Intra-observer and inter-observer repeatability of ocular surface interferometer in measuring lipid layer thickness

BACKGROUND Tear lipid morphology is important for normal tear function. Recently, there have been clinical studies using interferometry to assess lipid layer thickness (LLT). The aim of the study is to examine the repeatability of a commercially available interferometer. METHODS Two observers measured LLT in twenty Asian subjects (20 eyes) using an interferometer (LipiView® ocular surface interferometer, TearScience Inc, Morrisville, NC). Dry eye symptoms, tear break up time (TBUT) and corneal fluorescein staining were also prospectively evaluated. RESULTS Data for 20 participants are presented for either right or left eye (randomly selected). The mean LLT ± standard deviation of these participants was 53.53 ± 14.59 nm. When a single observer repeated the imaging on the same day, the coefficient of repeatability was 16 nm and the 95% limits of agreement were between -11 nm and 18 nm. When a different observer repeated the scan, the coefficient of repeatability was 13 nm and limits of agreement were -9 nm and 16 nm. LLT was not significantly associated with TBUT, presence of any corneal staining in any corneal zones, or symptomatic status. CONCLUSION With the repeatability of measurements being known, the significance of LLT changes measured by this interferometer may be better interpreted. In this small Asian study, the LLT was lower than previously reported studies.