Topical delivery into the cul-de-sac is, by far, the most common route of ocular drug delivery. Adsorption from this site may be corneal or noncor-neal. A schematic diagram of the human eye is depicted in Figure 1. The so-called noncorneal route of absorption involves penetration across the sclera and conjunctiva into the intraocular tissues. This mechanism of absorption is usually nonproductive, as drug penetrating the surface of the eye beyond the corneal-scleral limbus is taken up by the local capillary beds and removed to the general circulation (2). This noncorneal absorption in general precludes entry into the aqueous humor.
Recent studies, however, suggest that noncorneal route of absorption may be significant for drug molecules with poor corneal permeability. Studies with inulin (3), timolol maleate (3), gentamicin (4), and prostaglandin PGF2a (5) suggest that these drugs gain intraocular access by diffusion across the conjunctiva and sclera. Ahmed and Patton (3) studied the non-corneal absorption of inulin and timolol maleate. Penetration of these agents into the intraocular tissues appears to occur via diffusion across the conjunctiva and sclera and not through reentry from the systemic circulation or via absorption into the local vasculature. Both compounds gained access to the iris-ciliary body without entry into the anterior cham-
ber. As much as 40% of inulin absorbed into the eye was determined to be the result of noncorneal absorption.
The noncorneal route of absorption may be significant for poorly cornea-permeable drugs; however, corneal absorption represents the major mechanism of absorption for most therapeutic entities. Topical absorption of these agents, then, is considered to be rate limited by the cornea. The anatomical structures of the cornea exert unique differential solubility requirements for drug candidates. Figure 2 illustrates a cross-sectional view of the cornea. In terms of transcorneal flux of drugs, the cornea can be viewed as a trilaminate structure consisting of three major diffusional barriers: epithelium, stroma, and endothelium. The epithelium and endothelium contain on the order of 100-fold the amount of lipid material per unit mass of the stroma (6). Depending on the physiochemical properties of the drug entity, the diffusional resistance offered by these tissues varies greatly (7,8).
pavement epithelium 5 or 6 layers thick bowman's membrane stroma descemet's membrane pavement epithelium 5 or 6 layers thick bowman's membrane stroma descemet's membrane endothelium
Figure 2 Cross-sectional view of the corneal membrane depicting various barriers to drug absorption. (From Ref. 12.)
The outermost layer, the epithelium, represents the rate-limiting barrier for transcorneal diffusion of most hydrophilic drugs. The epithelium is composed of five to seven cell layers. The basement cells are columnar in nature, allowing for minimal paracellular transport. The epithelial cells, however, narrow distal to Bowman's membrane, forming flattened epithelial cells with zonulae occludentes interjunctional complexes. This cellular arrangement precludes paracellular transport of most ophthalmic drugs and limits lateral movement within the anterior epithelium (9). Corneal surface epithelial intracellular pore size has been estimated to be about 60 A (10). Small ionic and hydrophilic molecules appear to gain access to the anterior chamber through these pores (11); however, for most drugs, paracellular transport is precluded by the interjectional complexes. In a recent review, Lee (10) discusses an attempt to transiently alter the epithelial integrity at these junctional complexes to improve ocular bioavailability. This approach has, however, only met with moderate success and has the potential to severely compromise the corneal integrity.
Sandwiched between the corneal epithelium and endothelium is the stroma (substantia propia). The stroma constitutes 85-90% of the total corneal mass and is composed of mainly of hydrated collagen (12). The stroma exerts a diffusional barrier to highly lipophilic drugs owing to its hydrophilic nature. There are no tight junction complexes in the stroma, and paracellular transport through this tissue is possible.
The innermost layer of the cornea, separated from the stroma by Descermet's membrane, is the endothelium. The endothelium is lipoidal in nature; however, it does not offer a significant barrier to the transcorneal diffusion of most drugs. Endothelial permeability depends solely on molecular weight and not on the charge of hydrophilic nature of the compound (13,14).
Transcellular transport across the corneal epithelium and stroma is the major mechanism of ocular absorption of topically applied ophthalmic pharmaceuticals. This type of Fickian diffusion is dependent upon many factors, i.e., surface area, diffusivity, the concentration gradient established, and the period over which concentration gradient can be maintained. A parabolic relationship between octanol/water partition coefficient and corneal permeability has been described for many drugs (15-19). The optimal log partition coefficient appears to be in the range of 1-3. The permeability coefficients of 11 steroids were determined by Schoenwald and Ward (15). The permeability versus log partition coefficient fit the typical parabolic relationship, with the optimum log partition coefficient being 2.9. Narurkar and Mitra studied a homologous series of 5' aliphatic esters of 5-iodo-2'-deoxyuridine (IDU) (16,17). In vitro corneal permeabilities were optimized at a log partition coefficient of 0.88, as can be seen graphically in Figure 3 and in Table 1, where CMP represents the corneal permeability values as measured by in vitro perfusion experiments on rabbit corneas (I = IDU, II = IDU-propionate, III = IDU-butyrate, IV = IDU-isobutyrate,
Table 1 Physicochemical Properties of IDU and Its 5'-Ester Prodrugs
Solubilitya in pH 7.4 phosphate buffer, 25°C Ka ± SD m.p. (°C) (M/L ± SD [x103]) (octanol/water)
Was this article helpful?