Evaluation of Dry Eye Severity and Ocular Surface Inflammation in Patients with Pemphigus and Pemphigoid

ABSTRACT Purpose To evaluate ocular surface involvement, tear cytokine levels, and histopathological changes in pemphigus and pemphigoid patients. Methods A total of 22 patients (15 pemphigus and 7 pemphigoids) and 21 non-diseased controls were enrolled in our study. All participants underwent ocular surface evaluation, which included ocular surface disease index test, slit lamp observation, dry eye-related examination, tear multicytokine analysis, and conjunctival impression cytology. Results Pemphigus and pemphigoid patients presented much more severe conjunctivochalasis, corneal epithelial defects, corneal opacity, symblepharon and dry eye. Severe ocular surface squamous metaplasia and a significant increase of tear macrophage inflammatory protein-1beta, tumor necrosis factor-alpha, interleukin (IL)-1β, IL −6, and IL-8 occurred in pemphigus and pemphigoid patients. Conclusions Our results revealed that ocular surface inflammation and dry eye persist in most pemphigus and pemphigoid patients, and do not occur in parallel with the systemic course. Regular ophthalmological examinations and local anti-inflammatory should be provided for pemphigus and pemphigoid patients.

Autoimmune bullous diseases (AIBDs) are a group of rare and severe autoimmune diseases in which IgG autoantibodies destroy the epidermal cell junction and dermal-epidermal cell junction, resulting in skin and mucosa damage. 1 AIBDs mainly include pemphigus and pemphigoid subtypes, with various clinical manifestations, pathological characteristics, and complicated diagnoses. 2Pemphigus and pemphigoid are rare diseases, with an estimated annual incidence of 0.5 to 16.1 cases per million for pemphigus and 0.5 to 42.8 cases per million for pemphigoid. 1 Pemphigus and pemphigoid are characterized by the appearance of urticarial lesions and flaccid or tense blisters on the skin or mucous membranes in these areas that then develop into erosions and crusts when the blisters rupture, often resulting in systemic lesions affecting many organs, including the oral, nasal, pharyngeal, laryngeal, esophageal, and anogenital mucosa and the eyes. 3As a recurrent chronic disease, the course and progression of pemphigus and pemphigoid are often unpredictable and variable.Disease complications and drug-related side effects impact the quality of life of patients. 4ifferent types of pemphigus and pemphigoid have different severities of ocular involvement, among which mucous membrane pemphigoid (MMP) in pemphigoid is the most serious, with approximately 80% of MMP affecting the eyes and causing ocular cicatricial pemphigoid (OCP). 5Patients may suffer from redness, photophobia, foreign body sensation, pain, and other symptoms, among which dry eye is very common. 3,6,7In the chronic stage, there will be changes in the patient's ocular surface structure, such as conjunctival cicatrization, eyelid margin erosion, symblepharon, lacrimal duct obstruction, and corneal opacity, which will lead to a decrease in tear secretion, goblet cell loss, meibomian gland dysfunction, and further deterioration of appearance and vision. 3,5,80][11] However, few studies have focused on the ocular involvement of pemphigus and pemphigoid, and there is currently a lack of a comprehensive evaluation focusing on the ocular surface inflammation and dry eye of pemphigus and pemphigoid.In recent years, tear cytokine levels have attracted attention as potential biomarkers for understanding ocular surface inflammation and predicting disease prognosis or therapeutic effects. 12Studies have confirmed that cytokines such as interleukin (IL)-8, C-X-C chemokine receptor 1 (CXCR1), matrix metalloproteinase 8 (MMP-8), MMP-9, and myeloperoxidase (MPO) are significantly elevated in the tears of OCP patients. 13,14In addition, conjunctival impression cytology is a noninvasive, accurate, semiquantitative test that can assess the pathological changes of the ocular surface in patients by observing ocular surface cytological changes. 15Studies have shown that OCP patients have few goblet cells and severe ocular squamous metaplasia. 16,17herefore, in this study, we collected information on ocular symptoms and signs, detected multiple cytokine levels in tears, and performed conjunctival impression cytology evaluation of pemphigus and pemphigoid patients to further understand the ocular surface condition of pemphigus and pemphigoid patients and to explore the mechanisms underlying the occurrence of ocular involvement for corresponding treatment.

