Development of

Purpose: To develop in vitro methods to assess binding by sodium hyaluronate in eye drops to corneal surfaces. Methods: Two different, complementary corneal binding set-ups were developed. In a dynamic in vitro model, confluent corneal epithelial cells (HCE-T) were assembled in chamber slides and a declining channel. A static model was constructed with ex vivo porcine corneas clamped in Franz cells. To test the predictive capacity of models, four different eye drops containing sodium hyaluronate were spiked with tritium-labeled sodium hyaluronate to standardize quantification. In both settings, eye drops were applied for 5 min and physiological conditions were mimicked by flushing with artificial tear fluid. Spreading experiments on HCE-T next to synthetic membranes were used for further characterization. Results: Binding was more pronounced in dynamic HCE-T model. Three of the four eye drops demonstrated sigmoidal elution of sodium hyaluronate, suggesting pronounced binding. One solution eluted distinctly faster, likewise the buffer control. The static method produced a similar ranking but at lower levels. When eye drops in which phosphate buffer was replaced by citrate buffer (i.e., to prevent calcification) were used, binding was not influenced. All eye drops spread immediately when placed on HCE-T and at the same order of magnitude on glass and polyethylene terephthalate surfaces. Conclusion: Dynamic and static models performed on different corneal sources were used to determine sodium hyaluronate binding kinetics in solutions under physiological conditions. These methodologies resulted in a ranking of the capacity of sodium hyaluronate to bind in vitro to corneal surfaces.