In-vivo confocal microscopy (IVCM) allows in-vivo evaluation of ocular surface and provides information on the morphology features at the cellular level [1, 2]. Using laser light at a wavelength of 670 nm, well-contrast and high-quality images can be obtained [3]. The enface monochrome images at a magnification of approximately 600 times, covering a field of 0.16 mm2, with transversal optical resolution of 1 µm and longitudinal optical resolution of 4 µm [4]. Over the past three decades, it has been well-recognized that many ocular and systemic diseases are associated with varying degrees of conjunctival alterations, which can significantly impact the quality of daily life of those affected [5]. However, a significant proportion of substructural changes of the conjunctiva are overlooked because they cannot be observed with a slit-lamp biomicroscope. In addition, sampling of the conjunctiva by traditional histopathological methods, such as impression cytology, can cause unavoidable irritation. Repeated conjunctival sampling further exacerbates the vulnerability of the ocular surface. IVCM overcomes these disadvantages of traditional histopathological examination. It demonstrates objective evidence of structural changes and provides quantifiable data to help clinicians grade the disease severity and monitor disease progression [6]. It also allows for assessing therapeutic efficacy [7] or early detecting the side effects of drugs or surgeries [8]. Several types of confocal microscopes with different principles have been introduced for use in ophthalmology [4], including the tandem scanning confocal microscope, slit scanning confocal microscope, and laser scanning confocal microscope. Tandem scanning confocal microscope was based on the modified Nipkow spinning disc technology, which used a metal disc with multiple pinholes of 30 µm, allowing for true-color, real-time imaging. However, the system had limited light throughput, resulting in relatively low image quality and contrast [9]. The system is no longer commercially available. Slit scanning confocal microscope allows improved light output and faster acquisition time, but at the cost of axial resolution that ranged from 8 to 25 µm, in comparison to 9 to 12 µm in tandem scanning confocal microscope [10]. The laser scanning confocal microscope is currently the most advanced commercially available design. It offers a magnification of 800 times, providing a greater contrast than tandem or slit scanning confocal microscope with a lateral resolution of 1 µm and an axial resolution of 4 µm. Current laser scanning confocal microscopes used in clinical settings include the Heidelberg Retinal Tomograph (HRT), which will be the focus in our review. In addition, all IVCM examinations in this review were performed using the Heidelberg Retinal Tomograph 3 (HRT 3) Rostock Cornea Module (Heidelberg Engineering GmbH, Heidelberg, Germany).

In this review, we summarize and discuss current knowledge about the applications of IVCM on conjunctival diseases, covering conjunctival aging changes, dry eye disease, glaucoma and glaucoma treatments, conjunctival neoplasms, pterygium, allergic conjunctivitis, trachoma and trachomatous scarring, and conjunctiva-associated lymphoid tissue (CALT) changes in ocular diseases.

IVCM allows clear visualization of the conjunctiva from the superficial layer to the conjunctival substantia propria at a depth of approximately 130 µm [11]. Superficial epithelial cells of conjunctiva appear as 10–15 μm sized irregularly-shaped cells, with oval-shaped prominent nuclei, and sometimes hyperreflective desquamation. Intermediate epithelial cells are tightly-arranged cells with punctate, hyperreflective nuclei. Basal epithelial cells have a polygonal shape with hyperreflective cell borders, and the small bright nuclei are visible or absent (Fig. 1a–d). Goblet cells (GCs) are as highly hyperreflective, ovoid cells of uniform brightness with low reflectivity nuclei, typically two to three times larger than the surrounding epithelial cells (Fig. 1e). Dendritic cells (DCs) are typically observed as hyperreflective bodies with dendritic processes distributed in conjunctival epithelium (Fig. 1f). Non-typical morphologies of DCs such as lacking dendrites, long dendrites, or a wire-mesh pattern with long intertwined dendrites may be observed [11, 12]. The conjunctival microcysts can be seen in the intermediate epithelial layer, basal epithelial layer, or subepithelial sites. The microcysts are giant, round, or oval-shaped structures, commonly containing hyperreflective contents and surrounded by hyporeflective rings. With different scanning planes, they can manifest as small granular, highly reflective structures or even vacuoles. The conjunctival lamina propria is located beneath the basement membrane and consists of a complex network of hyperreflective overlapping fibers, and blood vessels with rich immune cells (Fig. 1g, h). The CALT diffuse lymphoid layer is composed of scattered or clustered distributed hyperreflective lymphocytes. CALT lymphoid follicles are well-defined, round-shaped structures that host hyperreflective lymphocytes within a collagenous network [13].