Patient characteristics
Patients who met the following conditions were included in the experimental group: (1) age ranging from 18 to 80 and (2) a diagnosis of pemphigus and pemphigoid by a dermatologist based on clinical manifestations and laboratory test findings.Patients were excluded based on the following criteria: (1) the presence of allergies, infection symptoms, or other systemic immune diseases; (2) suffering from an acute inflammatory disease of the ocular surface within the past three months; (3) a history of eye surgery or wearing contact lenses within the past three months.
Inclusion criteria for non-diseased controls: (1) age ranging from 18 to 80 and (2) did not have pemphigus or pemphigoid.The exclusion criteria for non-diseased controls were as follows: (1) age or sex mismatch with patients; (2) the presence of a severe ocular surface disease or autoimmune disease; (3) the use of a topical eye medication within the past 30 days; and (4) a history of eye surgery or wearing contact lenses within the past three months.
Twenty-two pemphigus and pemphigoid patients diagnosed in the dermatology clinic of the Second Xiangya Hospital of Central South University from March 2021 to July 2022 were enrolled in this study.Twenty-one age-and sex-matched non-diseased controls were selected as the control group.The duration of the patient's systemic and ocular involvement was recorded from the onset of the relevant symptoms.Patients and non-diseased controls were received by qualified doctors in the ophthalmology clinic for health education and the collection of information on ocular symptoms and signs, and nineteen patients who could cooperate well were sampled for tear cytokine detection and conjunctival impression cytology evaluation.

Ocular symptoms and signs
(1) Symptoms of the ocular surface: The OSDI score was used to assess the frequency of ocular symptoms and their impact on daily life in the past week.The OSDI score was calculated as follows: OSDI score ¼ ðtotal scores of answered questions � 100Þ=ðnumber of answered questions � 4Þ.OSDI scores range from 0 to 100, with higher scores indicating more symptoms.To facilitate the evaluation, we divided the OSDI score into 2 grades as reported: 22 points and less were considered normal, and higher than 22 points were considered abnormal. 18) Objective ocular examinations: We performed a series of eye examinations on all of our participants and collected the following data: (1) Corneal opacity: Participants' corneas were examined under a slit lamp, and a decrease in corneal transparency with a grayish-white color was defined as corneal opacity.(2) Meibomian gland secretion score: The meibomian gland is closely connected to tear film function, and the blockage or abnormal secretion of the meibomian gland causes evaporative dry eye.As a result, we assessed meibomian gland secretion in participants by turning their upper and lower eyelids, gently squeezing them to observe the nature of the secretions at the eyelid margins, and then scoring them.0 indicates that the secretions are clear and transparent liquid, 1 indicates cloudy liquid, 2 indicates cloudy granular secretions, and 3 indicates thick secretions similar to toothpaste.19 (3) LIPCOF score: LIPCOF was tested to evaluate conjunctivochalasis and dry eye.The number of paranasal and nasal bulbar conjunctival folds of the lower eyelid observed under the slit lamp in the natural open eye state can be classified into 4 levels (0 = no conjunctival folds, 1 = one permanent and clear parallel folds, 2 = two permanent and clear parallel folds, 3 = more than two permanent and clear parallel folds).20 (4) Schirmer I test score: The Schirmer I test evaluates the function of the lacrimal gland by measuring tear production.A tear secretion strip was placed in the conjunctival sac of both eyes under the eyelid, the participants were instructed to close both eyes gently for 5 minutes, and the amount of reflex tear secretion was recorded.21 (5) CFS score: Corneal fluorescein staining is a method for the examination of corneal defects and ulcers.The cornea was stained with fluorescein sodium and observed under cobalt blue light.The scores were scored from 0 to 4 using the Oxford scoring standard.22 (6) TBUT: Tear film break-up time examination is commonly used to estimate the stability of the tear film.Fluorescein sodium was dropped into the conjunctival sac of the lower eyelid, and participants were instructed to blink 3 times.After the last blink, the patient opened their eyes naturally.The time of the first black spot on the cornea was observed under cobalt blue light, which was the tear film rupture time.The average value was obtained in 3 consecutive measurements, with TBUT ≥ 10s as normal and TBUT < 10s as abnormal.21

