A Novel scoring system for distinguishing keratoconus from normal eyes

Abstract

Purpose: To evaluate the accuracy of a novel scoring system in differentiation of keratoconus (KC) eyes from normal eyes using a Scheimpflug camera system tomography. Setting: Marmara University Hospital, Istanbul, Turkey and Birinci Eye Hospital, Istanbul, Turkey. Design: Retrospective case-control study. Methods: The study included 624 keratoconus eyes and 512 healthy eyes. Thirty nine significant parameters obtained from the Scheimpflug imaging system (Pentacam-Oculus Optikgerate GmbH, Wetzlar, Germany) were studied. The cut-off value and area under receiver operating characteristic (AUROC) curve analysis for each studied parameter were established in the previous study. Minus three and plus three standard deviations of the cut-off value were scored after multiplication of AUROC for each parameter. The sum of all scores (TKS; Total Keratoconus Score) was compared between keratoconus and normal eyes. Results: Average TKS value was -29.57 +/- 5.65 (Range from -43.11 to -7.09) in normal eyes and 36.23 +/- 24.3 (Range from -16.82 to 97.45) in keratoconus eyes (p < 0.001). Receiver operating characteristic curve analyses showed perfect predictive accuracy for TKS (ROC 1.0 (95% CI: 0.999-1.0)). The TKS distinguished the keratoconus group from the normal group with 99% sensitivity and 99% specificity at the best cut-off point of -12.45. Conclusions: The new scoring system measured by the Scheimpflug imaging system provides perfect discrimination of keratoconus from normal corneas. (C) 2016 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved.