Methods for detecting age-related maculopathy: a comparison between photographic and clinical assessment

Purpose: To examine the sensitivity, specificity and overall agreement betw een photographic and clinical assessment in detecting age-related maculopat hy (ARM) features in the context of an epidemiological study, the Vitamin E , Cataract and Age-related Maculopathy Study (VECAT). Methods: A total of 1204 volunteers aged between 55 and 80 years of age, wh o were enrolled in the VECAT Study, had both slit-lamp biomicroscopy examin ation and fundus photos taken as part of the baseline ophthalmic examinatio n. The Nidek 3-DX fundus camera (Nidek, Gamagori, japan) was used to produc e paired, one-framed, coloured, 15 degrees stereoslides of the macular area at a fixed angle. An International Classification and Grading System for A ge-related Maculopathy and Age-related Macular Degeneration was used to gra de the stereoslides. Agreement in the detection of drusen, pigment abnormal ities, and late stage ARM features was assessed using unweighted kappa stat istic. Cases of disagreement were verified using clinical data records, gra ding documentation and the review of stereoslides. Results: Macula status was available for 2386 eyes. For drusen of size < 63 <mu>m, sensitivity was 47%, specificity was 68% with a kappa value of 0.20 . For drusen greater than or equal to 125 mum, sensitivity and specificity were greater than or equal to 81%. Kappa values ranged from 0.56 to 0.71. L evels of agreement For pigment abnormalities and late ARM were in the subst antial range (i.e. kappa values from 0.70 to 1.00). Conclusions: Slit-lamp biomicroscopy was found to be comparable to photogra ding (using the Nidek 3-DX fundus camera) for detecting features pertaining to ARM. However, given the objectivity and permanency of stereoslides, pho tograding is still the more reliable and the preferred system of assessing ARM in the context of an epidemiological study.