FormalPara Key Summary Points

Many patients diagnosed with Graves’ orbitopathy (GO) without corneal damage still complain of ocular surface irritation symptoms.

Previous studies for inactive GO focused on changes in the cornea, and fundus and orbital abnormalities, while this study aimed to analyze the risk factors for ocular surface irritation symptoms in non-corneal-damage inactive mild and moderate-to-severe GO.

For inactive mild GO, upper eyelid lag, the Clinical Activity Score (CAS), and extraocular muscle involvement were risk factors for ocular surface irritation symptoms.

In inactive moderate-to-severe GO, extraocular muscle involvement and marginal reflex distance 1 (MRD1) were risk factors for ocular surface irritation symptoms.

Introduction

Graves’ orbitopathy (GO) is the most common external manifestation of Graves’ disease. GO may cause changes in the appearance of the eyeball and vision or impair visual quality, significantly degrading patients’ quality of life and affecting psychosocial health [1,2,3,4]. Patients with inactive GO still suffer from symptoms of ocular surface irritation, such as spontaneous orbital pain, gaze-evoked orbital pain, grittiness, photophobia, and tearing. Previous studies have revealed that even patients with mild or minimal eyelid malposition can have ocular surface complaints [5]. We found that a huge number of patients with moderate-to-severe GO still complained of ocular surface irritation symptoms in the clinic, even after being treated with glucocorticoids or radiotherapy. However, those studies for inactive GO focused primarily on changes in the cornea, and fundus and orbital abnormalities. Therefore, we performed a retrospective analysis of patients in our hospital diagnosed with inactive mild and moderate-to-severe GO without concurrent corneal damage, aiming to explore more effective treatment recommendations through the analysis of risk factors for ocular surface irritation symptoms.

Methods

Subjects and Test Methods

A retrospective analysis was conducted on the clinical data for patients first diagnosed with inactive mild and moderate-to-severe GO cases without concurrent corneal damage in the Endocrinology and Ophthalmology Departments of Sun Yat-sen Memorial Hospital, Sun Yat-sen University, from April 2017 to September 2023. If a patient’s binocular conditions were inconsistent, the more serious one was selected; otherwise, an eye was randomly selected. The medical history, symptoms, and ocular examination results were all recorded. The diagnosis of GO was essentially based on the Bartley diagnostic criteria [6], and the exclusion criteria were as follows: (1) patients with a history of medication affecting the microenvironment of the ocular surface, such as use of eye drops, eye ointments, or systemic drugs for the treatment of meibomian gland disease, marginal blepharitis, rheumatoid arthritis, or antidepressants; (2) patients with a past history of ocular surgery and orbital radiotherapy; (3) patients wearing corneal contact lenses; (4) patients diagnosed with trichiasis or lagophthalmos; and (5) patients diagnosed with other ocular surface diseases, such as ocular allergy, meibomian gland disease, marginal blepharitis, or pterygium.

The European Group on Graves’ Orbitopathy (EUGOGO) classification [7] was used to assess the severity of GO, including mild, moderate-to-severe, and vision-threatening GO. The Clinical Activity Score (CAS) was used to evaluate the inflammatory activity of GO [7, 8], and CAS ≥ 3/7 was considered as active while CAS < 3/7 was considered as inactive GO. In this study, patients with inactive mild and moderate-to-severe GO were enrolled. The sample was divided into symptomatic and asymptomatic groups according to the ocular surface symptoms. The following indexes were observed: corneal staining evaluation (fluorescein staining), upper eyelid lag, Bell’s phenomenon, extraocular muscle involvement, tear film breakup time (TFBUT), Schirmer test II, marginal reflex distance 1 (MRD1), marginal reflex distance 2 (MRD2), palpebral fissure width (MRD1 + MRD2), proptosis, free triiodothyronine (FT3), free thyroxine (FT4), thyroid-stimulating hormone (TSH), and thyrotrophin receptor antibody (TRAB).

This study was performed strictly in accordance with the principles of the Declaration of Helsinki, and approved by the Ethics Committee of Sun Yat-sen Memorial Hospital, Sun Yat-sen University (SYSKY-2023-1257-01). Due to the retrospective nature of the study, the requirement for informed consent was waived.

