	1.3 (1.1)	4.3 (1.5)	3.0 (1.0)	5.3 (3.0)	6.6 (4.0)	14.0 (0)
	p =0.002*	p=0.02*	p=0.0009*	p=0.001*	p=0.02*	p=0.004*	p=0.06

Phenytoin
5 mg/kg/d (n=55)	4.9 (1.2)	2.4 (1.6)	6.9 (1.7)	4.8 (1.5)	6.7 (2.0)	9.9 (2.9)	7.8 (5.5)
6 mg/kg/d (n=8)	5.6 (0.7)	2.6 (1.3)	6.8 (1.6)	5.1 (1.1)	6.8 (1.5)	10.3 (2.4)	10.1(5.7)
7 mg/kg/d (n=8)	5.0 (1.40)	2.1 (1.5)	5.7 (2.7)	4.3 (2.3)	6.0 (2.6)	8.7 (4.3)	9.3 (6.6)
8 mg/kg/d (n=5)	3.2 (1.3)	0.4 (0.8)	5.0 (3.0)	4.0 (2.5)	6.0 (3.3)	8.8 (4.9)	8.4 (4.7)
	p = 0.011*	p=0.05*	p=0.06	p=0.56	p=0.71	p=0.65	p=0.67

Phenobarbitone
4 mg/kg/d (n=12)	5.0 (1.4)	2.5 (1.7)	6.5 (2.2)	4.7 (1.6)	7 (1.5)	10.4 (2.6)	7.2 (6.2)
5 mg/kg/d (n=2)	3.0 (1.4)	1.0 (1.4)	4.5 (2.1)	3.5 (3.5)	5 (4.2)	6.5 (4.9)	11.5 (6.3)
> 6mg/kg/d (n=3)	3.6 (2.0)	1.0 (1.7)	4.7 (3.8)	2.6 (3.0)	3.3 (4.1)	4.6 (6.4)	16.3 (3.5)
	p =0.17	p=0.25	p=0.72	p=0.28	p=0.21	p=0.05*	p=0.08

Carbamazepine
10 mg/kg/d (n=7)	5.0 (1.6)	2.6 (2)	8.0 (0)	4.6 (1.6)	7.3 (1.6)	10.1 (2.7)	9.5 (4.5)
15 mg/kg/d (n=1)	2.0 (0)	3.0 (0)	3.0 (0)	6.0 (0)	2.0 (0)	3.0 (0)	16.0 (0)
20 mg/kg/d (n=3)	3.0 (0	0.6 (1.6)	3.6 (1.1)	2.3 (1.5)	3.3 (4.1)	3.6 (3.7)	15.6 (5.1)
	p = 0.14	p=0.22	p=0.06	p=1.1	p=0.18	p=0.04*	p=0.14

Values are depicted as Mean (SD).

* p vlaue is statistically significant (<0.05).

Seventy four cases (80%) were on monotherapy and 16 (20%) cases were on polytherapy. Except one case on Phenytoin and another case on Phenobarbitone, with as onset >5 years (p <0.05) (Table I). With respect to the seizure types, the mean test scores for recent memory in cases with simple partial seizure was significantly lower as compared to cases with complex partial and generalized seizure (p <0.05) (Table I). The mean test scores for attention, concentration and imme-diate and remote memory were significantly lower in cases having poor seizure control (p <0.05) (Table I). The mean test scores of cognitive functions studied based on individual studied predicted behavior disturbance.

Discussion

Attention, concentration and memory are important areas of cognitive function and perform a vital role in academic achievement. About 16-50% of epileptic children have been reported to be academic underachievers due to various cognitive dysfunctions(6,7). Academic problems in children with epilepsy have been shown to arise mostly from specific cognitive deficiencies rather than generalized cognitive dysfunctions(3). Impairment of different areas of cognitive function like memory, attention, concentration, abstract reasoning, information processing, auditory perceptual abilities and language processing abilities studied through a wide variety of neuropsychological tests have been reported to be relatively common in epileptic children irrespective of cause(1,6,8). The present study showing impairment of attention, concentration and memory in epileptic children of schoolgoing age also confirms these findings.

Early age at onset of seizure and poor seizure control which were associated with significant impairment of all the test parameters, appear to be very important factors that affect concentration were significantly lower with increasing doses and with phenobarbitone and carbamazepine, the mean test scores for remote memory were significantly lower with increas-ing doses (Table II). Regarding the duration of AED therapy, no significant difference was observed amongst the different duration categories. Results of assessment of behavior showed that the mean test scores for Conner' Parent's Rating Scale in the epileptic children as a whole were significantly higher (hence disturbed) than the controls (p <0.05) (Table I). The mean test scores were also significantly higher in cases with age at onset below 5 years. Seizure type, seizure control, type of AED, dose of AED and duration of therapy were not sigificantly related to the test scores.

Results of multivariate analysis revealed that out of all the variables, early age at onset of seizure (<5 years) was associated with significant impairment of attention, concen-tration and memory (DF, KCIT, IMM) and Polytherapy was associated with significant impairment of attention and concentration (KCIT) (Table III).

Table III^__Best Predictors (Reduced) Cognitive Function Tests by Multivariate Analysis.

Test	Predictor variable	Co-efficient	Standard error	p	Adjusted R (squared)
DF	Constant	4.18	0.83
	Early Age at Onset	_0.01	0.003	0.001	0.24

KCIT	Constant	7.12	0.86
	Early age at onset	0.01	0.006	0.02	0.18
	No. of drugs	_1.20	0.43	0.006
	(Polytherapy)

IMM	Constant	5.89	0.83
	Early age at onset	0.01	0.004	0.03	0.21

For abbreviations of tests, please refer to footnote on Table I.

