Ocular Manifestations Related to Antibodies Positivity and Inflammatory Biomarkers in a Rheumatological Cohort.

Purpose: Ocular involvement is frequent in autoimmune diseases and even can be the first manifestation. There are multiple descriptions in the literature around the world regarding this topic. However, we evidenced a lack of studies analyzing the relationship between the ocular manifestations and systemic biomarkers, especially in Latinamerica. Therefore, this study aimed to examine the relationship between the positivity of inflammatory biomarkers and the ocular manifestations in a Colombian cohort of rheumatological patients. Patients and
Methods: We conducted an observational, descriptive, non-comparative cross-sectional study in a rheumatology center, in Bogotá, Colombia, from 2013 to 2019. We calculated a sample size of 797 patients to assess the prevalence of ocular manifestations and inflammatory biomarkers. We performed univariate analyses for categorical and continuous variables and bivariate analyses using the Chi-square and Fisher's exact test for categorical variables.
Results: Women represented 84% of the population, and the mean age was 54.61± 15.64 years. Of 797 patients, 21.45% reported one or more ophthalmological diagnoses, being keratoconjunctivitis sicca (KCS) the most common (15.93%), followed by uveitis, and cataract (1.38%, each one). Regarding ophthalmological symptoms, 35% presented at least one, being dry eye sensation (DE) the most common (30.86%), followed by ocular pain (2.76%), red eye, and decreased visual acuity (2.63%, each one). The antibodies or inflammatory biomarkers most frequently found were antinuclear antibodies (ANAs) (35.3%), C-reactive protein (28.7%), and rheumatoid factor (27.9%). We found statistical associations between consumption of complement 3, anti-CCP, anti-RO, and anti-LA antibodies with ocular manifestations such as photophobia, DE, conjunctivitis, KCS, uveitis, retinal vasculitis, and maculopathy.
Conclusion: Ocular manifestations are frequently found in patients with positive antibodies and inflammatory biomarkers. Our results suggest antibodies and inflammatory molecules could be biomarkers for ocular manifestations in patients with rheumatological diseases. This study provides the basis for future longitudinal studies.

Introduction

Autoimmune diseases (ADs) are chronic systemic disorders characterized by the activation of the immune system,1 mainly the adaptive response. This leads to the production of antibodies, which attack own tissues, generating destruction and dysfunction of different organs. One of the principal targets is the ocular tissue; this damage may present as the first manifestation of ADs.2

The prevalence and severity of ocular manifestations in ADs vary among diseases and are influenced by factors such as autoantibodies or inflammatory biomarkers positivity.3 Its prevalence range from 18% to 35% of patients.4 For example, in patients with rheumatoid arthritis (RA), the presence of anti-cyclic citrullinated peptide antibodies (Anti-CCP) and rheumatoid factor (RF) have been associated with vision-threatening complications, including scleritis, sclerosing keratitis, and peripheral ulcerative keratitis.5 Moreover, patients with anti-neutrophil cytoplasmic antibody-associated (ANCA) vasculitis may develop anterior segment or optic nerve involvement related to positive proteinase-3 ANCA (PR3-ANCA) and myeloperoxidase-ANCA (MPO-ANCA), respectively.6

Even though multiple studies have described the association between antibodies positivity and ocular manifestations in some ADs, more evidence is needed in most autoimmune and autoinflammatory diseases to guarantee early diagnosis and treatment.4,7 Additionally, there is a lack of studies in this field in Latinamerica.8 Therefore, this study aimed to examine the relationship between the positivity of inflammatory biomarkers and the ocular manifestations in Colombian patients with rheumatologic diseases.

Methodology

Design

We conducted an observational, descriptive, non-comparative cross-sectional study on patients who attended a rheumatology center in Bogotá, Colombia, from 2013 to 2019.

Study Population

Adult patients (>18 years old) who consulted the rheumatology center with a new or previous diagnosis of a rheumatological disease were included. A simple random sampling for finite populations was used to estimate the true proportion of adult patients with rheumatological diseases and ocular manifestations. Taking as reference what is reported in the literature, an expected proportion of 27%,2,9,10 a population of 13,763 patients treated at the rheumatology center, a confidence interval of 95% (taking a normal distribution critical point of 1.96), and an estimation error of 3%, a sample size of 793 medical records was obtained. Based on the information available, four additional cases were considered for a total of 797 medical records. The sample size calculation was done using R Software 4.0.4 sampling book package.11 All patients’ diagnoses included in this study were classified according to the International Classification of Disease, tenth edition (ICD-10).

