Dynamic Scheimpflug-based assessment of keratoconus and the effects of corneal cross-linking

Abstract

Purpose: To evaluate the ability of the Corvis ST (Oculus Optikgeräte GmbH, Wetzlar, Germany) to distinguish between normal and keratoconic corneas and detect the possible stiffening effect of corneal cross-linking (CXL) in vivo.

Method: A retrospective study including 86 eyes of 86 participants (27 patients with untreated keratoconus, 27 patients with CXL-treated keratoconus, and 32 normal patients). The prospective study included 9 keratoconic eyes measured 24 days before and 97 days after CXL.

Results: The following parameters from the Corvis ST showed a significant difference between keratoconic and normal eyes: intraocular pressure, corneal thickness, first applanation time, highest concavity (HC) radius, HC deformation amplitude, and HC deflection amplitude (analysis of variance, P < .0001). Receiver operating characteristic curve analysis showed the highest area under the curve for HC deflection amplitude (0.942, P < .0001). Prediction intervals for individual measurements within the two groups overlapped considerably. No significant difference was found between the untreated and CXL-treated keratoconus groups (analysis of variance, P > .05). Measurements performed before and after CXL in the same patients showed significant difference in HC deformation amplitude (P = .018), HC time (P = .018), and second applanation time (P = .027).

Conclusions: Although the corneal deformation characteristics differed between groups of keratoconic and normal eyes, this study showed that the standard parameters of the Corvis ST cannot readily be used for diagnosis of keratoconus in the individual patient or to document the effect of CXL in vivo.