MRI-Based Structural Development of the Human Newborn Hypothalamus
Background: Preclinical evidence suggests that intrauterine exposures can impact hypothalamic structure at birth and future disease risk, yet early human data are limited. Methods: Hypothalamic volumes were measured from 699 T1-weighted MRI scans from 631 newborns (54% female; 27-45 weeks postmenstrual age/PMA) in the Developing Human Connectome Project. Linear mixed-effects models tested associations with prenatal exposures: gestational age (GA) at birth, PMA at scan, sex, maternal body mass index (BMI), and smoking. Findings were partially replicated in the Adolescent Brain and Child Development (ABCD) Study (release 5.1) data (16,934 observations from 11,207 participants). Results: Absolute hypothalamus volume increased with PMA (+5.5%/week, t=39.9, p<10^-10), but not after adjusting for brain volume (t=1.2, p=0.24). Males showed larger absolute (+3.3%, t=3.2, p=0.002) but smaller relative hypothalamus volume (t=-2.8, p=0.005). Lower GA was linked to larger relative hypothalamus volume (t=-6.5, p<10^- 9), with evidence for sex moderation (t=-2.4, p=0.019). Smoking during pregnancy was associated with smaller hypothalamus volume in newborns (t=-2.05, p=0.04; dose dependence: t=-2.9, p<0.01). Smoking remained associated with reduced hypothalamus volume in adolescents (t=-2.8, p=0.005). Conclusions: The findings suggest that the hypothalamus is a crucial and underexplored target of perinatal influences for understanding the origins of long-term health and disease.