Structural alterations in the superior temporal gyrus associated with depressive symptoms in migraine without aura: a voxel-based morphometry and surface-based morphometry magnetic resonance imaging study.

Background: Migraine without aura (MwoA) frequently co-occurs with depressive symptoms, yet the neurobiological mechanisms underlying this comorbidity remain insufficiently characterized. This study integrated voxel-based morphometry (VBM) and surface-based morphometry (SBM) to compare gray matter volume (GMV) and cortical morphology between patients with MwoA and healthy controls (HC) and to further determine whether structural brain alterations vary across MwoA subgroups stratified by depressive symptom severity, thereby identifying potential structural correlates of depressive symptoms in this population. Methods: A total of 67 patients with MwoA {38 with lower depressive symptoms [MwoA-LS, 17-item Hamilton Depression Rating Scale (HAMD-17) <8] and 29 with higher depressive symptoms (MwoA-HS, HAMD-17 ≥8)} and 40 HC were included. High-resolution T1-weighted magnetic resonance imaging (MRI) data were processed using CAT12 to derive GMV, cortical thickness (CT), local gyrification index (LGI), sulcal depth (SD), and fractal dimension (FD). Between-group comparisons were conducted using two-sample t-tests within a general linear model framework, with sex, education, total intracranial volume (TIV) (for VBM), mean CT (for SBM), and Hamilton Anxiety Rating Scale (HAMA) scores included as covariates. Statistical significance was defined using a voxel-/vertex-level threshold of P<0.001 (uncorrected), followed by cluster-level false discovery rate (FDR) correction at q<0.05. Results: Compared with HC, patients with MwoA exhibited significant reductions in GMV in the left amygdala (cluster size =305; FDR-corrected q=0.046), left parahippocampal gyrus (cluster size =273; q=0.036), left lingual gyrus (two clusters, size =218, q=0.046 and size =177, q=0.046), and right angular gyrus extending into the middle occipital gyrus and precuneus (cluster size =427; q=0.036). No significant between-group differences were observed in any SBM measure. In the subgroup analysis, patients in the MwoA-HS group showed significantly lower LGI in the right superior temporal gyrus than those in the MwoA-LS group (cluster size =73; FDR-corrected q=0.019). Exploratory partial correlation analyses indicated that no imaging metric was significantly associated with HAMD-17 scores or other clinical measures after FDR correction (all q>0.05). Conclusions: Patients withMwoA exhibited widespread GMV reductions across brain regions involved in emotional and pain-related processing. In addition, patients with higher depressive symptom levels showed a distinct cortical morphological feature characterized by lower LGI in the right superior temporal gyrus, which was not detected by VBM. Given the exploratory nature of these findings, the lack of significant brain-clinical correlations after correction for multiple comparisons, and the cross-sectional study design, these results should be validated in larger longitudinal cohorts.