When lipid-deficient dry eye is suspected, clinicians typically examine meibomian gland atrophy and meibum quality to aid in the diagnosis, although lack of meibography image contrast may also be a strong indicator, according to a pilot study published in Optometry and Vision Science. Researchers were alerted to the impact of image contrast after finding that patients taking a course of isotretinoin for other conditions exhibited gland fading rather than shortening.

The retrospective analysis was conducted at the School of Optometry, University of California, Berkeley, with data and images from Berkeley’s Ocular Surface Study database. Of 345 potential patient charts from 2012 to 2018, 13 eyes of 12 patients met criteria for lipid-deficient dry eye. For the control group, 31 eyes of 22 patients were included.

An infrared light meiography device captured high-contrast infrared meibo-scans. The difference between mean pixel intensity — grayscale 0 to 255 — of meibomian glands (MGs) standing out from a darker background was defined as contrast units. Results demonstrated that MG image contrast was a meaningful clinical indicator for lipid-deficient dry eye, with “good sensitivity and excellent specificity.”

Continue Reading

Participants with lipid-deficient dry eye displayed an average of 11.9 units lower MG contrast than controls (P <.001), but not significantly different Schirmer I tear test outcomes (P <.77). In dry eye cases, there was a 9.5-second shorter fluorescein tear break-up time than for controls (P <.001), and 7.7 units of decreased meibum quantity (P =.02). Lower meibum quality of 7.6 units was found in dry eye cases, as compared with control patients (P =.04).

Importantly, data showed 18.9% higher gland atrophy in those with lipid-deficient dry eye, but after adjusting for mean age variations between the cohorts, “percent meibomian gland atrophy and meibum quantity were no longer different between groups.” After controlling for age, meibum quality expressed higher significance (P =.001), and MG contrast remained significant (P =.004).

Investigators examined decreased contrast as a classifier for this type of dry eye. The cut-off point of 28.3 contrast intensity units resulted in best possible patient classification, with 85.3% accuracy, sensitivity of 0.67, and specificity of 0.95. Using the 28.3 cut-off, 73% of eyes with dry eye demonstrated low-contrast meibography, and 85% of control eyes exhibited high-contrast images. Thus, participants with less than 28.3 MG contrast had a 14.9 times greater chance of lipid-deficient dry eye diagnosis than those with high contrast of more than 28.3, the data shows.

Previous studies have not yet rated contrast’s impact, mostly due to diverse measurement instruments. Limitations of this study include a small sample, and use of 1 meibography tool, restricting generalizability of results. Researchers suggest clinicians still evaluate MG atrophy in diagnosing dry eye, but propose that “meibomian gland contrast should also be considered, particularly for symptomatic patients with minimal percent meibomian gland atrophy.”

Disclosure: One study author declared an affiliation with the biotech or pharmaceutical industry. Please see the original reference for a full list of authors’ disclosures.


Yeh TN, Lin MC. Meibomian Gland Contrast Sensitivity and Specificity in the Diagnosis of Lipid-deficient Dry Eye: A Pilot StudyOptom Vis Sci. 2021; 98(2): 121-126. doi: 10.1097/OPX.0000000000001636