Abstract
Abstract: :
Purpose: To longitudinally evaluate cup-to-disc ratio in treated patients with primary open-angle glaucoma (POAG). Methods: Fifty eyes of 50 treated POAG patients with a minimum of 9-year longitudinal series of stereoscopic optic disc photographs were studied. Eyes with any other ocular disease except for mild cataract were excluded. Each set of stereoscopic photographs was digitized and viewed stereoscopically on computer screen using a hand-held stereoscope(Screen-VU, Portland,OR) . Computer-assisted planimetry was performed on each set of photographs with examiner-defined cup and disc margins using custom-made software. The software computed linear cup-to-disc ratios (LCDR) for the entire disc as well as 4 disc quadrants. Both inter-rater (3 raters) and intra-rater variabilities were evaluated in a masked fashion, using intraclass correlation. Changes in LCDR were estimated using linear regression over time. Results: We studied 26 men and 49 Caucasians (1 African-American). The mean age at the beginning of the study was 58.6 ± 10.0 years (range, 33 to 80). The mean follow-up period was 14.9 ± 4.0 years (range, 9.4 to 25.8;median,14.5). A total of 166 sets of stereo photographs were analyzed (3.3 ± 1.2 per patient). The initial and final LCDRs were 0.64 ± 0.16 and 0.70 ± 0.15 respectively. The intra-rater and inter-rater intraclass correlations were 0.97 (95% CI: 0.94-0.98) and 0.76 (95% CI: 0.61-0.87). Using linear regression the rate of LCDR change was +0.006 ± 0.006 per year (range, -0.006 to +0.023). The rates of LCDR change in the 4 disc quadrants were 0.005 ± 0.007 (SN), 0.006 ± 0.008 (ST), 0.007 ± 0.012 (IN), and 0.007 ± 0.010 (IT) per year. Conclusion: The intra-rater variability of estimating LCDR was very good (0.97) while inter-rater variability was substantial (0.76). In treated patients with POAG, the rate of progressive cupping was slow (LCDR change of +0.006 per year).
Keywords: 498 optic disc • 430 imaging/image analysis: clinical