Background and Purpose-It is unclear whether diets with high glycemic index (GI) and low cereal fiber (CF) are associated with greater risk of stroke. We aimed to assess the relationship between dietary GI and CF content, retinal microvasculature changes, and stroke-related mortality. Methods-The study consisted of a population-based cohort, 49ϩ years, examined at baseline (1992 to 1994). At baseline, participants completed validated food frequency questionnaires. Mean GI was calculated using an Australian database.Retinal arteriolar and venular diameters were measured from photographs. Mortality data were derived using the Australian National Death Index. Results-Over 13 years, 95 of 2897 participants (3.5%) died from stroke. Increasing GI (hazard ratio, 1.91; 95% CI, 1.01 to 3.47, highest versus lowest tertile) and decreasing CF (hazard ratio, 2.13; 95% CI, 1.19 to 3.80, lowest versus highest tertile) predicted greater risk of stroke death adjusting for multiple stroke risk factors. Persons consuming food in the highest GI tertile and lowest CF tertile had a 5-fold increased risk of stroke death (hazard ratio, 5.06; 95% CI, 1.67 to 15.22). Increasing GI and decreasing CF were also associated with retinal venular caliber widening (P trend Ͻ0.01). Adjustment for retinal venular caliber attenuated stroke death risk associated with high GI by 50% but did not affect the risk associated with low CF consumption. Conclusions-High-GI and low-CF diets predict greater stroke mortality and wider retinal venular caliber. The association between a high-GI diet and stroke death was partly explained by GI effects on retinal venular caliber, suggesting that a high-GI diet may produce deleterious anatomic changes in the microvasculature.

Background: Dietary factors are known risk factors for age-related macular degeneration (AMD)-the leading cause of visual loss among persons aged ͧ65 y. High-glycemic-index diets have been hypothesized as a risk factor for AMD, but prospective data are unavailable. Objective: The objective was to examine the association between dietary glycemic index and the 10-y incidence of AMD in the Blue Mountain Eye Study population. Design: This was a population-based cohort study with 3654 participants (ͧ49 y) examined at baseline (1992)(1993)(1994); 2335 patients were reexamined after 5 y and 1952 after 10 y. The Wisconsin System was used to grade 10-y incident early and late AMD from retinal photographs. A food-frequency questionnaire was used to collect dietary information at baseline, and an Australian database was used to calculate the mean glycemic index. Results: Over 10 y, 208 of 1810 persons (cumulative incidence: 14.1%) developed early AMD. After age, smoking, other risk factors, and dietary constituents were adjusted for, a higher mean dietary glycemic index was associated with an increased 10-y risk of early AMD in a comparison of quartiles 1 and 4 [relative risk (RR): 1.77; 95% CI: 1.13, 2.78; P for trend ҃ 0.03]. Conversely, a greater consumption of cereal fiber (RR: 0.68; 95% CI: 0.44, 1.04; P for trend ҃ 0.05) and breads and cereals (predominantly lower glycemic index foods such as oatmeal) (RR: 0.67; 95% CI: 0.44, 1.02; P for trend ҃ 0.03) was associated with a reduced risk of incident early AMD. No relation was observed with late AMD. Conclusions: A high-glycemic-index diet is a risk factor for early AMD-the recognized precursor of sight-threatening late AMD. Low-glycemic-index foods such as oatmeal may protect against early AMD.