Diagnostic accuracy of the Retinal Thickness Analyser: differentiation between normal eyes and eyes with glaucoma or macular pathologies

Abstract

Background: The Retinal Thickness Analyser (RTA) is intended to detect glaucomatous changes as well as macular pathologies at the posterior pole. We determined the diagnostic accuracy for eyes with manifest glaucoma or macular diseases.

Methods: We examined 71 eyes with long-term, established eye conditions. Included were 28 eyes with glaucoma, 21 with different macular diseases and 22 normal eyes. All examinations were evaluated in a blind-test by RTA experts without any clinical information on the patients. After comparison of the RTA interpretation with the clinical diagnosis, we determined sensitivity, specificity, positive and negative predictive values.

Results: Of 71 examinations, 15 (21%) were not interpretable. If these results are excluded, the following diagnostic accuracy values were calculated for glaucoma and macular disorders respectively: sensitivity 75 and 59%, specificity 55 and 97%, positive predictive value 48 and 90% and negative predictive value 80 and 84%. These values were not significantly different when both eyes of each patient were included in the final analysis (n=133).

Conclusion: The diagnostic values of the RTA determined in this case control study were not satisfactory. However, no clinical information was used in the assessment. The extent to which additional clinical information increases the diagnostic value remains to be determined.