Abstract: What makes tears stay on the eye

Viscosity ain't the answer. C'mon, folks, Dr. Holly and others figured that out a long time ago. How about somebody go back and read their studies???

Incidental findings: Reading between the ingredient lines in search of possible product names, it sounds to me like Refresh Plus and Soothe were dramatically outperformed by Systane and Refresh Celluvisc (or Liquigel - both are 1% CMC... or could be Theratears liquigel for that matter) in terms of "residence time".

Precorneal residence time of artificial tears measured in dry eye subjects.
Optom Vis Sci. 2008 Aug;85(8):725-31.
Paugh JR, Nguyen AL, Ketelson HA, Christensen MT, Meadows DL

PURPOSE.: The purpose of this investigation was to measure the precorneal residence time of saline and five marketed artificial tears in dry eye subjects using fluorometry.

METHODS.: FITC-dextran, 70 kDa molecular weight, was admixed under sterile conditions (0.1% wt/vol) into buffered saline and the marketed artificial tear formulations of varying viscosity. Precorneal residence time (RT) was measured directly in 16 mild to moderate dry eye subjects, classified by sub-type, in a six-way cross-over, masked and randomized study. FITC-dextran tracer decay with a scanning fluorometer was used to estimate the gross RT (i.e., the time in minutes for the signal to return to baseline).

RESULTS.: All subjects were classified as having non-inflammatory meibomian gland dysfunction except one, who had a mixture of aqueous deficiency and meibomian gland dysfunction. In two separate determinations, the saline RTs were 19.1 +/- 7.4 and 17.6 +/- 8.2 min. The RTs for the formulations varied to some degree by viscosity, with two higher viscosity formulations demonstrating the longest RTs of 36 to 41 min, approximately twice that of saline (p < 0.001 for both 0.4% polyethylene glycol/0.3% propylene glycol, and 1.0% carboxymethylcellulose). An oil emulsion, low viscosity carboxymethylcellulose and moderate viscosity hydroxypropylmethylcellulose-containing formulation were not statistically different from saline (RTs of 18, 22 and 24 min, p values = 0.983, 0.818 and 0.099, respectively).

CONCLUSIONS.: More than two-fold RT differences were found for the higher viscosity, more muco-adhesive formulations compared to saline. However, other formulations provided RTs close to saline, suggesting that RT is influenced by factors other than simple viscosity. Future studies should examine the interplay of spreading characteristics, pseudoplasticity and muco-adhesion relative to RT to determine the individual and cumulative effects on formulation retention.