Higher Glycemia Linked to Lower Cognitive Scores in Type 2 Diabetes

Although glucose-lowering medication choice is not associated with cognitive performance after 4 years, higher glycemic levels are modestly associated with poorer cognitive outcomes among adults with short-duration type 2 diabetes (T2D), according to study results published in JAMA Internal Medicine.

To assess whether different classes of second-line glucose-lowering medications or glycemic control affect cognitive performance among adults with T2D of less than 10 years’ duration, researchers conducted the GRADE trial (ClinicalTrials.gov Identifier: NCT01794143), a large, multicenter, open-label, randomized controlled trial (RCT) involving 5047 participants, of whom 3721 were included in the cognition analysis comprising baseline and 4-year data.

Participants were randomized to glargine, glimepiride, liraglutide, or sitagliptin in addition to metformin. Cognitive performance was assessed at baseline, year 4, and year 6. The Digit Symbol Substitution Test (DSST) was the primary outcome, with secondary measures of memory and verbal fluency.

Most participants were male (62.3%), non-Hispanic (81.6%), and White (65.9%), with a mean diabetes duration of 4.3 years and follow-up of 4.1 years. The researchers noted no significant baseline differences in cognitive or demographic characteristics across the 4 treatment groups (glargine, glimepiride, liraglutide, sitagliptin).

At year 4, adjusted mean cognitive scores (including DSST, Spanish English Verbal Learning Test [SEVLT] immediate/delayed, and animal/letter fluency tests) were nearly identical across treatment groups. However, older age, male sex, lower education, and longer diabetes duration were associated with lower cognitive scores.

In linear regression models adjusted for baseline cognitive scores, the researchers observed no significant associations between time-weighted hemoglobin A1c (HbA_1c) and cognition. However, after adjustments for age, sex, race, ethnicity, education, and diabetes duration, higher time-weighted HbA_1c was consistently linked to lower DSST, SEVLT (immediate and delayed), and animal fluency scores. Specifically, primary metabolic outcome (HbA_1c ≥7%) was associated with lower DSST (β, -0.6; 95% CI, -1.1 to -0; P =.03) and animal fluency (β, -0.3; 95% CI, -0.6 to 0; P =.03) performance. Secondary metabolic outcome (HbA_1c >7.5%) was similarly associated with poorer DSST (β, -0.8; 95% CI, -1.3 to -0.2; P =.004) and animal fluency (β, -0.4; 95% CI, -0.6 to -0.1; P =.01) scores.

Sensitivity analyses, including year 6 data, supported these findings, highlighting that while treatment assignment did not affect cognition, higher glycemic exposure was modestly but consistently correlated with worse cognitive outcomes.

Study limitations include inability to establish causality, limited generalizability due to the exclusion of participants with poor glycemic control and enhanced adherence during the trial, and the lack of brain imaging data, which may prevent assessment of subclinical changes that might precede cognitive effects.

The study authors concluded, “This RCT found that, in persons with early T2D, the choice of second-line glucose-lowering therapy added to metformin did not affect cognitive outcomes over 4 years of follow-up.”

Disclosure: Some study authors reported affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of authors’ disclosures.

This article originally appeared on Endocrinology Advisor