Influence of optic disc size on the sensitivity of the Heidelberg Retina Tomograph

Purpose: The aim of this study was to evaluate the influence of optic disc size on the ability of variables generated by the Heidelberg Retina Tomograph, software version 1.11, to distinguish glaucoma patients from normals in a cross-sectional study. Patients and methods: For evaluation of the optic disc with the Heidelberg Retina Tomograph, eyes of 87 normals and 61 glaucoma patients were selected. Disc area, volume above reference, rim area, cup area, cup volume, third moment in contour and the mean retinal nerve fibre layer thickness at the border of the optic disc were selected for evaluation. The sensitivity at 80% specificity of these variables was calculated under three conditions: (1) no correction for disc size; (2) calculation in the following disc size ranges: class I (1.0-2.0 mm(2)), class II (2.0-2.5 mm(2)), class III (2.5-3.15 mm(2)) and class IV (3.15-5.0 mm(2)); (3) calculation globally, taking into account the different disc sizes of the four classes. Results: Neuroretinal rim area, cup area and volume, third moment and nerve fibre layer thickness were significantly and positively correlated with optic disc size in normals. Neuroretinal rim volume measurements were statistically independent of optic disc size. Sensitivity was lowest when the size of the optic disc was not considered. Sensitivity increased when evaluated in four optic disc size classes. With growing optic disc size sensitivity of the retinal nerve fibre layer increased and third moment value decreased. Cup volume and area had their lowest sensitivity for large optic discs. The area of the neuroretinal rim was the only variable which showed similar sensitivity values under all three conditions. Conclusions: The sensitivities of HRT variables varied with optic disc size. Evaluation of their sensitivity requires the definition of optic disc classes or statistical correction for the size of the optic disc.