EDITORS
`
`Thom I. Zimmerman
`
`Eéarangii §.. K0932?
`
`@3‘°d&fi1a§§ha§°§3~ }
`
`Qbgfi
`
`Eegh ea"
`
` §.§pg§fi£:@%§§ ~ gmfim :4
`
`SENJU-MITSUBISHI 2007
`
`SENJU-MITSUBISHI 2007
`
`

`
`
`
`
`
`
`
`acology
`
`EDITOR-IN-CHIEF
`
`Thom J. Zimmerman, M.D., PH.D.
`Professor and Chairman
`Department of Ophthalmology and Visual Sciences
`Professor of Pharmacology and Toxicology
`University of Louisville School of Medicine
`Kentucky Lions Eye Center
`Louisville, Kentucky
`
`EDITORS
`
`Karanjit S. Kooner, M.D.
`Assistant Professor
`Department of Ophthalmology
`Southwestern Medical School
`Dallas, Texas
`
`ASSOCIATE EDITOR
`
`Robert D. Fechtner, MD.
`Associate Director
`
`Glaucoma Service
`
`Mordechai Sharir, M.D., PH.D.
`Chief of Glaucoma
`Department of Ophthalmology
`The Edith Wolfson Hospital
`Tel Aviv, Israel
`
`Associate Professor
`Department of Ophthalmology and Visual Sciences
`University of Louisville School ofMedicine
`Kentucky Lions Eye Center
`Louisville, Kentucky
`
`R lipfiincett - Raven
`
`P
`
`U
`
`B
`
`L
`
`I
`
`S
`
`H
`
`E
`
`R
`
`5
`
`Philadelphia - New York
`
`

`
`Acquisitions Editor: Vickie Thaw
`Developmental Editor: Delois Patterson
`Manufacturing Manager: Dennis Teston
`Associate Managing Editor: Kathy Bubbeo
`Production Editor: Nicholas Radhuber
`Cover Designer: Susan J. Moore
`Indexer: Jayne Percy
`Compositor: Compset
`Printer: Maple Press
`
`© 1997, by Lippincott~Raven Publishers. All rights reserved. This book is protected by copyright. No
`part of it may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means-
`electronic, mechanical, photocopy, recording, or otherwise—-without the prior written consent of the
`publisher, except for brief quotations embodied in critical articles and reviews. For information write
`Lippincott—Raven Publishers, 227 East Washington Square, Philadelphia, PA 19106-3780.
`Materials appearing in this book prepared by individuals as part of their official duties as US. Gov~
`ernment employees are not covered by the above-mentioned copyright.
`
`Printed in the United States of America
`
`9
`
`8
`
`7
`
`6
`
`5
`
`4
`
`3
`
`2
`
`1
`
`Library of Congress Cataloging-in-Publication Data
`
`Textbook of ocular pharmacology / editor—in~chief, Thom J. Zimmerman;
`editors, Karanjit S. Kooner,
`Mordechai Sharir, Robert D. Fechtner.
`p. cm.
`Includes bibliographical references and index.
`ISBN 0-781-70306-9
`l. Ocular pharmacology. I. Zimmerman, Thom J.
`[DNLM: 1. Eye Diseases—drug therapy. 2. Eye—drug effects. WW
`166 T355 1997]
`RE994.T49 l997
`6l7.7’O6l—dc21
`DNLM/DLC
`for Library of Congress
`
`97»l434
`CIP
`
`Care has been taken to confirm the accuracy of the information presented and to describe generally ac-
`cepted practices. However, the authors, editors, and publisher are not responsible for errors or omis-
`sions or for any consequences from application of the information in this book and make no warranty,
`express or implied, with respect to the contents of the publication.
`The authors, editors, and publisher have exerted every effort to ensure that drug selection and dosage
`set forth in this text are in accordance with current recommendations and practice at the time of publi—
`cation. However, in view of ongoing research, changes in government regulations, and the constant
`flow of information relating to drug therapy and drug reactions, the reader is urged to check the package
`insert for each drug for any change in indications and dosage and for added warnings and precautions.
`This is particularly important when the recommended agent is a new or infrequently employed drug.
`Some drugs and medical devices presented in this publication have Food and Drug Administration
`(FDA) clearance for limited use in restricted research settings. It is the responsibility of the health care
`provider to ascertain the FDA status of each drug or device planned for use in their clinical practice.
