Abstract
Purpose :
Measuring contrast between Meibomian glands (MGs) and the background on meibography can be helpful in monitoring longitudinal changes induced by intervention (e.g., isotretinoin), but currently, there is no method in place to do so. This study aimed to develop a method to measure contrast and to assess its repeatability and agreement with measurements taken under various conditions.
Methods :
Meibography of upper and lower eyelids from both eyes of study participants were captured at two separate visits under four conditions: face centered (C), face turned left (L), face turned right (R), and face centered with room lights off (CLO). Using ImageJ, mean pixel intensity (greyscale: 0-255) was measured of lines drawn along the central 5 MGs (Fig 1a) and along the background regions between the MGs (Fig 1b). The difference between mean intensity along the glands MGs (AvgIntG) and along background (AvgIntB) was defined as contrast (AvgIntG-AvgIntB). Repeatability of measurements at 2 visits and limits of agreement (LOA) between positions C and L, C and R, and C and CLO were determined using Bland-Altman plots.
Results :
Repeatability, assessed with 40 pairs of meibography images with contrast mean(SD) of 46.8(10.8) and range between 24.9 to 76.7, yielded a coefficient of repeatability of 13.2. On average, contrast for C was: 2.1 (95% CI: -0.7, 4.8) greater than for L, with LOA between -9.5 (95% CI: -14.3, -4.8) and 13.7 (95% CI: 9.0, 18.4); 1.4 (95% CI: -1.1, 4.0) greater than for R, with LOA between -9.4 (95% CI: -13.7, -5.0) and 12.2 (95% CI: 7.8, 16.6); and 0.1 (95% CI: -2.0, 2.2) greater than for NCL, with LOA between -8.8 (95% CI: -12.4, -5.2) and 9.0 (95% CI: 5.5, 12.7).
Conclusions :
Contrast appears to be a repeatable, objective measure for identifying subtle, clinically undetectable changes in MGs. Differences greater than ~15 are less likely due to head positioning or room conditions and more likely due to a change in MG condition. It is unclear what decreased contrast in meibography represents, but data from isotretinoin patients suggest that it may represent decreased meibum production (Fig 2).
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.