Factors Associated with Variability in Retinal Nerve Fiber Layer Thickness Measurements Obtained by Optical Coherence Tomography

Abstract

To identify factors associated with variability in retinal nerve fiber layer (RNFL) thickness measurements obtained by optical coherence tomography (OCT). Retrospective observational case series. One hundred fifty consecutive patients with known or suspected glaucomatous optic nerve damage undergoing OCT imaging for RNFL thickness measurement. One eye with known or suspected glaucoma from each patient was scanned twice within the same visit using the Stratus OCT's fast nerve fiber layer thickness acquisition protocol. For each scan, the average RNFL thickness, signal strength (a measure of the amount of light reflecting back from the retina), and analysis confidence (AC, a measure of the quality of the data as reported by the OCT software) were recorded. Retinal nerve fiber layer thickness measurements of the initial and repeat scans for each case were compared, and the difference in thickness measurements was correlated with difference in signal strength, AC (either low or normal), and the average RNFL thickness. Difference in RNFL thickness. The mean difference in RNFL thickness between the initial and repeat scans was 10 microns (12.5%; range, 0.04-86.15 microns). Difference in signal strength between initial and repeat scans positively correlated with the difference in RNFL thickness (Spearman correlation coefficient = 0.44; P<0.001), whereas the average nerve fiber layer thickness negatively correlated with the difference in RNFL thickness (Spearman correlation coefficient = -0.25; P = 0.002). The presence of a scan with low AC was associated with a lower average RNFL thickness and a greater difference in RNFL thickness between the initial and repeat scans. Signal strength variability, low AC, and low RNFL thickness are associated with variability in RNFL thickness measurements obtained by Stratus OCT. Comparability of RNFL thickness measurements between visits may be improved if scans of similar signal strengths without low AC are obtained. This is especially important for patients with moderate glaucomatous optic nerve damage and for patients from whom good quality scans are not obtainable.