Risk factors for choroidal neovascularization and geographic atrophy in the complications of age-related macular degeneration prevention trial

Objective: To determine risk factors for choroidal neovascularization (CNV) and of geographic atrophy (GA) in eyes with large drusen. Design: Cohort study within a multicenter, randomized clinical trial of laser treatment for the prevention of vision loss from advanced age-related macular degeneration. Participants: One thousand fifty-two participants with 10 or more large drusen (>= 25 mu m) and visual acuity of 20/40 or better in each eye. Methods: At baseline, participants provided a brief medical history. Trained readers evaluated baseline color photographs for drusen characteristics and pigmentary abnormalities. One eye of each participant was assigned to laser treatment and the contralateral eye was assigned to observation. The Complications of Age-Related Macular Degeneration Prevention Trial (CAPT) Reading Center readers identified CNV and endpoint GA from color photographs and fluorescein angiograms obtained during follow-up visits scheduled for 5 or 6 years. Estimates of relative risks (RRs) and 95% confidence intervals (CIs) were obtained from survival analyses of observed and treated eyes, considered separately and combined. Main Outcome Measures: Development of CNV and of endpoint GA. Results: Choroidal neovascularization developed in 141 observed eyes and 141 treated eyes, including 57 patients affected bilaterally. Statistically significant risk factors for CNV in the multivariate model for all eyes were older age (RR, 2.81 [95% CI, 1.33-5.941 for >79 years vs. 50-59 years), cigarette smoking (RR, 1.98 [95% CI, 1.16-3.39] for current vs. never), and focal hyperpigmentation (RR, 1.84 [95% CI, 1.22-2.76] for >= 250 mu m vs. none). Among eyes free of GA at baseline, endpoint GA developed in 61 observed eyes and in 58 treated eyes, including 29 patients affected bilaterally. Statistically significant risk factors for GA in the multivariate model for all eyes were older age (RR, 6.39 [95% CI, 1.64-24.9] for >79 years vs. 50-59 years), greater retinal area covered by drusen (RR, 5.10 [95% CI, 2.57-10.1] for >= 25% vs. <10%), retinal pigment epithelium (RPE) depigmentation (RR, 2.64 [95% CI, 1.26-5.53), and focal hyperpigmentation (RR, 10.4 [95% CI, 4.51-24.0] for >= 250 mu m vs. none). Conclusions: Among CAPT participants, increased age and focal hyperpigmentation were risk factors for the development of CNV and for GA. Cigarette smoking was significantly associated with CNV only, whereas retinal area covered by drusen and RPE depigmentation were associated significantly with GA only.