Abstract
Purpose: The purpose of this study was to examine the inherent precision and accuracy of TearLab Osmolarity System using salt solutions, including solutions of very high osmolarity (>360 mOsm/L).
Materials and methods: Ten salt solutions with osmolarity between 286 mOsm/L and 394 mOsm/L (increments of 12 mOsm/L) plus an additional solution of 400 mOsm/L were tested twice on both the TearLab osmometer and a freezing point depression osmometer. For precision, we compared the two repeated osmolarity measurements of 11 solutions obtained from TearLab. For accuracy, we compared the averaged osmolarity measurements obtained from TearLab to those from the freezing point depression osmometer. For both precision and accuracy, Bland–Altman test of agreement was used.
Results: For precision, the upper 95% limit of agreement was 4.7 mOsm/L, and the lower 95% limit of agreement was −7.1 mOsm/L. The repeatability coefficient was 5.9 mOsm/L. For accuracy, the upper 95% limit of agreement was 4.8 mOsm/L and the lower 95% limit of agreement was −5.3 mOsm/L.
Conclusions: The present study is the first study to demonstrate that the TearLab in situ osmometer can precisely and accurately measure osmolarity of salt solutions, including those with very high osmolarity. Future studies to evaluate the precision and the accuracy of the machine in measuring complex fluids, such as tears, need to be done, and the clinical significance of measuring tear osmolarity in patients needs to be further determined.