Purpose:
To compare Fourier-domain optical coherence tomography (FDOCT), optic nerve photography and confocal laser polarimetry (GDx-VCC) for the detection of retinal nerve fiber layer (RNFL) injury in glaucomatous and normal eyes.
Methods:
Consecutive patients undergoing FDOCT (Topcon 3D-1000, Topcon, Inc., Paramus, NJ), GDx-VCC (Carl Zeiss Meditec, Inc., Dublin, CA), optic nerve photography, and perimetry were enrolled. Glaucomatous eyes had an achromatic visual field defect consistent with glaucoma (SITA-SAP 24-2, Carl Zeiss Meditec, Inc., Dublin, CA). Normal subjects had no history of glaucoma and a normal SITA-SAP examination. Disc photographs and FDOCT thickness maps were visually inspected by 2 observers, masked to the appearance of the optic disc and visual field, and the superior and inferior parapapillary RNFL classified as normal or abnormal. A GDx defect was defined as a Nerve Fiber Index (NFI) greater than 30 and/or an arcuate cluster of superpixels (p<0.5%).
Results:
12 normal eyes (7 patients) and 29 glaucomatous eyes (21 patients) were evaluated. Mean deviation (-7.68 ± 6.73 dB vs. 0.23 ± 1.04 dB, p < 0.001, t-test) and pattern standard deviation (7.34 ± 4.33 dB vs. 1.35 ± 0.23 dB, p < 0.001) were greater in glaucomatous eyes. Sensitivity and specificity data can be found in the table. For eyes with visual field loss >5 dB, sensitivity was 100% for GDx and FDOCT and 60% for disc photography. Disagreements among observers were more common in review of optic nerve photographs (10/41 eyes, kappa=0.49) than GDx (4/41, kappa=0.81) or FDOCT (1/41, kappa=0.95).
Conclusions:
Visual inspection of FDOCT images showed a greater sensitivity in detecting RNFL defects than did a review of optic nerve photography.
Keywords: imaging/image analysis: clinical • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • nerve fiber layer