Test-retest variability of retinal nerve fiber layer thickness and macular ganglion cell-inner plexiform layer thickness measurements using spectral-domain optical coherence tomography.

PURPOSE: To evaluate the test-retest variability of spectral-domain optical coherence tomography (OCT) in measurement of retinal nerve fiber layer (RNFL) thickness and macular ganglion cell-inner plexiform layer (GCIPL) thickness. METHODS: A total of 65 eyes of healthy subjects were enrolled in this observational cross-sectional study. RNFL thickness and GCIPL thickness were measured using the repeat scan optic cube and macular cube protocol using Cirrus HD-OCT (software version 6.0). A single operator obtained 3 measurements during 1 session to determine test-retest variability. Intrasession repeatability was defined by intraclass correlation, limits of agreement, and coefficient of variation. RESULTS: The mean age of patients was 37.89±15.11 years (range, 10 to 70 y). The mean RNFL thickness readings as measured during 3 sessions were 93.89±9.73, 93.63±10.00, and 93.55±9.64 μm and average GCIPL thickness measurements were 82.90±4.61, 82.98±4.24, and 83.06±4.36 μm, respectively. Coefficient of variation was 1.2 for average RNFL thickness and 0.82 for average GCIPL thickness. The intraclass correlation coefficient showed a good correlation between repeat measurements for both average RNFL and GCC thicknesses (0.994 and 0.990, respectively). The limits of agreement (95% confidence interval) for the 3 sessions ranged from -3.61 to 4.13 μm for the average RNFL thickness and -2.55 to 2.40 μm for GCIPL thickness measurements. CONCLUSIONS: In healthy eyes, Cirrus HD-OCT shows excellent intrasession repeatability for RNFL and GCIPL thickness measurements.