Statistical Analysis

SPSS 22.0 software (Statistical Package for the Social Sciences; IBM Corp., Armonk, NY, USA) was used for the statistical analysis. Quantitative data were statistically expressed as mean ± standard deviation (SD), and the Kolmogorov–Smirnov (K–S) test was used to determine whether the data followed a normal distribution. If the data followed a normal distribution with homogeneous variance, the independent Student t test (or t′ test) was used to verify whether the differences across the two groups were statistically significant. Meanwhile, the frequencies and percentages were used to provide a statistical description of the qualitative data, and either the Pearson χ2 or Fisher’s exact probability test was used for comparison between the two groups. Multivariate logistic regression was used to analyze risk factors. A value of P ≤ 0.05 denoted statistical significance.

Results

Analysis of General Clinical Data

This study included a total of 307 eyes from 307 cases. Gender, age, CAS, and the frequency of ocular surface irritation symptoms are described in Table 1. Among the sample, 203 (66.1%) and 104 (33.9%) patients had mild and moderate-to-severe GO, respectively, with 217 cases (70.7%) scored as CAS 0, 57 cases (18.6%) as CAS 1, and 33 cases (10.7%) as CAS 2.

Table 1 General conditions of non-corneal-damage inactive GO
Full size table

Risk Factors for Ocular Surface Irritation Symptoms in Patients with Non-Corneal-Damage Inactive Mild GO

As shown in Table 2, significant differences in CAS were exhibited between the symptomatic group and the asymptomatic group (P < 0.001). Additionally, there was a significant difference in the frequency of ocular surface irritation symptoms between CAS 0 and CAS 1 groups (χ2 = 16.762, P < 0.001), as well as between CAS 0 and CAS 2 groups (P = 0.001). Symptoms in CAS 1 included eyelid swelling (ten cases, 32.2%) and chemosis (three cases, 9.7%); CAS 2 included eyelid swelling (six cases, 26.1%), chemosis (three cases, 13.0%), and conjunctival redness (one case, 4.3%). A significant increase in upper eyelid lag (P = 0.049) and extraocular muscle involvement (P = 0.019) was observed in the symptomatic group compared with the asymptomatic group. Multivariate logistic regression analysis showed that upper eyelid lag [OR = 3.485, 95% CI = 1.013–11.994, P = 0.048], CAS 1 [OR = 10.412, 95% CI = 2.889–37.527, P < 0.001], CAS 2 [OR = 6.036, 95% CI = 1.699–21.444, P = 0.005], and extraocular muscle involvement [OR = 2.036, 95% CI = 1.074–3.857, P = 0.029] were risk factors for ocular surface irritation symptoms in patients with non-corneal-damage inactive mild GO (Table 3).

Table 2 Related factors of non-corneal-damage in inactive mild GO with ocular surface irritation symptoms
Full size table
Table 3 Multivariate logistic regression analysis of the non-corneal-damage inactive mild GO with ocular surface irritation symptoms
Full size table

Risk Factors for Ocular Surface Irritation Symptoms in Patients with Non-Corneal-Damage Inactive Moderate-to-Severe GO

Compared with the asymptomatic group, CAS exhibited a significant difference in the symptomatic group (P = 0.004). Additionally, there was a significant difference in the frequency of ocular surface irritation symptoms between the CAS 0 and CAS 2 groups (P = 0.005). CAS 1 included eyelid swelling (five cases, 19.3%), chemosis (three cases, 11.5%), and conjunctival redness (one case, 3.8%); CAS 2 included eyelid swelling (three cases, 30.0%), chemosis (two cases, 20.0%), and conjunctival redness (one case, 10.0%). Among patients with non-corneal-damage inactive moderate-to-severe GO, extraocular muscle involvement (P < 0.001), MRD1 (P = 0.030), and TSH (P = 0.034) were more prevalent in the symptomatic group than in the asymptomatic group (Table 4). Multivariate logistic regression analysis showed that extraocular muscle involvement [OR = 19.917, 95% CI = 1.669–237.611, P = 0.018] and MRD1 [OR = 1.636, 95% CI = 1.030–2.600, P = 0.012] were statistically significant risk factors for ocular surface irritation symptoms (Table 5).

Table 4 Related factors of non-corneal-damage inactive moderate-to-severe GO with ocular surface irritation symptoms
Full size table
Table 5 Multivariate logistic regression analysis of non-corneal-damage inactive moderate-to-severe GO with ocular surface irritation symptoms
Full size table

Discussion

Patients with inactive GO suffer from ocular surface irritation symptoms such as spontaneous orbital pain, gaze-evoked orbital pain, grittiness, photophobia, and tearing. In the present study, more than half of the patients in the non-corneal-damage inactive mild and moderate-to-severe GO groups experienced ocular surface irritation symptoms. Thus, it is imperative to explore risk factors for ocular surface irritation symptoms in patients with non-corneal-damage inactive mild and moderate-to-severe GO.

