Abstract:

Objective To investigate the role of combining corneal visualization Scheimpflug technology (Corvis ST) and Pentacam HR in the diagnosis of keratoconus. Design Diagnosis test. Participants Twenty-one eyes with diagnosis of subclinical keratoconus (SKC group), forty-two eyes with keratoconus (KC group) and sixty normal eyes (control group) were included. Methods Tomography and biomechanical parameters of all eyes were obtained with the Pentacam HR and Corvis ST (Oculus Optikgerate GmbH, Wetzlar, Germany), respectively. BAD-D, CBI (Corvis biomechanical index) and TBI (tomographic and biomechanical index) were compared between the KC group, SKC group and control group. ANOVA and Kruskal-Wallis H rank sum test were used for statistical analysis. The receiver operating characteristic (ROC) curves were plotted to distinguish keratoconus and subclinical keratoconus from normal controls. Main Outcome Measures CBI, BAD-D and TBI. Results BAD-D and TBI were significantly different between the SKC group and the control group (P<0.05 for each parameters). The cut-off point of BAD-D, TBI was 2.44 and 0.58, respectively. For these two parameters, the AUC value was 0.759 (sensitivity 66.7%, specificity 95.0%) and 0.804 (sensitivity 71.4%, specificity 78.3%), respectively. Three parameters were significantly different between the KC group and the control group (P<0.001 for each parameters). The cut-off point of BAD-D, CBI and TBI was 3.77, 0.97 and 0.99, respectively. For these three parameters, the AUC value was 1.000 (sensitivity 100%, specificity 100%), 0.988 (sensitivity 92.9%, specificity 96.7%) and 0.950 (sensitivity 100%, specificity 90.0%), respectively. Conclusion Among the three parameters, TBI has the highest sensitivity for detecting subclinical keratoconus, followed by BAD-D. But CBI was not useful in detecting subclinical keratoconus. All the three parameters, BAD-D, TBI and CBI have good diagnostic power for keratoconus. （Ophthalmol CHN，2019, 28： 150-154）

Key words: keratoconus, biomechanics, tomography