In utero exposure to gestational diabetes (GDM) or hypertension (HYP) is associated with child obesity. These maternal conditions alter the in utero milieu and may predispose offspring to obesity via an effect on the brain. Cellular inflammation (gliosis) in the mediobasal hypothalamus (MBH) has been associated with child and adult obesity and type 2 diabetes in adults. We evaluated the brain MRIs of offspring exposed to GDM or HYP during gestation for evidence of MBH gliosis. This case-control study included a subsample of 3child participants (9-11y) in the NIH ABCD study. Baseline T2-weighted MRI scans were analyzed for evidence of MBH gliosis (high MBH T2 signal) based on the mean bilateral MBH/Amygdala (AMY) T2 signal ratio. Putamen (PUT) /AMY and MBH/PUT were negative and positive control ratios, respectively. Findings from 49 children exposed to GDM and 53 exposed to HYP (by pregnancy survey) were contrasted to 2age- and sex-matched controls (healthy pregnancy) . Baseline, 1y and 2y follow-up anthropometrics were available. At baseline, groups did not differ on age, sex, race, and socioeconomic status. GDM group had more Hispanic children and higher birth weight than controls and HYP. Offspring exposed to GDM or HYP had higher BMI z-score, % of BMI 95th percentile, and waist/height ratio than controls at all time points (all P<0.01) . Overall, there were significant group differences in MBH/AMY signal ratio (F (2,294) :355, P=0.03, adjusted for site, age, sex and ethnicity) . Specifically, compared to controls, GDM-exposed offspring, but not HYP, had higher MBH/AMY signal ratio (GDM β:0.05, P=0.02; HYP β:0.03, P=0.11) . The results persisted when further adjusted for baseline BMI z-score (GDM β:0.04, P=0.03) . Results were replicated with the positive control ratio, but not the negative control ratio. GDM-exposed offspring have greater evidence of MBH gliosis, independent of body adiposity. This could represent one pathway by which the in utero environment may impact offspring’s future metabolic health.


L.E.Sewaybricker: None. K.Olerich: None. S.L.Kee: None. S.J.Melhorn: None. S.Chandrasekaran: None. E.Schur: None.


National Institutes of Health (R01DK117623)

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at