Selective amygdala subregional atrophy in Parkinson’s disease: from subjective cognitive decline to mild cognitive impairment

Background Cognitive impairment is a common and disabling non-motor feature of Parkinson’s disease (PD). Subjective cognitive decline (SCD) has been proposed as an early clinical stage of cognitive dysfunction in PD; however, its neuroanatomical correlates remain poorly defined. The amygdala is an early target of PD-related pathology, yet subregional alterations across different cognitive stages of PD have not been systematically examined. Objective To characterize amygdala subregional atrophy patterns across the cognitive continuum of PD, specifically distinguishing between subjective cognitive decline and mild cognitive impairment (MCI). Methods High-resolution T1-weighted MRI was acquired in 68 PD patients classified as cognitively normal (PD-NC, n = 21), subjective cognitive decline (PD-SCD, n = 20), or mild cognitive impairment (PD-MCI, n = 27), and in 27 healthy controls. Amygdala subregional volumes were quantified using automated segmentation. Group differences were assessed using analysis of variance with correction for multiple comparisons. Associations between amygdala subregional volumes and cognitive performance were examined using correlation and multinomial logistic regression analyses. Results Compared to controls and cognitively normal PD patients, the PD-MCI group exhibited significant atrophy specifically in the medial and cortical nuclei of the amygdala. Volumes of the medial and cortical nuclei were positively correlated with global cognitive performance and visuospatial/executive function. Multinomial logistic regression identified larger medial amygdala volume as an independent correlate of cognitive preservation in PD. Conclusion Cognitive impairment in PD is associated with selective atrophy of specific amygdala subregions. In contrast, PD-SCD involves no detectable amygdala atrophy, suggesting it represents a structurally preserved stage preceding overt neurodegeneration. These findings suggest that PD-SCD represents a structurally preserved stage and highlight amygdala subregional volumetry as a potential biomarker for cognitive impairment in PD.