Abstract
Purpose :
The Reichert Ocular Response Analyzer (ORA) and the Oculus CorvisST (CST) analyse the dynamic corneal response to an air-puff to provide measures of corneal biomechanics. In isolation, the clinical value of the outputs provided by these devices have been questioned. Given the premise that there is correspondence between the biomechanical metrics provided by the two devices we examine first, the nature of the correspondence and second, whether cluster analysis would aid in interpretation and characterisation of the results.
Methods :
113 healthy adults (226 eyes; 71% females) aged 24.5±6.11 years were assessed with the ORA and CST. ORA metrics included corneal hysteresis (CH(mmHg)) and corneal resistant factor (CRF(mmHg)). CST metrics are based on evaluation of 3 phases during the air-puff application: first applanation (A1), highest concavity (HC) and second applanation (A2). Each phase was described by examining the: deformation amplitude (DA), time point (t), velocity (v) and length (l). Measures of central corneal thickness (CCT (mm)) were obtained from CST readings. The relationship between ORA and CST measurements was evaluated by Pearson's correlation coefficient. Data were assessed via cluster and discriminant analysis to group the eyes according to their ORA and CST metrics.
Results :
Strongest associations were observed between CH and A2DA (r=0.272) and CRF and A1t (r=0.474). Multiple correlation coefficient demonstrated that CH showed the highest association with the quantification of A1 (r=0.348) and CRF with A2 (r=0.228). Cluster analysis revealed that the data can be categorized into 3 distinct groups: Group 1 (n=33%): CCT (551±30.5), IOP (13.6±1.52), CH (11.4±1.58) and CRF (10.7±1.45); Group 2 (n=21%): CCT (586±34.7), IOP (17.1±2.17), CH (11.8±2.09) and CRF (12.4±1.96); Group 3 (n=46%): CCT (520±43.3), IOP (12.4±1.77), CH (10.3±1.23) and CRF (9.5±1.14). Aside from HCt and A2DA, all CST metrics were significantly different between groups. Discriminant analysis showed that A1t, A2v and A1DA had the highest impact on grouping.
Conclusions :
The results of the study suggest that the biomechanical characteristics of healthy corneas can be categorised into distinct groups. It is anticipated that these findings will aid in the assessment and interpretation of ORA and CST outputs particularly when attempting to discriminate between healthy eyes and those with corneal pathology.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.