Vivo Studies of Scleral Permeability

The diffusion of small molecules across the sclera has been well characterized. Various sulfonamides diffuse across rabbit sclera (40,41). In patients undergoing cataract surgery, methazolamide, a hydrophilic compound, penetrated the sclera much faster than the cornea, but ethoxzolamide, which is lipophilic, had similar diffusion across the sclera and cornea (40). Other studies also have found greater scleral permeability to hydrophilic molecules than lipophilic ones (42,43). 

The importance of the transscleral route in the penetration of topically administered macromolecules was demonstrated by denying corneal access by affixing a glass cylinder to the corneoscleral junction with cyanoacrylate adhesive (44). Significant intraocular levels of inulin (5 kDa) were obtained by topical administration outside the cornea in their paradigm. Neither reentry from the general circulation nor delivery by local vasculature accounted for the intraocular levels of the drug. 

Contralateral tissue drug levels were 1% of those in the treated eye. Further, intravenous administration achieved 5% of the tissue levels attained after topical dosing. These results imply that systemic absorption and intraocular reentry are not significant routes of delivery. In animals where topical dosage was applied after sacrifice, there was no significant difference in intraocular drug levels compared to live animals. This implied that delivery by local vasculature was not a major component. 

We studied the in vivo pharmacokinetics of transscleral delivery of IgG. We used an osmotic pump, the tip of which was secured flush against bare sclera in rabbits to facilitate unidirectional movement, to deliver fluorescently labeled IgG (150kDa) at rates on the order of pL/hr. Biologically relevant concentrations in the choroid and retina were attained for periods of up to four weeks with negligible systemic absorption (6). Levels in the vitreous and aqueous humors, and orbit were negligible. Although there was a spatial concentration gradient, the IgG concentration in the choroidal hemisphere distal to the footprint of the osmotic pump tip was half of that in the proximal hemisphere. The elimination of IgG from the choroid and retina followed first-order kinetics with half-lives of approximately two to three days. 

The high ocular selectivity in our experimental design may be due to the sponge-like nature of the sclera. While bolus injections of drugs into the subconjunctival space may overwhelm the absorptive capacity of the sclera, leading to systemic absorption, the gradual delivery employed in our experiments may permit more complete scleral absorption as the absorptive capacity of the sclera is more closely matched to the rate of drug delivery. 

---