Abstract
Purpose:
To determine the tear film osmolarity of patients with rosacea to further understand the pathophysiology of the disease and optimal management.
Methods:
This is a prospective, nonrandomized, clinical, single-center study designed to investigate the osmolarity in tear samples of patients with rosacea and control patients. Twenty patients [aged 64 years (31-89 years), 16 women and 4 men] with mild to severe rosacea, and twenty-five control patients [aged 45 years (20-76 years), 19 women and 6 men] were enrolled in the study. Rosacea patients were recruited from a cornea clinic where they previously had received the diagnosis. Tear samples were collected directly from the inferior lateral tear meniscus, and osmolarity was analyzed using the TearLabTM osmometer. This was accompanied by a Schirmer test with anesthesia as well as the completion of the Ocular Surface Disease Index questionnaire. TearLabTM osmolarity was analyzed as a function of patient-level and eye-level variables using tobit regression (Tobin, 1958) in which below-range values are addressed using survival analysis methodology but which is otherwise similar to linear regression. Correlations between data from eyes from the same patient were addressed by adding a random effect for patient to the model.
Results:
The rosacea group had a significantly higher mean age and OSDI scores compared to the control group. A significantly higher proportion of rosacea patients had punctal plugs and were using tetracycline or artificial tears relative to the control group. In the multivariate tobit regression analysis of osmolarity by patient characteristics, after adjusting for all other variables in the model, tear osmolarity between eyes of rosacea (mean 300 mOsm/L right eye, 295 mOsm/L left eye) and control patients (mean 298 mOsm/L right eye, 296 mOsm/L left eye) did not differ significantly. However, eyes of patients using tetracycline, which were only in the rosacea group, had a significantly higher tear osmolarity than eyes not using tetracyclines with an estimated difference in mean osmolarity of 16.0 mOsm/L.
Conclusions:
Tear film osmolarity in rosacea patients did not significantly differ from control patients, but we did observe a significant effect of tetracycline on tear film osmolarity. This may be due to the fact that rosacea patients using tetracycline had more advanced disease, characterized by worse evaporation.
Keywords: 486 cornea: tears/tear film/dry eye