Reproducibility and Repeatability of Cirrus™ HD-OCT Peripapillary Retinal Nerve Fibre Layer Thickness Measurements in Young Normal Subjects

Abstract

Purpose: To assess the reproducibility and repeatability of peripapillary retinal nerve fibre layer (RNFL) thickness measurements using a spectral-domain optical coherence tomography (SD-OCT) device in healthy subjects. Methods: In this observational study, 68 young Caucasian healthy volunteers (68 eyes) were subjected to Cirrus™ high-definition (HD) OCT (Zeiss) peripapillary RNFL thickness measurements by two experienced examiners in two different sessions. Average, 4-quadrant and 12-clock-hour sector RNFL thicknesses were analysed. For each option, intra-observer, intrasession repeatability and interobserver, intersession reproducibility were tested. To assess the repeatability of measurements, the Bland and Altman plots were used and the coefficient of repeatability was calculated. Interobserver and intersession reproducibilities were analysed by means of concordance correlation coefficients (CCCs). Results: The sample age ranged from 21 to 39 years (mean 29.09, standard deviation ±5.21). The average RNFL thickness ranged from 90.97 to 91.46 and from 91.34 to 91.78 µm, for the first and the second operator, respectively. The highest repeatability and reproducibility were obtained for average RNFL thickness with coefficients of repeatability of 5.30 and 6.05 µm for the first and the second operator, interoperator CCCs of 0.95 and 0.96 for the first and the second session, and intersession CCCs of 0.96 and 0.97 for the first and the second operator, respectively. Conclusions: Cirrus OCT peripapillary average RNFL thickness measurement in young healthy subjects showed high interoperator and intersession reproducibility. Intrasession repeatability as tested by coefficient of repeatability was next to the device resolution, with very similar results between the two operators. When analysing quadrant and clock hour sector RNFL thickness measurements, both repeatability and reproducibility tend to decrease.