Antibodies and Inflammatory Biomarkers Assessment

The database was validated with 38 antibodies and inflammatory biomarkers. All data were collected from the patients’ clinical records. We used the International Consensus on Antinuclear Antibodies (ANAs) Patterns recommendations to evaluate them (positive/negative, pattern, and antibody levels).12 For polyautoimmunity-associated antibodies, we considered the following as positive values: >300 for anti-double stranded DNA antibodies (anti-dsDNA), >20 IU for antibodies to extractable nuclear antigens (ENAs: Sm, U1-RNP, Ro/SS-A, La/SS-B), >40 IU for anti-cardiolipin (ACA IgG and IgM), and beta-2 glycoprotein 1 antibody (B2GPI IgG and IgM), >6 IU for RF, >60 IU for anti-CCP, >0.6 IU for autoantibodies directed against the thyroglobulin protein (anti-Tg), and >100 IU for anti-thyroperoxidase-antibodies (anti-TPO).13,14 For the other antibodies and biomarkers, laboratory reference values were used.

Data Collection

We elaborated and validated a database in Microsoft Excel (Microsoft Corp., Redmond, WA, USA) to record the information. Variables included were: type of rheumatologic diseases, antibodies, inflammatory biomarkers, ocular symptoms, and ocular diagnosis. Our trained personnel evaluated and extracted the electronic medical records information during four months.

Statistical Analysis

The results were reported as means and standard deviation for continuous variables and frequency distribution tables for categorical variables. In addition, to evaluate associations between ocular manifestations and positive antibodies and inflammatory biomarkers, categorical variables were assessed using the Chi-square independence test and Fisher’s exact test when indicated, with 95% and 99% confidence levels. All analyses were done in software R version 3.4.4.

Bias Control

Confusion bias was considered because some ocular manifestations could not be related to ADs but to other external causes (e.g. medications and systemic comorbidities). We controlled this bias by designing a temporality variable, which indicated the appearance of the ocular manifestation concerning ADs. After statistical analysis, we considered the onset moment to identify if ocular manifestations were explained by other systemic diseases or the use of medications, proposing hypotheses in the discussion section. Also, a selection bias was considered because patients who consulted were mainly from a high social level; nevertheless, some patients who attended were covered by the national health insurance. Finally, a random selection of the patients’ charts was performed to control the selection bias and ensure a heterogeneous sample.

Results

From a total of 13,763 medical records of patients with rheumatologic diseases, 797 patients were randomly included. Eighty-four percent of the population were women, and the mean age was 54.61 ± 15.64 years. Familial autoimmunity and smoking were the most frequent medical history. Demographic data are shown in Table 1.

Table 1

Demographic Data

Variable	Data n (%)
Female	670 (84.06)
Age (mean, SD)	54.61 SD: ± 15.64 years
Active smokers	49 (6.14)
Past smokers	178 (22.33)
Tattoos	16 (2)
Silicon prosthesis	22 (2.76)
Familial autoimunity	211 (26.47)
Mean age of autoimmune disease diagnosis	Female: 45.88 yearsMale: 50.65 years

Abbreviation: SD, standard deviation.

The most frequent positive biomarker was ANAs in 282 (35.38%) patients, followed by C-reactive protein in 229 (28.73%), RF in 223 (27.98%), and Erythrocyte Sedimentation Rate (ESR) in 193 (24.22%). More information is found in Figure 1.

Figure 1

Frequency of positive autoantibodies and inflammatory biomarkers found in the rheumatological cohort.

The distribution of antibodies by diseases had a usual pattern. However, we saw that in most cases, ANAs took an important role being significantly associated with most of the conditions (Figure 2) and []. In addition, ocular manifestations related to polyautoimmunity presented statistically significant associations with ANAs, anti-ribonucleoproteins antibodies (Anti RNP), IgM anti-cardiolipin antibodies (IgM ACA), IgG anti-cardiolipin antibodies (IgG ACA), anti-Sjögren syndrome-related antigen A (Anti-Ro), anti-Smith antibodies (Anti Sm) (Figure 2). More detailed information on diseases and laboratory findings is shown in [].

