Psychosocial distress during pregnancy is linked to adverse perinatal outcomes, but the biological correlates remain incompletely characterized. The authors investigated whether circulating levels of the neurosteroid allopregnanolone (ALLO) and related steroid hormones associate with prenatal psychosocial distress in a well-characterized pregnancy cohort, and whether these associations vary by race/ethnicity.
The analysis used data from 237 participants in the Healthy Start Study. Participants were selected into two groups according to self-reported symptoms on depression and anxiety screening tools: a high-distress group (Edinburgh Perinatal Depression Scale [EPDS] ≥13 or EPDS-3A ≥7; n = 57) and a low-distress group (EPDS <4 and EPDS-3A <2; n = 180). The study is reported as a preprint and has not been peer reviewed.
Maternal serum concentrations of ALLO, progesterone, pregnanolone, cortisol, and cortisone were measured using high-performance liquid chromatography–tandem mass spectrometry (HPLC-MS/MS). Each participant contributed samples from two timepoints in pregnancy with median gestational ages of approximately 17 and 27 weeks (range 10–34 weeks).
The investigators modeled the relationship between high versus low prenatal psychosocial distress and repeated measures of steroid concentrations using linear mixed-effects models. Primary adjustment included maternal age, fetal sex, smoking status, and gestational age at blood draw. Additional models included sociodemographic covariates to assess the influence of social and structural factors. The potential for effect modification by race/ethnicity was assessed.
After adjustment for maternal age, fetal sex, smoking, and gestational age at blood draw, mean circulating ALLO concentration was 20.6% lower in individuals with high prenatal distress compared with those with low distress (95% CI: -30.9%, -8.7%). There was no evidence that race/ethnicity modified this association.
The association between high distress and lower ALLO attenuated when sociodemographic characteristics were added to the models, indicating that social and structural variables influence circulating ALLO concentrations during pregnancy.
Several ALLO-related ratios were lower in the high-distress group. However, when results were examined across adjusted models, only the ALLO-to-progesterone ratio remained statistically significant. The persistence of the ALLO-to-progesterone ratio suggests that relative measures of neurosteroid conversion may more specifically capture metabolic pathways linked to prenatal psychosocial distress than absolute ALLO concentrations.
The authors report no evidence of effect modification by race/ethnicity for the primary ALLO association. The attenuation of the ALLO association after adding sociodemographic covariates highlights the influence of social and structural determinants on circulating steroid levels and on observed associations between distress and neurosteroids.
These results indicate that measures of neurosteroid metabolism, particularly the ALLO-to-progesterone ratio, may be more robust markers of associations with prenatal psychosocial distress than absolute circulating ALLO concentrations, which are sensitive to sociodemographic confounding. The findings point to an interplay between psychosocial stressors, steroid metabolism, and social determinants during pregnancy.
Given the preprint status, the authors caution that the findings have not been peer reviewed and should not be used to guide clinical practice without further validation.
This report is a preprint posted to medRxiv and has not undergone peer review. The article provides limited detail on some analytic specifics beyond the summary presented; where details are not reported in the source, they are not available for this summary. The observational design and attenuation of associations after sociodemographic adjustment indicate residual confounding and the need for cautious interpretation.
De-identified participant data and analytic code supporting the study findings are available from the corresponding author on reasonable request and with approval from the Healthy Start Study and the Colorado Multiple Institutional Review Board. Ethical approval for the Healthy Start Study and this secondary analysis was provided by the Colorado Multiple Institutional Review Board at the University of Colorado, Denver/Anschutz (COMIRB #09-0563 and COMIRB #24-0224). The authors declared no competing interests. The research was funded by the National Institute of Diabetes and Digestive and Kidney Diseases.