Wearable Continuous Glucose Monitoring and Cognitive Performance in Older Adults With Type 2 Diabetes

TL;DR

This study explores how glucose levels in older adults with type 2 diabetes relate to cognitive performance using wearable glucose monitors.

Contribution

It introduces a wearable CGM protocol in older adults with T2D and links glucose metrics to specific cognitive functions.

Findings

Less time in low glucose correlated with better visuospatial performance.

Higher glucose variability showed trends with worse language performance.

Time in high glucose was linked to slower executive function.

Abstract

Limited work has evaluated continuous glucose monitoring (CGM) in relation to cognitive performance, and this has been conducted largely in type 1 diabetes. We explore associations between CGM and cognition among older adults with type 2 diabetes (T2D). Data are from community-residing adults over age 60 enrolled in the Einstein Aging Study, who self-reported T2D. Participants wore blinded Abbott FreeStyle Libre 3 sensors for 14 days and completed the WAIS III Digit Span, Trails A and B, Multilingual Naming(MINT), and WAIS III Block Design. CGM metrics include mean glucose, variability(coefficient of variation), %time in range(70-180 mg/dL), %time low(<70 mg/dL), %time high(181-250 mg/dL), and %time in very high(>250 mg/dL) glucose. Linear regression models were adjusted for age, sex, education and race/ethnicity. Interim analysis of the first 30 participants(73.9±7.4yrs, 56% female) indicate that they spent 73.8% of their time in range, 16.4% in high, 8.5% in very high, and 1.3% in low glucose. Less time in low glucose was associated with better Block Design scores(β = 0.35,p=0.041). More variability showed a trend with worse MINT scores(β = 0.36,p=0.058) and more time in high showed a trend with slower Trail B performance(β = 0.37,p=0.097). Metrics were not associated with Digit Span or Trail A. A wearable CGM protocol is feasible in older adults and identified clinically relevant metrics that were associated with cognitive performance in T2D. Time in hypo- and hyperglycemia, and variability, were most predictive, with each linked to different cognitive functions including visuospatial processing, language, and executive function. Analyses will be extended to the larger sample.