Purpose:
Compare Zernike representation of corneal thickness and biomechanics of keratoconus (KC) to normal (N) eyes
Methods:
Both eyes from 40 KC and 51 N patients were analyzed retrospectively. KC diagnosis was based on clinical data including Placido Topography. Corneal thickness was measured by Pentacam HR. The matrix of corneal thickness from limbus to limbus was analyzed using a linear form of Zernike polynomials up to 9th order. The root mean square error (RMSE) between the in vivo and the Zernike estimated corneal thickness was used as a goodness of fit metric. The RMS of C3, C4 and C5 was reported as LORMST (lower order polynomials for thickness). Similarly, the RMS of C6 to C14 was reported as HORMST (higher order polynomials for thickness). Biomechanical measurements were made by Ocular Response Analyzer (ORA) and reported as CH (corneal hysteresis), CRF (corneal resistance factor) and Hysteresis loop area (HLA). Since two eyes per patient were used, a non-parametric, clustered and correlated receiver operating characteristics curve (ROC) was constructed for LORMST, HORMST, CH, CRF and HLA. Z test was used to compare the area under the ROC curve (AUROC) between variables. A p value < 0.05 was considered statistically significant.
Results:
The mean RMSE was less than 0.5 + 0.1 micron in KC and N groups. LORMST and HORMST were higher in KC than N. HLA, CH and CRF were lower in KC than N. HLA had the highest AUROC followed by HORMST and CRF. The sensitivity and specificity was the highest for HLA. Cut-off values (the optimal sensitivity and specificity) were 79023, 4.05 and 8.3 for HLA, HORMST and CRF, respectively. Comparing the AUROC’s, HLA was superior to all variables (p < 0.05) but not significantly better than HORMST (p = 0.12). HORMST was significantly better than LORMST (p=0.002), but similar to CH (p=0.16) and CRF (p = 0.4) in distinguishing KC from N eyes.
Conclusions:
HLA (biomechanical index) and HORMST (3-D thickness index) can improve the diagnostic precision of current KC screening.
Keywords: cornea: clinical science • keratoconus • imaging/image analysis: clinical