Predictors of Discordance between Dry Eye Symptoms and Signs: Insights from the Sjögren’s International Collaborative Clinical Alliance

Abstract

Purpose : The diagnosis and management of dry eye syndrome (DES) could be complicated by the discordance between DES-related symptoms and signs. We performed a cross-sectional study to investigate the factors of and develop predictive models for the discrepancy in DES symptomatology.

Methods : We used data from 3455 participants aged 21 to 89 years from the Sjögren’s International Collaborative Clinical Alliance study. A wide spectrum of potential indicators, including demographic, mental and physical measures, rheumatology diseases, symptoms and signs, ocular traits and medications, were analyzed. We performed a multivariable stepwise linear regression model with backward elimination and Bayesian information criteria to select predictors for the discrepancy in DES symptomatology, which was defined as the difference between the rank score of ocular surface disease index-6 and the rank score of ocular staining score (OSS). Model performance was assessed using R².

Results : 11 predictors, such as “eye discomfort in places with low humidity”, “Sjögren’s disease (SjD) status”, “eye irritation while watching TV or working on a computer”, and “light sensitivity”, remained in the final models, achieving an R² of 0.68 (confidence interval (CI): 0.66 – 0.69). Notably, ocular symptoms emerged as the most influential contributors to the variance in the outcome (R² = 0.54, 95% CI: 0.52 - 0.56), substantially outperforming the collective predictive power of all other factors (R² = 0.33, 95% CI: 0.30 - 0.36). Conversely, demographics exerted the least influence on the predictive performance (R² = 0.10, 95% CI: 0.08 - 0.12). Upon stratifying the analysis by SjD status, the selected factors showed reduced predictive performance in the non-SjD participants compared to that in the SjD group.

Conclusions : Certain ocular symptoms, such as eye irritation, discomfort and light sensitivity, might be more subjective and less indicative of ocular signs. Additionally, our model, even excluding ocular symptoms, showed an improved prediction compared to the previous study, suggesting that OSS could be a more sensitive indicator of ocular surface disturbance compared to other common ophthalmologic tests.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.