Early Development of Proliferative Retinopathy Increases Risk of Cognitive Dysfunction in Middle-Aged Adults with Type 1 Diabetes (P5.032)

OBJECTIVE: Investigate the relationship between markers of retinal vascular health (RVH) and cognitive dysfunction (CD) in middle-aged adults with type 1 diabetes (T1D) followed over 25 years. BACKGROUND: CD is a common complication of T1D, with suspected microvascular etiology. Identifying biomarkers that detect early cognitive changes will help in designing and implementing prevention and intervention strategies. The retina, an extension of the CNS, may be such a biomarker. While RVH measures show promise as CD biomarkers in elderly and type 2 diabetes populations, they have not been characterized in T1D. DESIGN/METHODS: Participants were from the Pittsburgh Epidemiology of Diabetes Complications (EDC) Cohort, an on-going prospective study with 25 years of follow-up. Only those having measures of retinal arteriolar and venular diameter and free of proliferative retinopathy (PR) at baseline (1986-88) who also completed an extensive neurocognitive battery in 2010-2013 were included (N=75). Having two or more cognitive test scores 1.5SD or worse than published normative means met the definition of CD. Logistic regression models, controlling for education, glycemic control and T1D duration, were used to calculate odds ratios for CD from: 1) baseline arteriolar and venular diameter; 2) incident PR over 25 years of EDC; and 3) duration of incident PR. RESULTS: 40[percnt] of patients met definition of CD (mean age=48±7 years, mean duration=40±6 years). Smaller baseline retinal venular caliber increased the odds of CD 25 years later (adjusted OR [95[percnt]CI]:1.02 [1.00, 1.05], p=0.06). Incident PR over 25 years increased the odds of CD (3.38 [1.13, 10.08; p=0.03). Longer PR duration increased the odds of CD but the association was not independent of glycemic control (1.05 [0.98 1.12]; p=0.18). CONCLUSIONS: Non-invasive and inexpensive measures of RVH could serve as very early indicators of CD in T1D patients, aiding clinicians in early identification of T1D patients at increased risk of CD. Disclosure: Dr. Nunley has nothing to disclose. Dr. Ryan has nothing to disclose. Dr. Saxton has received personal compensation for activities with Forest Laboratories and Eisai Inc. as a consultant. Dr. Orchard has nothing to disclose. Dr. Costacou has nothing to disclose. Dr. Aizenstein has nothing to disclose. Dr. Ryan has nothing to disclose. Dr. Miller has nothing to disclose. Dr. Rosano has nothing to disclose.