Neuroimaging study on brain structural network topological properties affected by obesity in children and adolescents.

OBJECTIVE

To examine alterations in the topological properties of brain structural networks in obese adolescents through graph theory analysis.

METHOD: A total of 86 obese adolescents admitted to the Department of Child Health Care at the Children's Hospital of Nanjing Medical University from January 2023 to April 2024 were prospectively enrolled. This cohort included 60 patients in the simple obesity group (Simple OB), 26 patients in the obese group with metabolic syndrome (MS), and 24 healthy volunteers in the control group (CON). All participants underwent DTI scans of the head, anthropometric assessments, and blood tests. Using graph theory, a whole-brain structural network was constructed, and global and nodal topological properties were calculated. Modular analysis was also conducted, and the differences among the three groups were compared using one-way ANOVA. Post-hoc corrections were applied after ANOVA.

RESULTS

(1) There were no significant differences in gender, age or fasting blood glucose (FBG) among the three groups. However, there were significant differences in body height(p < 0.001), body weight (p < 0.001), BMI (p < 0.001), Waist circumference (p = 0.006), DBP (p = 0.005), systolic blood pressure (SBP) (p = 0.004), HDL-C (p = 0.002) among the three groups. Compared with the CON group, there were significant differences in TBF (p < 0.001), TG (p < 0.001), TC (p = 0.029), ALT (p < 0.001), AST (p = 0.020), FINS (p = 0.019) between the OB group and the MS group. (2) The small-world attribute (σ), characteristic path length (Lp) and global efficiency (Eg) did not differ significantly among the three groups (p > 0.05). However, the clustering coefficient (Cp) and local efficiency (Eloc) varied among the three groups, with both Cp and Eloc values being lower in the OB group than in the CON group(p = 0.039 and 0.041 respectively). Node level analysis revealed differences in node properties in the left frontal lobe, right insula, right anterior central gyrus, left anterior cingulate gyrus, left anterior cuneus and bilateral occipital lobes. The modular analysis revealed that the connection strength between the Association and Limbic modules in the OB group was significantly higher than that in the CON group (p = 0.033). (3) The Eloc was negatively correlated with TBF(β = - 0.227, p = 0.036), The modularity was negatively correlated with FINS (β = - 0.293, p = 0.006).

CONCLUSION

Adolescent obesity results in aberrant topological properties of brain's structural network, leading to abnormalities in brain regions associated with self-awareness, cognitive control, emotional regulation and sensorimotor functions in adolescents with obesity. At the level of the brain's structural network, these findings may provide insights into the underlying neural mechanisms of obesity and offer imaging evidence to support clinical interventions.