Application of Keratograph and Fourier-Domain Optical Coherence Tomography in Measurements of Tear Meniscus Height.

Abstract

To compare the interoperator repeatability of tear meniscus height (TMH) measurements obtained with a keratograph and Fourier-domain optical coherence tomography (FD-OCT) and to assess the agreement between the methods.Forty-seven eyes with DED and 41 healthy eyes were analyzed using the Schirmer test I and tear breakup time test (TBUT). The TMH was measured three times with each device. The repeatability of measurements was assessed by within-subject standard deviation (Sw), repeatability (2.77 Sw), coefficient of variation (CoV) and intraclass correlation coefficient (ICC). Efficacy in detecting DED was evaluated in terms of the area under the curve (AUC). The TMHs obtained with the keratograph were 0.03 mm lower than those obtained with FD-OCT in both groups (p < 0.001 for the DED group and p = 0.0143 for the control group, respectively). The intraexaminerICCs of the keratographic TMH were 0.789 and 0.817 for the DED and control groups, respectively, and those of the FD-OCT TMH were 0.859 and 0.845, respectively. Although a close correlation was found between the TMHs measured with the keratograph and FD-OCT by the Spearman analysis in both groups (both p < 0.001), poor agreement between the devices was shown in both groups using a Bland–Altman plot. The AUCs of the keratography and FD-OCT results were 0.971 (p < 0.001) and 0.923 (p < 0.001), respectively. Both devices had excellent diagnostic accuracy in differentiating normal patients from DED patients. FD-OCT TMH measurements were more reliable than the keratograph data in the DED group. Agreement between the devices was poor in both groups.