Anterior chamber angle measurement with optical coherence tomography:: Intraobserver and interobserver variability

PURPOSE: To assess intraobserver and interobserver variability of anterior segment optical coherence tomography (AS-OCT) as an objective diagnostic tool to quantify the anterior chamber angle and opening width. SETTING: University Eye Clinic, Lubeck, Germany. METHODS: The anterior chamber angle and opening width were assessed in 18 eyes of 9 healthy volunteers by 2 observers. Intraobserver reproducibility was evaluated by calculating an intraclass correlation coefficient (ICC) in a mixed model. Each observer had a separate model using information from 5 scans. Interobserver variability was determined by Bland-Altman analysis. The ICC was calculated in a mixed model using a residual maximum likelihood method. The results of 3 and 5 repeated scans were evaluated to indicate the change to 1 measurement application. RESULTS: The mean anterior chamber angle measurement was 35.9 degrees +/-5.7 (SD) for observer A and 36.2 +/- 5.7 degrees for observer B. The ICC was 0.94 and 0.91, respectively. The mean opening width was 315 +/- 62 mu m for observer A and 317 +/- 60 mu m for observer B. The ICC was 0.97 and 0.93, respectively. Interobserver comparisons showed a mean difference between anterior chamber angle measurements of -0.27 +/- 1.6 degrees, a limit of agreement (LOA) interval from -3.52 to 2.98 degrees, and an ICC estimate of 0.96. The mean difference in opening width measurements was 2.40 +/- 12.40 mu m, the LOA from -27.20 to 22.40 mu m, and the estimated ICC 0.96. Using I instead of the mean of 5 measurements, the LOA range increased by 3.46 degrees for the anterior chamber angle and 30.0 mu m for the opening width. CONCLUSION: Anterior chamber angle and opening width measurements by OCT showed low intraobserver and interobserver variability, indicating OCT is a valuable technique for quantitative assessment that provides reproducible measurements and objective documentation by different examiners.