Reliability of Optical Devices in Terms of Measurements of Corneal Parameters in Cross Linked Keratoconic Eyes: Cross-Sectional Study

Abstract

To determine the reliability and agreement of different optical devices in terms of keratometry and pachymetry measurements in cross-linked (CXL) keratoconic eyes. Material and Methods: Thirty-eight CXL-treated keratoconic eyes were evaluated. Three repeated measurements were performed with Topcon KR-1 autorefractokeratometry, Sirius topography, Nidek AL-scan optical biometry and RTVue anterior segment optical coherence tomography (AS-OCT). The devices were compared in terms of pachymetry and keratometric values. Limits of agreement (LoA) between the devices were detected using Bland-Altman analysis. Intra-examiner reliability was obtained using the intraclass correlation coefficient (ICC). Results: While the mean central corneal thickness (CCT) and minimum corneal thickness (MCT) measured in AS-OCT were 456.37±41.52 μm and 431.89±43.37 μm, respectively, these values were 432.40±48.97 μm and 415.67±47.69 μm in topography (p<0.001). The 95% LoA between the devices were -21.7 to 69.7 μm for CCT and -17.7 to 50.1 μm for MCT. Intra-examiner reliability was excellent for both devices in relation to the CCT and MCT measurements (ICC>0.97). When devices were compared in terms of the keratometric values and corneal astigmatism measurements, there was a significant difference among devices except between autorefractokeratometry and optical biometry (p<0.05; for all). The lowest LoA among the devices was found to be 3.3 D for steep keratometry and 2.9 D for flat keratometry and 2.3 D for corneal astigmatism. Intra-examiner reliability was excellent in all three devices in terms of the keratometric measurements (ICC>0.99). Conclusion: Pachymetry and keratometry can be performed independently in each device with excellent reliability in CXL-treated keratoconic eyes. However, we consider that measurements made with different devices cannot be used interchangeably in these eyes.