Validation of a New Questionnaire to Screen Dry Eye Patients for Sjögren's Syndrome: preliminary findings

Abstract

Purpose : Ophthalmologists currently lack an evidence-based screening tool to aid in the identification of Sjögren’s Syndrome (SS) among dry eye patients, which leads to delays in diagnosis and treatment. We evaluated the validity of a previously developed screening algorithm in a new dry eye cohort.

Methods : Participants included patients who were over the age of 18, had no prior autoimmune work up, and who had complaints of dry eye for at least three months. Study procedures included the completion of a questionnaire to assess both ocular and systemic questions, an ocular surface examination, measurement of unstimulated whole salivary rate and serological testing. Lip biopsies were performed when indicated to confirm the diagnosis of SS. The screening algorithm score was determined using questionnaire responses, as well as the results of the assessment of tear film break-up time and conjunctival staining with lissamine green. A participant was diagnosed with SS if they met the 2016 American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) criteria. Patients were stratified into having a low, medium or high likelihood of having SS based on their screening algorithm score.

Results : To date, 65 patients have been enrolled, with 83% of them being female. The mean age (standard deviation) was 56 (17.5) years. The percentage of SS diagnoses increased with the likelihood score (6.3% for score ≤3, 18.2% for score of 4 to 6, 22.2% for score of 7, and 44.4% for score 8 or above, linear trend p=0.02). The area under the ROC curve for screening for SS using the likelihood score was 0.70 (95% CI: 0.55-0.84). A cut-off of greater than or equal to 7 points for the likelihood score had a sensitivity of 67% (95% CI: 42%-85%) and a specificity of 66% (95% CI: 52%-78%).

Conclusions : We demonstrated that a higher likelihood score is associated with a higher risk of having SS. Once validated, this algorithm has the potential to serve as an effective screening tool for eye care providers to determine the likelihood of Sjogren’s in their dry eye patients, allowing for a diagnosis and treatment to take place sooner. This likelihood scoring system relies on a combination of a brief series of questions combined with routine ocular surface signs and would thus be easily implemented.

This is a 2020 ARVO Annual Meeting abstract.