Lid Margin Score Is the Strongest Predictor of Meibomian Area Loss

PURPOSE: Although meibography provides direct evidence of gland dropout in meibomian gland dysfunction, this specialized technique is not available in most clinics. The primary aim was to determine which clinical ocular marker was most related to meibomian area loss. A secondary aim was to determine associations with confocal microscopy imaging of the lid margin. METHODS: One hundred participants from age 18 to 65 years were recruited. Measurements of the right eye and its upper eyelid, where relevant, included noninvasive tear break-up time, bulbar and limbal redness scores, blepharitis score, lipid layer thickness, number of parallel conjunctival folds, tear osmolarity, corneal fluorescein staining, phenol red thread test, lid margin score, meibography, and in vivo confocal microscopy. Participants also completed the Ocular Surface Disease Index questionnaire. The relationships between the measurements were determined using the Spearman correlation. The receiver operating characteristic curve and area under the receiver operating characteristic curve were used to determine the cutoff value of clinical markers. RESULTS: Significant correlations were found between meibomian area and lid margin score (r = -0.47, P < 0.01), and meibomian tortuosity and lid signs of blepharitis (r = -0.32, P < 0.01). Area under the receiver operating characteristic curve analysis revealed that a lid margin score of ≥1.70 detected meibomian area loss with a sensitivity of 0.58 and a specificity of 0.86. There were significant correlations between meibomian area and orifice area at 30 μm depth (r = -0.25, P = 0.01). CONCLUSIONS: The lid margin score was most related to the meibomian area and thus the best predictor of undiagnosed meibomian area loss.