Dry eye
According to TFOS DEWS II, if dry eye is suspected, the first step is to rule out diseases that have symptoms similar to those of dry eye.If there are multiple positive symptoms as assessed by the OSDI score (OSDI score ≥ 13), then based on their ocular surface signs, including tear film break-up time (TBUT < 10s) and ocular surface fluorescein staining (CFS positivity), as long as one of the test results is abnormal, the disruption of tear film homeostasis is proven, and the diagnosis of dry eye is made. 23

Tear collection and testing
After slightly wiping the eyelid and removing any fluid with a clean cotton swab, 10 μl of normal saline was added into the inferior conjunctival sac of participants, participants were instructed to blink three times, and the disposable Microcapillary Fluid collector (Seinda, Guangdong, China) was placed at the outer 1/3rd of the lower eyelid margin.The tear fluid entered the collector due to the siphoning effect.Tears (2.2 μl) in one eye were collected at 10-second intervals, three times per eye.The collected tears were collected into a 200 μl EP tube and stored at −80°C.These tears were then were uniformly selected for cytokine detection, following the instructions of the Milliplex Map Human High Sensitivity T-Cell Panel-Immunology Multiplex Assay Kit (Millipore, Billerica, MA, USA).In brief, 2 μl of each tear sample was diluted with normal saline to 25 μl.Tear samples were added to the corresponding sample wells.The positive control group was added to the standard substance with gradient concentration, while the negative control group was only added to the same amount of solvent.After incubation and washing, all the test wells were supplemented with microbeads and antibodies.MAGPIX (Luminex, Austin, TX, USA), a liquid-phase chip detector with xPONENT® software (Luminex, Austin, TX, USA), was used to detect the fluorescence intensity of cytokines.The cytokine concentration of each sample was calculated by fitting the standard curve drawn by the standard substance.

Evaluation of conjunctival impression cytology
After topical anesthesia of the participant's eyes with oxybuprocaine hydrochloride eye drops, the exfoliated cells of the participant's right superior temporal conjunctiva were collected using a 10 mm diameter semicircular sterile acetate membrane (Advantec, Tokyo, Japan), and then the samples were fixed in 95% ethanol and stored at 4°C.Periodic acid Schiff (PAS) staining was performed using a glycogen staining kit (G1360, Solarbio, Beijing, China).Fixed specimens were washed with distilled water, and periodic acid was added to oxidize the ethylene glycol in the cell to dialdehyde.Then, Schiff's reagent was used to react with aldehydes to make the glycogen-rich mucins in goblet cells purplish red, followed by gradient dehydration with ethanol, a transparent acetic acid membrane with xylene, and finally sealed with neutral resin.All specimens were observed using the Invitrogen EVOS M7000 fully automatic live-cell fluorescence microscopy imaging system (Thermo Fisher Scientific, Massachusetts, US).Five images were randomly captured from each membrane at a multiple of 20, and all images were imaged using Image-Pro Plus 6.0 (Media Cybernetics, Maryland, US).The total goblet cell density (GCD, cells/mm 2 ) of conjunctival impression specimens was counted and graded by Nelson grading. 25

Statistical methods
All statistical analyses in this study were performed using GraphPad Prism 8 (GraphPad Software Inc., San Diego, USA) and SPSS 20 (SPSS Inc., Chicago, USA).Independent t-tests were used to analyze the normally distributed data, and these data are presented as the mean ± SD.Median ± quartile spacing and the Mann-Whitney U test were used to analyze the nonnormally distributed data.The Chi-square test or Mann-Whitney U test was used for the comparison of classified data.Correlation analysis was performed using Spearman's test.Differences with P < .05were considered significant.