Representative in-vivo confocal microscopy (IVCM) images from normal human conjunctiva. a Superficial epithelial cells: large irregular-shaped cells with prominent oval-shaped nuclei. b Superficial epithelial cells with hyper-reflective desquamation (arrow). c Intermediate epithelial cells manifested as tightly arranged cells with punctate and hyper-reflective nuclei. d Basal epithelial cells: polygonal shaped cells with hyper-reflective cell borders and the small bright nuclei visible or not. e Goblet cells: large hyper-reflective, ovoid cells of uniform brightness with hypo-reflectivity nuclei. f Dendritic cells: hyper-reflective cell bodies, with long dendrites (arrows) or lacking dendrites (arrowhead). g Conjunctival lamina propria: a network of hyper-reflective overlapping fibers. h Conjunctival lamina propria: blood vessels with hyper-reflective immune cells (arrows)

Aging is a systemic process, and eyes are no exception to age-related changes. Several ocular diseases are age-related, such as cataract [14], age-related macular degeneration, corneal nerve health [15], and conjunctiva-related disorders such as dry eye and conjunctivochalasis, which commonly coexist in the elderly [16, 17]. IVCM helps to better understand conjunctival changes with age.

Previous studies have reported a decreasing trend in the circularity, size and density of the conjunctival epithelial cells in age-matched healthy individuals although there were no significant differences [11, 15]. However, electron microscopy revealed more pronounced morphological differences, showing a flatter and more elongated shape of the epithelial cells in people over 80 years old compared with the homogeneous polygonal shape of the superficial cells in people aged 50 to 79 years [18]. This may be due to differences in the sensitivity of various methods to detect the ultrastructural structures. Electron microscopy reveals changes in the intercellular space, which was thought to change with age [18].

GCs cluster or scatter throughout the superficial layers of the conjunctival epithelium on IVCM scans. It is generally accepted that the density of GCs continues to increase in early childhood and then remains at a stable level [19] in the healthy population. GCs dysfunction is associated with several age-related ocular surface diseases. However, debate continues about the relationship between GCs density and aging. Zhu et al. [11] evaluated the GCs in the bulbar conjunctiva in different age groups and reported no differences in cell morphology or density among different age groups. Therefore, they hold the view that the changes in GCs were not age-dependent, which is in agreement with the findings of a study on conjunctival biopsy [20]. Nevertheless, this is contrary to the studies that showed a significant decrease in GCs density of palpebral conjunctiva in the aged population [18]. These findings suggest that changes in GCs density may be unsynchronized or independent between bulbar and palpebral conjunctiva with aging, and it has led to research for imaging markers of conjunctival aging on IVCM.

Hyaline bodies, characterized by a central or an eccentric granular mass surrounded by a lucent zone, were regarded to be composed of occluded GCs [18]. It was observed in 25% of people over 79 years old [18]. A conjunctival microcyst is another feature that may be age-related (Fig. 2a, b), as previous studies showed that the detection rate of conjunctival microcysts on IVCM was greatly increased with age [11]. Whether conjunctival microcysts are degenerated GCs or normal intermediates that arise during the development and maturation of GCs remains unclear [21]. The age-dependent detection rates of microcysts suggest that the quality of GCs may change with age (Fig. 2c, d). In addition to focusing on GCs density, it may be a more pertinent approach to understand the conjunctival pathological changes with aging by examining changes in the quality of GCs as well as their proper function in mucin secretion.

Representative in-vivo confocal microscopy (IVCM) images for young and elderly individuals. a Conjunctival microcyst detected in a 64-year-old subject, manifested as a large, oval-shaped structure with hyperreflective contents and surrounded by a hyporeflective ring. b Conjunctival microcysts detected in a 71-year-old subject, manifested as dark round vacuoles in various size with or without hyperreflective granules (arrows). c Conjunctival epithelium of a 27-year-old subject. Goblet cells are scattered at a high cell density, whereas conjunctival microcysts are absent. d Conjunctival epithelium of a 74-year-old subject, showing lower density of goblet cells

Several studies have also reported that the density of DCs in the bulbar or the central inferior palpebral decreases with age [20, 22], which is consistent with the results of conjunctival biopsies [23]. Those findings may explain the compromised immune function and increased susceptibility to ocular surface inflammation in elderly. Nevertheless, Wei et al. [24] found no correlation between the density of DCs in the superior palpebral conjunctiva and age. The inconsistent results on the relationship between the DCs density and age could be due to the variation in the scanning area of the conjunctiva.

