PURPOSE: To compare the peripapillary retinal nerve fiber layer (RNFL) thickness measured by spectral and time domain optical coherence tomography (OCT). PATIENTS AND METHODS: One hundred and eighty-eight subjects (normal tension glaucoma, n = 80; primary open angle glaucoma, n = 49; healthy controls, n = 59) were included in this study. Peripapillary RNFL thickness was measured by an Optic Disc Cube 200 × 200 scan of the spectral domain Cirrus HD OCT and by a Fast RNFL scan of the time domain Stratus OCT. They were compared using linear regression (R(2)), Bland-Altman plots, weighted Kappa (K(w)) coefficients, and areas under the receiver operating characteristic curve (AROC). RESULTS: Except for the inferior quadrant, RNFL thickness measurements by Cirrus HD OCT were thinner than Stratus OCT (all p < 0.001); for the inferior quadrant, this tendency was vice versa (p = 0.043). Average RNFL thicknesses of the two OCTs showed a good correlation (R(2), 0.852) but not excellent agreement (mean of difference, 4.0 μm; +1.96 SD, 10.7 μm; -1.96 SD, -18.6 μm). Regarding the probability codes, only the average RNFL thickness had almost perfect agreement (K(w), 0.983). In addition, Cirrus HD OCT had the largest AROC for the superior quadrant (0.795), but Stratus OCT had the largest AROC for average RNFL thickness (0.764) (p = 0.281). CONCLUSIONS: Except for the inferior quadrant, the average and quadrants RNFL thickness measured by the Cirrus HD OCT were thinner than the Stratus OCT. Regarding discriminating power for glaucoma detection, the Cirrus HD OCT was better than the Stratus OCT for the superior and inferior quadrant RNFL thicknesses but not for the average RNFL thickness.
PURPOSE: To compare the peripapillary retinal nerve fiber layer (RNFL) thickness measured by spectral and time domain optical coherence tomography (OCT). PATIENTS AND METHODS: One hundred and eighty-eight subjects (normal tension glaucoma, n = 80; primary open angle glaucoma, n = 49; healthy controls, n = 59) were included in this study. Peripapillary RNFL thickness was measured by an Optic Disc Cube 200 × 200 scan of the spectral domain Cirrus HD OCT and by a Fast RNFL scan of the time domain Stratus OCT. They were compared using linear regression (R(2)), Bland-Altman plots, weighted Kappa (K(w)) coefficients, and areas under the receiver operating characteristic curve (AROC). RESULTS: Except for the inferior quadrant, RNFL thickness measurements by Cirrus HD OCT were thinner than Stratus OCT (all p < 0.001); for the inferior quadrant, this tendency was vice versa (p = 0.043). Average RNFL thicknesses of the two OCTs showed a good correlation (R(2), 0.852) but not excellent agreement (mean of difference, 4.0 μm; +1.96 SD, 10.7 μm; -1.96 SD, -18.6 μm). Regarding the probability codes, only the average RNFL thickness had almost perfect agreement (K(w), 0.983). In addition, Cirrus HD OCT had the largest AROC for the superior quadrant (0.795), but Stratus OCT had the largest AROC for average RNFL thickness (0.764) (p = 0.281). CONCLUSIONS: Except for the inferior quadrant, the average and quadrants RNFL thickness measured by the Cirrus HD OCT were thinner than the Stratus OCT. Regarding discriminating power for glaucoma detection, the Cirrus HD OCT was better than the Stratus OCT for the superior and inferior quadrant RNFL thicknesses but not for the average RNFL thickness.
Authors: A Perdicchi; M Iester; D Iacovello; A Cutini; M Balestrieri; M G Mutolo; A Ferreras; M T Contestabile; S M Recupero Journal: J Ophthalmol Date: 2015-12-15 Impact factor: 1.909