Maternal pregnancy folate status and association to language impairment in children of women with epilepsy after prenatal antiseizure medication exposure

Abstract

Background: 2-8 out of 1000 pregnancies occur in women with epilepsy. Most women use antiseizure medication (ASM) during pregnancy to avoid potentially harmful epileptic seizures. Fetal exposure to ASM is associated with an increased risk of congenital malformations and adverse neurodevelopment. It is important to identify factors that modulate the risk of ASM-associated fetal harm. Folate is a B vitamin important for normal brain development and associated with favorable neurodevelopmental outcome in the children. Many ASM interact with folate metabolism causing reduced folate concentrations, in addition to the folate-lowering effect of the pregnancy itself.

Aims: The aims of this research project were to examine the risk of language impairment in children of women with epilepsy aged 5 and 8 years and associated risk factors, and to examine the association between maternal folate status during pregnancy, plasma ASM concentrations, and ASM-associated language impairment in children of women aged 1.5-8 years. The overall aim was to find ways to improve the outcome in children of women with epilepsy after prenatal ASM exposure.

Material and methods: The data source was The Norwegian Mother, Father and Child Cohort Study (MoBa). MoBa is a prospective, population-based pregnancy cohort study conducted by the Norwegian Institute of Public Health and linked to the Medical Birth Registry of Norway (MBRN). Pregnant women from all over Norway were invited to participate during the years 1999-2008. 41% of the invited pregnancies consented to participate. Data on medical background, social background, epilepsy diagnosis, ASM use, vitamin supplement use, and language impairment in the children were collected from parental-reported questionnaires during the pregnancy and after birth when the child was at age 1.5, 3, 5 and 8 years. We measured plasma ASM concentrations and vitamin and metabolite concentrations in maternal samples from gestation week 17-19 and ASM concentrations in umbilical cord samples collected immediately after birth. Language impairment was examined based on the following parental-reported screening instruments: The Ages and Stages Questionnaire (ASQ); a one-item question regarding expressive language delay; the Speech and Language Assessment Scale (SLAS); and the Norwegian instrument The Twenty Statements about Language-Related Difficulties (Language 20).

Results: The maternal epilepsy cohort in MoBa consisted of 346 ASM-exposed children of 297 women and 388 ASM-unexposed children of 323 women with epilepsy. The control group consisted of 113,674 children of 94,338 women without epilepsy. For ASM-exposed children, the adjusted odds ratio (aOR) for language impairment at age 5 years was 1.6, 95% confidence interval (CI) 1.1-2.5 compared to children of women without epilepsy. At age 8 years, the corresponding aOR for language impairment was 2.0, CI 1.4-3.0. Maternal use of periconceptional folic acid was associated with decreased risk of ASM-associated language impairment at ages 1.5, 3, 5, and 8 years. High maternal valproate concentrations correlated with poor language score at ages 1.5 (ASQ: Spearman’s rho (r) = -0.50, n = 17, p-value (p) = 0.04) and 5 years (ASQ: r = -0.77, n = 9, p = 0.02; Language 20: r = 0.82, n = 9, p = 0.01). High maternal ASM concentrations correlated with high concentrations of unmetabolized folic acid (UMFA; n = 199, r = 0.22, p = 0.002), and with low concentrations of riboflavin (n = 188, r = -0.32, p <0.001) and metabolically active pyridoxine (PLP; n = 188, r = -0.19, p = 0.01). There was no association between ASM and plasma niacin status.

Conclusions and implications: Fetal ASM exposure in utero may have long term consequences for language outcome in children of women with epilepsy. Use of folic acid in the periconceptional period was associated with better language outcome in ASM-exposed children. Maternal ASM concentrations in the second trimester interacted with folate metabolism and non-folate B vitamins associated with folate function, and with language score. Folate may play a role in ASM-associated risk of language impairment. Periconceptional folic acid supplementation seems to improve the outcome in ASM-exposed children of women with epilepsy.