Abstract
Purpose:
To determine if typical measures of the circumpapillary retinal nerve fiber layer (cpRNFL) thickness obtained with frequency-domain optical coherence tomography (fdOCT) can detect glaucomatous damage of the macula.
Methods:
One eye of 130 open-angle glaucoma patients or suspects was tested with the 10-2 visual field (VF) and fdOCT macular cube and disc scans (3D-OCT 2000, Topcon, Inc). Eyes were included if they had glaucomatous optic neuropathy on fundus exam and a MD (24-2 VF) better than -6 dB. Probability maps of the retinal ganglion cell plus inner plexiform layers (RGC+) were generated. To minimize false positives, an eye was classified as “Macula Abnormal” if both the RGC+ macular, and the 10-2, probability plots of at least one hemifield were abnormal based upon cluster criteria. A subset of these eyes was considered “Only Macula Abnormal” if the 24-2 VF was normal outside of the 12 points within the central ±10°. The fdOCT machine’s clinical report was used to assess the abnormality of the cpRNFL thickness within clock hour and quadrant regions of the disc.
Results:
Of the 130 eyes, 40 were classified as Macula Abnormal and 9 of these as Only Macula Abnormal. The cpRNFL thickness of the temporal quadrant, typically associated with the macula, was a poor indicator of macular damage. 24 (60%) of the 40 eyes were not significant at the 5% (yellow) level, and 29 (72.5%) were not significant at the 1% (red) level. The 3 clock hours (OD: 8, 9, 10; OS: 2, 3, 4) of the temporal quadrant were also a poor indicator of macular damage. 15 (37.5%) of the 40, and 5 (55.6%) of the 9 eyes did not have a single clock hour abnormal at the 5% or 1% level; and 25 (62.5%) of the 40, and 6 (66.7%) of the 9 eyes did not have a single clock hour abnormal at the 1% (red) level. Based upon a recent model [1], clock hour 7 OD (clock hour 5 OS) also receives input from the RGCs in the macula. Including this clock hour improved the results, although 7 (17.5%) of the 40 eyes and 2 of the 9 eyes still did not have a single clock hour at 1%.
Conclusions:
Standard analysis of the temporal quadrant of cpRNFL thickness can miss glaucomatous macular damage. While detection was substantially improved by including the inferior clock hour nearest to the temporal quadrant, cpRNFL thickness still missed macular damage confirmed on fdOCT macular cube scans and 10-2 VFs. 1. Hood, Raza et al. PRER. 2013.
Keywords: 550 imaging/image analysis: clinical •
610 nerve fiber layer •
758 visual fields