School of Medicine

To assess the performance, in an independent population, of previously published confocal scanning laser ophthalmoscopy Topographic Change Analysis (TCA) parameter cut-offs ofr discriminating between progressing glaucoma, stable glaucoma, and healthy eyes. Five published TCA cut-offs were applied to the following four groups: 54 glaucomatous eyes (at study baseline examination) progressing by optic disc stereophotograph assessment, 79 glaucomatous eyes progressing by standard automated perimetry guided progression analysis (GPA), 72 stable glaucoma eyes (patients tested 5 times over 5 weeks), and 135 healthy eyes. All eyes were imaged at least four times by Heidelberg Retina Tomograph (HRT; Heidelberg Engineering, Heidelberg, Germany) as part of the Diagnostic Innovations in Glaucoma Study (DIGS) and African Descent and Glaucoma Evaluation Study (ADAGES). Sensitivity and specificity for classifying progressed and stable eyes, respectively, were reported. The two TCA parameters providing the best sensitivity/specificity trade-off were the 95% cut-off for the largest clustered super-pixel area within the optic disc margin10 (sensitivity of 0.922 in stereophotograph progressors and specificity of 0.778 in stable glaucoma eyes) and the Moderate Criteria (largest clustered superpixel area within the optic disc margin > or equal to 1% of the disc area with > or equal to 50 [mu]m mean depth change). These cut-offs detected progression over a similar time frame. Specificity in healthy eyes was lower than in stable glaucoma eyes. Previously published HRT TCA parameters can discriminate between progressing and stable glaucoma eyes in an independent population with good sensitivities and specificities. Low specificity of TCA in healthy eyes might be due to the effects of aging on optic disc topography, evidenced by the long follow-up in this group.