Previous studies have generally focused on ocular surface inflammation in active GO [9,10,11,12,13,14,15,16,17]. According to those studies, the etiologic factors of ocular surface irritation symptoms and ocular surface damage include tears, cornea, conjunctiva, glands, palpebral fissure width, proptosis, upper eyelid retraction, lagophthalmos, and poor Bell’s phenomenon [3, 12, 14,15,16,17]. However, our study found that for non-corneal-damage inactive mild GO cases, CAS, upper eyelid lag, and extraocular muscle involvement in the symptomatic group were more serious than those in the asymptomatic group. As we know, upper eyelid lag describes the condition in which the upper eyelid lags behind eyeball movement on vertical downward pursuit and remains high, constituting a combination of eyelid edema and erythema, proptosis, and lagophthalmos, ultimately resulting in upper eyelid retraction [18]. Furthermore, in the symptomatic group with inactive mild GO, MRD1 was higher than the asymptomatic group, and there was no statistically significant difference between the groups, which was contrary to the findings reported previously [13]. Extraocular muscle involvement is related to periocular muscle inflammation, which can result in poor Bell’s phenomenon. As multivariate logistic regression analysis showed that a higher CAS, upper eyelid lag, and extraocular muscle involvement were risk factors for ocular surface irritation symptoms in inactive mild GO, we infer that ocular surface irritation symptoms may be caused by the inflammation of Müller’s muscle, the levator muscle, and the periocular muscle. Ocular surface irritation symptoms may be caused by muscle inflammation, which leads to abnormal eye position or eyelid position, resulting in visual fatigue and imbalance of the ocular surface environment, and then the release of inflammatory factors, damaging the lacrimal gland, conjunctiva, cornea, and other structures—and a vicious circle. In contrast to EUGOGO clinical practice guidelines [7], for patients with inactive mild GO who suffer from ocular surface irritation symptoms, we determined that artificial tear solutions, eyelid taping, and downward massage of the upper eyelid have limited effects, and anti-inflammatory eye drops or immunosuppressive agents (cyclosporine [CsA] eye drops [17]) may be another treatment.

Among patients with non-corneal-damage inactive moderate-to-severe GO, CAS, extraocular muscle involvement, MRD1, and TSH were increased in the symptomatic group relative to those in the asymptomatic group, which was consistent with previous studies demonstrating upper eyelid retraction as one of the factors for ocular surface damage [12, 13, 15]. Multivariate logistic regression analysis revealed that extraocular muscle involvement and MRD1 were risk factors for ocular surface irritation symptoms in inactive moderate-to-severe GO cases. We speculate that the cause of symptoms of extraocular muscle inflammation is the same as that for mild GO.

Our results indicated that even in non-corneal-damage inactive mild or moderate-to-severe GO, eyelid malposition and periocular muscle inflammation play significant roles in the manifestation of ocular surface irritation symptoms. Thus, in order to decrease those symptoms, pharmacotherapy (anti-inflammatory eye drops, CsA eye drops, botulinum toxin injection, etc.) [16,17,18,19,20] or surgical intervention such as orbital decompression, correction of upper eyelid retraction, or lateral tarsoconjunctival flap surgery [21,22,23] should be implemented as soon as possible.

This study had several limitations. First, our study only compared an asymptomatic and symptomatic group with inactive mild or moderate-to-severe GO, and the lack of a control group with no signs of Graves’ disease affects the accuracy of the results. Secondly, this study only used fluorescein staining to assess the presence of ocular surface damage and did not use the Ocular Surface Disease Index (OSDI) questionnaire and graded comparisons of the severity of ocular surface damage.

In previous studies of inactive GO, decreased cell density, squamous metaplasia of bulbous conjunctival epithelial cells, corneal neurodegeneration, and decreased sensitivity of corneal nerves were found using impression cytology examination and confocal microscopy [10, 17, 24,25,26]. Moreover, studies of ocular surface inflammation have demonstrated that inflammatory factors can be expressed in tears, cornea, conjunctiva, and lacrimal glands [27,28,29]. Next, we will explore the characteristics of ocular surface irritation symptoms in inactive GO at the cellular and molecular levels, and provide further evidence to support this study's findings through treatment.

Conclusion

Our results suggested that eyelid malposition and periocular muscle inflammation are risk factors for the ocular surface irritation symptoms in patients with non-corneal-damage inactive mild and moderate-to-severe GO.