Congenital adrenal hyperplasia : pre-and postnatal treatment effects on cognition, behavior and brain resting-state functional connectivity

Abstract

Congenital adrenal hyperplasia (CAH), an autosomal recessive disorder affecting adrenal steroid synthesis, is linked to impaired adrenal synthesis of cortisol and aldosterone. The increased production of androgen precursors in the adrenal cortex during fetal life leads to the virilization of external genitalia in girls with CAH already in utero. Prenatal virilization can be reduced or alleviated entirely by dexamethasone (DEX) treatment in affected girls, but with potentially adverse effects on growth parameters, cognition, behavior, brain structure and brain networks and metabolism. The effects of synthetic glucocorticoids (GCs) on fetal development are time- and dose-dependent, with different outcomes in early vs. late gestational treatment. Patients with CAH are also treated postnatally with life-long GC replacement therapy to mimic the physiological levels following the circadian rhythm. Because of the inherent difficulties in mimicking the natural rhythm of cortisol secretion, there is a risk of under- or over-treatment during the individual’s life span, with potentially adverse effects on brain function and structure, metabolism and quality of life.

In this thesis the long-term effects of pre-and postnatal DEX treatment were investigated in a cohort of 206 individuals: patients with CAH not prenatally treated with DEX (n=71), a small cohort of patients with CAH prenatally treated with DEX (n=13), individuals at risk of CAH who were prenatally treated during the first trimester of fetal life (n=18) and population controls (n=104). Patients with CAH were diagnosed through the national neonatal screening program. We hypothesized that GC treatment could impact cognitive performance, behavior and resting-state functional connectivity of the brain. Compared to controls, we did not identify significant differences in cognitive performance in children and adolescents aged 7-17 years (mean age 11.5 years) with CAH compared with the population controls. However, patients with the salt-wasting (SW) genotype performed worse on a subtest assessing visuospatial working memory compared to patients with simple-virilizing (SV) CAH. The scores on this subtest were in the average range for the general population. In summary, early diagnosis may optimize treatment and benefit cognitive development.

The small cohort of CAH cases prenatally treated with DEX (6 females, 5 males) showed poorer performance in a subtest assessing verbal intelligence than patients not treated with DEX. The behavioral outcomes evaluated using parental and self-rated questionnaires reflected a good overall adjustment in children and adolescents with CAH compared to controls. The parental questionnaires suggested more social problems in CAH patients. Moreover, the parents of children with CAH prenatally treated with DEX (8 girls, 5 boys) scored their children as having more social problems (males) and more withdrawn/depressed problems (girls). Additional studies in larger cohorts are needed to draw more definitive conclusions. To our knowledge there are no studies on resting-state functional connectivity in patients with CAH or persons at risk of CAH prenatally treated with DEX. When we performed whole-brain analyses to investigate the functional connectivity of the brain during rest in 31 patients with CAH (18 females), aged 16-33 years (mean age 23.7 years), we found increased functional connectivity in the precuneus in patients with CAH compared to controls.

Post hoc analyses within the precuneus revealed that patients with SV CAH had stronger connectivity. The altered functional connectivity may reflect a reorganization of the brain in patients with CAH. Looking at the resting-state functional connectivity in 18 participants (8 females), aged 16-33 years (mean age 20.8 years) exposed to DEX treatment during the first trimester of fetal life because of the risk of having CAH, we used two approaches: an exploratory whole-brain analysis and a seed-based analysis. We chose three brain regions (amygdala, hippocampus, superior frontal gyrus) for the seed-based analysis based on our previous findings and literature evidence. We did not find any differences in resting-state functional connectivity between DEX-exposed individuals and controls.

In conclusion, this thesis extends the existing literature on GC effects on functional connectivity in the brain and cognitive and behavioral outcomes in patients with CAH.