Article Text

Download PDFPDF

Prevalence and predictors of Sjögren's syndrome in a prospective cohort of patients with aqueous-deficient dry eye
  1. Melissa (Shiao Hui) Liew1,
  2. Min Zhang1,
  3. Elizabeth Kim2,
  4. Esen K Akpek3
  1. 1Pfizer Inc., San Diego, California, USA
  2. 2Pfizer Inc., New York, New York, USA
  3. 3Department of Cornea and External Disease Service, Ocular Surface Diseases and Dry Eye Clinic, The Wilmer Eye Institute, Baltimore, Maryland, USA
  1. Correspondence to Dr Esen K Akpek, Department of Cornea and External Disease Service, Ocular Surface Diseases and Dry Eye Clinic, The Wilmer Eye Institute, Johns Hopkins Medical Institutions, 600 N. Wolfe Street, Maumenee #317, Baltimore, MD 21287-9238, USA; esakpek{at}


Aims To assess the prevalence and determine predictors of Sjögren's syndrome (SS) in patients with clinically significant aqueous-deficient dry eye.

Methods Patients enrolled in an industry-sponsored, multicentre clinical trial (NCT00784719) were assessed prospectively for the presence of SS. Ocular testing included Schirmer test, corneal fluorescein staining, conjunctival lissamine green staining, and tear-film breakup time. Review of systems questionnaire, medical history, dry eye questionnaire and laboratory work-up (Sjögren-specific antibody A (SSA), Sjögren-specific antibody B (SSB), rheumatoid factor (RF) and antinuclear antibody (ANA)) were obtained.

Results Of 327 patients, 38 (11.6%) had SS: 21 (6.4%) with primary SS (pSS), and 17 (5.2%) with secondary SS. Nine patients (3%) were newly diagnosed using the applied diagnostic criteria based on American–European consensus criteria. Patients with SS had significantly worse conjunctival and corneal staining, Schirmer test (with and without anaesthesia), and symptoms compared with patients without SS. pSS Was significantly more likely to occur in patients with positive ANA (OR: 13.9) and RF (OR: 4.8).

Conclusions Ophthalmologists caring for patients with clinically significant dry eye should have a high index of suspicion for underlying SS and low threshold for serological work-up. RF and ANA are recommended as useful tests in SSA/SSB-negative patients for further diagnostic referral.

  • Conjunctiva
  • Cornea
  • Diagnostic tests/Investigation
  • Lacrimal gland
View Full Text

Statistics from


Dry eye is one of the most common problems encountered by eyecare practitioners globally. In the USA alone, an estimated 10 million individuals have clinically significant dry eye syndrome. This may be an underestimation if individuals who self-treat are considered.1 The Dry Eye Workshop Definition and Classification Subcommittee has identified two classes of dry eye: aqueous–deficient and evaporative.2 Aqueous-deficient dry eye (ADDE) is associated with decreased lacrimal secretion, and is subcategorised into Sjögren's syndrome (SS)-related dry eye and non-SS dry eye.3 ,4 Two forms of SS are recognised: primary SS (pSS), consisting of dry eye in combination with dry mouth, presence of certain autoantibodies and a positive minor salivary gland biopsy,5 and secondary SS (sSS), which has the features of both pSS and an autoimmune connective tissue disease, most commonly rheumatoid arthritis.2

SS is a multisystem autoimmune disease with a prevalence of 0.1%–3% in the general population,6 ,7 characterised by lymphocytic infiltration of exocrine glands and other organs. Involvement of lacrimal and salivary glands results in ocular and oral dryness, the main features of the disease.8 Other system involvement is common;9 approximately one-third of patients with pSS have extraglandular systemic findings, including non-visceral and visceral manifestations.10 Patients with SS are likely to experience greater pain, fatigue and disability, as well as depressed mood and cognitive symptoms, compared with individuals without SS.11 Lymphoma is a concern in pSS, occurring in approximately 5% of patients, and the risk increases with disease duration.12 ,13 One study reported that patients with SS had a 44-fold higher relative risk of lymphoma compared with age-, sex- and race-matched controls.14 Lymphoma is one of the most serious complications of SS and a source of significant mortality. Approximately one in five deaths in patients with pSS is caused by lymphoma.13 Although SS is a multisystem syndrome with organ-specific and systemic manifestations, symptoms do not always present concurrently, making diagnosis difficult. Early recognition is important to reduce morbidity and healthcare costs, and improve quality of life.11

A small-scale, retrospective study at a tertiary centre estimated that approximately one-quarter of patients with dry eye have underlying SS.12 Thus far, no prospective studies have been published assessing the rate of SS in large dry eye cohorts. A study carried out prospectively by the Sjögren's International Collaborative Clinical Alliance (SICCA) assessed the SS rate in a different patient population, including patients who presented with dry eye OR dry mouth.13 Here, we report the prevalence of SS in a large, prospective cohort of patients at eyecare practices in the USA, including optometry, primary, secondary and tertiary ophthalmologic centres, with clinically significant ADDE, and a comparison of clinical and serologic parameters between patients with and without SS to identify its clinical predictors.

