In this study, to determine the correlation between meibomian gland loss (dropout) and meibum quality, we evaluated the correlation between meibomian gland dropout in the central eight meibomian glands of the eyelid and meibum quality in the same central eight meibomian glands of the eyelid as accurately as possible. Furthermore, we analyzed the direct correlation between meibomian gland dropout and meibum quality by analyzing 31 eyes of 23 patients with focal dropout (loss of 1 or 2 adjacent meibomian glands) in meibography. When squeezing the eyelid in the region of the focal dropout, we recorded the meibomian glands and the secretion with an infrared camera. This procedure allowed us to grade the meibum secretion from the orifices of 1 or 2 adjacent meibomian glands.
Analysis of the central eight meibomian glands of the eyelids in 68 eyes revealed no significant correlation between meibomian gland dropout and meibum quality in the upper and lower eyelids (Tables 2 and 3). This is similar to what we often observe in the clinic. On the basis of this finding alone, therefore, clinicians may devaluate the clinical value of infrared meibography. The lack of a significant correlation between these 2 measurements, however, may be explained as follows. First, we evaluated only the central eight meibomian glands of the eyelid to assess the correlation as accurately as possible. Even within this specific region, dropout appears differently depending on the location, and the quality of the meibum secreted by each orifice also varies depending on the location. For example, in an eyelid with less than one-third dropout (meiboscore 1) at the temporal side, but clear meibum secreted from the middle and nasal side, the overall meibum quality would be graded ‘Clear liquid secretion’ even if the meiboscore is 1. Second, when there is no dropout, the meibum quality will be normal if there is no MGD. If the MGD is hypersecretory, however, meibum secretion will be increased and the meibum may be clear or yellow with or without increased viscosity.[1] If there is obstructive MGD, either meibum with toothpaste consistency or no meibum at all will be secreted.[1] As such, when there is no dropout, the meibum quality will vary depending on the state of the meibomian glands. Third, even if there is dropout, if a part of the meibomian gland remains near the orifices (shortening), the meibum will exhibit various qualities.
Analysis of the direct correlation between meibomian gland dropout and meibum quality in 31 eyes of 23 patients with focal dropout (i.e., loss of 1 or 2 adjacent meibomian glands) in meibography revealed that, among the 31 eyelids, 26 (84%) did not secrete meibum from the orifice(s), 4 eyes showed a normal volume (just covers orifices) with cloudy liquid, and 1 eye showed a volume 2 to 3 times greater than normal with an inspissated consistency. That is, in meibomian glands with dropout, there was little to no secretion of meibum from the meibomian gland.
Eom et al. reported that meibomian gland dropout observed in meibography correlated with the quality of the meibum secreted after eyelid expression in 26 eyes with obstructive MGD.[12] The meibum quality was observed in 8 meibomian glands located in the middle third of the upper and lower eyelids, whereas infrared meibography was used to measure the meibomian gland dropout area for the middle two-thirds, and the ratio was calculated. They reported that the greater the loss of the meibomian glands in both the upper and lower eyelids, the higher the value of the meibum quality grade (i.e., increased viscosity or toothpaste consistency). One difference between our study and that of Eom et al. is that we selected the same central eight meibomian glands for both measurements, and not the middle two-thirds region for infrared meibography and the central eight meibomian glands for meibum quality. Because the dropout is not uniform across the entire eyelid and the quality of the meibum secreted by squeezing is not uniform across the entire eyelid, it is preferable to analyze the same small area. We did our best to accurately analyze the direct correlation between dropout and meibum quality by choosing same central eight meibomian glands. Second, we analyzed the direct correlation between meibomian gland dropout and meibum quality in 23 patients with focal dropout in meibography. Third, unlike Eom et al., our results did not demonstrate a significant correlation between dropout and meibum quality in 68 eyes of 68 patients. Our findings are consistent with our observations in the clinic. Fourth, Eom et al. analyzed 26 eyes of patients with obstructive MGD, and we analyzed 99 eyes of 91 patients who visited our clinic for dry eye disease examination.
Our findings will facilitate interpretation of the lipid layer thickness measurement. The difference is that lipid layer thickness measured by a tearfilm interferometer is physiologic, unlike the meibum secreted by eyelid squeezing. In the present study, when each meibomian gland was examined in more detail, we observed that meibomian glands with dropout secreted little to no meibum, while when observed macroscopically, there appeared to be no relationship between meibomian gland dropout and meibum quality. The lipid layer thickness is the result of meibum secreted from all meibomian glands in the upper and lower eyelids. Therefore, meibomian gland dropout and lipid layer thickness are likely to appear unrelated.
Treatment should be determined according to the MGD type. In the case of obstructive MGD with little dropout and little meibum, thermal pulsation treatment is required.[13] In the case of hypersecretory MGD, in which there is little meibomian gland dropout and excessive secretion of meibum, treatment with systemic antibiotics such as tetracycline, doxycycline, and minocycline is necessary.[14] In cases with little dropout, normal meibum quality and quantity, and a dominant lack of aqueous tears, it is important to increase tear volume with artificial tears, punctal plug insertion, or diquafosol. For hyposecretory MGD with severe dropout and little to no meibum secretion, LipiFlow® therapy has little effect[15] and lipid-containing eye drops are required. Therefore, infrared meibography is one of the most important tests for selecting the appropriate treatment for MGD or dry eye disease.
We found that meibomian glands are more often obstructed in the lower eyelids than in the upper eyelids. Table 1 shows that the upper eyelids had a higher meiboscore, but the lower eyelids had a lower meibum quality grade (inspissated/toothpaste consistency or no meibum). Therefore, the obstructive type of MGD is considered to be more prevalent in the lower eyelid. Eom et al. reported similar results in their study, in which the quality of expressed meibum was assessed on a scale of 0 to 3 for each gland: 0 = clear; 1 = cloudy; 2 = cloudy with debris; 3 = inspissated (toothpaste-like).[12] Complete obstruction without meibum expression when applying firm digital pressure was graded as a 3. Meibum quality was assessed in 8 glands in the central third area of the upper and lower eyelids (total score range, 0–24). The mean expressed meibum grade (± SD) of the lower eyelids was 16.5 ± 5.1, which was significantly greater than that of the upper eyelids (11.2 ± 5.2).[12] In another study, Eom et al. reported that the mean number of glands expressible among the central 8 glands in the upper eyelids was 3.9 ± 2.6, significantly higher than that for the lower eyelids (2.2 ± 2.4, p < 0.001).[16] The upper eyelids move more than the lower eyelids during blinking. Therefore, meibum in the upper eyelids appears to be more easily and continuously secreted by the mechanical action of the pretarsal orbicularis muscle, making it less clogged than the lower eyelids.[17, 18]
We performed the eyelid squeezing last because it is an invasive procedure. Eyelid manipulation during meibography may inadvertently lead to the expression of meibum and confound the meibum quality assessment. Because we applied firm pressure for eyelid squeezing instead of gentle pressure, however, we do not believe that eyelid manipulation during meibography prior to eyelid squeezing confounded the observer’s ability to assess the meibum quality.