Letters ization of testing techniques would remove these variables. In- We believe the high variability that we found in our study, tersession variability was noted, while blink-to-blink intervals, which was consistent with results reported previously by other 3,5 2,3 incomplete blinking, and open-eye timing during instru- research groups, was likely due to a combination of fac- ment application were not held constant. Thus, alterations in tors, including some of those described by Aquavella as well the tear film dynamics may have varied during the time in which as other factors such as the disease process (in dry eye groups) osmolarity measurements were obtained and we cannot as- and measurement error from the instrument. Regardless of the sume that the reestablishment of the tear film and its associ- source of this variation, the resulting high variability of osmo- ated osmolarity do not vary with incomplete or erratic blinks. larity measurements with the TearLab system in patients with Was there never a time when the probe unintentionally im- dry eye and other patients makes it difficult to interpret tear pacted the eyelid margin, conjunctiva, or both, stimulating re- osmolarity results in clinical practice. flex tears that altered the osmolarity reading obtained?
JAMA Ophthalmology – American Medical Association
Published: Dec 1, 2015