Materials and methods

Adult patients with dry eye who participated in a Phase I/II, prospective, randomised, multicentre clinical trial (A Prospective, Randomised, Placebo and Active Comparator Controlled Study of CP-690,550 in Subjects With Dry Eye, sponsored by Pfizer Inc, and registered with #NCT00784719) were enrolled at 23 centres in the USA. The study was carried out in accordance with the ethical principles in the Declaration of Helsinki, and the International Conference on Harmonisation good clinical practice guidelines. The final protocol was reviewed and approved by the institutional review board and the independent ethics committee of the investigational centres. Patients provided informed written consent.

Clinical trial primary endpoints are reported elsewhere.15 A secondary aim was to assess the SS rate in patients with clinically significant ADDE, and determine its clinical and serologic predictors. An evaluation of the prevalence of SS, type of SS (primary vs secondary), proportion of newly versus previously diagnosed cases, and a comparison of ocular and serologic findings at baseline between patients with and without SS was conducted at entrance visit.


Adult patients with a diagnosis of dry eye for ≥6 months and the following criteria were enrolled: Schirmer test without anaesthesia: ≥1 and ≤7 mm; sum of corneal fluorescein staining score: ≥4 (National Eye Institute (NEI) Scale)16; moderate-to-severe symptom score on the modified Ocular Comfort Index (OCI) questionnaire.

Key exclusion criteria included evidence or history of haematopoietic, lymphoproliferative or malignant disorders, other ocular disorders not caused by dry eye, prior/current use of systemic immunosuppressive or immunomodulatory therapies, use of topical cyclosporine or topical steroids within 1 month of screening, contact lens wear within 2 weeks of screening and/or during study participation, ocular surgery within 4 months of screening, and significant eyelid disease with anterior and/or posterior blepharitis with primarily evaporative dry eye.

Diagnosis of SS

Diagnosis of SS was based on the American–European Consensus Group 2002 revised criteria,5 requiring at least four out of six criteria, or three out of four objective criteria, to be present. The six criteria include: subjective and objective ocular dryness; subjective and objective oral dryness; presence of Sjögren-specific antibody A (SSA)/Ro and/or Sjögren-specific antibody B (SSB)/La; and positive minor salivary gland biopsy. Patients were prospectively evaluated for subjective and objective dry eye, subjective dry mouth and SS serology, but were not required to have assessments for objective signs of dry mouth (salivary flow rate, sialography or scintigraphy) or undergo minor salivary gland biopsy.

Patients defined as having pSS were those who had a previously established diagnosis of SS and no other autoimmune connective tissue disease, or patients with positive dry eye signs and symptoms, subjective dry mouth symptoms, and a positive test for SSA and/or SSB. Patients defined as having sSS were those who had a known underlying autoimmune connective tissue disease (eg, rheumatoid arthritis, systemic lupus erythematosus, scleroderma, Wegener's granulomatosis, primary biliary cirrhosis, mixed connective tissue disease) with either a previously established diagnosis of SS, or objective dry eye signs, subjective dry eye symptoms, subjective dry mouth symptoms, and a positive autoantibody test (SSA, SSB, rheumatoid factor (RF), or antinuclear antibody (ANA)).

Ocular assessments

Assessments were performed in the following order: Ocular Surface Disease Index (OSDI) and OCI patient symptom questionnaires, tear breakup time (TBUT), corneal fluorescein staining, Schirmer test without anaesthesia, conjunctival lissamine green staining and Schirmer test with anaesthesia.

TBUT was measured using 5 μl of 2.0% non-preserved sodium fluorescein instilled into the lower forniceal conjunctiva. Corneal staining was graded using a 0–3 scale in five corneal zones (central, inferior, nasal, temporal and superior) identified by the NEI/Industry workshop (total score range: 0–15).17 Schirmer test without anaesthesia was performed ≥2 min after corneal staining assessment with eyes closed using a Schirmer filter paper strip (EagleVision 5×35 mm Whatman No. 1, Kent, UK), and read at 5 min.

Conjunctival staining was assessed using impregnated paper strips (Akorn Lissamine Green Strips, Northeast Medical Products, Old Saybrook, Connecticut, USA); the Oxford grading system was used for scoring, which divides the bulbar conjunctiva into two zones (nasal and temporal) graded independently using a 0–5 scale (total score range: 0–10).18 Schirmer test with anaesthesia was performed using proparacaine 0.05% eyedrops using the test strips mentioned above.

