Sex differences in associations between white matter microstructure and gonadal hormones in children and adolescents with prenatal alcohol exposure

Abstract

Despite accumulating evidence from animal models demonstrating that prenatal alcohol exposure (PAE) results in life-long neuroendocrine dysregulation, very little is known on this topic among humans with fetal alcohol spectrum disorders (FASD). We expected that alterations in gonadal hormones might interfere with the typical development of white matter (WM) myelination, and in a sex-dependent manner, in human adolescents with FASD. In order to investigate this hypothesis, we used diffusion tensor imaging (DTI) to assess: 1) whether or not sex moderates the impact of PAE on WM microstructure; and 2) how gonadal hormones relate to alterations in WM microstructure in children and adolescents affected by PAE.

Methods: 61 youth (9 to 16 yrs.; 49% girls; 50% PAE) participated as part of the Collaborative Initiative on Fetal Alcohol Spectrum Disorders (CIFASD). DTI scans and passive drool samples were obtained to examine neurodevelopmental associations with testosterone (T) and dehydroepiandrosterone (DHEA) levels in boys and girls, and estradiol (E2) and progesterone (P) levels in girls. Tract-based spatial statistics were utilized to generate fractional anisotropy (FA) and mean diffusivity (MD) for 9 a priori WM regions of interest (ROIs).

Results: As predicted, alterations in FA were observed in adolescents with PAE relative to controls, and these differences varied by sex. Girls with PAE exhibited lower FA (Inferior fronto-occipital and Uncinate fasciculi) while boys with PAE exhibited higher FA (Callosal body, Cingulum, Corticospinal tract, Optic radiation, Superior longitudinal fasciculus) relative to age-matched controls. When gonadal hormone levels were examined in relation to DTI measures, additional group differences in FA were revealed, demonstrating that neuroendocrine factors are associated with PAE-related brain alterations.

Conclusions: These findings provide human evidence that PAE relates to sex-specific differences in WM microstructure, and underlying alterations in gonadal hormone function may, in part, contribute to these effects. Determining PAE-effects on neuroendocrine function among humans is an essential first step towards developing novel clinical (e.g., assessment or intervention) tools that target hormone systems to improve on-going brain development among children and adolescents with FASD.

Fig. 1

Hormone levels in: A) boys and B) girls. Salivary hormone levels by group and sex. Control in red; Alcohol-exposed in blue. Cohen’s d is represented to indicate expected size of a group (PAE vs Control) effect within same-sexed participants. No significant group differences were detected while controlling for age and time of saliva collection, with the exception of a statistical trend for testosterone:DHEA ratios among boys. Abbreviations: AE: prenatal alcohol exposed; DHEA: dehydroepiandrosterone; pg/mL: Picogram/milliliter.

Fig. 2

Regions of Interest exhibiting significant group differences. ROIs demonstrated along the X, Y and Z planes.

Fig. 3

Associations between FA and hormone levels in: A) boys and B) girls. Control in red; Alcohol-exposed in blue. Raw regression coefficients (Beta) ± standard error and p values are presented from post-hoc analyses for each group to represent effect size. Abbreviations: AE: prenatal alcohol exposed; CST: corticospinal tract; DHEA: dehydroepiandrosterone; FA: fractional anisotropy; IFO; inferior fronto-occipital tract; SFO: superior fronto-occipital tract; SLF; superior longitudinal fasciculus; pg/mL: Picogram/milliliter.