Maternal Epilepsy and Cognitive/Psychiatric Status of Sons

Abstract

To the Editor: Developmental impairments have been reported among children and adolescents exposed in utero to antiepileptic drugs, particularly valproic acid.1,2 We examined whether mother's history of epilepsy was associated with IQ and psychiatric status in early adulthood in a cross-sectional analysis of prospectively collected data (described in detail elsewhere3). The study population included 21,051 men born as singletons in 1977-1983 in northern Denmark, presenting for the mandatory medical examination and intelligence testing to determine army fitness. Evidence of a mental illness was considered present among men with diagnoses F.XX per 10th Revision of the International Classification of Diseases. Low IQ was defined as a score below 85 on the conventional IQ scale. Draft evaluation data were linked, via birth registration, to mother's hospitalization records in the Danish National Patient Registry.4 We identified maternal diagnoses of epilepsy recorded since the Registry's inception, in 1977, until each draftee's date of birth. Analyses were adjusted for mother's age at delivery, mother's marital status, and birth order. The median age at conscription was 19.0 years (range, 18.0-25.0 years). Of the 21,051 men, 2505 (12%) were exempt from full evaluation on the basis of medical history, and did not have IQ measurements. Of the 21,051 men, 72 (0.3%) had been born to mothers with a prior hospitalization for epilepsy. A mental-illness diagnosis was recorded among 1451 men (6.9%), including 8 of 72 men (11.1%) whose mothers had been hospitalized for epilepsy. The adjusted prevalence ratio (PR) was 1.6 (95% confidence interval [CI] = 0.8 to 3.0) compared with unexposed men. Of the 18,546 men, 62 with IQ measurements (0.3%) had a record of maternal epilepsy hospitalization, and 2853 of them (15.4%) had a low IQ, producing an adjusted PR of 1.2 (95% CI = 0.7 to 2.0). Mean IQ scores were 95.3 and 100.0 for men with and without a prenatal record of maternal epilepsy, respectively. The adjusted mean difference associated with prenatal history of maternal epilepsy was −4.2 points (95% CI = −8.0 to −0.4). The associations tended to be stronger for maternal epilepsy diagnosed during the pregnancy (Table).TABLE: Association of Prenatal Exposure to Maternal Epilepsy With Mental Illness and IQ at Age 19 Among 21,051 Danish MenWe found greater prevalence of mental illness and mild cognitive impairment among men with prenatal exposure to maternal epilepsy. Men undergoing intelligence testing during army conscription represent a relatively healthy subgroup of the underlying birth cohort; therefore, the observed associations are not likely to be mediated by severe disability or congenital malformations. The 0.3% prevalence of maternal epilepsy in our data is low relative to the reported 0.6% for women of reproductive age in Denmark.5 Epileptic mothers identified here had exacerbation of disease requiring hospitalization. The stronger association seen for epilepsy hospitalization recorded during gestation could represent the effect of disease exacerbation, or of pharmacotherapy in women who were required to start or resume medication while pregnant. Both seizure type and medication type may affect offspring IQ.6 Earlier studies in children and adolescents showed 6-9-point lower mean IQ scores at age 3 for prenatal exposure to valproic acid as compared with other antiepileptic drugs.2 Confounding by indication is pervasive in studies of these medications, because therapy depends on the type and severity of the disease.1,7 Although we had no data on maternal pharmacotherapy, valproic acid has been on the market for more than 40 years, and some mothers in our study probably received that drug.8 Despite the low power of this analysis, our results are unlikely due to chance in the context of existing evidence. Barring potential confounding by maternal IQ, our results suggest that mild developmental deficits after maternal epilepsy—or its treatment—persist into adulthood even in relatively healthy men. Vera Ehrenstein Department of Clinical Epidemiology Aarhus University Hospital Aarhus Denmark [email protected] Henrik Toft Sørensen Department of Clinical Epidemiology Aarhus University Hospital Aarhus, Denmark Department of Epidemiology Boston University School of Public Health Boston, MA Lars Pedersen Department of Clinical Epidemiology Aarhus University Hospital Aarhus, Denmark