Demographic data and disease duration
A total of 22 patients were included in this study: 15 patients were diagnosed with pemphigus, and 7 patients were diagnosed with pemphigoid.The age (53.64 ± 12.37 vs. 50.33± 15.28 years, P = .439)and sex ratio (P = .650)were comparable between patients (12 men and 10 women) and non-diseased controls (11 men and 10 women).The mean duration of pemphigus and pemphigoid patients' general diseases was 43.23 ± 32.33 months (range from 4-108 months).All patients were in their chronic period and receive systemic therapies.Three patients were receiving ocular medication and three patients have past ophthalmic history including hordeolum and cataracts.Eighteen patients had ocular surface involvement (81.8%), with a mean duration of 20.66 ± 21.46 months (range from 0-72 months) (Table 1).

Analysis of ocular surface condition
The ocular symptoms and binocular surface signs of the pemphigus and pemphigoid patients and non-diseased controls are summarized in Table 1.The differences between the patient group and non-diseased controls in the ratio of corneal opacity, symblepharon, LIPCOF score, CFS positivity, and Schirmer I test score were significant (P < .05).However, the differences in the OSDI score, OSDI level, meibomian gland secretion score, TBUT, and TBUT level between the patient group and non-diseased group were not significant (P > .05).
Representative images of ocular and systemic changes in patients are shown in Figure 1.

Analysis of dry eye severity and conjunctival impression cytology
The OSDI score, reflecting subjective symptoms of dry eye, as well as the TBUT, Schirmer I test score, and CFS score, reflecting objective indications of dry eye, were used to assess dry eye status. 23Dry eye was present in 63.6% of pemphigus and pemphigoid patients and 52.4% of non-diseased controls.Patients had a higher frequency of dry eye.However, there was no significant difference in dry eye frequency between the two groups (P > .05).
The type of dry eye (EDE, ADDE, MDE) in patients was statistically different from that in non-diseased controls (P = .015),with the mixed dry eye being the most common type, followed by evaporative dry eye, and aqueous-deficient dry eye did not appear.As shown in Figure 2, the patient group had a significantly lower goblet cell density than the non-diseased group (94.53 ± 56.85 vs. 144.62± 71.72 cells/mm 2 , P = .020)and a significantly more severe Nelson's classification (P = .030,Table 1).

Comparison of pemphigus group and pemphigoid group
The twenty-two patients included 15 cases of pemphigus and 7 cases of pemphigoid.There was no significant difference in age, sex, or the duration of disease between the two subtypes (P > .05,Table 1).All pemphigoid patients (100%) and 11 pemphigus patients (73.3%) had ocular surface involvement, but the difference in ocular involvement duration was not statistically significant (P > .05,Table 1).The ocular surface symptoms and signs (OSDI score, OSDI level, ratio of corneal opacity, symblepharon, meibomian gland secretion score, LIPCOF score, CFS positivity, Schirmer I test score, TBUT, and TBUT level) showed no significant differences between the two subtypes (P > .05,Table 1).There was no statistically significant difference in goblet cell density and Nelson's classification system between the two groups (P > .05,Table 1), suggesting that the ocular surface squamous metaplasia was similar in both groups.Figure 2(a,b) show representative images of PAS staining for conjunctival impression cytology in the pemphigus group and pemphigoid group.The results of the multicytokine analysis showed differences in tear cytokine levels between the two groups were not significant (P > .05,Table 2).

