Abstract】Dry eye is one of the most common ocular disorders. Although a lot of progress has been made in basic and clinical research related to it in the past few years, its pathophysiological mechanisms are still not well understood. In vivo confocal microscopy is a non-invasive ocular imaging technique that makes it possible to observe ocular surface epithelial cells, immune and inflammatory cells, corneal nerves, corneal stromal cells, and lid gland structures at the cellular level. Thus, it can help to better understand the pathogenesis as well as the pathophysiology of dry eye and thus assist in the diagnosis and treatment of the disease. The use of confocal microscopy allows the evaluation of ocular surface structures in dry eye related conditions and the quantification of their alterations, which allows the disease to be identified at an early stage and facilitates the stratified treatment of patients. In addition, dynamic observation of changes in the ocular surface confocal images allows monitoring of the clinical outcome of dry eye, facilitating timely adjustment of treatment regimens and more accurate assessment of prognosis. Keywords: dry eye syndrome; in vivo confocal microscopy; ocular surface; lid gland According to the 2007 International Dry Eye Symposium, dry eye syndrome (DES) is a “multifactorial disease of the tear and ocular surface that causes discomfort, visual disturbances, tear film instability with underlying ocular surface damage; it is also associated with increased tear film osmolarity and Ocular surface inflammation” [1]. Due to the lack of uniform diagnostic tests and criteria, the exact prevalence of dry eye is not known; however, it is estimated that the prevalence of dry eye is approximately 5-30% in people over 50 years of age and will certainly continue to increase as the population ages [2]. In addition, the pathophysiological mechanisms of dry eye still need to be studied in depth, and all clinical trials of new therapies for dry eye in the United States have failed since 2002, posing a problem for ophthalmologists. The in vivo confocal microscopy (IVCM) is a new non-invasive ocular surface imaging instrument with high resolution. Its unique physical properties make it possible to observe the layers of the living cornea at the cellular level, and thus is increasingly used for the study of a variety of corneal diseases, such as epithelial lesions, various stromal degenerative and dystrophic diseases, endothelial lesions, corneal deposits, infections, and traumatic lesions [3]. Using IVCM to observe corneal epithelium, corneal stromal cells, corneal nerves, and immune and inflammatory cells within the cornea in patients with dry eye can not only further elucidate the pathophysiological mechanisms of dry eye, but also monitor the changes in the patient’s condition and thus evaluate the therapeutic effects. In this paper, we present a summary of the cellular changes in the cornea, conjunctiva, and lid gland observed in patients with dry eye using IVCM, with the aim of shedding light on future studies. I. Cornea Increased production/reduced clearance of inflammatory cytokines and proteolytic enzymes in the ocular surface of patients with dry eye affects structures such as cells and nerves in all layers of the cornea; moreover, it has been demonstrated that corneal thickness is generally significantly thinner in patients with dry eye [4]. (i) Corneal epithelium Under normal conditions, the corneal epithelium plays an important role in ocular surface homeostasis as a barrier against pathogens and other harmful substances; however, in patients with dry eye, the microstructure of the corneal epithelium is significantly altered. The morphology of corneal surface epithelial cells in patients with sjogren’s syndrome (SS) dry eye is irregular with patchy changes as observed by IVCM [5]; and Chen et al [6] found that the mean area of surface corneal epithelial opaque cells was significantly increased in patients with dry eye (P<0.0001< span="">), and the area of opaque cells area correlated significantly with clinical dry eye indicators, such as symptoms of blurred vision (r=0.86, P=0.0001), conjunctival lissamine green staining score (r=0.4, P=0.026), and corneal fluorescein staining score (r=0.5, P=0.002). In addition, some patients with dry eye had positive corneal epithelial fluorescein staining, a sign that was confirmed by IVCM observations to be due to increased uptake of fluorescein by surface corneal epithelial cells due to abnormalities such as impaired tight junction integrity, increased epithelial permeability, and cell death, rather than staining at corneal epithelial defects as previously thought [7]. Quantitative observation of the corneal epithelium using IVCM revealed that the density of cells in the various layers of the corneal epithelium may change in patients with dry eye [8-11]: it is generally accepted that the density of cells in the surface and intermediate epithelium is significantly reduced in affected eyes, but the density of cells in the basal epithelial layer is controversial, with different studies showing that it may increase or decrease, or may not change significantly. (ii) Corneal stroma Some IVCM observational studies of the cornea have found that in addition to microstructural changes in the corneal epithelium, morphological and quantitative changes in corneal stromal cells occur in patients with dry eye. Subsequent studies have shown a significant increase in abnormal hyper-reflective cells in the corneal stroma of patients with hyperthyroidism-related dry eye [12] and rheumatoid arthritis with SS [13] compared to healthy controls; although many researchers have suggested that these hyper-reflective cells are corneal stromal cells in an “activated” state [14], it