In vivo confocal microscopy assessment of meibomian glands microstructure in patients with Graves' orbitopathy

Abstract

Background: To evaluate microstructural changes in the meibomian glands (MGs) in patients with active and inactive Graves' orbitopathy (GO), using in vivo confocal microscopy (IVCM), and to investigate the correlations between clinical and confocal findings.

Methods: Forty patients (80 eyes) with GO (34 eyes with active GO, 46 eyes with inactive GO), and 31 age- and sex-matched control participants (62 eyes) were enrolled consecutively. A researcher recorded the clinical activity score (CAS) for each patient. A complete ophthalmic examination was then performed, including external eye, ocular surface and MGs. IVCM of the MGs was performed to determine the MG acinar density (MAD), MG longest and shortest diameters (MALD and MASD), MG orifice area (MOA), MG acinar irregularity (MAI), meibum secretion reflectivity (MSR), acinar wall inhomogeneity (AWI), acinar periglandular interstices inhomogeneity (API), and severity of MG fibrosis (MF).

Results: All confocal microscopy assessments of MGs significantly differed among groups (all P = 0.000). Compared to controls, GO groups showed lower MOA (1985.82 ± 1325.30 μm² in active GO and 2021.59 ± 1367.45 μm² in inactive GO vs. 3896.63 ± 891.90 μm² in controls, all P = 0.000) and MAD (87.21 ± 32.69 /mm² in active GO and 80.72 ± 35.54 /mm² in inactive GO vs. 114.69 ± 34.90 /mm² in controls, P = 0.001 and 0.000, respectively); greater MALD (118.11 ± 30.23 μm in active GO and 120.58 ± 27.64 μm in inactive GO vs. 58.68 ± 20.28 μm in controls, all P = 0.000) and MASD (44.77 ± 19.16 μm in active GO and 46.02 ± 20.70 μm in inactive GO vs. 27.80 ± 9.90 μm in controls, all P = 0.000); and higher degrees of MAI, MSR, and MF (all P<0.05). Eyes with active GO had higher degrees of MAI (P = 0.015), AWI (P = 0.000), and API (P = 0.000), while eyes with inactive GO had higher degrees of MSR (P = 0.000) and MF (P = 0.017). In GO groups, AWI and API were positively correlated with CAS (r = 0.640, P = 0.000; r = 0.683, P = 0.000, respectively), and MF was negatively correlated with CAS (r = - 0.228, P = 0.042).

Conclusions: IVCM effectively revealed microstructural changes of MGs in eyes with GO and provided strong in vivo evidence for the roles of obstruction and inflammation in the ocular surface disease process. Furthermore, it revealed discernible patterns of MG abnormalities in eyes with active GO and inactive GO, which are not easily distinguishable by typical clinical examinations.

Keywords: Graves’ orbitopathy; In vivo confocal microscopy; Meibomian glands; Microstructure.

Fig. 1

Infrared meibography images of meibomian gland (MG) dropout. Compared with controls (A), MG dropout was higher in active (B) and inactive Graves’ orbitopathy (C), which was most pronounced in active Graves’ orbitopathy (B)

Fig. 2

In vivo confocal microscopy images of meibomian gland (MG) acinar density (MAD), MG acinar longest diameter (MALD), and MG acinar shortest diameter (MASD). The multipoint tool was used to evaluate the number of clearly visible acinar unit (asterisk). The straight-line selection tool was used to trace and measure the MALD (L) and MASD (S). Compared with controls (A), the acini were obviously dilated and fused in active (B) and inactive Graves’ orbitopathy (C)

Fig. 3

In vivo confocal microscopy images of meibomian gland orifice. The polygon tool was used to trace and measure meibomian gland orifice area (MOA). The orifice (highligted by a circle) was round, and the interior was uniform, presented low reflection in controls (A). However, the orifice was irregular in shape, with obvious blockage of high-reflective materials (arrow) in active Graves’ orbitopathy (B), and partial fibrosis (arrow) around the orifice in inactive Graves’ orbitopathy (C)

Fig. 4

In vivo confocal microscopy images of meibomian gland acinar irregularity (MAI), meibum secretion reflectivity (MSR), acinar wall inhomogeneity (AWI), and acinar periglandular interstices inhomogeneity (API). The acini were oval, and arranged neatly, with connective tissues of interstice of the acini distributing homogeneously in controls (A). However, the acini were obviously irregular in shape and arranged disorderly in active (B) and inactive Graves’ orbitopathy (C). Particularly, the reflectivity of the acinar wall, as well as the intra- and extra-space of the acini were inhomogeneous in Graves’ orbitopathy (arrows)

Fig. 5

In vivo confocal microscopy images of meibomian gland fibrosis (MF). Compared with controls (A), excessive fibrosis (arrows) could be noted in active (B) and inactive Graves’ orbitopathy (C), which was most pronounced in inactive Graves’ orbitopathy (C)