Alternative Measures of Hyperglycemia and Risk of Alzheimer's Disease in the Community: The Hisayama Study.

Context and Objective We investigated the associations of hemoglobin A1c (HbA1c), glycated albumin (GA), GA/HbA1c ratio, and 1,5-anhydroglucitol (1,5-AG) with the development of Alzheimer's disease (AD). Design and Participants A total of 1187 community-dwelling Japanese subjects aged ≥65 years without dementia were followed up for an average of 4.8 years. Results The age- and sex-adjusted incidence of AD increased significantly with higher quartiles of GA/HbA1c ratio, and a similar tendency was seen for GA, whereas no such association was observed for HbA1c and 1,5-AG. After adjusting for potential confounding factors, positive association of GA/HbA1c ratio with the risk of AD remained significant: the multivariable-adjusted hazard ratio (HR) was significantly higher in the third [HR = 2.11, 95% confidence interval (CI) = 1.16 to 3.82] and fourth (HR = 2.01, 95% CI = 1.09 to 3.68) quartile than in the first quartile. Among subjects with normal glucose tolerance, those with high GA/HbA1c ratio had a higher risk of AD than those with low GA/HbA1c ratio (HR = 1.82, 95% CI = 1.05 to 3.16), and a similar tendency was found in those with glucose intolerance (HR = 1.73, 95% CI = 0.96 to 3.13). No such associations were observed for HbA1c, GA, and 1,5-AG, regardless of glucose tolerance status. Conclusions Our findings suggest that elevated GA/HbA1c ratio-but not HbA1c, GA, or 1,5-AG level-is significantly associated with the risk of AD in subjects both with and without glucose intolerance. GA/HbA1c ratio may be a useful biomarker for predicting incident AD.