Data increasingly support evidence of gliosis in the mediobasal hypothalamus (MBH) in the setting of obesity and insulin resistance. It is if other regions of the brain are similarly affected. We tested if MBH gliosis is related to radiologic evidence of gliosis in selected subcortical, cortical, and hypothalamic regions.

We recruited 40 obese adults (aged 25-61 y), with variable glucose tolerance by OGTT (n=24 normal glucose tolerance, n=11 impaired glucose tolerance and n=5 with T2DM). Individuals underwent a multi-echo T2 brain MRI. Two analysts placed regions of interest (ROIs) on the scans and determined the T2 relaxation time as a measure of gliosis. ROIs included the MBH, amygdala, dorsomedial hypothalamus (DMH; area of the paraventricular nucleus), hippocampus, white matter (WM), cortex, caudate, ventral striatum (VS), and putamen. Left and right sides were averaged. High (n=13) and low (n=14) T2 groups were derived from the tertiles with the highest (strongest evidence for gliosis) and lowest T2 relaxation times. The groups were compared by linear regression adjusted for age and sex.

Mean BMI in the low T2 group was 35.7 ± 4.1 kg/m2 and high T2 group was 37.1 ± 4.9 kg/m2. The groups did not differ in T2 relaxation time of WM (P=0.29), cortex (P=0.15), caudate (P=0.49), putamen (P=0.79), amygdala (P=0.09), or hippocampus (P=0.44). In the high T2 group, T2 relaxation time was shorter in the VS (adjusted mean 80.9±2.9 vs. 83.1±3.0, P=0.04) and longer in the DMH (adjusted mean 108.9±5.3 vs. 103.7±5.4, P=0.02), compared to the low T2 group. When subjects with T2DM (n=1 in low and n=3 in high T2 group) were excluded, the differences were attenuated (VSP=0.13; DMHP=0.10).

The findings reveal preliminary evidence of longer T2 relaxation times in the DMH of obese individuals with signs of MBH gliosis and no evidence for gliosis in cortical or subcortical regions. Further study of gliosis in the DMH and its relation to blood glucose control and autonomic function in humans, particularly in those with T2DM, is warranted.


J.L. Rosenbaum: None. S.J. Melhorn: None. M. De Leon: None. M. Webb: None. E. Schur: None.


American Diabetes Association (1-17-ICTS-085 to E.S.); National Institutes of Health (T32HL007028); University of Washington Institute of Translational Health Sciences (UL1TR002319)

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