Distinct brain morphometric similarity network phenotypes are associated with mild cognitive impairment in type 2 diabetes mellitus
Year: 2026
Type: article
Study design: Observational Study
Abstract: Introduction Patients with type 2 diabetes mellitus (T2DM) exhibit heterogeneous brain alterations and cognitive dysfunction, complicating the characterization of their associations with mild cognitive impairment (MCI). Objective To identify imaging-defined subtypes of T2DM-related brain alterations and investigate their associations with MCI. Methods A total of 490 patients with T2DM and 288 healthy controls were included. Individual-level morphometric similarity networks (MSNs) were constructed from cortical morphological features, including gray matter volume, surface area, cortical thickness, and mean curvature. Heterogeneity through discriminative analysis was applied to identify subtypes of T2DM-related brain alteration based on MSNs. Group differences were assessed using general linear models. Multivariable logistic regression evaluated the associations between brain alteration subtypes and MCI. Linear regression and mediation analyses assessed associations among MSN alterations, metabolic markers, and cognitive function. Results Two distinct subtypes of T2DM-related brain alterations were identified, revealing divergent patterns of MSN reorganization and cognitive impairment. T2DM1 was characterized by increased MSN strength and relatively preserved cognition, whereas T2DM2 exhibited widespread MSN reductions and higher odds of MCI. The visual, somatomotor, and ventral attention networks contributed to subtype differentiation and showed distinct associations with glycated hemoglobin, glucose response, and insulin-related measures. Reduced limbic network MSN strength emerged specifically in T2DM2 and was associated with poorer cognition and higher odds of MCI, while mediating the association between hyperglycemia and cognitive function. Conclusions T2DM-related brain alterations exhibit two distinct MSN phenotypes with different odds of MCI. In the phenotype marked by widespread MSN reductions and lower MMSE scores, reduced limbic network MSN strength may serve as an imaging marker of the association between metabolic dysfunction and cognitive impairment.
Source: Frontiers in Endocrinology
Authors: Xiaomei Yue, Zhao Dai, Zifeng Qiu, Shijun Qiu, Gang Li
Language: English
Subfield: Psychiatry and Mental health
Field: Medicine
Domain: Health Sciences
Keywords: type 2 diabetes, mild cognitive impairment, morphometric similarity networks, cortical morphology, brain network reorganization, limbic network, visual network, sensorimotor network, ventral attention network, HbA1c, hyperglycemia, insulin, gray matter volume, cortical thickness, cognitive impairment
Open Access status: gold
Estimated APC paid: $3,801
OpenAlex ID: W7225895653