Serologic testing

Blood was collected at baseline for serologic testing, which was performed at a single commercial/independent clinical laboratory (Quintiles Laboratories North America, Marietta, Georgia, USA), and included SSA, SSB, RF and ANA antibodies.

Statistical analysis

Analysis of variance compared baseline sign and symptom endpoints between patients with SS versus patients without SS. Logistic regression evaluated the association between status of SS and baseline RF and ANA serologic testing results.



The majority of patients were female (non-SS: 75.5%; pSS: 100%; sSS: 94.1%), and Caucasian (table 1). Patient disposition is displayed in figure 1.

Table 1

Baseline demographic features of patients with clinically significant aqueous-deficient dry eye

Figure 1

Study design and patient disposition in patients with clinically significant aqueous-deficient dry eye syndrome. *Patients completing treatment through Week 8. BID, twice daily; Other, lost to follow-up, no longer willing and ‘other’; QD, once daily.

Of 327 patients enrolled, 38 (11.6%) had SS: 21/38 (55.3%) with pSS and 17/38 (44.7%) with sSS. In this population, an additional 3% (n=9) of patients were newly diagnosed with SS during the study, based on the applied diagnostic criteria. The proportion of established SS diagnoses prior to enrolment versus new diagnoses were 17/21 (81.0%) versus 4/21 (19.0%) for pSS, and 12/17 (70.6%) versus 5/17 (29.4%) for sSS.

Patients with SS (both pSS and sSS) had statistically worse mean conjunctival lissamine green staining (p<0.0001), corneal fluorescein staining (p<0.0001), Schirmer test (with and without anaesthesia (both p<0.01)) and total OSDI scores (p<0.001), compared with patients without SS (table 2). There was no significant difference in TBUT or total OCI scores between patients with and without SS.

Table 2

Summary of dry eye signs and symptoms in patients with clinically significant aqueous-deficient dry eye

For conjunctival staining in patients with SS, severity scores <3 and ≥3 were seen in 63.2% and 36.8% of patients, respectively, in the temporal conjunctival zone, and in 47.4% and 52.6% of patients, respectively, in the nasal zone (see supplementary figure S1). There was no indication of conjunctival staining concentrated in any specific zone in patients with SS. Corneal staining in patients with SS revealed that in the central and superior zones, the greatest proportion of patients had a severity score of 1 (47.4% in each of the two zones) (figure 2A). More severe staining scores were observed in the inferior corneal zone, where 81.6% of patients had a severity score ≥2, compared with the superior zone, where 71.1% of patients had a severity score <2 (figure 2B).

Figure 2

Severity of corneal staining by location in patients with clinically significant aqueous-deficient dry eye syndrome. (A) 0–3 and (B) 2–≥2.

A significant association was observed between SS status (pSS vs non–SS) and RF (<10 vs ≥10) test result (OR: 4.8 (95% CI 1.90 to 11.95); p=0.009) (figure 3A) and ANA (<1 : 160 vs ≥1 : 160) test result (OR: 13.9 (95% CI 5.23 to 36.82); p<0.0001) (figure 3B).

Figure 3

Proportion of patients with non-Sjögren's syndrome (SS), primary SS and secondary SS in patients with clinically significant aqueous-deficient dry eye syndrome by (A) rheumatoid factor score and (B) antinuclear antibody score. *One patient with primary SS did not have an ANA score at baseline. ANA, anti-nuclear antibody; RF, rheumatoid factor; SS, Sjögren's syndrome.


Studies assessing the prevalence of SS among patients with dry eye are scarce.1 ,19 One prospective study by SICCA assessed the rate of SS in a different patient population than the present study, including those who presented with dry eye or dry mouth symptoms (n=1208).19 Although a large proportion (85%) of patients reported dry eye symptoms, they did not necessarily have signs of dry eye. In that cohort, 510 patients had SS; however, the study did not report whether diagnosis was made prior to, or as a result of the study.19 Another small-scale, retrospective study reported that one-tenth of patients presenting with dry eye to a tertiary centre had pSS, one-third of whom carried the diagnosis at presentation, while half were newly diagnosed by initial evaluation.1 One-sixth required further investigation, including minor salivary gland biopsy, for definitive diagnosis. In comparison, our study found that 6.4% of patients with ADDE had pSS; one–quarter of these were newly diagnosed as part of the study, although all had clinically significant dry eye. The prevalence of sSS was 5.2%, and 29.4% of those were newly diagnosed. The difference in prevalence between these studies could be due to several reasons. The previous research was carried out at a tertiary centre and may have included more severe cases as the majority were referrals. Our study was performed in a patient population entering a clinical trial conducted at 23 sites ranging from optometry practices to tertiary care ophthalmological centres. However, rates of pSS and sSS here might not necessarily be applicable to the general population. Furthermore, assessments of objective dry mouth findings, or minor salivary gland biopsy, were not performed, which might have led to underdiagnosis of SS. Nevertheless, the aim of the study was to determine whether additional patients without history of SS could be diagnosed in an ophthalmologic setting using parameters that could be easily assessed.

