Topical ocular drug delivery: recent developments and future challenges

Rate limiting barrier to the penetration of ocular hypotensive beta blockers across the corneal epithelium in the pigmented rabbit

The objective of this study was to determine the rate-limiting layer(s) in the penetration of ocular hypotensive beta blockers across the corneal epithelium in the pigmented rabbit. The beta blockers studied were, in order of increasing lipophilicity, atenolol, timolol, levobunolol, and betaxolol. Corneal drug penetration was evaluated over 240 min using the isolated pigmented rabbit cornea in the modified Ussing chamber. Reversed phase HPLC was the analytical methodology. The cornea was preexposed to 20-100 microM digitonin for 15 min in an attempt to strip off selective layers of the corneal epithelium. The corneal epithelium offered no resistance to the penetration of the very lipophilic betaxolol. The major resistance to the corneal penetration of the moderately lipophilic timolol and levobunolol was in the superficial cell layers of the corneal epithelium. For the hydrophilic atenolol, the resistance to corneal drug penetration appeared to extend across all corneal epithelial cell layers. The above findings suggest that the number of corneal epithelial cell layers limiting the corneal penetration of ocularly administered drugs is inversely related to drug lipophilicity.

Comparison of different models for the testing of pilocarpine eyedrops using conventional eyedrops and a novel depot formulation (nanoparticles)

An objective in the development of ophthalmic formulations is the use of in vitro or animal models that closely resemble the clinical situation. For this reason, experiments with conventional pilocarpine nitrate eyedrops and a depot formulation of pilocarpine nitrate sorbed to poly (butylcyanoacrylate) nanoparticles were carried out. In vitro, the diffusion of pilocarpine through bovine cornea was measured using Edelhauser cells. In vivo, the rabbit aqueous humor concentration of pilocarpine and miosis were determined after application of the above formulations. In addition, intraocular pressure was measured. Since pilocarpine has little influence on intraocular pressure in healthy rabbits, the pressure had to be increased artificially. Three models were employed that are described in the literature, namely, the betamethasone model, the alpha-chymotrypsin model, and the water-loading model. Pilocarpine could be loaded onto nanoparticles by 15% but was rapidly released from the nanoparticles based on the bovine corneal experiment. Nanoparticles only enhanced the aqueous humor concentration at 30 min; this increase, however, led to a considerably extended period of miosis as well as a reduction in intraocular pressure. The duration of the action and the intensity of the response were different among the three models tested. According to the present results, the betamethasone model seems to represent the best correlation to the clinical situation.

Influence of ophthalmic ointments on the penetration of pilocarpine drops

To evaluate the influence of prior administered ointments upon the intraocular penetration of drops, we applied either a combination antibiotic ointment, a combination antibiotic-steroid ointment, or a lubricant ointment at different time periods before instillation of 2% pilocarpine drops into rabbit eyes. The level of pilocarpine measured in aqueous humor at one hour was compared to the level achieved in eyes without ointment application. No change was found in pilocarpine levels following application of the lubricant ointment. Pilocarpine concentrations were significantly lower (38%-48%) in animals receiving the antibiotic ointment at 5 or 15 minutes before pilocarpine drops. Pilocarpine levels also were decreased significantly when the steroid-antibiotic ointment was applied 15, 30 or 60 minutes before the drops. To verify these effects at other time periods, ointments were applied 15 minutes before pilocarpine drops and aqueous humor samples aspirated at 30 minutes or two hours. A 22% decrease in aqueous humor pilocarpine was found at 30 minutes with both the antibiotic and steroid-antibiotic combination; no effect was observed at two hours. To determine the influence of each drug component within the ointment on the penetration of pilocarpine, solutions of dexamethasone, neomycin, bacitracin and polymyxin B were mixed individually with pilocarpine drops and applied into the animal eyes. Bacitracin and polymyxin B significantly reduced one-hour pilocarpine levels in aqueous humor; however, no significant effect was observed with either neomycin or dexamethasone.

Influence of corneal epithelial integrity on the penetration of timolol prodrugs

The interplay of corneal epithelial integrity and prodrug lipophilicity in the corneal penetration of a homologous series of timolol prodrugs was studied. A reversed phase HPLC procedure was used to assay for timolol and its prodrugs. The optimal distribution coefficient for corneal penetration on a logarithmic basis was found to be 2.5, and for two prodrugs of equal lipophilicity the enzymatically labile one was better absorbed. The corneal permeability coefficient of O-acetyl, propionyl, and butyryl timolol was unaffected by deepithelizing the cornea, whereas that of the other prodrugs was reduced. This suggests that these three prodrugs diffused across the corneal epithelium proper as readily as across its basement membrane, which was not removed by the deepithelizing procedure. In contrast to the in vitro results, deepithelizing the cornea did not affect the aqueous humor concentration of timolol at 20 min afforded by any of the prodrugs except the O-propionyl and octanoyl esters. This suggests that factors other than corneal penetration exerted a more important role in affecting the aqueous humor drug concentration when the cornea was deepithelized.

Relative effectiveness of prodrug and viscous solution approaches in maximizing the ratio of ocular to systemic absorption of topically applied timolol

The effectiveness of the prodrug approach in maximizing the ratio of ocular to systemic absorption of topically applied timolol was compared with that based on viscous solutions. The pigmented rabbit was the experimental animal. o-Butyryl timolol, a lipophilic ester prodrug of timolol, was found to be twice as effective as hyaluronic acid and poly(vinyl alcohol) solutions in enhancing the ocular absorption of timolol. In addition, it was slightly more effective than retaining the instilled dose in the conjunctival sac for 240 min via nasolacrimal blockade. On the other hand, due to its inability to minimize contact of the instilled dose with the nasal mucosa, o-butyryl timolol was less capable than the viscous solutions in reducing systemic drug absorption. But this could be corrected by reducing the instilled solution volume from 25- to 5 microliter, resulting in a nine-fold reduction in plasma timolol levels while still providing the same amount of drug in the aqueous humor as a five times higher dose of timolol. The net result was a 15-fold improvement in the ratio of ocular to systemic drug absorption, which was also achieved upon retaining the instilled dose in the conjunctival sac for 240 min. It thus appears that the approach based on improved corneal absorption would be as effective as that based on improved retention in the conjunctival sac in maximizing the ratio of ocular to systemic drug absorption, so long as a smaller instilled dose volume of the prodrug is used.