Cracking the Code: Which Ocular Symptoms Predict Dry Eye Signs? Insights from a large international sicca registry

PURPOSE: The study aimed to identify symptom-based predictors of dry eye disease (DED) signs in the Sjögrens International Collaborative Clinical Alliance (SICCA) cohort. METHODS: Retrospective analysis examining 16 ocular symptoms (most graded 0-4) and artificial tear (AT) use (graded 0-3) as predictors of DED signs [abnormal ocular surface staining (OSS) and Schirmer test (ST), graded yes/no]. Symptoms included descriptors of spontaneous and evoked pain, visual complaints, and other metrics (e.g., redness). Logistic regressions were used for univariable and multivariable analyses, with OSS and ST as outcome variables. Hierarchical analyses were performed to identify symptom-based clusters. RESULTS: Mean age of the population (n=3514) was 53±13 years; 91% were female, 54% were Caucasian, and 44% met 2016 criteria for SjD. On multivariable analysis, OSS was positively related to AT use (odds ratio (OR)=1.51, 95% confidence interval (CI) 1.36-1.67), eye redness (OR=1.17, 95% CI 1.08-1.26), and blurred vision (OR=1.11, 95% CI 1.02-1.20). Similarly, eyes that felt dry, AT use, and blurred vision positively related to ST. Cluster analysis revealed three groups with varied symptom burden (low, mild-moderate and high). Mild-moderate cluster had higher frequencies of abnormal ocular signs and positive serologies (anti-SSA/Ro antibody, elevated Immunoglobulin-G) compared to the high symptom-burden cluster. CONCLUSION: AT use and blurred vision were stronger indicators of ocular surface and tear film abnormalities than pain symptoms, and their presence merits an ophthalmology referral. Symptom-based clustering revealed varied symptom-burden groups, likely driven by underlying neurotrophic and neuropathic mechanisms, highlighting heterogeneity within sicca presentations and the need to develop phenotype-based targeted therapies.