In this study, mean conjunctival staining, corneal staining, Schirmer test (with and without anaesthesia), and total OSDI scores were statistically worse in patients with SS (both pSS and sSS) compared with patients with non-SS. However, there was no correlation between temporal conjunctival staining and diagnosis of SS. A previous study suggested that temporal bulbar conjunctival staining with Rose Bengal is important for diagnosis of SS and differentiation of pSS from non-SS dry eye.20 Non-invasive Rose Bengal staining of the temporal conjunctiva had a detection sensitivity of 96.1%, with a specificity of 56.2%.20 However, in this study, temporal conjunctival staining score with lissamine green was found not to correlate with a diagnosis of SS; instead, inferior cornea staining was more common in patients with pSS. Constant bathing of the inferior corneal zone with the lower tear meniscus is well recognised. A previous study demonstrated no difference in the height of the lower tear meniscus in normal subjects and patients with ADDE, in contrast with significantly decreased upper-tear meniscus in patients with ADDE.21 It is also known that there is an imbalance in tear fluid cytokines, and significantly increased levels of IL-1 α, IL-6, IL-8, TNF-α and TGF-β1 RNA transcripts have been reported in patients with SS compared with controls.22 One hypothesis may be that increased staining in the corneal inferior zone in patients with SS could be the result of constant direct contact between the tear meniscus, containing inflammatory mediators, and the inferior portion of the cornea. If this hypothesis is valid, one would expect conjunctival staining in the area of the inferior meniscus to be greater; unfortunately, inferior conjunctival staining was not scored separately in this study. It is also common to observe inferior staining in patients with blepharitis; therefore, perhaps the increased contact with the lower lid margin over the inferior cornea also plays a role. Although this is less likely in our study, as patients with clinically significant blepharitis were excluded to not confound the data. Nevertheless, many patients with clinically significant aqueous tear deficiency do have a degree of meibomian gland dysfunction, and a combination of ADDE and evaporative dry eye is common.

The results of this study support calling for an increased index of suspicion for ophthalmologists caring for dry eye, and the recommendation that the presence of SS should be assessed in all patients with clinically significant ADDE. SSA and SSB antibodies are reportedly the hallmark antibodies in pSS, but are only present in 60%–70% of cases presenting to rheumatology clinics.23 This rate decreases to 50% in patients who present to ophthalmology clinics primarily with dry eye symptoms.1 A study in patients with SS presenting to rheumatology clinics concluded that RF and ANA were predictive and had prognostic value for patients who did not fulfil the 2002 revised criteria for pSS.24 Our results suggest that ANA positivity is associated with an approximately 14-fold increase in the likelihood of pSS versus non-SS, suggesting that, although not clinical practice, patients with significant dry eye signs and symptoms plus a positive RF or ANA result should be offered diagnostic testing even if their SSA or SSB results are negative. This finding was also noted in a study by Whitcher and colleagues, which reported that patients with SS were more likely to be positive for RF and ANA as well as having higher IgG levels.19

In conclusion, SS should be assessed in patients with clinically significant ADDE. Using a portion of the current diagnostic criteria that can easily be assessed in an ophthalmologic setting, we were able to newly diagnose an additional nine patients (n=9/39, >30%) with SS in a large cohort of patients with clinically significant ADDE. These results suggest that as well as SSA and SSB, ANA and RF antibodies are also important in helping to determine SS diagnosis, and the necessity for further evaluations in patients who have tested negative for SSA and SSB.


The authors would like to thank the A3921034 investigators and study team.


View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Contributors EKA: made substantial contributions to study design, and was involved in the acquisition of data. ML, EK and MZ: made substantial contributions to study conception and design, and data analysis. All authors were involved in the interpretation of the data. All authors discussed and agreed the content of the manuscript before writing took place. All authors reviewed and approved the manuscript's content prior to submission, and jointly agreed to submit the final version.

  • Funding This study was sponsored by Pfizer Inc. Pfizer Inc participated in the design of the study, conduct of the study, data collection, data management, data analysis, interpretation of the data, and the preparation, review and approval of the manuscript. Editorial support, under the direction of the authors, was provided by Karen Irving of Complete Medical Communications, and funded by Pfizer Inc.

  • Competing interests MZ and EK are employees of Pfizer Inc. SH(Melissa)L was employed by Pfizer Inc at the time of the study. EKA has no conflicts of interest.

  • Ethics approval The final protocol was reviewed and approved by the institutional review board and independent ethics committee of the investigational centres.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • At a glance
    Harminder Singh Dua Arun D Singh