Figure 2

Antibodies and inflammatory markers associated with the most frequent rheumatological disease. (A) Rheumatoid Arthritis 266 (33.37%); (B) Fibromyalgia 181 (22.71%); (C) Sjögren Syndrome 157 (19.72%); (D) Systemic lupus erythematosus 79 (9.91%); (E) Polyautoimmunity 55 (6.9%). P-value is based on Chi-square Independence test and Fisher’s exact test with a 95% *and 99% **Confidence level. ***P-value asymptotically significant for a 95% confidence level.

Thirty-five percent of the patients reported one or more ophthalmological symptoms, being dry eye (DE) sensation the most common in 246 (30.86%) patients, followed by ocular pain in 22 (2.76%), red eye, and decreased visual acuity (VA) in 21 (2.63%) each. We found statistically significant associations between DE and IgG B2GP (P=0.01), ANAs (P=0.002), Anti-Ro (P<0.0001), and anti-Sjögren syndrome-related antigen B (Anti-La) (P=0.0008). In the same way, we found statistically significant associations between photophobia and consumption of C3 (C3) (P=0.05) (Table 2). Data about ophthalmological symptoms and their association are shown in [].

Table 2

Ophthalmological Symptoms and Statistically Associated Antibodies Compared with the Literature

Ocular Symptoms	Associated Antibodies in our Population	Associated Antibodies in the Literature
Ocular pain n=22 (2.76%)	–	Anti-MOG in patients with Optic neuritis48AQP4 in patients with Optic neuritis48TRab in patients with Graves orbitopathy49
Photophobia n=14 (1.75%)	C3 (P=0.05***)	Anti-TRPM1 in patients with paraneoplastic retinopathy40
Foreign body sensation n=5 (0.62%)	-	Lack of information
Tearing n=5 (0.62%)	-	Lack of information
Red-eye n=21 (2.63%)	-	PR3-ANCA and MPO in patients with AAV6
Decreased VA n=21 (2.63%)	-	Anti-MOG in patients with Optic neuritis48AQP4 in patients with Optic neuritis48Anti-phospholipid antibodies17,28
Pruritus n=4 (0.5%)	-	Lack of information
Burning n=6 (0.75%)	-	Lack of information
Floaters n=9 (1.12%)	-	Lack of information
Dry eye n=246 (30.86%)	IgG B2GP(P = 0.01**)ANA (P =0.002**)Anti-La (P =0.0008***)Anti-Ro (P <0.0001***)	Anti-Ro in patients with SS44Anti-La in patients with SS44RF in patients with SS44ANAS > o = 1:320 in patients with SS44Anti-CCP in patients with AR3Anti-dsDNA in patients with LES38C3 in patients with LES38SP1 in patients with SS36Anti-CA6 in patients with SS37M(3) mAChRs in patients with SS50Anti-hK13 in patients with SS34
Diplopia n=2 (0.25%)	-	LA; ACA; B2GP in patients with antiphospholipid syndrome17PR3-ANCA and MPO-ANCA in patients with AAV6mAChRs in patients with Myasthenia Gravis51

Notes: p-value is based on Chi-square independence test and Fisher’s exact test when indicated with a 99% **Confidence level. ***P-value asymptotically significant for a 95% confidence level.

Abbreviations: Anti-MOG, anti myelin oligodendrocyte glycoprotein antibody; AQP4, anti aquaporin-4 antibody; TRab, thyrotropin receptor autoantibodies; Anti-TRPM1, autoantibody against transient receptor potential cation channel, subfamily M, member 1; PR3-ANCA, Proteinase 3- antineutrophil cytoplasmic antibodies; MPO-ANCA, myeloperoxidase anti-neutrophil cytoplasmic antibodies; VA, visual acuity; LAC, lupus anticoagulant; ACA, anticardiolipin; B2GPI, anti-beta-2-glycoprotein I; ANAS, antinuclear antibody; Anti-Ro, anti–Sjögren’s syndrome-related antigen A autoantibodies; Anti-La, anti–Sjögren’s syndrome-related antigen B autoantibodies; RF, rheumatoid factor; Anti-CCP, anti-cyclic citrullinated peptide antibodies; Anti-dsDNA, anti-double-stranded DNA antibody; C3, consumption of complement component 3; SP1, Salivary protein 1; Anti-CA6, anti-carbonic anhydrase 6; mAChRs, autoantibodies against muscarinic acetylcholine receptors; Anti-hK13, autoantibodies against kallikrein 13; mAChRs, autoantibodies against muscarinic acetylcholine receptors; RA, rheumatoid Arthritis; SS, Sjögren Syndrome; SLE, systemic lupus erythematosus; ANCA, antineutrophil cytoplasmic antibodies.

