Tear osmolarity may function as a biomarker of dry eye severity

Study finds strong correlation with disease severity score, especially in patients with mild or moderate dry eye disease.

Issue: July 2010

Clinical testing of an osmolarity screening device designed for in-office use showed strong correlation with a diagnosis of dry eye based upon composite clinical signs and symptoms of dry eye disease, particularly in mild to moderate disease.

In a recent study, tear osmolarity tested with the TearLab Osmolarity System more accurately predicted the severity of a patient’s dry eye disease compared with tests traditionally used to make a diagnosis.

According to a poster presented at the joint meeting of the American Academy of Ophthalmology and Pan-American Association of Ophthalmology, “osmolarity was found to have superior dynamic range and resolution compared to the other signs, especially in the mild to moderate cohort.”

“In that mild to moderate area, any one test becomes less reliable for both diagnosis and for serving as an indicator of severity,” Gary N. Foulks, MD, FACS, lead author of the poster, said. “If you looked for the correlation, the best correspondence was with tear osmolarity.”

According to the poster, the TearLab Osmolarity System is designed to measure osmolarity of a tear sample, approximately 50 nL collected from the inferior lateral tear meniscus of the ocular surface, using electrical impedance. In clinical testing, tear sampling has been performed in as few as 5 to 6 seconds, with calculation of osmolarity taking less than 20 seconds, Dr. Foulks said. While tear osmolarity testing is currently performed in laboratories, the TearLab system is intended to be an office-based modality for in vitro diagnostic tear testing.

The TearLab Osmolarity System is currently cleared by the U.S. Food and Drug Administration as a 510(k) device.

Strong correlation with clinical assessment

In the study, 299 patients were assigned a disease severity score based on a symptom score of 5 or greater on the Ocular Surface Disease Index (OSDI), as well as one eye exceeding a preselected score on two of five classically used diagnostic tests: tear film breakup time of 7 or less, Schirmer test score of less than 7, corneal staining value greater than 0, conjunctival staining value greater than 0 and/or meibomian score of greater than 5. Osmolarity score and each individual diagnostic score were then tested against the severity score for degree of correlation.

In patients with severe dry eye, tear film breakup time showed the strongest correlation with the severity score, with 98.7% sensitivity, while osmolarity had 94.7% sensitivity in this category of patients. However, osmolarity outperformed each diagnostic test in normal eyes and in eyes with mild or moderate dry eye.

While previous studies on tear osmolarity have used a diagnostic cutoff score of below 316 mOsms/L to designate normal eyes, the current study found a stronger correlation when a score of 308 mOsms/L was used as a cutoff. The 316 mOsms/L score produced specificity of 98.7%, and sensitivities of 47.0% and 89.3% in mild/moderate and severe eyes, respectively; however, when the 308 mOsms/L cutoff score was used, the test produced specificity of 88.0%, and sensitivities of 74.5% and 94.7% in mild/moderate and severe eyes, respectively.

Of particular note, according to Dr. Foulks, is the strong correlation in mild or moderate patients, “where the variability of the standard tests, like staining, Schirmer and tear breakup, is very great.” By comparison, sensitivity for other diagnostic tests in mild/moderate eyes were: conjunctival staining, 69.1%; tear film breakup time, 67.8%; meibomian grading, 49.7%; corneal staining, 41.6%; and Schirmer test, 27.5%.

Beyond the score

While the adjusted cutoff value for normal eyes provided greater overall accuracy in identifying disease severity for eyes in the study, the use of a single particular value from an osmolarity test as an absolute diagnostic cutoff point may be misleading, according to Dr. Foulks.

Patients in the study with normal or severe disease severity scores were more likely to have a repeatable value on their osmolarity test, but patients with mild or moderate dry eye typically had greater variability on retest, Dr. Foulks said. That variability, he said, is the hallmark of disease and may be due to the decreased ability of the eye with mild or moderate dry eye to respond to environmental stimuli or insult, such that “a stress that would disturb the tear osmolarity in a mild to moderate patient would be compensated for in a normal patient with adequate reflex tearing.”

“That complicates things a bit in terms of using this test as a diagnostic instrument,” Dr. Foulks said, “because it means you need to be evaluating values from one eye vs. the other, and also in those patients that are suspected of having mild dry eye, it may take more than one measurement to prove your point.”

That variability in diagnostic acumen, however, may reflect greater understanding of the dry eye disease process. According to Dr. Foulks, more recent research on dry eye has revealed the various compensatory mechanisms for different types of dry eye and how they fail, that these compensatory mechanisms may change the osmolarity of the tear, and that these compensatory mechanisms may persist to some degree even in patients with dry eye.

In actuality, then, tear film osmolarity may function as a more precise diagnostic tool, because it may be able to detect dry eye in patients before they reach the level of severity that would be detectable by ocular staining. As well, because tear osmolarity correlates strongly with disease severity, change in test scores may function as a marker of disease progression in response to therapy regimens as patients are followed longitudinally.

“It means we can use the instrument and the measurement earlier in the course of the disease to establish the diagnosis but also to see how patients are responding to any particular management scheme,” Dr. Foulks said.

Future directions

If tear osmolarity can function as a surrogate biomarker of disease severity, and if it can give clinicians an accurate portrayal of response to therapy over time, the testing modality may also be important for clinical trials involving new therapies for dry eye. Currently, the FDA does not recognize surrogate biomarkers as an endpoint upon which to base approval of new drugs or therapies.

“Some of the things that are being done right now are to provide validation of osmolarity as a reliable clinical endpoint not only for the diagnosis of dry eye, but also for monitoring the treatment of dry eye,” Dr. Foulks said.

Additionally, while osmolarity testing may indicate a dry eye state, it does not indicate the underlying cause of the dry eye. As the technology evolves, testing may be adapted to detect the presence and/or level of a protein produced by the lacrimal gland, for example, which could indicate aqueous tear deficiency, he said.

TearLab is also in the process of applying for a waiver categorization under the Clinical Laboratory Improvement Amendments that would allow the device to be used in the clinical setting with the physician serving as the laboratory director. – by Bryan Bechtel

Gary N. Foulks, MD, FACS, can be reached at University of Louisville Department of Ophthalmology and Visual Science, 301 E. Muhammad Ali Blvd., Louisville, KY 40202, U.S.A.; +1-502-852-7665; e-mail: gnfoul01@gwise.louisville.edu.