remains to be confirmed whether these cells may be of bone marrow mesenchymal origin. In addition, Benítez del Castillo et al [8] suggested that the cell density of the anterior corneal stroma was significantly higher in patients with SS-associated dry eye (1348±220/mm2) than in patients without SS-associated dry eye (1183±273/mm2) and in the N<60 years (1107±210/mm2) and N≥60 years groups (1075±201 /mm2); however, in terms of posterior stromal layer cell density, there were no statistically significant differences between the groups of SS-associated dry eye patients (808 ± 117/mm2), non-SS-associated dry eye patients (795 ± 150/mm2), N < 60 years (741 ± 142/mm2), and N ≥ 60 years (768 ± 119/mm2). Similarly, a study by Villani et al [13] suggested that the anterior and posterior corneal stromal layer cell densities were significantly higher in the rheumatoid arthritis with or without SS group than in the healthy control patients (p < 0.001). (iii) Corneal nerves Since corneal nerves play an important role in regulating corneal sensation, epithelial integrity, cell proliferation, and wound healing, they have received a lot of attention from ophthalmologists. Some studies have shown decreased corneal sensitivity in dry eye patients [15,16], but others have found corneal sensory hypersensitivity in dry eyes [17,18]. Similarly, studies on the effects of dry eye on subbasal nerve density have yielded mixed results. Some investigators observed significantly lower nerve density in the subbasal layer of the cornea in both SS and non-SS associated dry eye than in controls [10,16,18]. In contrast, other studies have not found changes in subbasal nerve density [5,15], and there are even studies demonstrating increased corneal nerve density in SS patients [19]. These studies on the correlation between corneal sensitivity and subbasal nerve density may have yielded different results because of the different grading and severity of dry eye disease in the patients included in each study [20]. Nevertheless, in terms of morphological changes of corneal nerves in patients with dry eye, all studies suggest increased curvature and reflectivity of subbasal nerves [10]. In addition, some investigators have found a significant increase in the number of nerve bead formation in the subbasal layer of the cornea in patients with SS and non-SS associated dry eye. These special structures may be nerve fibers containing metabolically active transmitters, which contribute to the improvement of nutritional abnormalities in the corneal epithelium [5,8]; or the formation of nerve beads may represent nerve damage and thus the need to secrete nerve growth factors to promote repair through an inflammatory response [10]. (iv) Corneal immune and inflammatory cells Langerhans cells (LCs) are epithelial dendritic cells (DCs), which are dedicated antigen-presenting cells of the cornea. They control inflammation by inducing tolerance or activating T lymphocytes and recruiting other immune and inflammatory cells to the cornea to fight pathogens and are an important component of immunity. The distribution of epithelial DCs in normal subjects decreases gradually from the periphery to the center of the cornea, where they are generally located near the subbasal plexus [21,22].Lin et al [22] used IVCM to observe the density and distribution of DCs and other immune and inflammatory cells in the corneas of patients with dry eye. They found that the density of DCs in the central cornea was 127.9 ± 23.7/mm2 and 89.8 ± 10.8/mm2 in SS and non-SS-associated dry eye patients, respectively, both of which were significantly higher than in normal subjects (34.9 ± 5.7/mm2) (P < 0.05); in addition, they demonstrated that the density of DCs in the peripheral cornea was significantly higher in SS patients (157.2 ± 29.7/mm2) (P < 0.05), whereas the group of patients with non-SS-associated dry eye (106.9 ± 10.5/mm2) showed only a mild increase (P > 0.05) compared with normal controls (90.7 ± 8.2/mm2). They also found few non-dendritic leukocytes among the corneas of healthy subjects (1.6±0.6/mm2 centrally and 4.3±1.3/mm2 peripherally); however, non-dendritic leukocytes were significantly increased in both the central and peripheral corneas of SS patients, 49.0±12.9/mm2 and 84.2±36.8/mm2, respectively; whereas the central corneas of patients with non-SS-associated dry eye The leukocyte density in the central cornea of non-SS-associated dry eye patients was only mildly increased (4.6±1.0/mm2), but there was no significant change in the periphery (8.4±3.1/mm2). These findings demonstrate that the density of epithelial DCs and other inflammatory cells in the central and peripheral cornea is increased in SS and non-SS-associated dry eye patients, especially in SS patients; this also suggests that dry eye can be considered as a localized chronic inflammation, where an imbalance between protective immune regulatory mechanisms and pro-inflammatory responses in the ocular surface further leads to tissue damage [23]. II. Conjunctiva The goblet cells (GCs) within the conjunctival epithelium secrete mucus to moisten the cornea and conjunctiva and provide protection, in addition to the presence of paracrine lacrimal glands in the lid conjunctiva, making it clinically relevant to study the conjunctiva in patients with dry eye by IVCM. Conjunctival GCs appear as larger, highly reflective oval cells with relatively uniform brightness under IVCM and are therefore easier to identify [22]. Quantitative observation of other cells in the conjunctiva revealed that the density of conjunctival inflammatory cell infiltration was significantly higher in SS and non-SS-associated dry eye patients than in controls (P < 0.001), and the density of conjunctival inflammatory cells was negatively correlated with tear film