Abstract: Staining... what color, or does it matter?

OK so rose bengal and lissamine green basically performed the same except rose bengal was a little harder on the patients. But the really intriguing, if not new, finding here is that neither of them had any correlation to disease severity. And yes, thank you authors for using OSDI to rank disease severity. Maybe someday we really can end this silly business of telling patients they're fine because their corneal surface looks OK with little or no staining.

Staining Patterns in Dry Eye Syndrome: Rose Bengal Versus Lissamine Green.
Cornea. 2009 Jul 1. [Epub ahead of print]
Machado LM, Castro RS, Fontes BM.
From the *Department of Ophthalmology, State University of Campinas (UNICAMP), Campinas, Brazil; and daggerDepartment of Ophthalmology, Federal University of São Paulo (UNIFESP-EPM), São Paulo, Brazil.

PURPOSE:: To evaluate and compare corneal staining patterns of lissamine green (LG) versus rose bengal (RB) in patients with dry eye syndrome. Secondary objectives included addressing patient's comfort after instillation and to correlate disease severity with staining patterns.

METHODS:: Randomized, comparative, crossover series. Patients with previous diagnosis of mild to moderate dry eye syndrome were divided in 2 groups regarding dye instillation order (group A: RB first; group B: LG first). Both dyes were applied in regular intervals, and a staining score (van Bijsterveld scale) was used to correlate and compare the results. Disease severity was determined by the Ocular Surface Disease Index. Comfort was evaluated by patient's answer in an objective questionnaire.

RESULTS:: Sixty eyes of 30 consecutive patients (24 females and 6 males) were included. There was no statistical difference between groups regarding disease severity, sex, or age. LG and RB showed good clinical correlation in both groups (group A: r = 0.939, P < 0.001; group B: r = 0.915, P < 0.001). LG was better tolerated than RB (P = 0.003 in both groups). Overall, we found a low statistical correlation between disease severity and staining scores.

CONCLUSIONS:: Both LG and RB showed similar staining patterns. RB was found to provide greater patient discomfort. There was no correlation between disease severity (addressed by the ocular surface disease index questionnaire) and staining patterns (measured by the van Bijsterveld scale).