Cardiovascular risk factors and the long-term incidence of age-related macular degeneration - The Blue Mountains Eye Study

Purpose: To assess the relationship between cardiovascular disease and cardiovascular risk factors, other than smoking, and risk of long-term incident age-related macular degeneration (AMD). Design: Population-based cohort study. Participants: There were 3654 baseline (1992-1994) participants aged >= 49 years included in the Blue Mountains region, west of Sydney, Australia. Of these, 2335 (75% of survivors) were reexamined after 5 years (1997-1999) and 1952 (76% of survivors) after 10 years (2002-2004). Methods: Stereoscopic color fundus photographs were graded using the Wisconsin Age-related Maculopathy Grading System. History, physical examination, and fasting blood samples provided data on possible risk factors. Age-related macular degeneration incidence was calculated using the Kaplan-Meier survival approach. Discrete linear logistic models were used to assess risk of incident AMD. Relative risks (RR) and 95% confidence intervals (CI) are presented after adjusting for age, gender, smoking, and other risk factors. Main Outcome Measure: Incident early and late AMD. Results: Increasing high-density lipoprotein (HDL) cholesterol was inversely related to incident late AMD (RR per standard deviation [SD] increase, 0.74; 95% CI, 0.56-0.99). Elevated total/HDL cholesterol ratio predicted late AMD (RR per SD increase, 1.35; 95% CI, 1.07-1.70) and geographic atrophy (GA; RR per SD, 1.63; 95% CI, 1.18-2.25). Diabetes predicted incident GA (RR, 3.89; 95% CI, 1.36-11.08), but not neovascularAMD. History of stroke (RR 2.01; 95% CI, 1.12-3.58), or any cardiovascular disease (stroke, myocardial infarction, or angina; RR, 1.57; 95% CI, 1.13-2.16) predicted incident early AMD and incident indistinct soft or reticular drusen (RR, 2.38; 95% CI, 1.33-4.27 for stroke; RR, 1.80; 95% CI, 1.28-2.52 for any cardiovascular disease). Neither pulse pressure, systolic or diastolic blood pressure, or presence of hypertension at baseline were associated with incident AMD. Conclusions: Our findings provide some evidence of links between cardiovascular risk factors and AMD. Further prospective evaluation of these relationships is warranted, as these findings could have therapeutic implications.