Previous research has shown that all conjunctival lymphoid structures, including the CALT, undergo age-related changes [25]. Among the IVCM parameters of CALT, the density of lymphocyte, follicular density, and perifollicular lymphocytes density sharply decrease with age following a cubic regression. A significant reduction in follicular area with age, accompanied by a marked increase in follicular reflectivity was also observed [13]. Moreover, the diameter of the fibers network of the conjunctival lamina propria decreased significantly with age [26], which appears to account for the increased conjunctivochalasis in older individuals.

Dry eye disease (DED) is a chronic, multifactorial ocular surface disease with tear film instabilities, hyperosmolarity, and inflammatory damage being major etiological factors [27, 28]. The incidence of DED is increasing worldwide each year and represents a growing burden on those affected by vision and life quality [29]. Aqueous deficient type of DED can be further classified as either Sjögren’s syndrome dry eye (SSDE) or non-Sjögren’s syndrome dry eye (NSSDE) [30]. Primary Sjögren's syndrome is a complex autoimmune disorder with multisystemic effects. It is characterized by lacrimal hyposecretion and permanent inflammation of the ocular surface, leading to severe tear deficiency and gradual epithelial damage on ocular surface [31, 32]. Sustained exposure to hypertonicity and inflammatory stimuli of the ocular surface has been shown to cause the loss of GCs [33]. Conventional cytological examinations of the conjunctiva, such as vital staining, impression cytology, and brush cytology, are important in assessing the density of conjunctival cells and inflammatory infiltrate in DED. Although impression cytology and rose bengal staining are the most sensitive and specific methods of detection [34, 35], frequent impression cytology sampling exacerbates ocular surface vulnerability in SSDE patients. Hong et al. [36] characterized the morphology and density of GCs in SSDE patients under IVCM and confirmed the observations were consistent with those obtained by impression cytology, suggesting that IVCM is a viable alternative to impression cytology in such cases. In addition, the assessment of mid-epithelial or subepithelial inflammatory infiltrates can be difficult with impression cytology. Therefore, IVCM is an effective alternative to assess ocular surface structures at multiple sublayers.

IVCM examination of SSDE patients showed expanded conjunctival epithelial cells with nuclear pyknosis, decreased nucleocytoplasmic ratio, and even epithelial cell loss (Fig. 3a, b) [37]. Several studies have shown reduced cell density of the bulbar conjunctival epithelium via IVCM in both SSDE and NSSDE patients than normal subjects [37,38,39]. Both SSDE and NSSDE patients revealed lower basal epithelial cell densities in the bulbar and tarsal conjunctiva compared to controls [40]. This contradicts with another study where the authors reported higher density of corneal basal epithelial cells and anterior stromal cells in the dry eye group compared to the controls [39, 41]. The conflicting findings may result from the fact that DED disrupts the cell renewal but promotes the cell repairment at the same time. Therefore, the apoptosis-proliferation status is very dynamic in corneal and conjunctival epithelial cells. Furthermore, Tais et al. [39] identified a significantly lower conjunctival epithelium cell density in SSDE than in NSSDE. In contrast, Villani et al. [42] detected a higher density of inferior tarsal conjunctival epithelial cells in SS patients. Differences in the site of examination may account for this discrepancy. Yet currently, there are no proven specific IVCM signs to differentiate SSDE from NSSDE.

Representative in-vivo confocal microscopy (IVCM) images from a 67-year-old female with non-Sjögren’s syndrome dry eye. a Enlargement of conjunctival epithelial cells with nuclear pyknosis. b The conjunctival epithelial layer with arrows shows the area of epithelial cells dropping out. c Polymorphic inflammatory cells infiltration within the conjunctival epithelium. d Wire netting distributed dendritic cells with long dendrites

Inflammatory cell infiltration in the conjunctival epithelium of dry eye patients is observed on IVCM scans. The three major types of inflammatory cells are polymorphs, dendritic cells, and lymphocytes [38]. The density of inflammation cells was higher in DED patients than in controls (Fig. 3c, d), and the SSDE group presented a higher level than NSSDE. Notably, the inflammatory cell density was linked to several dry eye clinical parameters. There was a significant positive correlation between the inflammatory cell density and vital staining scores as well as a significant negative correlation between the inflammatory cell density and the Schirmer test scores and tear breakup time [39]. Hence, inflammatory cell density may provide value in assessing the severity of DED and serve as a potential indicator of treatment efficacy.

