[A comparison of device-based diagnostic methods for keratoconus].

PURPOSE: Due to the modern device-assisted diagnosis of keratoconus by topography, tomography and biomechanical properties of the cornea, a large number of parameters and indices are obtained as a result of clinical examinations. The aim of the present study was to investigate how modern screening methods support the diagnosis of keratoconus.

PATIENTS AND

METHODS

In this prospective study, 93 eyes of 93 keratoconus patients and 107 eyes of 107 healthy subjects (control group) were included. The keratoconus group contained 85 % males, whereas the distribution in the control group was balanced. The mean age was 35 ± 12 years in the keratoconus group and 27 ± 7 years in the control group. Exclusion criteria for both groups were previous eye surgery, cross-linking therapy, glaucoma, uveitis or other inflammatory diseases of the eye. All patients with a thyroid disorder were also excluded from the control group. All eyes were examined using the TMS-5 topographer, Pentacam and Ocular Response Analyzer (ORA). Based on receiver operator characteristics (ROC), the performance of various keratoconus indices was determined by means of the area under the curve (AUC).

RESULTS

All parameters showed statistically highly significant differences between the keratoconus and control group (p ≤ 0.0001). The Surface Asymmetry Index (SAI) and the Keratoconus Severity Index (KSI) of TMS performed well with (mean value keratoconus group/mean value control group/AUC) SAI (2.43/0.36/0.969) and KSI (50.87/0.37/0.912). Pentacam parameters Index of Surface Variance (ISV) and Topographic Keratoconus Classification (TKC) were comparable to TMS parameters with ISV (90.05/15.77/0.969), TKC (2.23/0.00/0.940). ORA indices Corneal Hysteresis (CH), Corneal Resistance Factor (CRF) and Keratoconus Match Index (KMI) showed slightly poorer performance with CH (8.22/11.48/0.909), CRF (7.25/11.20/0.951), KMI (0.31/1.05/0.909).

CONCLUSION

In this study, tomography and topography was more reliable in diagnosing keratoconus than evaluating the biomechanical properties of the cornea. SAI and KSI (TMS) as well as TKC and ISV (Pentacam) showed improved recognition rates compared to the KMI (ORA). However, individual parameters alone are not sufficient for the diagnosis of keratoconus.