Thalamic subregional signatures of cognitive dysfunction in type 2 diabetes mellitus

Introduction/Objective Type 2 diabetes mellitus (T2DM) is associated with brain structural changes and cognitive decline, yet its impact on thalamic subregions remains unclear. This study aimed to identify volume changes in thalamic subregions in individuals with T2DM and to examine their associations with cognitive and diabetes-related clinical parameters. Methods In this cross-sectional case-control study, the thalamus was segmented into 25 subregions per hemisphere using FreeSurfer, and the volume of each subregion was calculated. Subregional volumes were compared between patients with T2DM and healthy controls (HC) using a generalized linear model (GLM), and partial correlation analysis was used to examine the associations between these volumetric measures and cognitive indicators. Results We identified five subregions of the thalamus that exhibited differences between the normal control and T2DM groups after correction for multiple comparisons. The left mediodorsal medial magnocellular (MDm) subregion showed volumetric reduction in the T2DM group, whereas four subregions, the right and left limitans/suprageniculate (L-Sg) and the right and left lateral pulvinar (PuL), exhibited volumetric enlargement in the T2DM group. Within the T2DM group, left PuL volume was negatively correlated with AVLT immediate recall (r = -0.35, P = 0.022) and right L-Sg volume was positively correlated with the Digit Span Test (DST) forward score (r = 0.31, P = 0.044); neither survived correction for multiple comparisons. Discussion In this cross-sectional, single-centre sample, T2DM was associated with nucleus-specific, bidirectional thalamic volume differences in the absence of a whole-thalamus difference: smaller left MDm and larger bilateral L-Sg and PuL volumes. Because only T1-weighted images were analysed, the biological substrate of the larger volumes cannot be determined here; neuroinflammation, oedema and gliosis are plausible but untested hypotheses. The imaging-cognition associations were modest and did not survive correction for multiple comparisons. Conclusion Individuals with T2DM showed cross-sectional, region-specific differences in thalamic subnuclear volumes that were not detectable at the level of the whole thalamus. These are observational associations from a modest single-centre sample; their diagnostic, screening or predictive value is not established and requires longitudinal validation with clinically defined cognitive outcomes and independent replication.