Factors associated with ocular surface status in the pemphigus and pemphigoid patients
To further understand the influence of disease duration on ocular surface status in pemphigus and pemphigoid patients, we analyzed the correlation between disease duration and ocular surface parameters in the patient group.The dermatological duration was not significantly associated with any ocular surface indexes (OSDI score, TBUT, Schirmer I test score, CFS score, LIPCOF score, GCD) (P > .05,Table 3).There was a moderate positive correlation between the ocular involvement duration and the OSDI score (R 2 = 0.557, P = .007).Ocular involvement duration was not significantly correlated  with other ocular surface parameters (TBUT, Schirmer I test score, CFS score, LIPCOF score, GCD) (P > .05,Table 3).
To gain a better understanding of the influence of tear cytokine levels on ocular surface status in pemphigus and pemphigoid patients, we also analyzed the correlation between the significantly increased cytokines in tear and ocular surface indexes in the patient group.IL-6 was negatively correlated with TBUT (R 2 = −0.523,P = .022).Other tear cytokines (MIP-1β, TNF-α, IL-1β, IL-8) were not significantly associated with any ocular surface parameter (P > .05,Table 3).

Discussion
Pemphigus and pemphigoid are autoimmune diseases that seriously affect the skin and mucous membranes, which can be divided into multiple subtypes based on antigens and clinical manifestations. 26Pemphigus and pemphigoid frequently involve multiple organs and manifest in a variety of ways, with the eyes being a well-known target.][29] Early symptoms of OCP include chronic progressive conjunctival fibrosis, as well as late corneal opacity and neovascularization, which can result in severe vision loss. 30As reported, other subtypes of pemphigus and pemphigoid also affect the eyes, with patients progressing from simple conjunctival hyperemia and edema to abnormal changes in the conjunctiva, cornea, and eyelid tissue, eventually leading to corneal ulcer and blindness. 3,5,31,32As a result, it is critical to improve the understanding of pemphigus and pemphigoid ocular involvement and to carry out timely clinical ophthalmic examination and treatment.In our study, we collected information on chronic ocular symptoms and signs in 22 patients with pemphigus and pemphigoid of different types.According to our findings, 36.4% of pemphigus and pemphigoid patients had no obvious subjective symptoms (OSDI score < 22).A few pemphigus and pemphigoid patients suffered more severe ocular surface lesions, including corneal opacity in 13.6% of patients and symblepharon in 18.2% of patients.This has been discovered in previous studies as well, 33,34 but the majority of them have focused on patients with MMP or OCP, which involves eye disease more frequently and severely.In particular, conjunctivochalasis was first found to be more common and severe in pemphigus and pemphigoid patients in our study, possibly due to the loss of conjunctival epithelial cohesion and increased collagenolytic activity caused by ocular inflammation. 35Conjunctivochalasis can have an impact on tear secretion, tear film stability, and the density of conjunctival goblet cells, eventually leading to severe dry eye. 36ry eye is a multifactorial ocular surface disease characterized by a loss of tear film homeostasis, which causes an unstable tear film, hyperosmolarity, ocular surface inflammation and injury, and nerve paresthesia. 37Many studies have shown that dry eye is a common complication of OCP or MMP and that patients have higher corneal fluorescent staining scores, less tear secretion, and worse tear film stability. 13,34,38,39In recent years, only two studies have focused on dry eye in pemphigus and pemphigoid.Kiyat, P et al. 40 evaluated dry eye severity and meibomian gland secretion in 20 patients with pemphigus.The ocular surface staining score, OSDI score, and meibomian gland score were significantly higher than those in non-diseased individuals, while the Schirmer I test score and TBUT were significantly lower.Tan, J. C et al. 41 evaluated the ocular manifestations of 22 pemphigus and bullous pemphigoid patients.Of the 42 eyes tested, 39 (93%) had dry eye, with abnormal Schirmer I test results, low TBUT values, and severe fluorescein staining.Similar to the above studies, the occurrence of dry eye in our study was common in pemphigus and pemphigoid patients, but the prevalence may vary due to the scope of the study and the diagnostic criteria used for dry eye evaluation.It should be noted that the types of dry eye in our pemphigus and pemphigoid patients were mainly mixed dry eye.These findings could be the result of chronic ocular inflammation, which damages some ocular surface structures, such as lacrimal glands, accessory lacrimal glands, and meibomian glands. 