Dry eye disease symptom change independent of tear osmolarity

Oct 9 2013

By Lynda Williams, Senior medwireNews Reporter

Tear osmolarity is a poor indicator of treatment response in patients with dry eye disease (DED), suggests research published in the American Journal of Ophthalmology.

Changes in tear osmolarity did not significantly correlate with changes in Ocular Surface Disease Index (OSDI) scores for 186 DED patients assessed on two occasions a mean of 137 days apart. Nor did tear osmolarity significantly predict corneal fluorescein staining.

“The preponderance of currently available data suggest that tear osmolarity measurement can in fact aid in the diagnosis of DED with good sensitivity and specificity, and our data do not refute this,” explain Reza Dana (Harvard Medical School, Boston, Massachusetts, USA) and co-workers.

“However, our data suggest that tear osmolarity measurements have significant limitations in prospective follow-up of patients as a marker of whether patients are doing ‘clinically better’ or not.”

By contrast, there was a modest but significant correlation (R = 0.31) between changes in OSDI score and corneal fluorescein staining, which the researchers note is stronger than was previously believed.

The relationship between OSDI score and corneal fluorescein staining persisted when assessing only eyes with the highest osmolarity or staining, and when assessing patients with an OSDI score over 12 indicating ocular surface disease. But there was still a lack of correlation between tear osmolarity and OSDI score and corneal fluorescein staining in these groups.

And multiple regression analysis confirmed that changes in corneal fluorescein staining but not tear osmolarity significantly predicted symptom alteration.

Nevertheless, Dana et al say that the modest correlation between symptom change and corneal fluorescein staining should not be overstated.

“Dry eye disease, like many other chronic diseases (eg, arthritis, inflammatory bowel disease, multiple sclerosis, etc), is multifactorial, and it is likely unrealistic to expect any one clinical sign, or biomarker, to correlate highly with patient symptoms,” they comment.

“It very well may be that a combination of clinical variables including measurement of surface epitheliopathy/staining, along with various biomarkers, will be the most reliable prognosticator for response to therapy.”

Licensed from medwireNews with permission from Springer Healthcare Ltd. ©Springer Healthcare Ltd. All rights reserved. Neither of these parties endorse or recommend any commercial products, services, or equipment.