On the other hand, 171 (21.45%) patients presented one or more ophthalmological diagnoses, being keratoconjunctivitis sicca (KCS) the most common, present in 127 (15.93%) of them, followed by cataract in 11 (1.38%), uveitis in 11 (1.38%), and scleritis in 10 (1.25%). We found statistically significant associations between KCS and Anti CCP (P= 0.03), ANAs (P= 0.03), lupus anticoagulant (LAC) (P=0.01), Anti-Ro (P<0.0001), and Anti-La (P<0.0001). Similarly, conjunctivitis was statistically significantly associated with C3 (P=0.03) and consumption of C4 (C4) (P=0.05). Also, relationships between retinal vasculitis and MPO-ANCA (P=0.05), uveitis and PR3-ANCA (P=0.04), and maculopathy and Anti CCP (P= 0.04) were found (Table 3). Ophthalmic diagnoses with laboratory findings are shown in [].

Table 3

Ophthalmological Diagnosis and Statistical Positive Association Antibodies Compared with Literature

Ocular Diagnosis	Associated Antibodies in Our Population	Associated Antibodies in the Literature
Keratitis 8 (1%)	-	ANA in a patient with SLE52,53
Conjunctivitis 4 (0.5)	C3 (P =0.03*)C4 (P = 0.05*)	β4-integrin antibody in patients withOCP54,55ANCA in patients with Granulomatosis with Polyangiitis33
Keratoconjunctivitis Sicca 127 (15.93%)	Anti CCP (P = 0.03*)ANAs (P =0.03*)LAC (P =0.01*)Anti-Ro (P < 0.000***)Anti-La (P < 0.000***)	Anti-Ro/SSA and Anti-La/SSB in patients with SS21RF in patients with SS22RF in patients with RA56Anti-dsDNA and C3 in patients with SLE38ANA in patients with SS22anti-CA6 in patients with SS37
OSSN 2 (0.25%)	-	Lack of information
PUK 2 (0.25%)	-	PR3-ANCA in patients with AAV6,7,27
Corneal perforation 1 (0.12%)	-	Anti-dsDNA and ANA in patients with SLE57
Cataract 11 (1.38%)	-	IAAs in a patient with T1DM58
Uveitis 11 (1.38%)	PR3-ANCA (P = 0.04*)	HLA-B27 in patients with JIA31HLA-B27 in patients with uveitis32,59 ANA in patients with JIA15ARA were significantly associated in patients with uveitis30
Anterior uveitis 7 (0.87%)	-	PR3-ANCA in patients with AAV7HLA-B27 in patients with anterior uveitis60
Posterior uveitis 1 (0.12%)	-	ACA in a patients with SLE28Antiphospholipid antibodies in patients with APS17
Panuveitis 1 (0.12%)	-	Anti-UACA in patients with VKH29
Exophthalmos 2 (0.25%)	-	TRab in patients with Graves’ disease61–63EMAb in hyperthyroid Graves’ disease64
Glaucoma 7 (0.87%)	-	ACA and B2GP in patients with APS and normal-tension glaucoma65,66Anti-Ro/SS-A in normal-pressure glaucoma67Anti-rhodopsin antibody in normal-pressure glaucoma68ARA in patient with optic nerve, and optic nerve head antigens in primary open-angle glaucoma and normal-tension glaucoma69
Optic neuritis 5 (0.62%)	-	AQP4 and Anti-MOG48
Retinal vasculitis 1 (0.12%)	MPO-ANCA (P =0.05*)	Antiphospholipid antibodies in patients with APS and LES17,28ARA in patients with systemic inflammatory diseases70P-ANCA positivity71
Macular edema 1 (0.12%)	-	Antiretinal antibodies in retinitis pigmentosa72Antiphospholipid antibodies in patients with APS17
CRAO 1 (0.12%)	-	Lupus anticoagulant and ACA in SLE73Antiphospholipid antibodies in APS17,74,75MPO-ANCA in EGPA76
Maculopathy 4 (0.5%)	Anti CCP (P =0.03*)	Anti-RPE antibodies in AEPVM77Antiphospholipid antibodies in APS17
Scleritis 10 (1.25%)	-	FR and Anti-CCP in RA78MPO-ANCA and PR3-ANCA in AAV6,7
Episcleritis 3 (0.37%)	-	Anti-CCP in RA3MPO-ANCA and PR3-ANCA in patients with AAV6,7
Blind eye 1 (0.12%)	-	Lack of information
Retinal detachment 1 (0.12%)	-	ACA in IDO28Antiphospholipid antibodies in patients with APS17