42Dry eye is a potentially sight-threatening condition that is often overlooked in pemphigus and pemphigoid patients and may progress to severe ocular involvement, which can affect patients' quality of life.Studies have shown that keeping the environment moist, using liquid or gel artificial tears, using anti-inflammatory eye drops, or performing punctual occlusion are all effective in treating dry eye in OCP. 7Therefore, our view is that early control of ocular surface inflammation, with particular attention to meibomian glands, is essential to reduce dry eye in pemphigus and pemphigoid patients.
Goblet cells in the conjunctiva can secrete mucin and form a natural barrier for the ocular surface, which has functions such as secreting mucus, removing ocular surface debris, immune regulation, and maintaining environmental homeostasis on the ocular surface. 43Based on past research, OCP patients have a great loss of ocular surface goblet cells.However, few studies have described ocular surface epithelial cells and goblet cells in pemphigus and pemphigoid. 16,17In this study, it was discovered that goblet cell density was significantly reduced in most pemphigus and pemphigoid patients without obvious ocular symptoms and signs.In addition, the average volume of conjunctival epithelial cells was increased, nuclear pyknosis was present in more epithelial cells, and a higher degree of ocular surface squamous metaplasia occurred.5][46] Ocular surface pathological changes in pemphigus and pemphigoid patients may also be related to chronic ocular surface inflammation.
To further investigate the role of ocular inflammation in the development of pemphigus and pemphigoid, the levels of 20 cytokines in tears were detected.The results showed that the pro-inflammatory factors TNF-α, IL-1β, and IL-6 and the chemokines MIP-1β and IL-8 were significantly elevated in the tears of pemphigus and pemphigoid patients.The continued abnormal synthesis of these cytokines may play a pathological role in chronic inflammation and autoimmunity. 47TNF-α is mainly secreted by macrophages and T cells and can kill or inhibit the proliferation of certain tumor cells, recruit immune cells and induce an inflammatory response leading to tissue destruction. 48IL-1β, whose main source is hematopoietic cells, triggers innate immunity and the inflammatory response via the IL-1 receptor family. 49IL-6 is a pleiotropic cytokine that promotes antibody production by B cells and regulates the maturation, proliferation, and differentiation of T cells and other nonimmune cells during the acquired immune response. 50IL-8 has a chemotactic effect on neutrophils, basophils, and T cells. 51MIP-1β acts on T cells, natural killer cells, monocytes, and several other immune cells by binding to the CCR5 receptor. 52Currently, some studies have shown higher expression of IL-6, IL-8, IL-12, and IL-17 in the ocular surface of OCP patients. 13,53Similar to the above study, IL-8 and IL-6 were significantly upregulated in the tears of pemphigus and pemphigoid patients.Through correlation analysis, we found that TBUT was significantly correlated with the tear cytokine IL-6 level in pemphigus and pemphigoid patients, suggesting that increased IL-6 levels in tears are closely associated with chronic ocular surface complications.These findings imply that ocular inflammation persists in pemphigus and pemphigoid patients and that changes in these cytokines provide insight into the occurrence of local immune responses.
Our study included 22 patients which were divided into two main groups according to the pathology of blister formation: pemphigus, causing intraepidermal blisters, and pemphigoid, causing subepidermal blisters. 1When comparing the eye conditions of the two subgroups, we discovered that there was no significant difference in symptoms, signs, goblet cells, and the degree of ocular surface squamous metaplasia and local inflammation.Studies have shown that approximately 80-90% of MMP cases involve eyes, while ocular involvement in other subtypes has been rarely reported. 5,6,8,54In our study, 11 out of 15 cases of pemphigus (73.3%) and all 7 cases of pemphigoid (100%) had ocular changes of varying degrees.Although ocular involvement in some subtypes was traditionally considered uncommon, its incidence may be underestimated because of the prominent cutaneous and oral symptoms in these subtypes.Some pemphigus and pemphigoid patients present early with conjunctivitis and delayed or incorrect treatment can lead to serious ocular complications.

