Keratoconus diagnosis using Corvis ST measured biomechanical parameters

Abstract

Purpose: To assess the diagnostic power of the Corneal Visualization Scheimpflug Technology (Corvis ST) provided corneal biomechanical parameters in keratoconic corneas.

Methods: The following biomechanical parameters of 48 keratoconic eyes were compared with the corresponding ones in 50 normal eyes: time of the first applanation and time from start to the second applanation [applanation-1 time (A1T) and applanation-2 time (A2T)], time of the highest corneal displacement [highest concavity time (HCT)], magnitude of the displacement [highest concavity deformation amplitude (HCDA)], the length of the flattened segment in the applanations [first applanation length (A1L) and second applanation length (A2L)], velocity of corneal movement during applanations [applanation-1 velocity (A1V) and applanation-2 velocity (A2V)], distance between bending points of the cornea at the highest concavity [highest concavity peak distance (HCPD)], central concave curvature at the highest concavity [highest concavity radius (HCR)]. To assess the change of parameters by disease severity, the keratoconus group was divided into two subgroups, and their biomechanical parameters were compared with each other and with normal group. The parameters' predictive ability was assessed by receiver operating characteristic (ROC) curves. To control the effect of central corneal thickness (CCT) difference between the two groups, two subgroups with similar CCT were selected, and the analyses were repeated.

Results: Of the 10 parameters compared, the means of the 8 were significantly different between groups (P < 0.05). Means of the parameters did not show significant difference between keratoconus subgroups (P > 0.05). ROC curve analyses showed excellent distinguishing ability for A1T and HCR [area under the curve (AUC) > 0.9], and good distinguishing ability for A2T, A2V, and HCDA (0.9 > AUC > 0.7). A1T reading was able to correctly identify at least 93% of eyes with keratoconus (cut-off point 7.03). In two CCT matched subgroups, A1T showed an excellent distinguishing ability again.

Conclusions: The A1T seems a valuable parameter in the diagnosis of keratoconic eyes. It showed excellent diagnostic ability even when controlled for CCT. None of the parameters were reliable index for keratoconus staging.

Keywords: Biomechanics; Cornea; Corvis ST; Keratoconus.

Fig. 1

(A–G) Box-and-whisker plots of Corneal Visualization Scheimpflug Technology (Corvis ST) parameters for two keratoconus subgroups and normal group.

Fig. 2

Receiver operating characteristic (ROC) curves of first applanation time (A1T) for distinguishing between keratoconus and normal eyes in central corneal thickness (CCT) controlled subgroups.