Notes: p-value is based on the Chi-square independence test and Fisher’s exact test when indicated with a 95% *Confidence level. ***P-value asymptotically significant for a 95% confidence level.

Abbreviations: ANA, antinuclear antibodies; ANCA, antineutrophil cytoplasmic antibodies; PR3-ANCA, Proteinase 3- antineutrophil cytoplasmic antibodies; MPO-ANCA, myeloperoxidase anti-neutrophil cytoplasmic antibodies; Anti-Ro, anti–Sjögren’s-syndrome-related antigen A autoantibodies; Anti-La, anti–Sjögren’s-syndrome-related antigen B autoantibodies; RF, rheumatoid factor; Anti-dsDNA, anti double-stranded DNA antibody; LAC, lupus anticoagulant; C3, consumption of complement component 3; anti-CA6, anti-carbonic anhydrase 6; P-ANCA, perinuclear antineutrophil cytoplasmic antibodies; IAAs, insulin autoantibodies; HLA-B27, human leukocyte antigen B27; ACA, anticardiolipin antibodies; Anti-UACA, anti-uveal autoantigen with coiled coil domains and ankyrin repeats; EMAb, eye‐muscle antibody; APSA, antiphosphatidylserine antibody; APL, antiphospholipid antibodies; ARA, antiretinal antibodies; anti-RPE, anti-retinal pigment epithelium; Anti-CCP, anti-cyclic citrullinated peptide antibody; Anti-TSH, anti-thyroid stimulating hormone receptor antibodies; SLE, systemic lupus erythematosus; OCP, ocular cicatricial pemphigoid; SS, Sjögren syndrome; RA, rheumatoid arthritis; AAV, ANCA-associated vasculitis; T1DM, type 1 diabetes mellitus; JIA, Juvenile Idiopathic Arthritis; JRA, Juvenile Rheumatoid Arthritis; VKH, Vogt-Koyanagi-Harada; APS, antiphospholipid syndrome; EGPA, eosinophilic granulomatosis with polyangiitis; AEPVM, acute exudative polymorphous vitelliform maculopathy; OSSN, Ocular Surface Squamous Neoplasia; PUK, peripheral ulcerative keratitis; JIA, Juvenile Idiopathic Arthritis.

Discussion

Although ocular manifestations and poor prognosis of ADs have been associated with antibodies positivity, most investigations focus on a common group of ADs; generally, ANCA-associated vasculitis (AAV), RA, antiphospholipid syndrome (APS), juvenile idiopathic arthritis (AIJ), Sjögren syndrome (SS), and Systemic lupus erythematosus (SLE).5,7,15–17 Therefore, we analyzed the relationship between ocular manifestations and the positivity of antibodies and inflammatory biomarkers in 45 immune-mediated diseases, as well as in polyautoimmunity phenomena, finding some novel and interesting associations that lead to new hypotheses about the role of these molecules as biomarkers in the phenotype of the ADs.

Keratoconjunctivitis Sicca

Currently, there is no validated definition or classification for KCS. Therefore, this varies depending on the study, making comparisons difficult.18 This condition strongly relates to SS and its autoimmune mechanism, especially in those with severe forms of the disease.19,20 Chung et al found that Anti-Ro and Anti-La autoantibodies concentrations were significantly higher in patients with primary SS who had moderate to severe KCS than those with mild KCS.21 Likewise, a cohort study in patients with primary SS found that patients with KCS with positive RF and ANAs were more likely to have abnormal Schirmer test values.22 Our results agree as we found a statistically significant association between KCS and seropositivity for Anti-Ro and Anti-La.

