Diagnostic Ability of Corneal Shape and Biomechanical Parameters for Detecting Frank Keratoconus

Abstract

Purpose: To assess the diagnostic capability of corneal shape and biomechanical parameters for distinguishing normal eyes from frank keratoconus (KCN).

Methods: This is a retrospective analysis of eyes of 137 control subjects and 145 patients with KCN, with one of their eyes randomly selected as the study eye. Corneal biomechanical parameters were assessed using the Corvis ST (Oculus Optikgeräte GmbH) and Oculus Response Analyzer (Reichert Ophthalmic Instruments). Corneal tomography was performed using the Oculus Pentacam. The clinical diagnosis of KCN was based on slit-lamp findings (eg, Fleischer ring, Vogt striae) and abnormal topographic patterns on the sagittal (axial) front curvature map, disregarding tomographic and biomechanical findings. The discriminative ability for each parameter was tested using the Mann-Whitney U test. The accuracy of each parameter with statistically significant differences was determined using receiver operating characteristic curves, which were compared using the DeLong method.

Results: Statistically significant differences were observed for all corneal shape and biomechanical parameters tested (P < 0.05) except peak distance at the highest concavity (P = 0.504). The area under the curve (AUC), sensitivity, and specificity for corneal hysteresis (CH) and corneal resistance factor (CRF) were 0.894, 80.7%, and 84.7% and 0.946, 85.5%, and 89.1%, respectively. The dynamic corneal response parameters from the Corvis ST, stiffness parameter at first applanation, integrated radius (IR), and deformation amplitude ratio at 2 mm (DA-2 mm) had AUC, sensitivity, and specificity of 0.965, 86.2%, and 94.9%; 0.961, 87.6%, and 93.4%; and 0.950, 80.7%, and 98.5%. The Corvis biomechanical index had an AUC of 0.998 (95% confidence interval, 0.983-1.000), with 96.6% sensitivity and 99.3% specificity. The Belin-Ambrósio enhanced ectasia deviation index (BAD-Dv3) and the tomographical/biomechanical index (TBI) had an AUC of 1.0 (95% confidence interval, 0.987-1.000), with sensitivity and specificity of 100%. The combined indices, Corvis biomechanical index, BAD-D, and TBI, had a statistically higher AUC than that of all corneal biomechanical parameters (DeLong, P < 0.001).

Conclusions: The Scheimpflug-derived shape and biomechanical parameters are able to accurately distinguish normal corneas from frank (clinical) keratoconic corneas. However, the combined parameters were more effective. Further studies should test milder ectasia cases.