Effect of Adjusting Retinal Nerve Fiber Layer Profile to Fovea-Disc Angle Axis on the Thickness and Glaucoma Diagnostic Performance

Abstract

To compare optical coherence tomography (OCT) retinal nerve fiber layer (RNFL) thickness, specificity, and glaucoma discriminating ability prior and following adjustment to the fovea-disc angle axis. Cross-sectional observational study. Fovea-disc angle was measured in 282 healthy subjects (normative database), 46 nonglaucomatous myopic subjects, and 86 glaucomatous subjects. The fovea-disc angle was determined using the Cirrus PanoMap feature after registering the macular ocular fundus image to the peripapillary ocular fundus image. RNFL prediction limits, specificity, and glaucoma discrimination ability were determined before and after adjustment to fovea-disc angle axis. The mean fovea-disc angles were 7.4 ± 4.1 degrees (normative database), 7.1 ± 4.3 degrees (myopic subjects), and 7.6 ± 4.2 degrees (glaucomatous subjects) (all P > .05). Standard and fovea-disc angle-adjusted RNFL thicknesses showed small significant differences in quadrants and larger differences in clock-hour sectors (all P < .05), without consistent patterns in the direction of the change. The specificity remained unchanged for all parameters in normative database subjects after adjustment to fovea-disc angle. The number of sectors with improved specificity and those with worsened specificity did not differ significantly in the first (8 vs 6), second (5 vs 8), and random (6 vs 7) normative database half (all P > .05). In myopic subjects, the specificity improved and worsened in 4 sectors in each case. The glaucoma discriminating ability significantly increased in 6 sectors and decreased in 5 others. Correcting the RNFL profiles using the fovea-disc angle axis had no consistent effect on the thickness, specificity, and glaucoma diagnostic performance of RNFL parameters. This phenomenon may be specific to different OCT platforms.