Learning disability: occurrence and long-term consequences in childhood-onset epilepsy

Abstract

This study analyzed the occurrence of learning disability (LD) in adults with childhood-onset epilepsy and the impact of LD on medical and social outcome. Any LD occurred in 76%: in 57% of mentally normal (IQ > 85), in 67% of mentally near-normal (IQ = 71–85), and, self-evidently, in all mentally retarded (IQ < 71) adults. Half of the patients (51%) with LD had mental retardation. In multivariate analysis, mental retardation and subsequent LD were predicted by occurrence of cerebral palsy (odds ratio [OR] = 3.83; 95% confidence interval [CI] = 1.77–8.28, P = 0.0006), onset of epilepsy before the age of 6 years (OR = 3.63, 95% CI = 1.57–8.42, P = 0.0026), and poor early effect of drug therapy (OR = 2.78, 95% CI = 1.43–5.39, P = 0.0025). Among mentally normal or near-normal subjects, a symptomatic etiology of epilepsy was the only predictor (OR = 7.72, 95% CI = 3.02–19.76). The degree of LD significantly affected medical, social, and educational long-term outcomes.

Introduction

People with epilepsy have for centuries been believed to be notorious underachievers at school or incapable of attending school due to a cognitive deficit. The belief has invariably been based on studies about epilepsy in special schoolchildren and institutionalized patients and, thus, biased by selection [1]. Indeed, a WHO report found the main problem to be the lack of truly unbiased samples for testing [2]. Early data from institutions showed 35–60% of the mentally retarded to have epilepsy [3], [4]. More recent, population-based studies [5], [6], [7] found epilepsy in 23–33% of mentally retarded people. On the other hand, among those with epilepsy, 31–41% are reportedly retarded [8], [9], [10], [11], [12].

There are still only few epidemiological research-based data on cognitive deficits causing learning disturbances in people who have epilepsy. One fundamental problem is the inconsistent use of the terms mental retardation (MR) and learning disability (LD). No consensus so far exists, because, in a very recent discussion, some would like to keep the terms separated [13], some others want to have the term LD as an umbrella term for all conditions with learning disability [14], and the remaining authors simply do not discuss the problem. Most of the prevalence studies are cross-sectional and give only rates of prevalence of MR [8], [10], [11], [12], [15], [16], [17], [18], [19], omitting those who have LD but normal intelligence (IQ > 70) [20]. In the National Child Development Study of 15,496 children, 64 children had developed epilepsy by the age of 11. At that age, 43 were in normal schools, 20 were in special schools, and 1 child was at home [10]. In a Scandinavian study of 36,500 children, 155 had epilepsy. Of them, 20% had both MR and cerebral palsy and an additional 21% had MR only [11]. In another Scandinavian population-based study, MR occurred in 39% of 198 children with epilepsy who were derived from an unselected population of 38,500 children [17]. Still another Scandinavian population study of 49,000 children [12] reported MR in 38% (mild in 14% and severe in 24%). Cerebral palsy was much more common among children with severe MR (49%) than among those with mild MR (14%). From the same patient series, Beckung et al. [21] later recruited those who had no other neurological deficits, such as MR, cerebral palsy, and similar deficits, for the assessment of their sensory and motor function. Thirty percent of their study subjects with epilepsy but without additional neurological impairments exhibited gross motor functional deficits in balance, coordination, and speed. The study did not consider the intelligence level among these nonretarded subjects.

Brorson and Wranne [22] followed 194 of 195 children for 12 years with respect to survival and seizure prognosis. An initial “neurodeficit” or abnormal neurological examination, MR, or frequent seizures or any combination proved to be a negative prognostic factor. Forty-one percent of patients with a neurological deficit and 79% of those without any neurodeficit were in 3-year terminal remission at the end of follow-up.

Camfield et al. [23] classified their patients with respect to intelligence level into four groups: normal, mild to severe mental retardation, learning disorder, and learning disorder with behavioral disorder. The latter two groups were not defined. Twelve percent were clinically assessed as mentally retarded and 16% had a learning disorder. No data were given on the remaining 4%. A Japanese clinic-based follow-up study [18] found a significantly higher remission rate in the mentally normal than in the mentally retarded (IQ < 70) (76% vs 37%). Again, no distinct difference was observed between MR and LD, but 72% attended regular classes in an ordinary school. Of 49 mentally retarded, 14% attended regular classes at ordinary schools; 27% attended special classes at ordinary schools, and the remaining 41% were at schools for handicapped, in institutions, or invalids at home.

Our purpose was to study the occurrence of LDs and their effects on medical and social outcomes and factors that might affect outcome. We hypothesized that, even in the long run, the outcome is not as favorable in the learning disabled as in patients with no LD.