`Pharmacology is an ever changing field of medicine. All efforts have been made by the Editors, con-
`tributors and publisher to ensure accuracy of drug dosages presented in this textbook. Some of the indi»
`cations may not yet have been approved by the Food and Drug Administration (FDA). Therefore, the
`package inserts for each drug should be consulted for use and dosage as approved by the FDA. It is ad-
`visable to keep abreast of revised recommendations of drugs as discussed in peer-reviewed journals or
`at scientific meetings.
`
`

`
`Textbook of Ocular Pharmacology,
`edited by TI. Zimmerman, et al.
`Lippincott—Raven Publishers, Philadelphia © 1997.
`
`CHAPTER W
`
`Easic Considerations of Qcular
`
`Drug-Delivery §§/stems
`
`Amir Bar—llan and Ron Neurnann
`
`
`
`Drugs topically applied to the eye are subjected to the nor-
`mal (or pathological) physiology of the conjunctival sac,
`which in turn regulates the rate of clearance (drainage) of
`drugs from the sac into the nasal cavity. Most drugs pene-
`trate the eye through the cornea, protected by a complex
`tear—washing mechanism that easily washes away most
`compounds in a few minutes. The rate of the secretion and
`clearance of tears determines variable contact time for any
`given drug with the ocular surface. Radioactive tracers ap-
`plied to the eyes are readily washed by the tears, with elim-
`ination of approximately two thirds of the radioactive sig-
`nal within the first 2 minutes (1). The drug volume is
`readily cleared to the nasal cavity, where systemic absorp-
`tion takes place. It follows that, despite variable intraocular
`penetration, systemic absorption may be effective, and
`nearly all drug dosages applied to the eye are absorbed sys-
`temically, leading sometimes (e.g., timolol, atropine, epi-
`nephrine) to substantial systemic drug effects.
`
`DRUGS AS CHEMICAL COMPOUNDS
`
`Commonly used drugs are chemical compounds with
`distinct physicochemical properties that
`largely control
`their penetration into active sites within the eye. High mol-
`ecular weight and size may limit efficient corneal pene-
`tration. lonic/nonionic partition and acid/base balance in
`solution, as well as
`the solubility constant and the
`“octanol/water partition coefficient,” all determine the
`drug dissolution rate in distinct solvents and, hence, the ef-
`fective drug concentration in solution. As detailed in the
`sections to follow, the physicochemical properties of the
`preparation (formulation) control corneal contact time and
`thus may largely affect bioavailability.
`
`A. Bar-llan and R. Neumann: Pharmos Limited, Kiryat Weiz-
`mann, Rehovot, Israel.
`
`139
`
`Tear Film Physiology and Corneal Contact Time
`
`The conjunctival sac and external epithelium are washed
`constantly by a basic tear secretion averaging 1.2 ul per
`minute. The total volume of fluids held by the conjunctival
`sac is approximately 10 ul (~20% the amount delivered by
`a typical eyedropper) (2). Ocular irritation, however, in-
`duces a reflex secretion of tears that may increase tear se-
`cretion up to 400 [L1 per minute (3) and may easily wash
`out the irritating compound (i.e., the drug). In addition,
`pathological conditions that disrupt the normal tear physi-
`ology (e. g., dry eyes, external inflammation with excessive
`tearing, eyelid malpositioning, and others) may exert a ma-
`jor effect on topical ocular drug delivery.
`The corneal surface is densely innervated with sensory
`nerve fibers, causing extreme sensitivity to any change in
`the composition of the tear solution bathing the cornea.
`Eyedrops may easily destroy the delicate interactions be-
`tween the corneal surface and the tear film, resulting in
`corneal irritation in mild cases and toxic epitheliopathies in
`severe cases. Among the basic parameters to be considered
`are osmolality (266 mOsm/kg to 445 mOsm/kg are accept-
`able to the eye) and pH. Basic pH is especially irritating to
`the ocular surface. Buffer capacity is also important be-
`cause tears can neutralize a wide range of pH levels with
`low buffer capacity. Thus, pH in some pilocarpine com-
`mercial solutions may be as low as 3.5 to 4.0 without pro-
`ducing excessive irritation. The mere instillation of a drop
`of fluid may induce some reflex tearing, and high levels of
`viscosity may cause an especially unpleasant sensation.
`Moreover, the chemical compound itself (the active drug)
`may induce either lacrimation (e.g., with pilocarpine and
`antazoline) or reduced tear flow (e.g., with timolol and
`anaesthetics). Generally,
`the more irritating a chemical
`compound, the more difficult it is to extend its corneal con-
`tact time. (More details are found in the section entitled
`Ophthalmic Aqueous Solutions.)
