Associations of long-term PM2.5 exposure and tear muramyl dipeptide with dry eye disease in children with diabetes: a prospective cohort study

This prospective cohort study aimed to evaluate the independent associations of long-term PM 2.5 exposure and tear muramyl dipeptide (MDP) levels with dry eye disease (DED) risk in children with diabetes and to explore their potential interaction. A total of 57 children with diabetes from the ongoing Shanghai Children and Adolescent Diabetes Eye Study (SCADE) were included in this prospective cohort analysis, and 54 completed the one-year follow-up. DED was diagnosed using standardized criteria. Tear samples were collected with test strips, and MDP levels were quantified by mass spectrometry. Air pollutant data were obtained from the Air Quality Historical Data Platform, and individual exposure levels were estimated using questionnaire-assisted exposure duration assessment. Normality testing, Receiver operating characteristic (ROC) analysis, multivariate logistic regression, and interaction modeling were performed to evaluate the associations between long-term exposure to major air pollutants (PM 2.5 , PM₁₀, SO₂, O₃, NO₂, and CO), tear MDP levels, and risk of DED. Multicollinearity assessment was conducted before interaction analysis. Tear MDP levels were significantly higher in the DED group (2.355 ng/mL (1.281–3.727 ng/mL) ) compared with the non-DED group (0.991 ng/mL (0.581–1.473 ng/mL), p < 0.0001). Among pollutants, only PM 2.5 showed a significant association with DED. Each standard deviation (SD) increase in PM 2.5 was associated with an increased risk of DED (odds ratio [OR] = 1.072, 95% confidence interval [CI]: 1.023–1.123, p = 0.004). Similarly, each SD increase in tear MDP level was associated with an elevated risk of DED, with OR = 2.219, 95% CI: 1.190–4.134, p = 0.012. Long-term PM 2.5 exposure and elevated tear MDP levels were both independently associated with an increased risk of DED in children with diabetes. In addition, tear MDP may serve as a potential biomarker for disease monitoring.