Differential association of cardiovascular risk factors with neovascular and geographic AMD in the EUREYE study

Abstract

Abstract: : Purpose: To investigate the association of cardiovascular risk factors by type of AMD Methods:4760 randomly sampled people aged 65 and over participated in an eye examination and risk factor assessment in seven study centres across Europe. Two non–simultaneous 35⁰stereoscopic colour fundus images centred were taken for each eye and graded at a single reading centre by the International ARM Epidemiological Study Group classification using the grade in the worst eye. Data on smoking, cardiovascular (CVD) risk factors and hormone replacement therapy (HRT) were collected by interview. Multinomial modelling was used to assess the associations (relative risk ratios, RRR) of risk factors with type of AMD : neovascular AMD (NVD) and geographic atrophy (GA). Results:There were 109 cases of NVD, 49 of GA and 2261 Stage 0. In confounder adjusted analyses, current smokers were associated with NVD (RRR= 2.62, 95% CI 1.38 to 4.97) and GA (RRR= 4.80, 95% CI 2.04 to11.30); for ex–smokers the RRRs were 1.8 for both types. Increased RRRs for NVD were found for diabetes (RRR= 1.89, 95% CI 1.06 to 3.39), CVD history (RRR= 2.04, 95% CI 0.96 to 4.27) and weekly use of aspirin (RRR= 1.92, 95% CI 1.41 to 2.63). RRRs for GA for these factors were around 1 although the confidence limits were wide. HRT was associated with reduced RRRs for both types of AMD but only significantly for NVD (RRR= 0.24, 95% CI 0.06 to 0.99). Conclusions:Our results suggest that the adverse effects of smoking are observed for both types of AMD. Other CVD risk factors are associated only with NVD although the small sample size for GA limits conclusions. The association with weekly aspirin use may reflect residual confounding due to unreported CVD but an adverse effect of aspirin cannot be excluded.

Keywords: age–related macular degeneration • clinical (human) or epidemiologic studies: risk factor assessment