Air-pulse corneal applanation signal curve parameters for the characterisation of keratoconus

Aims To test the use of parameters obtained from the ocular response analyser (ORA) to distinguish between normal and keratoconic eyes, to determine the severity of keratoconus (KC), and to evaluate changes after treatment with cross-linking (CXL) and intrastromal corneal ring (ICR) implantation. Methods In total, 42 parameters were derived from the ORA's applanation response curve. Data on 119 subjects were included in the study. We determined the power of discrimination between control subjects (n=48) and KC patients (stages I and II, n=54) using receiver operating characteristic (ROC) curves. We then tested the correlation between all KC stages (n=71) and the ORA parameters. Finally, we evaluated the changes at 4 months after CXL (n=22) and ICR (n=39). Results The ROC curves for 12 parameters showed excellent results; 24 parameters generated moderate results and six parameters generated poor results. The correlations between the KC stages and 14 parameters were significant, with good to moderate results. The corneal resistance factor and the area under the second peak of the signal curve produced the best results in distinguishing between normal and KC eyes. Two parameters after CXL (p2area and time1) and six parameters after ICR implantation (aplhf, uslope11, w11, path11, time1 and deltatime) showed significant differences with respect to the preoperative conditions. Conclusion Most of the new waveform parameters demonstrated good ability to distinguish between KC and normal eyes. The changes in the parameter values after CXL and ICR treatments were smaller than expected.