Type 2 diabetes mellitus, brain atrophy and cognitive decline in older people: a longitudinal study.

AIMS/HYPOTHESIS: The aims of the study were to examine whether type 2 diabetes mellitus is associated with greater brain atrophy and cognitive decline, and whether brain atrophy mediates associations between type 2 diabetes and cognitive decline.

METHODS: Participants without dementia aged 55-90 years from the Cognition and Diabetes in Older Tasmanians (CDOT) study underwent brain MRI (ventricular and total brain volume) and neuropsychological measures (global function and seven cognitive domains) at three time points over 4.6 years. Mixed models were used to examine longitudinal associations of type 2 diabetes with cognitive and MRI measures, adjusting for covariates. A test of mediation was used to determine whether brain atrophy explained associations between type 2 diabetes and cognitive decline.

RESULTS: A total of 705 participants (diabetes: n = 348, mean age 68.2 years [SD 7.0]; no diabetes: n = 357, mean age 72.5 years [SD 7.1]) were available at baseline. Adjusting for age, sex, education and vascular risk factors, there were significant diabetes × time interactions for verbal memory (β -0.06; 95% CI -0.09, -0.02) and verbal fluency (β -0.03; 95% CI -0.06, -0.00). Although people with diabetes had lower brain (β -14.273; 95% CI -21.197, -6.580) and greater ventricular (β 2.672; 95% CI 0.152, 5.193) volumes at baseline, there were no significant diabetes × time interactions (p > 0.05) or evidence of mediation of the diabetes-cognition relationship by brain atrophy.

CONCLUSIONS/INTERPRETATION: In older community-dwelling people, type 2 diabetes is associated with decline in verbal memory and fluency over ~5 years. The effect of diabetes on brain atrophy may begin earlier (midlife).