Brain structural-functional connectivity coupling and glucose metabolism in adolescents with bipolar disorder and major depressive disorder.

AIM: To explore potential neuroimaging features derived from [18F]FDG PET/MRI imaging that may help characterize differences between adolescent with bipolar disorder (BD) and those with major depressive disorder (MDD).

MATERIALS AND

METHODS

Patients were grouped based on the Hamilton Depression Scale (HAMD) and Hypomania Check List (HCL), and the severity of suicidal ideation and behavior was assessed. 14 BD and 14 MDD adolescent patients were enrolled. Simultaneous PET and MRI data (including 3D T1-weighted imaging, Diffusion Tensor Imaging (DTI), and resting-state functional MRI (fMRI)) were acquired. Functional connectivity (FC) and structural connectivity (SC) were derived from fMRI and DTI respectively. A multiple linear regression model was employed to predict FC from SC. SC-FC coupling was defined as Pearson correlation coefficient between predicted FC and original FC. Using cerebellar gray matter uptake as reference region, standardized uptake value ratio (SUVr) of each brain region was calculated as metabolic parameter.

RESULTS

Compared with the MDD group, the BD group showed significantly decreased SC-FC coupling and glucose metabolism in the thalamus. ‌In the BD group, SC-FC coupling in the paracentral lobule showed a significant positive correlation with metabolism (r = 0.70). In the MDD group, SC-FC coupling in cingulate gyrus showed a significant negative correlation with metabolism (r=-0.73). ‌In the BD group, SC-FC coupling in superior parietal lobule, postcentral gyrus, and precentral gyrus correlated significantly with HAMD, HCL, and suicide scores, respectively. In the MDD group, SC-FC coupling in precentral gyrus, orbital gyrus, and hippocampus showed the highest correlations with these clinical indicators. The middle temporal gyrus and parahippocampal gyrus were specifically associated with suicide scores in the BD group. ‌Notable intergroup differences between BD and MDD were thalamus (AUC = 0.86) and inferior temporal gyrus (AUC = 0.76), with SC-FC coupling features in these regions showing more prominent differences than metabolic features.

CONCLUSION

This study identifies the thalamus as a key region in adolescent BD. Differences in SC-FC coupling in thalamus and inferior temporal gyrus reflect notable exploratory features between adolescent BD and MDD patients, suggesting that SC-FC coupling parameters may represent potential neuroimaging signatures for differential diagnosis. The differences in correlations among SC-FC coupling, metabolism, and clinical parameters between the groups indicate distinct pathogenic mechanisms for BD and MDD.