However, the associations were not strong as only 18-24% of the variation could be explained. None of the parameters cognitive functions adversely in epileptic children. Several earlier studies also have shown impairment of different areas of cognitive function in epileptic children in association with early age at onset of seizure(9) and poor seizure control(10). These two factors appear to be interrelated as it has been observed that the longer duration needed for seizure control is frequently associated with early age at onset of seizure(11). The impairment of cognitive function (memory) in simple partial seizure as observed in this study has probably not been reported earlier although there are reports of such impairments in association with temporal lobe epilepsy(12). The basic pathology in temporal lobe epilepsy, now classified as complex partial seizures, lies in the temporal or frontotemporal lobes which are also responsible for several important cognitive functions. Since simple partial seizure can progress to complex partial seizures, the same basic pathology may be operative in both these types of seizures which may explain the impairment of memory in simple partial seizure as observed in this study.

Impairment of cognitive function in association with polytherapy has been reported in adults and recently in children also(13). Impairment of cognitive functions in association with higher doses of AEDs has also been reported earlier(14,15). Such correlations have also been reported with higher blood levels of AEDs(13,16). Our observation however needs further confirmation as the sample size of cases receiving higher doses of AEDs were small.

The significant disturbance of behavior documented in epileptic children in this study has been reported earlier(17). The significant association of early age at onset of seizure with disturbance of behavior has also been reported by some earlier researchers mostly in temporal lobe epilepsy(17).

An early age at onset of seizure was found to be associated with significant cognitive impairment as well as behavior disturbance. Early age at onset of seizure thus appears to be a important factor in determining the development of cognitive function and behavior in epileptic children probably through adverse influences on functional brain maturation, which is at its peak in the early years of life.

In conclusion, impairment of cognitive function and behavior disturbance are important problems in epileptic children of school-going age in our country. The cognitive function in them is adversely influenced by early age at onset of seizure, poor seizure control, partial seizure, AED polytherapy and probably increasing doses of AEDS, while behaviour is adversely influenced by early age at onset. Most of these influences are beyond our control, however, AED therapy is within our control. Avoidance of polytherapy and using AEDs in the lowest possible effective doses should be the rational way of treating epileptic children.

Acknowledgement

The authors are grateful to Mr. L. Satya Narayana, Assistant Director, Institute of Cytology and Preventive Oncology, Maulana Azad Medical College and Professor S. Ramji, Department of Pediatrics, Maulana Azad Medical College, New Delhi 110 002, for help with statistical analysis.

References

1. Holdsworth L, Whitemore K. A study of children with epilepsy attending ordinary school. Dev Med Child Neurol 1974; 16: 746-758.

2. Rutter M, Graham P, Yule WA. Neuro- psychiatric study in childhood. Clinics in Developmental Medicine, London, Heinemann 1970.

3. Rugland AL. Neuropsychological assessment of cognitive functioning in children with epilepsy. Epilepsia 1990; 31 (Suppl 4): S41-S44.

4. Wechsler D. A standardized intelligence and memory scale for clinical use. J Psychol 1945; 19: 87-95.

5. Goyette CH, Conners CK, Ulrich RF. Normative data on revised Conner's parent and teacher rating scale. J Abn Child Psychol 1978; 8: 471-490.

6. Mitchell WG, Chavez JM, Lee H, Guzman BL. Academic underachievement in children with epilepsy. J Child Neurol 1991; 6: 65-72.

7. Seidenberg M, Beck N, Goisser M. Academic achievement of children with epilepsy. Epilepsia 1986; 27: 753-759.

8. Stores G. Studies of attention and seizure disorders. Dev Med Child Neurol 1973; 15: 376-382.

9. O' Leary DS, Loveil M, Sackellares JC, Berents GB, Seidenberg M, Boll TJ. Comparison of age of onset of effects for partial and generalized seizures. J Nerv Ment Dis 1983; 171: 624- 629.

10. Farwell JR, Dodrill CB, Batzel LW. Neuro-psychological abilities of children with epilepsy.

Epilepsia 1985; 26: 395-400.

11. Dodrill CB, Troupin AS. Neuropsychological effects of carbamazepine and phenytoin. A reanalysis. Neurology 1991; 41: 141-143.

12. Stores G. Memory impairement in children with epilepsy. Acta Neurol Scand 1981; 89: 21-29.

13. Trimble Mr. Antiepileptic drugs, cognitive functions and behavior in children: Evidence from recent advances. Epilepsia 1990; 31 (Suppl 4): 30-34.

14. Sommerfield-Zuskind E, Zuskind E. Effect of Phenobarbital on the mentality of epileptic pa-tients. Arch Neurol Psychiatr 1940; 43: 70-76.

15. Aman MG, Werry JS, Paxton JW, Turbott SH. Effect of sodium valproate on psychomotor performance in children as function of dose, fluctuations in concentration and diagnosis. Epilepsia 1987; 28: 115-124.

16. Mitchell WG, Zhou Y, Charez JM, Guzman BL. Effects of anti-epileptic drugs on reaction time, attention and impulsivity in children. Pediatrics 1993; 91: 101-105.

17. Ounsted C, Lindsay J, Norman R. Biologic factors in temporal lobe epilepsy. In: Clinics in Developmental Medicine, No. 22. London, Heineman Medical, 1966.