A novel noninvasive ocular surface analyzer for the assessment of dry eye with Meibomian gland dysfunction

Meibomian gland dysfunction (MGD) is a major cause of dry eye, and the purpose of the present study was to evaluate the differences between dry eye patients with MGD and controls using a novel noninvasive ocular surface analyzer. A total of 33 dry eye patients with MGD and 31 controls were enrolled in the present study. Testing included administration of the Ocular Surface Disease Index (OSDI), followed by Keratogragh 5M (Oculus, Wetzlar, Germany), measurements of the noninvasive tear film break-up times (NITBUTs), the first tear film break-up point and the morphology of meibomian gland. Meibomian gland loss (MGL) on the upper lid was evaluated using noncontact meibography. The first NITBUT (NITBUTf) was significantly shorter than the average NITBUT (NITBUTav; Mann-Whitney U-test, P< 0.01). There was a weak negative Spearman correlation between NITBUTf and OSDI (rs= -0.251, P= 0.046) as well as between NITBUTav and OSDI (rs= -0.250, P= 0.046). Furthermore, MGL showed a statistically significant Spearman correlation with OSDI (rs= 0.562; P< 0.01). In the MGD group, NITBUTf was significantly shorter (P< 0.01), and MGL and OSDI score were significantly greater (P< 0.01 and < 0.01, respectively) than in the control group. In the MGD group, the first tear film break-up point was most frequently located in the inferonasal quadrant, while it was most frequently located in the supertemporal quadrant in the control group, and there was a significant difference with this regard (chi(2)= 3.937, P= 0.047). In the MGD group, there were significantly more breakups in the lower quadrant than in the upper quadrant (P= 0.011), whereas no differences in the number of breakups in each quadrant were observed in the control group (P= 0.429). Furthermore, there was no significant difference in the breakups in the lower quadrants of the two groups (chi(2)= 0.525, P= 0.469). In the MGD group, there were significantly more breakups in the lower quadrant than in the upper quadrant (P= 0.011), whereas no differences in the number of breakups in each quadrant were observed in the control group (P= 0.429). Furthermore, there was no significant difference in the breakups in the lower quadrants of the two groups (chi(2)= 0.525, P= 0.469). As the NITBUTs showed only a weak correlation with OSDI, they may not be a strong indicator of the OSDI. However, MGL indicated changes of meibomian gland morphology in the MGD group and the first tear film break-up point was likely located in the inferonasal quadrant in the MGD group.