Purchase this article with an account.
Benjamin Sullivan, Stephen Zmina, Matthew Zmina, Michael Berg; Real World Analytical Performance of the TearLab Osmolarity System with an Enhanced Temperature Sensor. Invest. Ophthalmol. Vis. Sci. 2013;54(15):4336. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Tear film instability is an inherent feature of dry eye disease. The blink-to-blink variability that characterizes the disease, but disappears in normal subjects or treated patients, has caused many to confuse expected eye-to-eye differences for analytical variation of the TearLab Osmolarity System. The purpose of this study was to report real world performance of first generation and recently updated TearLab Osmolarity System, in order to quantify the relative difference between analytical and biological variability.
Prior to FDA CLIA waiver, 113 TearLab systems were installed under the moderately complex program, which required each doctor to be independently audited by state inspectors to ensure that the systems matched manufacturer claims. Sites performed 20 Normal and 20 High Control tests on one day, followed by ten sequential days of tests at each level, with expected results of 297 mOsm/L and 338 mOsm/L at each level. For the enhanced systems, 240 High Control tests were performed on three units at temperatures of 20°C, 23°C, 25°C and 30°C using three batches of test cards at one site.
First generation controls averaged 295.4 ± 4.0 mOsm/L and 338.6 ± 4.8 mOsm/L respectively for the 20/20 tests, with an average CV from the sites of 1.4% and 1.6% at each level. Over ten days, the controls averaged 294.1 ± 4.8 mOsm/L and 336.6 ± 5.2 mOsm/L respectively, with an average CV of 1.6% and 1.6% at each level. Enhanced systems reported a mean osmolarity of 340.3 ± 3.5 mOsm/L, with an average CV of 0.8% across each temperature on High Controls. Using these standard deviations, a Gaussian error model suggests that 1.6±0.4 tests are needed to produce a measurement within 3 mOsm/L (< 1% error) of the true value, validating the requirement that both eyes should be tested as part of a standard TearLab osmolarity examination, and that more than two tests give little additional information.
Preliminary results of the enhanced TearLab system suggest it meets or exceeds the precision of the first generation system. The analytical precision of the TearLab Osmolarity System is a minor component of the expected variability in dry eye disease, and eye-to-eye difference can be considered a clinical indicator of dry eye rather than the result of analytical imprecision.
This PDF is available to Subscribers Only