Assessment of lower tear meniscus measurements obtained with Keratograph and agreement with Fourier-domain optical-coherence tomography

Abstract

Aims: To compare intraobserver repeatability and interobserver and intersession reproducibility of lower tear meniscus height measurements (LTMH) obtained with the new Keratograph 5M and a Fourier-domain optical-coherence tomography (OCT; Spectralis OCT) and to assess the agreement between the two devices.

Methods: This is a observational cross-sectional study. Thirty eyes of 30 normal subjects were randomly imaged twice with both devices in our setting. Two examiners randomly made LTMH measurements using the devices. To assess intraobserver repeatability and interobserver and intersession reproducibility within-subject SD (Sw), test-retest repeatability, coefficient of variation (CoV) and intraclass correlation coefficients (ICCs) were calculated. Agreement between both devices was also determined.

Results: Average LMTH for Keratograph was 235.46±57.61 μm, whereas the mean LTMH measured by OCT was 245.84±67.18 μm. Intraobserver repeatability and interobserver and intersession reproducibility were moderate with Keratograph (CoV ≥0.16%, ≥0.19%, ≥0.17% and ICC ≤0.83, ≤0.76, ≤0.83, respectively) and high for the Spectralis OCT (CoV ≤0.13%, ≤0.12%, ≤0.11% and ICC ≥0.9, ≥0.92, ≥0.92, respectively). No statistically significant difference was detected in mean LMTH measurements made with both devices, although correlation was low (CoV 0.25%, ICC 0.55).

Conclusions: Spectralis OCT LTMH measurements were more reliable than the new Keratograph data. Agreement between the devices was poor.