stability and tear volume and positively correlated with ocular surface staining score; however, the density of conjunctival epithelial cells was significantly lower in SS and non-SS-associated dry eye patients than in controls (P <0.05); in addition, there were significantly more conjunctival epithelial vesicles in SS patients, and these structures corresponded to areas of positive conjunctival Tiger Red staining [25].Kojima et al [26] also found that the values of conjunctival epithelial cell nucleoplasm ratio and mean individual epithelial cell area in conjunctival blot cytology correlated well with the corresponding index measurements of IVCM, so they concluded that IVCM can be a useful non-invasive tool for evaluating conjunctival epithelial cell alterations in patients with dry eye. The quality and quantity of lipid secretion in the meibomian gland dysfunction (MGD) is altered, resulting in lipid-deficient dry eye. In patients with dry eye, the structure and quantity of the meibomian gland were found to be altered to varying degrees using IVCM. The mean glandular follicle diameters in the primary SS (SSI) and secondary SS (SSII) groups were 53±31um and 70±42um, respectively, which were not significantly different from those in the control group (53±14um); however, the lid gland opening diameters in SSI and SSII patients were 27.8±5.9um and 20.6±5.1um, respectively, both of which were significantly smaller than those in the control group (34.7±4.3um); in addition The density of the lenticular vesicles was significantly higher in SSI patients (138±69/mm2), but significantly lower in SSII patients (97±43/mm2); and in terms of the reflectivity of the ocular surface lipid layer (classified into grades 1-4), it was 1.7±0.6 and 2.2±0.8 in the SSI and SSII groups, respectively, both of which were significantly higher than in the control group (1.1±0.7). In addition to the above findings, IVCM imaging of the lid glands of MGD patients has also been performed, and an analytical study revealed that dysfunctional lid glands exhibit marked inhomogeneous changes in the periglandular mesenchyme, dilatation of the glandular vesicle wall caused by concentrated lipid secretion and glandular atrophy with periglandular fibrosis, hyperkeratosis of the glandular duct epithelium, and extensive infiltration of periglandular inflammatory cells [28,29]. Therefore, it has been suggested that periglandular inflammatory cell counts can be used as an effective indicator for monitoring MGD treatment [30]. IV.DISCUSSION AND CONCLUSION Recent developments in IVCM have made possible the real-time study of living cell morphology and the evaluation of various cells of the ocular surface at microscopic resolution. This optical tomography technique allows the observation and analysis of fine structures such as epithelial cells, corneal stromal cells, corneal nerves, endothelial cells, cupped cells, lid glands, and immune and inflammatory cells that were previously invisible with slit lamp examination. Thus, it can be used not only for diagnosis but also as a useful tool for monitoring the condition and measuring the effectiveness of treatment in patients with dry eye. The changes in the corneal epithelium of patients with dry eye observed by IVCM are mainly due to the exfoliation of the surface cell layer and the action of inflammatory mediators [31]. Corneal epithelial damage may be associated with increased tear film osmolarity due to excessive tear evaporation, which may lead to morphological and pro-inflammatory changes in the corneal epithelium, as well as disruption of ocular surface homeostasis [11]. IVCM studies have demonstrated significant changes in the density of subbasal nerves in patients with dry eye and abnormal morphology, such as bead-like structure formation, budding, distortion, and irregular branching. outgrowth, twisting, irregular branching forms, and increased neuroma-like structures, which can be explained by degenerative changes and regeneration of the nerve [5]. In the last few years, the role of inflammation in dry eye has been gradually recognized, and the increased density of epithelial DCs observed in IVCM may suggest an elevated corneal immune status in patients with dry eye [22]. Therefore, dynamic evaluation of the density of inflammatory cells in the central cornea can be used to grade the severity of dry eye and assist in the evaluation of the efficacy of anti-inflammatory drugs, thus guiding clinical treatment. According to the International Symposium on Meibomian Gland Dysfunction 2011 definition [32], MGD is primarily caused by terminal ductal keratosis and obstruction; meibomian gland obstruction may further result in a cystic dilatation of the gland, glandular cell atrophy, glandular reduction, and a hypersecretory state, but generally without an inflammatory response. However, IVCM images reveal not only altered glandular morphology but also extensive periglandular inflammatory cell infiltration in patients with MGD [29]; the clinical significance of this phenomenon remains to be clarified. Although corneal morphological alterations in patients with dry eye have been extensively studied using IVCM, few studies have been performed on conjunctival alterations in the presence of dry eye. Studies of IVCM on the conjunctiva are still in the beginning stages, and more adequate experimental results are urgently needed. In conclusion, IVCM is a non-invasive test that not only provides diagnostic information related to dry eye, which can be used to grade the degree of dry eye patients and thus better assist in treatment, but also evaluates the effectiveness of treatment and more accurately determines the prognosis of the disease. Therefore, the clinical value of IVCM should be valued and explored through further in-depth studies.