Conclusions
In summary, ocular inflammation and systemic lesions of pemphigus and pemphigoid do not develop in parallel; therefore, patients should be encouraged to have regular ophthalmological examinations accompanied by systemic therapy.At the same time, in addition to focusing on some signs, such as corneal opacity and symblepharon, which seriously affect visual acuity, ophthalmologists should also pay attention to conjunctivochalasis and dry eye, which have a high incidence in pemphigus and pemphigoid patients, and provide anti-inflammatory and targeted treatment to improve the patient's quality of life.

Figure 1 .
Figure1.Ocular and systemic lesions in pemphigus and pemphigoid patients.(a) One pemphigus patient had ocular involvement for 24 months, and a typical symblepharon was present.The patient's bulbar conjunctiva was congested, and the lower palpebral conjunctiva could not be separated from the bulbar conjunctiva, resulting in limited eye movement.(b) One pemphigus patient had ocular involvement for 6 months.Typical corneal fluorescein staining results showed that the patient had punctate or lamellar damage to the corneal epithelium.(c) One patient with pemphigoid had ocular involvement for 3 months with typical conjunctivochalasis, where the bulbar conjunctiva is excessively loose, causing the conjunctiva to form folds between the lower eyelid and the eyeball.(d) One pemphigoid patient had ocular involvement for 72 months with serious complications, including upper and lower eyelid adhesions, inability to open properly, meibomian gland dysfunction, eyelid margin hypertrophy, disorganized eyelashes, bald eyelashes, difficulty moving the eyeball, mixed conjunctival congestion, and massive neovascularization extending into the cornea, causing partial clouding.(e) and (f) One pemphigus patient had the systemic disease for 4 months, and the eyes were involved for 2 months, presenting typical skin lesions on the trunk and face of patients, with recurrent erythema, blisters, erosion, and scar development.

Figure 2 .
Figure 2. Conjunctival impression cytology specimens.(a) Representative specimen of pemphigus patients.Goblet cells were very rare and small, epithelial cells were large, and showed an abnormal appearance of polygonal squamous metaplasia with small and pyknotic nuclei.(b) Representative specimen of pemphigoid patients.Goblet cells were significantly reduced, epithelial cells were enlarged, and showed an abnormal appearance of squamous metaplasia with small and pyknotic nuclei.(c) Representative specimen from non-diseased controls.Goblet cells were abundant, and epithelial cells were small, round, and evenly distributed.Arrowheads indicate goblet cells.(Scale bar = 50 μm).

Table 1 .
Demographic data and ocular condition in pemphigus and pemphigoid patients and non-diseased controls.Continuous variables were compared using the independent t-test and shown as mean ± SD.The Chi-square test or Mann-Whitney U test was used for the comparison of classified data.P 1 : compared between the patient group and the non-diseased controls.P 2 : compared between the pemphigus group and the pemphigoid group.

Table 2 .
Tear cytokine levels (pg/mL) in pemphigus and pemphigoid patients and non-diseased controls.Tear cytokines were compared by Mann-Whitney U test and their concentrations were expressed as median ± interquartile range.*P < .05were shown in bold: compared between the patient group and the non-diseased controls.

Table 3 .
Correlation of duration of disease and tear cytokine levels with ocular surface indexes in pemphigus and pemphigoid patients.surface disease index, TBUT: tear film break-up time, CFS: corneal fluorescein staining, LIPCOF: lid-parallel conjunctival fold, GCD: goblet cell density, MIP-1β: macrophage inflammatory protein-1beta, TNF-α: tumor necrosis factor-gamma, IL: interleukin.The mean values of the left and right eye surface parameters of patients were used for Spearman's correlation test, which was represented by R-Squared and P values, *P < .05were shown in bold.