The reproducibility of glaucoma probability score (GPS) parameters of Heidelberg retina tomograph-III (HRT-III)

Purpose:To evaluate the reproducibility of Glaucoma Probability Score (GPS) parameters of Heidelberg Retina Tomograph-III (HRT-III). Methods:20 normal subjects and 20 glaucomatous patients underwent 10 HRT-III examinations within the same session in a randomized eye. All tests were performed by the same operator. Reproducibility of the GPS algorithm parameters was addressed by means of Intraclass Correlation Coefficient (ICC) for Glaucoma Probability, Rim Steepness, Cup Size, Cup Depth, Horizontal Retinal Nerve Fiber Layer (RNFL) Curvature, Vertical RNFL Curvature. Standard deviations were calculated for each parameter of each patient; comparison between normal and glaucomatous was done by means of t-test. Results:According to the HRT classification, the mean quality of the examinations was Good (22,62 ± 6,62 microns), ranging from Excellent to Poor ( from 10 to 44,70 microns); images obtained in glaucomatous patients had lower quality (P = 0.0018, t-test). ICC and standard deviations for each parameter are reported on the table. Conclusions:GPS parameters were highly reproducible. Difference in standard deviation for cup size between normal subjects and glaucomatous patients was statistically significant (P = 0.0017, P< 0.01), as well as for horizontal RNFL curvature (P= 0.0142, P<0.05) and vertical RNFL curvature (P= 0.0107, P<0.05).