Second-Generation Corneal Deformation Signal Waveform Analysis in Normal and Manifest Keratoconic Corneas After Statistical Correction for Potentially Confounding Factors
Narrative Responses:
Purpose
To evaluate and compare the difference in corneal biomechanical waveform parameters between manifest keratoconic (KCN) and normal corneas with a second generation ocular response analyzer (ORA), after controlling for potentially confounding factors, and to develop a new keratoconus match index.
Methods
A second generation ORA was used to obtain the corneal hysteresis (CH), corneal resistance factor (CRF), and 37 biomechanical waveform parameters in 61 KCN eyes (45 patients) and 257 healthy eyes (136 patients). Parameter differences between the two groups were assessed using T tests. A multivariable logistic regression model with stepwise variable selection was used to select the parameters most useful in distinguishing between the two groups, and was adjusted for central corneal thickness (CCT), gender and age to control for potential confounding effects.
Results
After controlling for confounding factors, there was a statistically significant difference between the CRF in normal group (10.0±1.8 mm Hg) and KCN group (6.3±1.9 mmHg)(p=0.01). There was no statistically significant difference between the CH in normal group (10.6±1.5 mm Hg) and KCN group (8.3±1.7 mmHg). Aside from CRF, 4 other parameters were found to be the most useful in distinguishing between groups:h1 (height of p1; p=0.00), dslope1 (down-slope of P1; p=0.01), uslope2 (upslope of the P2; p=0.00), and dslope21 (downslope of P2; P=0.01). The area under the receiver operating characteristic curve for the model using these parameters was 0.997.
Conclusion
Differences in multiple biomechanical waveform parameters including CRF, h1, dslope1, uslope2, dslope2 between keratoconus and normal eyes suggested that waveform analysis might be useful to differentiate between healthy and diseased biomechanical conditions. The second generation ORA system may have improved KCN detection abilities relative to the first generation system.