Contrarily to our study, no studies have shown a statistically significant association between LAC seropositivity and KCS. However, some studies have described a relationship between retinal involvement and LAC in patients with SLE due to immune complexes deposition in vessel walls.23 Based on our findings, more studies are needed to identify a possible mechanism for the development of KCS in the presence of LAC.

Conjunctivitis

The conjunctiva is the most immunologically active tissue in the external eye due to its anatomical location next to lymph nodes and vascular characteristics.24,25 Therefore, it can be affected by immunological responses mediated by antibodies or inflammatory biomarkers.25,26

Antibodies against the cytoplasmic domain of β4 integrin of the hemidesmosomes and anti-integrin antibodies have been associated with conjunctivitis in ocular cicatricial pemphigoid. In the same way, an association between conjunctivitis and ANCAs antibodies, especially in severe and recurrent cases, has been reported.6,27 Contrarily, we did not find any of these associations which could be attributed to the characteristics of our population. However, we found associations between C3 and C4 consumption and conjunctivitis, a finding that has not been previously described. Further studies are required to evaluate if this association is affected by the activity of diseases such as SLE or APS.

Maculopathy

Maculopathy can have different origins, among which the one with the most descriptions and associations with antibodies is cystoid macular edema, where it has been reported a possible relationship between LAC and ACA and macular edema in patients with LES and APS.17,28

Interestingly, our results showed an association between maculopathy and anti-CCP. All the patients with maculopathy had RA and anti-CCP with levels over 300 u/mL. However, this association could also be explained due to the previous use of hydroxychloroquine. Future longitudinal studies are necessary to clarify this possible association.

Uveitis and Retinal Vasculitis

A relationship between uveitis and HLA-B27, ANAs, anti-ACA, Uveal autoantigen with coiled-coil domains and ankyrin repeats (anti-UACA), PR3-ANCA, and anti-retinal antibodies (ARA) has been previously described. That is in line with our results; however, we emphasize that given the low number of patients with JIA in our population, the association with ANAs may not be seen.7,15,29–32 On the other hand, a close relation with antiphospholipid antibodies has been described when we talk about retinal vasculitis.17 In our study, we found an additional association with MPO-ANCAS. Although it is considered that these antibodies are related to scleral manifestations, in this case, we saw an interesting association with uveal manifestations.33 Nevertheless, this association could not be compared neck to neck with other studies cited due to the different methodologies and types of analyses.

Dry Eye

Dry eye is an ocular surface disease that presents as a consequence of chronic inflammation. This disease has been related to the presence of activated CD4+ T cells in ocular surface tissues, contributing to the idea that this is a self–antigen-driven autoimmune disease and that other autoreactive lymphocytes may be involved.34 Considering the difficulties in KCS and dry eye differentiation, we compared our results with those of patients with dry eye reported in the literature.18

Dry eye has been related to different systemic diseases and antibodies. For example, Anti-Ro positivity, ocular staining score, and Schirmer’s test are included as items in the 2016 American College of Rheumatology (ACR) Classification Criteria for SS.16,35 In the same way, Stern et al described the presence of autoantibodies against Klk13 and Klk1 in the serum of mice with dry eye and its contribution to ocular surface inflammation.34 Similarly, the “DREAM” Clinical Trial by Bunya et al described that patients with positive traditional SS autoantibodies had worse corneal and conjunctival staining and those with salivary protein-1 (SP-1) autoantibodies had a significantly higher dry eye prevalence.36 On the other hand, several autoantibodies for SS diagnosis, especially for early stages, have been studied. For example, carbonic anhydrase-6 antibody (CA6) is commonly abundant in lacrimal glands and helps to maintain tear film pH homeostasis to protect the epithelial cells of the cornea and conjunctiva from damage.37 Anti-CA6 was found to be associated with severe corneal and conjunctival staining in patients with aqueous-deficient dry eye.37 In our study, we did not evaluate anti-CA6.

