Tear osmotic differences across the ocular surface.

We used a freezing point depression method to determine the osmolalities of 502 non-stimulated 200-nanoliter tear samples. The samples were collected hourly at 9 times on each of 6 days. Four reference (tear prism) sites on the ocular surface were used: mid-superior, mid-inferior, and the tear prisms at the medial and lateral canthi. While 256 of the samples were from a young (25 year old) male volunteer having no complaints of dry eyes, the remaining 246 samples were taken from a second male volunteer of the same age range, but who had occasional complaints of dry eyes. The mean osmolalities for all sites and all times for both subjects were found to be 315 and 331 mOsm/kg, respectively, and were significantly different (P = 0.0004). Two of the six time-averaged intersite comparisons investigated here were found to be significantly different for each of the subjects, with the inferior tear prism and the medial canthus being the only shared pairs present. The magnitudes and patterns of these osmotic site differences were found to shift for both subjects over time, although this was more prominent and with a greater hypertonic bias for the subject with dry eye complaints. While these data do demonstrate statistically different osmolalities across the ocular surfaces of the two subjects examined in this report, these findings must be considered a preliminary view of broader population patterns yet to be studied.