DMN function changes on resting state fMRI in perimenopausal women.

Objectives: To investigate functional changes of the default mode network (DMN) in resting-state functional magnetic resonance imaging (rsfMRI) of perimenopausal women, and explore the association between these changes and estrogen levels. Methods: A total of 16 women in the perimenopausal period and 15 women in the premenopausal period underwent general health status, menopausal rating scale (MRS), and depression screening scale assessment, cognitive function tests (Stroop task) and measurement of sex hormone levels including prolactin (PRL), follicle stimulating hormone (FSH), estradiol (E2), testosterone (T), progesterone (P), and luteinizing hormone (LH). The resting state fMRI data were acquired using a 3.0 T magnetic resonance scanner, and the differences in DMN functional connection between these two groups were evaluated by independent component analysis (ICA). The amplitudes of low frequency fluctuations (ALFF) was analyzed using independently defined regions of interest (ROIs). Correlations between DMN connectivity and E2 were tested with Bonferroni/FDR correction; post hoc power analysis verified sample size adequacy. Results: Post hoc power analysis showed 89 % power for the primary outcome (DMN connectivity differences). After AlphaSim (cluster ≥74, p<0.05) and threshold-free cluster enhancement (TFCE, FWE p<0.05) correction, perimenopausal women exhibited significantly enhanced DMN connectivity in the bilateral middle frontal gyri (MFG, functional extension nodes of the anterior DMN), left insula, and posterior cingulate cortex (PCC, core hub of the posterior DMN) (all p<0.05). No significant group differences were detected in the canonical medial prefrontal cortex (MPFC) core (MNI coordinates: x=0, y=50, z=20). ALFF values of these regions were also significantly higher in perimenopausal women (all p<0.05). Correlation analysis revealed no significant linear correlations between E2 levels and DMN connectivity of the four regions in perimenopausal women (all p>0.05), while a significant positive correlation between E2 and left insula connectivity was observed in premenopausal women (r=0.489, Bonferroni-corrected p=0.032). No group differences were found in cognitive function (Stroop task: p=0.495-0.519) or PHQ-9 scores (p=0.067); adjusting for these variables did not alter DMN results. Conclusions: Estrogen reduction and fluctuation are associated with functional reorganization of DMN-associated regions in perimenopausal women. Enhanced connectivity of the bilateral middle frontal gyri, left insula, and PCC may reflect a compensatory mechanism to maintain cognitive and emotional stability amid E2 decline. These findings provide preliminary neuroimaging evidence for brain function adaptations during the perimenopausal transition, with E2 dynamics (decline and fluctuation) as key drivers.