Abstract
BACKGROUND: White matter abnormalities can cause network dysfunction that underlies major depressive disorder (MDD). Diffusion tensor imaging (DTI) is used to examine the neural connectivity and integrity of the white matter. Previous studies have implicated frontolimbic neural networks in the pathophysiology of MDD. Approximately 30% of MDD patients demonstrate treatment-resistant depression (TRD). However, the neurobiology of TRD remains unclear. METHODS: We used a voxel-based analysis method to analyze DTI data in young patients with TRD (n = 30; 19 males, 11 females) compared with right-handed, age- and sex-matched healthy volunteers (n = 25; 14 males, 11 females). RESULTS: We found a significant decrease in fractional anisotropy (FA) (corrected, cluster size >50) in the left middle frontal gyrus (peak coordinates [-18 46-14]), left limbic lobe uncus (peak coordinates [-18 2-22]), and right cerebellum posterior lobe (peak coordinates [26-34 -40]). There was no increase in FA in any brain region in patients. We also found a significant negative correlation between mean regional FA values in the three areas and Beck Depression Inventory symptom scores. CONCLUSIONS: We found significant differences in white matter FA in the frontal lobe, limbic lobe and cerebellum between TRD patients and controls. These data suggest that abnormalities of cortical-limbic-cerebellar white matter networks may contribute to TRD in young patients.