Free Papers

Reproducibility of the Corvis ST® measurements in healthy volunteers

Free Paper Details

First Author: S.Rodrigo Rey SPAIN

Co Author(s):    R. Cañones Zafra   E. Arranz Marquez   S. Ketabi Shadvar   G. Bolívar de Miguel   M. Teus        

Abstract Details

Purpose:

To assess the precision, in terms of repeatability and reproducibility, of the biomechanical parameters obtained by a Scheimpflug technology (Corvis ST, Oculus), mainly the latest set of indexes included in the new Corvis ST software.

Setting:

Department of Ophthalmology of Príncipe de Asturias University Hospital, Alcalá de Henares, Madrid, Spain.

Methods:

This is an observational, prospective study. 60 healthy eyes from 30 subjects were analyzed. Two consecutive valid examinations with a 5 minutes-pause between them were recorded with the Corvis ST device. Measurements were repeated at the same hour, in three different days. For each eye we analyzed the standard and the newest (bIOP, DA ratio, Integrated radius, ARTh, SPA1 and CBI) biomechanical parameters. The within-subject standard deviation (Sw) and coefficient of variation (CoV) were calculated to estimate the intra-session repeatability and inter-day reproducibility.

Results:

The original biomechanical parameters showed good repeatability and reproducibility, obtaining a CoV lower than 5% and intraclass correlation coefficient (ICC) higher than 0.6, except for A1 and A2 velocity and length with a higher CoV (between 12% and 15% in all cases) and ICC close to 0.3. The new parameters showed a similar good profile, the best ones were the bIOP, DA ratio, integrated radius and SPA-1 (CoV < 5.50% and ICC > 0.75 in all cases). The CBI revealed poor accuracy with a CoV of 41.54% and 55.36% and ICC of 0.826 and 0.720 for repeatability and reproducibility, respectively.

Conclusions:

A good reproducibility and repeatability were found for bIOP, DA and time parameters, moderate for velocity parameters and suboptimal for the length ones. This could be attributed to the differences in the proper location of the landmarks used in the automatic measurements. The CBI obtained a poor precision, what may be explained by the fair reliability of some of the parameters included in its logistic regression formula. In conclusion, caution should be taken when interpreting an isolated CBI value.

Financial Disclosure:

-