Cognitive Functions in Children at Risk of CAH Treated Prenatally with Dexamethasone

« Back to Volume 23, Issue 2, June 2007 - Table of Contents

Female fetuses with congenital adrenal hyperplasia (CAH) are at risk for masculinization of the reproductive structures due to increased androgen synthesis by the adrenals. Given perceptions of stigma attached to atypical genital appearance, as well as functional problems (eg, urogenital sinus) associated with cases of severe virilization, expectant mothers at risk of having a child affected by CAH are offered a potent corticosteroid, dexamethasone (DEX). This treatment is provided early in the pregnancy, before the gender and CAH status of the developing child are known. DEX administered early in the pregnancy has shown efficacy in reducing and/or eliminating masculinization; however, there are questions about harmful long-term effects given: (1) animal research showing adverse effects on somatic development and cognitive function and (2) a dearth of data on effects in humans. Adding to this concern is that only one in 8 children of parents who are carriers is affected by CAH; this means that 87.5% of fetuses are treated without the potential of benefit. The purpose of the present study was to assess children and adolescents cognitive development and social anxiety among children who were prenatally treated with DEX.

Twenty-six children (46% male; age range 7 to 17 years, mean ± SD 10.95, ± 2.33) of a population of 40 who were at risk of being affected with CAH and treated with DEX prenatally comprised the study cohort (35% refusal rate). Treatment was initiated at gestational week 6 to 7, was continued to term for the 4 CAH-affected females, but was discontinued at the end of the first trimester for CAH unaffected females and all males. Two control groups were recruited: 10 siblings to children treated with glucocorticoids and 25 children selected from the Swedish National Registry matched for age and sex; data from the 2 control groups were combined in analyses. Similar results were obtained for the 2 control groups on all measures, and data for the 2 groups were therefore combined. Participants were administered neuropsychological measures and they and their parents completed several questionnaires addressing psychological and educational areas.

Statistically significant differences were not found between:

  • DEX-treated CAH-affected, DEX-treated CAH-unaffected, and control groups for measures of Verbal Comprehension, Perceptual Organization, Full Scale IQ, and Processing Speed Index.
  • DEX-treated (CAH-affected and unaffected combined) and control groups for measures of handedness, subjective experience of social anxiety, placement in special class/special school or repeating a class, having any academic or other problems in school, and parental ratings of academic grades.
  • DEX-treated CAH-affected and DEX-treated CAH-unaffected groups from immediate and long-term memory, visuospatial working memory, and impulse inhibition.

Significant differences were found between:

  • DEX-treated and control groups for working memory and Stroop subtests assessing verbal processing speed such that those in the treated group performed poorer. With regard to Scholastic Competence, DEX-treated CAH-unaffected children rated themselves lower than those in the control group; DEX-treated CAH-affected children’s ratings did not differ from control group scores.
  • DEX-treated CAH-unaffected and control groups on the social anxiety measure such that the treated group rated themselves as more anxious.

The authors highlighted the finding that DEX-treated CAH-unaffected children performed poorer than controls on measures of verbal working memory, which was supported by patient-reported difficulties on the Scholastic Competence questionnaire. They also pointed to the finding of increased social anxiety for these short-treated children compared with controls. In their conclusion, they discussed the ethical dilemma regarding future use of DEX, especially when its short-term use contributes to adverse effects in CAH-unaffected children (ie, the majority of cases). Noting this concern and the paucity of data on the metabolic and neuropsychological effects of prenatal DEX treatments, the authors recommended that all future prenatal treatments be performed in the context of multicenter clinical trials. Further, they stated the importance of informing parents about the potential risks, benefits, and uncertainties regarding this treatment.

Hirvikoski T, Nordenstrom A, Lindholm T, et al. Cognitive functions in children at risk for congenital adrenal hyperplasia treated prenatally with dexamethasone. J Clin Endocrinol Metab. 2007;92:542-8.

Editor’s Comment

The objective of prenatal DEX for CAH affected females is to prevent or minimize the masculinization of the reproductive structures. Nevertheless, based on these neurocognitive findings, the authors call for caution in implementation of this treatment protocol; a warning echoed by basic animal researchers (including those conducting studies in nonhuman primates1 ) and clinicians.2 In general, assessment of risks of any treatment should be evaluated in the context of potential benefits. Given the lag time between confirmation of diagnosis at week 10 to 12, and initiating DEX treatment beginning postmenstrual week 6 to 7, 7 of 8 children are treated without any benefit. Furthermore, not all affected girls would exhibit the same degree of genital masculinization if left untreated.3 In light of evidence that parents in larger numbers are deferring early surgery for their daughters,4 and because orgasmic function and erectile sensation can be compromised by clitoral surgery, it has been recommended that surgery should only be considered in cases of severe masculinization (Prader III–V).5 Accordingly, even fewer than 1 in 8 offspring would “benefit” from the avoidance of surgery prenatally because it would only be the parents of a subgroup among those affected who would perceive a need for surgery.

Regarding the research methods and statistical analyses of this study, there are good reasons for replicating and extending these findings: the sample size is small limiting statistical power; the assessment of school performance was self-reported and based on a measure of limited utility in detecting subtle learning problems; the generalizability of the findings are limited to children younger than 18 years, ie, the pattern of neurocognitive effects may depend on chronologic age.

In addition to conducting additional studies on neurocognitive outcomes, it would be important to examine what parents are currently being told about the risks and benefits. Are they given information that minimizes long-term side effects while emphasizing short-term safety? This would be a fertile area for research in shared decision-making between parents and healthcare professionals.

David E. Sandberg, PhD

References - (linked to Pubmed Links)

  1. Coe CL, Lubach GR. Developmental consequences of antenatal dexamethasone treatment in nonhuman primates. Neurosci Biobehav Rev. 2005;29:227-35.
  2. Seckl JR, Miller WL. How safe is long-term prenatal glucocorticoid treatment? JAMA. 1997;277:1077-9.
  3. Nimkarn S, New MI. Prenatal diagnosis and treatment of congenital adrenal hyperplasia. Horm Res. 2007;67:53-60.
  4. Lee PA, Witchel SF. Genital surgery among females with congenital adrenal hyperplasia: changes over the past five decades. J Pediatr Endocrinol Metab. 2002;15:1473-7.
  5. Lee PA, Houk CP, Ahmed SF, Hughes IA, in collaboration with the participants in the International Consensus Conference on Intersex organized by the Lawson Wilkins Pediatric Endocrine Society and the European Society for Paediatric Endocrinology. Consensus statement on management of intersex disorders. Pediatrics. 2006;118:488-500.

« Back to Volume 23, Issue 2, June 2007 - Table of Contents


Last Updated: 04/30/2008

Copyright © 2003-2008 Prime Health Consultants, Inc.