Aim The aim of the study was to examine the possibilities of measuring tear osmolarity in a general clinical setting, and to identify the barriers preventing the uptake of new methodologies for its measurement.

Methods Five non-contact-lens wearers were recruited to evaluate the diagnostic capability of the TearLab. Three osmolarity measurements were taken at 1 min intervals in the morning at 09:00, midday between 12:00 and 13:00 and afternoon at 16:00 for two consecutive days. Forty more osmolarity measurements were carried out at different times on one subject with low and one subject with high tear osmolarity over 4 months. The osmolarity of a standard solution, 290 mOsm/l, was measured 19 times alternatively with the TearLab by two examiners.

Results Consecutive tear osmolarity readings in an individual varied up to 35 mOsm/l, but an average over three readings was found to be a reliable indicator of tear osmolarity at 95% confidence level. For population studies, a power analysis based on the variability of the data showed that three repeat measurements would be required to obtain reliable data for a study with <50 subjects, whereas one measurement would suffice for 490 or more subjects. There were no interobserver or interinstrumental differences, but readings obtained for the standard solution varied up to 89 mOsm/l.

Conclusion Three consecutive readings are required with the TearLab to obtain a reliable measure of tear osmolarity. The variation in recorded tear osmolarity makes it difficult to use the technique for the diagnosis of mild dry eye.