Abstract
Purpose::
To describe refractive error (RE) at baseline, examine RE as a risk factor for late age-related macular degeneration (AMD), and describe 5-year changes in RE among patients with bilateral large drusen.
Methods::
1052 participants with ≥10 large (>125µ) drusen and visual acuity ≥20/40 in each eye were enrolled into CAPT. One eye of each participant was randomly assigned to treatment, and the contralateral eye was assigned to observation. RE was measured following a standardized protocol at baseline, 6 months and annually. The association of baseline spherical equivalent (SE) and incident choroidal neovascularization (CNV) and endpoint geographic atrophy (GA) was evaluated through survival analysis. Endpoint GA was defines defined as a total of >1 MPS disc area of atrophy when all areas of geographic atrophy were combined. Associations of demographic and clinical factors with 5-year changes in SE were examined with multiple linear regression analysis. The generalized estimating equations (GEE) approach was used to account for the correlation between eyes of the same subject.
Results::
Of the 2084 eyes (excluding 20 eyes with baseline CNV), the mean SE was 0.72± 1.81 D; 1125 (54.0%) were hyperopic (>0.5D), 566 (27.2%) were emmetropic, and 393 (18.9%) were myopic (<-0.5D) at baseline. After adjustment for other risk factors, the relative hazard {95% CI} associated with +1 D was 1.04 ({0.97, 1.12}; p=0.26) for CNV and 0.96 ({0.87, 1.06}; p=0.39) for GA. Among the 1502 eyes with RE measured at 5 years, the mean change was 0.02±0.95D and did not differ between treated (0.04±0.96D) and observed (-0.005±0.95) eyes (p=0.27). Overall, 5 year-changes were more towards hyperopia in males (mean +0.10D) than in females (mean -0.03D; p<0.01). The relation between age and 5-year change depended on lens status (p for interaction <0.005). Among eyes phakic at baseline, older baseline age was associated with an increased myopic shift, while among pseudophakic eyes older baseline age was associated with increased hyperopic shift. For example, among patients older than 75, phakic eyes had SE decrease by 0.24D, but pseudophakic eyes increased by 0.29D.
Conclusions::
Among CAPT participants, baseline SE was not associated with incidence of CNV or endpoint GA. SE increased on average by 0.02D in 5 years. Older eyes with a natural lens experienced a myopic shift. Pseudophakic older eyes experienced a hyperopic shift.
Keywords: refractive error development • age-related macular degeneration