Abstract
Purpose :
To evaluate the relationship between eye diseases-diabetic retinopathy (DR), age-related macular degeneration (AMD), and glaucoma-and the development of Alzheimer’s disease and related dementia (ADRD) and the extent to which these might be explained by poor glycemic control and retinal microvascular disease burden.
Methods :
The Adult Changes in Thought (ACT) is a 25-year-old ongoing longitudinal study of dementia-free adults aged >65 followed for the development of ADRD. Among participants with diabetes, we used information from the electronic health record and ACT study visits to ascertain time-varying exposures of DR, AMD, and/or glaucoma at study entry or during follow-up and how established these conditions were (onset <5 vs ≥5 yrs ago). Using estimated glomerular filtration (eGFR) as a marker for retinal microvascular disease burden, we also computed time-varying estimates of mean level, trajectory, and variability of eGFR, as well as mean blood glucose (BG), using fasting/random glucose and HbA1C. We fit Cox models to estimate the relationship between each eye disease and risk of ADRD after adjusting for eGFR, BG, sex, education, >1APOE ε4, race, and smoking. The potential modifying effect of BG and eGFR were assessed by changes in the magnitude of each association from adjusted models.
Results :
Over 3342.4 person-years of follow-up among 526 ACT participants with diabetes, 131(25%) developed dementia including 98 (19%) who developed AD. A total of 85,439 eGFR and 72,960 GB measures were available for analysis. DR was associated with the development of both AD and dementia (the association was similar in magnitude to APOE ε4). (Figure) The hazard ratios (HR) for established DR vs. no DR and established AMD vs. no AMD changed very little after adjustment for eGFR and BG (from 1.65 [p=0.04] to 1.67 [p=0.03] for DR), (from 1.66 [p=0.07] to 1.74 [p=0.045]). There was no association between glaucoma and either outcome.
Conclusions :
An association between diabetic retinopathy and the development of Alzheimer’s and dementia in people with diabetes did not appear to be explained by markers of glycemic control and microvascular disease.
This is a 2021 ARVO Annual Meeting abstract.