Abstract
Abstract: :
Purpose: To examine which early age-related maculopathy lesions predict the development of visual impairment and blindness after 5-years in an Australian population-based cohort. Methods: The Blue Mountains Eye Study (BMES I) examined 3654 participants aged 50+ years during 1992-4, and re-examined 2335 (75.1% of survivors) during 1997-9 (BMES II). Visual acuity of each eye was measured on a Logmar chart, after refraction, using the ETDRS method. Stereo macular photographs were taken and graded using a modification of the Wisconsin Age-Related Maculopathy Grading System. We defined incident visual impairment in eyes with best-corrected visual acuity <20/40 (0-38 letters read correctly) at follow-up, where visual acuity was at least 20/40 at baseline. Incident blindness was defined in eyes with best-corrected visual acuity less than or worse than 20/200 (0-5 letters read) at follow-up, where visual acuity was better than 20/200 at baseline. Graders identified early age-related maculopathy lesions (drusen, pigmentary changes) from the baseline photos. Eye-specific data were analyzed accounting for the correlation between the two eyes, and after adjusting for age, gender and smoking. Results: Over the 5-year period, 5.0% of eyes developed visual impairment and 1.0% became blind. Incident visual loss was more frequent in women than men, in smokers than non-smokers and was strongly age-related. Increasing severity of drusen type, and increasing size and area of involvement by drusen strongly predicted 5-year incident visual impairment and blindness. The presence of any indistinct drusen more than doubled the likelihood of incident visual impairment over the period (OR 2.7) and increased the likelihood of incident blindness six-fold. The odds for incident visual impairment increased stepwise with increasing drusen diameter. However, the strongest predictor was the area involved by drusen ≷125µm diameter. The odds for incident visual impairment increased from 1.9 to 11.1 as the area involved by drusen increased from less than half that of the optic disc to greater than or equal to an optic disc. Corresponding odds for incident blindness increased from 3.2 to 24.0. Presence of macular pigment also predicted incident visual loss, while the co-presence of macular drusen and pigment magnified this effect. Conclusion: Many macular lesion characteristics predicted incident visual impairment and blindness in this population, including drusen size, the area involved by large drusen, and presence of macular pigment.
Keywords: 308 age-related macular degeneration • 354 clinical (human) or epidemiologic studies: prevalence/incidence • 355 clinical (human) or epidemiologic studies: risk factor assessment