SUBCORTICAL GRAY MATTER AND HIPPOCAMPAL SUBFIELD ALTERATIONS IN TYPE 2 DIABETES MELLITUS: A FREESURFER-BASED NEUROIMAGING STUDY WITH CORRELATION TO MOCA MEMORY INDEX SCORE

Abstract

Background: Cognitive decline in patients with type 2 diabetes mellitus is a growing clinical concern; however, it remains poorly recognized in routine diabetes care. The hippocampus and surrounding subcortical gray matter structures are particularly vulnerable to the metabolic and vascular damage that characterizes longstanding T2DM; however, most existing studies have relied on conventional MRI reporting rather than precise automated volumetric analysis. The Montreal Cognitive Assessment (MoCA) is widely used for cognitive screening in this population, but its standard Delayed Recall component has a limited scoring range that may miss early hippocampal-driven memory failure. Whether the Memory Index Score, a more comprehensive memory measure, better reflects hippocampal volume loss in patients with diabetes remains an open question. Methods: A cross-sectional study was conducted at Saveetha Medical College and Hospital, Chennai, enrolling 100 participants with Type 2 diabetes mellitus. Brain MRI was performed for all participants, with subcortical gray matter and hippocampal volumes derived using FreeSurfer’s automated segmentation. Cognitive function was assessed using the Montreal Cognitive Assessment, and both the Delayed Recall and memory index scores were recorded. Clinical variables, including HbA1c, diabetes duration, treatment type, lipid profile, and comorbidities, were documented. Group comparisons were performed using independent t-tests and chi-square tests, correlation analysis was performed using Pearson coefficients, and multivariate logistic regression was used to identify independent predictors of cognitive impairment. Results: Cognitive impairment was observed in 62 of the 100 participants. It was more common in patients with diabetic complications (67.3%) than in those without (56.3%), although the difference was not significant (p=0.255). MoCA scores fell below the normal cutoff in both groups (23.46 and 22.67, respectively) and did not differ significantly between them (p=0.427). Hippocampal volumes were comparable among the groups (p=0.233). Age was the only variable that significantly predicted complication prevalence (p=0.027), with patients aged 50–69 years carrying the greatest burden. No significant correlations were found between MoCA scores, hippocampal volume, and the complication status. Conclusion: Cognitive impairment in T2DM does not depend on the presence of diabetic complications; it runs across the board and starts earlier than most clinicians expect. Age is more important than glycemic control or complication status in determining who is at the highest risk. These findings make a straightforward case for routine cognitive screening in all patients with diabetes, not just those who have already developed complications, and for using more sensitive tools, such as the Memory Index Score and FreeSurfer-based volumetric analysis, to detect brain changes before they become irreversible.