Another promising feature of IVCM for the assessment of DED severity could be the conjunctival epithelial microcyst density. Studies on SSDE have demonstrated a significantly higher density of conjunctival microcysts than in the normal subjects although there appears to be no statistically significant difference between SSDE and NSSDE [39].

Glaucoma affects more than 70 million people worldwide and this number is expected to rise to 112 million by the year 2040 [43]. The reduction of intraocular pressure (IOP) with specific modalities including medications, laser therapy, and a variety of surgical interventions [44] has proven to be the most effective management for glaucoma. In addition to monitoring IOP and optic nerve damage, it is becoming increasingly important to monitor the health of the ocular surface. IVCM provides a perspective to elaborate glaucoma-associated conjunctival changes, non-invasively and quantifiably. By providing more morphological details, IVCM also assists in understanding the aqueous humor (AH) drainage pathway [45,46,47].

Ciancaglini et al. [45] examined the bulbar conjunctiva using IVCM and reported microcysts in the conjunctival epithelial layer in both untreated ocular hypertension patients and untreated primary open-angle glaucoma (POAG) patients. These microcysts appeared as round or oval hyporeflective extracellular structures between 10 and 200 µm in size [45]. Similarly, microcysts have also been observed in untreated low-tension glaucoma patients with no significant differences in microcysts mean density (MMD) and microcysts mean area (MMA) as compared to POAG [48]. Nevertheless, the presence of microcysts cannot be used as a distinct feature of glaucoma because microcysts have also been detected in normal subjects without morphological differences [26, 49]. Agnifili et al. [48] showed that microcysts in normal eyes represent the final stage of physiological uveo-scleral AH outflow. The same group also found that although microcysts showed significantly higher MMD and MMA in POAG and low-tension glaucoma patients than in normal subjects, there was no correlation between microcyst parameters and IOP levels [48]. Therefore, it has been proposed that microcyst formation may be activated by hyperbaric ocular conditions at early stages of the disease but hardly changes markedly in response to IOP-induced mechanical forces [45].

Topical monotherapy is the first-line option for glaucoma. However, over half of the patients require more than two topical medications to keep their IOP within a manageable range [50]. The use of anti-glaucoma medications leads to gradual eye discomfort as well as tear film instability [51]. The conjunctiva undergoes various tissue modifications, and such changes are associated with several factors, including the preserving agents, duration of medication, and combination of medications [52, 53]. Benzalkonium chloride (BAK) is the most common preservative in topical anti-glaucoma medications, which is responsible for most of the ocular surface side effects during long-term therapy [54]. It has been reported that the GCs density, as determined by IVCM, was significantly lower in glaucoma patients treated with BAK-preserved eye drops than in those treated with preserved-free eye drops and controls. Moreover, the BAK-preserved group showed a higher grade of epithelium irregularity and worse ocular surface parameters, such as Schirmer I test and tear breakup time [53, 55].

The conjunctiva can be affected by long-term (more than 12 months) topical treatment from epithelium to stroma. On IVCM examinations, conjunctival epithelium present with squamous metaplasia, cellular desquamation and keratinization, as well as loss of GCs [8, 56]. The activated DCs were observed in the epithelial layer and basal membrane of the conjunctiva [8]. Additionally, the use of multiple (more than 2) therapeutics results in greater negative impact on the conjunctiva, especially the loss of GCs [57]. Loss of GCs could result from the inflammatory state and the accumulation of preservatives on the ocular surface, further destabilizing the tear film. The poor quality of tears makes it difficult for the toxic or inflammatory cytokines of the ocular surface to be flushed, which in turn exacerbates the loss of GCs. The IVCM examination assists in monitoring drug-induced conjunctival damage in patients receiving long-term medication, therefore contributing to the management of glaucoma patients.

For patients with IOP uncontrolled by topical medications and with progressive visual field loss, surgical treatment is indicated [58]. Trabeculectomy has been the most effective filtering procedure to lower IOP [58]. The surgical technique typically drains AH from the anterior chamber to the subconjunctival space via a new pathway of intrascleral fistula and involves the elevation of the conjunctiva over the scleral flap, known as filtering blebs [59]. However, the long-term prognosis of surgery can be challenging due to filtering scarring or chronic inflammatory response [