Long–term reproducibility of retinal nerve fiber layer measurements in healthy eyes using a scanning laser polarimeter with variable corneal compensation

Abstract

Abstract: : Purpose: To assess the long–term reproducibility of RNFL measurements in normal eyes using a scanning laser polarimeter incorporating individualized corneal compensation. Methods: Five eyes of normal volunteers were imaged three times each at baseline and one year later to assess short–term and long–term reproducibility of RNFL measurements. Measurements were performed using a scanning laser polarimeter (SLP) incorporating individualized variable corneal compensation (GDx–VCC^TM, Laser Diagnostic Technologies, Inc., San Diego, CA). Reproducibility was defined as the coefficient of variation for the parameters generated by the GDX–VCC software. A repeated measures ANOVA was conducted on the parameters to assess the significance of long–term differences. Variance components estimates were also calculated to determine the proportion of total variance accounted for by the short–term and long–term effects. Results: The repeated measures ANOVA did not find significant differences for the long–term effect for any of the parameters. The coefficient of variations were similar for the short–term and long–term effects (see table). The variance component analysis found the subject differences accounted for 85.6% of the total variance on average (averaged across parameters), the short–term variability accounted for 13.8% of the total variance, and the long–term variability accounted for 14.4% of the total variance (long–term variability added only .7% to short–term variability). Conclusion:The long–term reproducibility of the GDx VCC parameters are similar to the short–term reproducibility. 

Keywords: imaging/image analysis: clinical • nerve fiber layer