Materials and Methods: Inter-device concordance was investigated using the intra-class correlation coefficient (ICC) for average and four quadrants peripapillary RNFL thicknesses, and the mixed-design two-way analysis of variance for color-coded graphical formats. The performance of RNFL thicknesses in discriminating early glaucoma from suspected glaucoma was assessed using the area under the receiver operating characteristic (AUROC) curve analysis.

Results: Thirty eyes with suspected glaucoma (Group I) and 30 with early glaucoma (Group II) were enrolled in the study. All RNFL measurements obtained by the Cirrus OCT were thicker than those obtained by the Copernicus OCT (p <0.001). On the 95% confidence interval, the ICC assessed a ?moderate to excellent? inter-device concordance for the average and inferior RNFL thicknesses in both groups. Except for the temporal quadrant, the color code concordance ranged from 46.7% to 70% in Group I and from 30% to 40% in Group II. The Cirrus OCT displayed more optimal color codes compare to the Copernicus. The values of AUROC for RNFL thickness ranged from 0.586 to 0.682 (p< 0.05, for only the average RNFL) for the Cirrus and from 0.703 to 0.769 (p<0.05, for all) for the Copernicus.

Conclusion: The Cirrus and Copernicus OCT yielded different peripapillary RNFL thickness measurements, and differences also existed in their color code interpretations and diagnostic powers. Keywords : Glaucoma, Optical Coherence Tomography (OCT), retinal nerve fiber layer (RNFL), Color code