Purpose: We tested the hypothesis that dietary intake of lutein is inversely associated with prevalence of diabetic retinopathy (DR) due to its antioxidant and anti-inflammatory properties and location within the retina.
Methods: We used logistic regression to examine the association between prevalent DR and energy-adjusted lutein intake by quartile (Q) using data collected from 1430 Atherosclerosis Risk in Communities Study (ARIC) participants with diabetes (n = 994 white, n = 508 black). DR was assessed from 45° non-mydriatic retinal photographs of one randomly chosen eye taken at visit 3 (1993-1995). Dietary lutein intake was estimated using a 66-item food frequency questionnaire at visit 1 (1987-1989).
Results: Median estimated daily lutein intake was 1370 µg/1000 kcals and prevalence of DR was ~21%. We found a crude association between lutein and DR (odds ratio, OR, 2.11, 95% confidence interval, CI, 1.45-3.09 for Q4, high intake, vs. Q1, low intake; p for trend <0.0001), which was attenuated after adjustment for ethnicity, duration of diabetes, glycosylated hemoglobin levels, field center and energy intake (OR 1.41, 95% CI 0.87-2.28; p for trend = 0.01). In analyses limited to persons with short diabetes duration (<6 years), the association no longer persisted (OR 0.94, 95% CI 0.31-2.16; p for trend =0.72) compared to the association in those with longer diabetes duration (≥6 years; OR 1.58, 95% CI 0.91-2.75; p for trend = 0.01).
Conclusion: Contrary to our hypothesis, we found that the odds of higher lutein intake were greater among those with DR than those without DR. However, after adjusting for confounders, intake of lutein was not associated with DR.
Keywords: Cohort studies; diabetic retinopathy; diet; epidemiology; lutein.