Abstract
Abstract: :
Purpose: To examine the association of dietary lipids with incident neovascular (NV) AMD and central geographic atrophy (CGA) in AREDS Methods:Baseline nutrient intake was estimated from a validated food frequency questionnaire in this multicenter clinic–based prospective cohort study. We ascertained incident NV AMD and CGA from baseline and annual stereoscopic color fundus photographs (median follow–up=6.3 y). Among individuals with early or intermediate AMD at baseline, 788 were at risk of developing advanced AMD in one eye; 2,506 were at risk in both eyes ("bilateral drusen"). We applied repeated measures logistic regression using generalized estimating equations in a 2–stage modeling process for each of 4 outcome–participant combinations (2 outcomes: incident NV AMD and CGA, and 2 baseline AMD category groups: bilateral drusen (BD) and unilateral advanced AMD (UAMD)). Stage–1 models were adjusted for age, sex, and AREDS treatment. Lipid variables associated with progression to an outcome (p < 0.15) were modeled in Stage 2 with a set of statistically–specified non–nutritional demographic, behavioral, medical, and non–retinal ocular risk factors for advanced AMD (including those from Stage–1). For NV AMD, additional variables included: race, smoking (pack years), and antacid use; for CGA: pack years, antacid use, education, body mass index, and iris color. Results: When relationships existed in Stage–2 models, they were between people reporting highest vs. lowest intakes. The likelihood of progression from baseline BD to CGA was reduced among people reporting highest consumption of eicosapentaenoic acid (EPA) (OR = 0.5; 95% CI = 0.3–0.9) and total omega(n)–3 long–chain polyunsaturated fatty acids (LCPUFA) (OR = 0.5; 95% CI = 0.3–1.0). People with BD reporting highest consumption of monounsaturated fatty acids (MUFA) had an increased risk of incident CGA (OR = 2.4; 95% CI = 1.2–4.8) and NV AMD (OR = 1.8; 95% CI = 1.1–2.9). Among people with UAMD at enrollment, those reporting highest cholesterol intake were more likely to progress to CGA (OR = 2.7; 95% CI = 1.1–6.4) in the fellow eye. Conclusions: Lipid intake is one modifiable factor that may influence likelihood of developing sight–threatening forms of AMD. While our findings suggest that certain dietary lipids are associated with NV AMD and CGA, we cannot rule out the possibility that results may be explained by predictors or correlates of lipid intake not measured in the study.
Keywords: age-related macular degeneration • clinical (human) or epidemiologic studies: risk factor assessment • nutritional factors