dry eye dx and tx
Diagnosing
Dry Eye: 100+ Years Post-Schirmer
If you could design a new diagnostic test for dry eye, what characteristics would it have? Some of our existing tests have shown the test of time (Schirmer, circa 1903) while others were ahead of their time (Touch tear microassay).
Ideally, a new test would be sensitive enough to detect small yet meaningful differences between dry eye and non-dry eye patients. It would be valid and repeatable and wouldn't result in false positives or incorrect diagnoses. Of course, the test would be inexpensive, foolproof, fast and billable to any insurance plan. Sounds perfect, right? How close are we to this ideal test?
Gold Standard Diagnostic Tests
No single clinical test is used to diagnose dry eye. In general, dry eye is a gestalt diagnosis based largely on patient-reported symptoms. Clinicians often perform a battery of diagnostic tests as part of an ocular surface disease work-up, including symptom assessment, measurement of aqueous production (Schirmer or phenol red thread test), assessment of corneal/conjunctival integrity (fluorescein and lissamine vital dye staining) and evaluation of tear stability (tear breakup time [TBUT]).
How do we interpret the results? Most clinicians weigh the burden of evidence: More abnormal tests indicate a greater likelihood of disease or more severe disease. I believe clinicians give symptoms more weight when clinical tests are normal or borderline (no staining, borderline TBUT, normal Schirmer). If the tests are obviously abnormal, treatment is easier to monitor.
Screening Tests
Clinicians perform tests to screen for and diagnose disease. Any new test is compared to truth — a new test to diagnose cancer is evaluated on patients known to have cancer by biopsy. Ideally, you want the screening test (faster, easier, less invasive) to reproduce the result of the more difficult and costly gold standard test.
A screening test with high sensitivity can correctly identify those with known disease. High specificity results if a test correctly classifies patients who don't have the disease. Sensitivity and specificity are trade-offs — increasing sensitivity by selecting a different abnormal cutpoint results in reduced specificity. With most dry eye tests, specificity is higher than sensitivity for the cutpoints generally reported in the literature. In other words, most dry eye tests are better at ruling out dry eye.
Are the Tests Worth the Effort?
So why do we perform these tests? Patient-reported symptoms tell us what's wrong, in the patient's own words. Improving or removing symptoms is important to both patients and clinicians.
Corneal and conjunctival staining is an indicator of overall ocular surface health — so if the ocular surface is damaged, the proverbial Pandora's Box of possible corneal infections is open.
While the Schirmer test can be wildly unrepeatable, it is so only for those who have mild or no dry eye disease. It's highly sensitive and repeatable for diagnosing aqueous deficient dry eye.
What Will the Future Bring?
Ultimately, research will identify the underlying biochemical and physiological etiology of specific symptoms, resulting in clinical screening tests that are microassay-based. We'll develop foolproof methodology of collecting tiny volume tear samples and the technology to easily evaluate critical biochemical tear biomarkers in the office.
In the future we'll need to keep up with emerging technology related to ocular surface disease as well as other ocular conditions. Are you ready?
Dr. Nichols is an associate professor at The Ohio State University College of Optometry in the area of dry eye research.