Dry eye has also been described as an ocular manifestation of RA. Vignesh et al described the association between anti-CCP antibodies and the severity of ocular manifestations in RA. Additionally, anti-CCP-related dry eye was the most common manifestation, with a prevalence of 28%.3 Regarding antibodies and inflammatory biomarkers in patients with SLE and dry eye, Chen et al found that dry eye severity was correlated with positive anti-dsDNA and C3 consumption. However, there was no relationship with ESR, ANAs, and C4 consumption, previously associated with SLE activity.34,38 We describe a new association between IgG B2GP antibodies and dry eye. Furthermore, our study results support the association between Anti-Ro, Anti-La, ANAs, and dry eye.

Photophobia

Photophobia has been associated with different conditions.39 In 2019, Ueno et al studied photophobia through electroretinography in patients with paraneoplastic retinopathy; they found a relation between autoantibody transient receptor potential cation channel, subfamily M, member 1 (Anti-TRPM1), and photophobia.40 Due to the limited number of studies reporting this association, we present the first study describing a statistically significant association between inflammatory biomarkers with photophobia.

Polyautoimmunity and Antibodies

Rheumatological and autoimmune diseases are associated with multiple antibodies.41 It has been described that some antibodies can predispose to polyautoimmunity, as is the case of Anti-Ro antibodies in SLE and ANAs antibodies in SS.42,43 In 2001, Gilboe et al described that patients with poliautoimmunity (SLE and SS) and Anti-Ro/SSB antibodies presented worse outcomes on the visual analog fatigue scale (VAS).42

Although we corroborated previously described associations between antibodies and polyautoimmunity, we also present new associations []. Therefore, we suggest early ophthalmology referral for patients who present an autoimmune profile with these characteristics to prevent complications. However, more analytical studies are needed to support this suggestion and the use of antibodies in personalized medicine.44

Fibromyalgia and Antibodies

Although fibromyalgia (FM) is not classified as an autoimmune disease, it is considered a rheumatologic disease whose etiology remains unknown.45 In 2008, even though anti-polymer antibodies presented a low sensitivity in patients with FM, they were associated with more significant pain and fatigue.46 A recent study showed that FM pain was associated with IgG autoantibodies that sensitize peripheral nociceptive afferent neurons.47 Our results showed new associations between FM and antibodies and inflammatory biomarkers []. That could be explained by the mechanism of latent autoimmunity in these patients or also related to this disease. Further studies are needed to evaluate this atypical association.

Limitations

Information bias could be a limitation because the extensive autoantibodies panel was not ordered in all cases, as antibodies are generally requested on a case-by-case basis according to the clinical manifestations and the clinically suspected diagnosis, which could generate an under-recording and underestimation. In fact, in some cases, patients could have positive autoantibodies in a latent manner without overt poliautoimmunity clinical signs. In addition, a measurement bias could be present as antibody data were gathered by various specialists and analyzed retrospectively. However, to control this bias, we re-interpreted antibodies positivity based on interpretation guidelines and articles. Moreover, interviewer bias could also occur since the medical center doctors did not have the training to perform a detailed ophthalmological examination, but all of them were trained in rheumatology and internal medicine; therefore, high quality of information was ensured, and detection bias was prevented.

Also, a confusion bias may occur due to the study’s design, considering it is a cross-sectional study, where outcome and exposure are present simultaneously. In some cases, patients with systemic diseases developed ocular manifestations after diagnosing the autoimmune diseases that required treatment and, therefore, it could have an implication in the ocular manifestation. Further prospective studies should consider this bias. It is important to emphasize some of the associations that we found statistically significant were assessed with a low number of cases which implies that the determination is not conclusive.

Conclusion

Ocular manifestations are common in ADs. However, currently, there is weak evidence on the relation between ocular manifestations and autoantibodies and inflammatory biomarkers positivity. Nevertheless, searching these relations is a fundamental stage in personalized medicine that could allow prompt diagnosis, preventing severe complications and costs. We found that the most frequently positive antibody was ANAs. Also, statistically significant associations were found between consumption of complement 3, anti-CCP, anti-RO, and anti-LA antibodies with ocular manifestations such as photophobia, DE, conjunctivitis, KCS, uveitis, retinal vasculitis, and maculopathy. These results allow generating hypotheses about the biomarker role of these molecules in the ocular phenotype of patients with ADs. Nevertheless, this association was based on a few cases, so it must be interpreted carefully. More studies of these characteristics should be conducted to evaluate new antibodies and new patterns of association not previously described.