For more than two decades, the fetuses of pregnant women carrying a fetus at risk of virilizing congenital adrenal hyperplasia (CAH) have received dexamethasone (DEX). While this measure makes traumatizing genital surgery unnecessary by preventing virilization, some have voiced concern about how glucocorticoid therapy may affect fetal programming. This is the first study assessing long-term cognitive development by direct examination of children exposed prenatally to DEX. Twenty-six children at risk (12 boys and 14 girls), along with 35 age- and gender-matched control children, underwent a wide range of neuropsychological tests at 7 to 17 years of age. DEX was administered from postmenstrual week 6 to 7. Women with a previous severely affected child received 20 μg/kg daily up to a maximum daily dose of 1.5 mg.
DEX-exposed and control children performed equally well on tests of verbal comprehension and perceptual organization. Mean full-scale IQ scores were comparable in the two groups. No differences were noted on tests of immediate and long-term memory. In evaluating working memory using a freedom-from-distractibility index, scores on the Digit Span, a subtest measuring verbal working memory, were lower in DEX-exposed girls and boys than in control subjects. There were no differences in performance on tests of impulse inhibition or processing speed. On child-completed questionnaires, DEX-exposed children reported having trouble figuring out the answers at school, independently of psychometric intelligence scores. Children without CAH who received short-term treatment had increased self-rated social anxiety. Parental ratings of their child’s performance in reading, writing, spelling, and mathematics disclosed no significant differences after adjusting for full-scale IQ.
Deficits in some aspects of cognitive function apparently are present in children and adolescents exposed in utero to maternal glucocorticoid therapy, given to prevent virilization when the fetus is at risk of CAH. Larger-scale retrospective studies are needed to confirm or question these findings.