Abstract
Purpose :
1. To experimentally measure the permeability of a model drug Ganciclovir through the “iris-hyaloid membrane-vitreous humor” (Iris-HM-V) composition.
2. To develop more precise experimental model based on our analytical data.
3. To obtain values for vitreous diffusion coefficient and permeability through “Iris-HM-V”.
Methods :
Since the hyaloid membrane is very thin and difficult to separate from the vitreous humor, instead of using a conventional Ussing chamber, we created an experimental diffusion model for hyaloid membrane permeability and vitreous diffusion coefficient measurements. Our technique involves extracting the entire vitreous with the hyaloid membrane attached. To keep the HM intact and to avoid potential damage, the iris was not peeled off in this study. The composition of our biological system is iris-vitreous humor with the hyaloid membrane intact. Thus the whole system is immersed in 20 ml of DMEM solution. In the next step, Ganciclovir was added to DMEM solution to a concentration of 10µM/l and used as a drug to study the permeability from the solution through the HM and vitreous humor. In this model 10 fresh bovine vitreouses have been used for each experiment. A different iris-HM-V is used for every time point. At every hour one bovine “Iris-HM-V” was removed from the solution medium and subsequently rinsed with distilled water. The donor samples were taken at the same time for concentration analyses measurements performed by HPLC. This process involves liquefaction of the vitreous bovine samples.
Results :
For mathematical analyses we assumed that vitreous has a spherical shape in a nearly buoyant-neutral surrounding. The diffusion transport equation was solved analytically by the Laplace transform technique. From each experimental measurement the average concentration of Gancyclovir has been measured over 10 hours, giving 10 data points with two unknown parameters, the hyaloid permeability (h) and the vitreous diffusion coefficient (D).
Conclusions :
Theoretical analyses are performed by least-squares while the values of h and D are floated. With two unknowns and ten time points, it was possible to obtain a measurement of these unknown parameters. The best fit from experimentally measured average spatial concentration yields the values of D and h approximately, 1.25 x 10-5 cm2/s and 1.44 x 10-5 cm/s correspondingly.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.