Key points are not available for this paper at this time.
PURPOSE: The authors determine which optic disc topographic parameters obtained by the Heidelberg Retina Tomograph (HRT, Heidelberg Engineering, Heidelberg, Germany) are most useful in detecting individuals with early glaucomatous visual field loss. METHODS: Ninety-nine eyes of 49 healthy individuals and 50 age-matched individuals with early glaucomatous visual field loss were included. Three images were obtained and the mean topography image was created and used in the analyses. The HRT discriminant analysis function (software version 2.01) was applied and compared to the Fisher linear discriminant function developed in this population. Analysis was repeated after stratifying by disc area (< 2 mm2 or 2-3 mm2). RESULTS: There were statistically significant differences between the healthy and glaucomatous groups for all optic disc topographic parameters (p < 0.05) measured. These differences remained after the analysis was repeated controlling for disc size, except for height variation contour. Applying the HRT discriminant function to this study population resulted in sensitivity and specificity of 62% and 94%, respectively. The sensitivity was 83% while specificity remained high (91%) for larger disc sizes. Using this data, additional discriminant functions that differentiated similarly between the two groups were found. The best formula used cup-shape measure (third moment), rim area, height variation contour, and retinal nerve fiber layer thickness and had a sensitivity and specificity of 78% and 88%, respectively. CONCLUSIONS: Several different discriminant analysis formulas are capable of detecting early glaucomatous visual field loss in a comparable manner. The characteristics of the study population are likely to influence the discriminating power of these various formulas.
Bathija et al. (Wed,) studied this question.