Diagnostic Capability of a New Objective Method to Assess Meibomian Gland Visibility.

Abstract

The diagnosis of dry eye disease and meibomian gland dysfunction (MGD) is challenging. Measuring meibomian gland visibility may provide an additional objective method to diagnose MGD. This study aimed to evaluate the ability of new metrics to better diagnose MGD, based on measuring meibomian gland visibility. One hundred twelve healthy volunteers (age, 48.3 ± 27.5 years) were enrolled in this study. Ocular surface parameters were measured using the Oculus Keratograph 5M (Oculus GmbH, Wetzlar). Subjects were classified according to the presence or absence of MGD. New metrics based on the visibility of the meibomian glands were calculated and later compared between groups. The diagnostic ability of ocular surface parameters and gland visibility metrics was studied through receiver operating characteristic curves. Logistic regression was used to obtain the combined receiver operating characteristic curve of the metrics with the best diagnostic ability. Statistically significant differences were found between groups for all ocular surface parameters and new gland visibility metrics, except for the first noninvasive keratograph breakup time and gland expressibility. New gland visibility metrics showed higher sensitivity and specificity than did current single metrics when their diagnostic ability was assessed without any combination. The diagnostic capability increased when gland visibility metrics were incorporated into the logistic regression analysis together with gland dropout percentage, tear meniscus height, dry eye symptoms, and lid margin abnormality score (P < .001). The combination of median pixel intensity of meibography gray values and the aforementioned ocular surface metrics achieved the highest area under the curve (0.99), along with excellent sensitivity (1.00) and specificity (0.93). New meibomian gland visibility metrics are more powerful to diagnose MGD than current single metrics and can serve as a complementary tool for supporting the diagnosis of MGD.