Repeatability and Agreement of Central Corneal Thickness and Keratometry Measurements between Four Different Devices

Abstract

Background. To estimate repeatability and comparability of central corneal thickness (CCT) and keratometry measurements obtained by four different devices in healthy eyes. Methods. Fifty-five healthy eyes from 55 volunteers were enrolled in this study. CCT (IOLMaster 700, Pentacam HR, and Cirrus HD-OCT) and keratometry readings (IOLMaster 700, Pentacam HR, and iDesign) were measured. For statistical analysis, the corneal spherocylinder was converted into power vectors (J0, J45). Repeatability was assessed by intraclass correlation coefficient (ICC). Agreement of measurements between the devices was evaluated by the Bland-Altman method. Results. The analysis of repeatability of CCT data of IOLMaster 700, Pentacam HR, and Cirrus HD-OCT showed high ICCs (range 0.995 to 0.999). The comparison of CCT measurements revealed statistically significant differences between Pentacam HR versus IOLMaster 700 (p<0.0001) and Pentacam HR versus Cirrus HD-OCT (p<0.0001), respectively. There was no difference in CCT measurements between IOLMaster 700 and Cirrus HD-OCT (p=0.519). The repeatability of keratometry readings (J0 and J45) of IOLMaster 700, Pentacam HR, and iDesign was also high with ICCs ranging from 0.974 to 0.999. The Pentacam HR revealed significantly higher J0 in comparison to IOLMaster 700 (p=0.009) and iDesign (p=0.041); however, no significant difference was between IOLMaster 700 and iDesign (p=0.426). Comparison of J45 showed no significant difference between IOLMaster 700, Pentacam HR, and iDesign. These results were in accordance with Bland-Altman plots. Conclusion. In clinical practice, the devices analyzed should not be used interchangeably due to low agreement regarding CCT as well as keratometry readings.