Abstract
Purpose: :
To determine if differences in automated classification procedures of optic disc structure with the Heidelberg Retinal Topographer II (HRT II) exists between Blacks and Whites.
Methods: :
Data from 96 eyes from Black subjects and 54 eyes from White subjects without detectable ocular disease were obtained using the Heidelberg Retinal Topographer II (HRT II). Previously developed automated classification procedures that model the optic disc as a smooth three–dimensional surface were applied to the topographic data. Glaucoma Probability Score (GPS), Vertical and Horizontal RNFL curvature, rim shape, cup depth, and cup size were compared between racial groups in linear models to adjust for differences in disc area obtained from conventional HRT II analysis.
Results: :
Black normal subjects had a significantly deeper CD than Whites (Black Group = 0.61 ± 0.1 mm, White Group = 0.5 ± 0.2 p < 0.013). Differences in vertical RNFL curvature (p < 0.089), cup depth (p < 0.058), and overall GPS score (p < 0.093) were of borderline significance. After adjustment for disc area, no significant differences in GPS score or global parameters were seen. Only cup size in the inferior temporal region remained significantly different between racial groups following adjustment for disc area (Black Group = 0.07 ± 0.03 mm, White Group = 0.06 ± 0.02 p < 0.002).
Conclusions: :
Racial variation in automated classification of optic disc structure seen with the HRT II is largely due to differences in disc area. However, small residual differences in regional cup size may persist independent of disc area. Normative databases used in the development of diagnostic strategies with automated classification procedures with the HRT II should include subjects with a wide range of disc area.
Keywords: optic disc • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound)