Purchase this article with an account.
Michele M. Iester, Gadi Wollstein, Richard A. Bilonick, Juan Xu, Jessica Nevins, Hiroshi Ishikawa, Larry Kagemann, Joel S. Schuman; Agreement Among Graders on Heidelberg Retina Tomograph (HRT) Topographic Change Analysis (TCA) Glaucoma Progression Interpretation. Invest. Ophthalmol. Vis. Sci. 2012;53(14):654.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
HRT TCA reports the topographic change from baseline in the optic disc (ONH) and the peripapillary area. The purpose of this study was to evaluate agreement among 3 glaucoma experts of TCA printout interpretations of glaucoma progression and explore methods for improving agreement.
100 eyes of glaucoma, glaucoma suspect, and healthy subjects with ≥5 visits and 2 good quality HRT scans acquired at each visit were enrolled to this study. Three experts independently graded randomly ordered TCA printouts as progressors or non-progressors. Each grader was presented with 2 sets of tests: a randomly selected single test from each visit and both tests from each visit. Furthermore, the TCA printouts were classified with and without pre-defined criteria. The criteria were: changes should be considered if reproducibly occurring within the ONH, while disregarding changes along blood vessels or at steep rim locations. Sign of image distortion (alternating red and green pixels) should be graded with caution. The agreement was modeled using common latent factor measurement error structural equation models for ordinal data.
Plots showing the probability of declaring progression were constructed for each expert (Figure). A shallower slope indicates lower precision of the assessment and a lower intercept indicates higher probability of classifying a subject as progressing. Assessing 2 scans per visit without using the pre-defined criteria produced significant differences in slope and intercept among the 3 experts. The agreement improved when only one scan per visit without criteria was used with significant difference still present among the experts. However, when the pre-defined criteria were used, no significant difference was noted for either slope or intercept for any of combination of experts.
Using a set of simple pre-defined criteria substantially improved agreement among experts in assessing TCA progression while using additional tests in each visit did not improve agreement.
This PDF is available to Subscribers Only