`
`

`
`140 / CHAPTER 10
`
`The Barriers
`
`The cornea is composed of a complex network of barri-
`ers comprising distinct layers of fluid and tissues that con-
`front any molecule that “dares” to cross. A drug molecule
`dissolved in the tear film may be washed easily from the
`corneal surface. As soon as the compound penetrates the
`corneal epithelium, however, the probability of intraocular
`penetration becomes much higher. It follows that rapid
`movement of chemical compounds from the tear film to the
`corneal epithelium is essential for efficient ocular drug
`delivery.
`The tear film is composed of approximately 10 ul of
`lipophilic, aqueous, and mucin layers totalling up to 7 /rm
`thick. The molecular trip through the tear film involves
`crossing through hydrophobic and hydrophilic environ-
`ments. Moreover, in the mucin layer, a significant fraction
`of the drug molecules may bind to “carrier” proteins, al-
`lowing only the “free,” unbound molecules to cross into
`the epithelium.
`Once in the corneal matrix, the drug should cross dis-
`tinct layers of corneal tissue, constituting a variety of barri-
`ers. The external corneal epithelium, composed of five to
`seven layers of a nonkeratinized squamous epithelium with
`an external net of tight junctions, allows only selective
`penetration of compounds. Because of the high concentra-
`tion of lipophilic membranes, nonpolar molecules (as well
`as nonpolar molecular moieties) have easier access into the
`epithelium. On the other hand,
`the hydrophilic corneal
`stroma has a low cellular content and may easily be
`crossed by polar molecules. The endothelium is a cellular
`monolayer that allows much better penetration of hydro-
`philic compounds compared with the epithelium. A pro-
`found description of the implications of this complex
`structure on pharmacokinetics can be found in Chapter 9. It
`has been estimated that under optimal conditions only 1%
`to 10% of the total drug applied directly onto the cornea is
`absorbed into the eye (4).
`
`GOALS OF THE OCULAR DRUG-
`DELIVERY SYSTEM
`
`The self—explanatory goal of the ocular drug-delivery
`system (DDS) is extended corneal contact time, resulting
`in increased drug penetration and a higher intraocular
`level. Nevertheless, sometimes the goals of a DDS may be
`more subtle, that is, reduced irritation leading to increased
`corneal contact time on one hand and increased patient
`compliance on the other hand. In other instances, aqueous
`levels may suffice but only with such dosages that induce
`systemic side effects. An efficient DDS may enable re-
`duced dosages, thus bypassing significant systemic side ef-
`fects without compromising the ocular therapeutic effect.
`Another use of an “unorthodox” delivery system may be
`reduced anterior chamber penetration for drugs needed for
`
`external ocular indications with potential intraocular ad-
`verse effects, such as topical steroids. Any system that can
`reduce the penetration of a topical corticosteroid to the an~
`terior chamber without compromising drug presence on the
`ocular surfaces would allow safer administration of the
`
`drug for external indications, such as allergic conjunctivi-
`tis. Similarly, minimizing systemic absorption of topically
`administered ophthalmic drugs is sometimes a major goal
`of ocular DDS.
`
`The definition of optimal drug delivery differs with the
`drug involved. For example, pilocarpine exerts its intraoc—
`ular pressure (IOP)—lowering effect with minimal side ef~
`fects (miosis and increased accommodation) when intro-
`duced continuously to the anterior chamber in constant
`lower levels. It follows that for pilocarpine the best mode
`of delivery should abolish the initial high peak of intraocu-
`lar penetration, thereby avoiding unpleasant ocular side ef-
`fects without compromising the desired therapeutic effect.
`For other medications, an initial high peak of intraocu~
`lar drug concentration may be advantageous in reaching
`higher therapeutic levels (i.e., antibiotics and steroids). In
`the following sections, we review the delivery aspects of
`ophthalmic aqueous solutions, suspensions, ointments, in-
`serts, collagen shields, liposomes, emulsions, and gels.
`
`OPHTHALMIC AQUEOUS SOLUTIONS
`
`Ophthalmic solutions are sterile solutions, essentially
`free from foreign particles, suitably compounded and pack-
`aged for instillation into the eye (5). Aqueous solutions
`are the most commonly used ocular DDS and are the least
`expensive to formulate compared with other types of DDS.
`All the ingredients are completely dissolved; so there is
`only minimal interference with vision. The major draw-
`back of aqueous solutions is their relatively short Contact
`time with the drug~absorbing surfaces, namely, the cornea,
`conjunctiva, and sclera.
`In addition to the active drug, ophthalmic solutions con-
`tain other ingredients, or excipients, that are added to con-
`trol various characteristics of the formulation, such as the
`tonicity, buffering and pH, viscosity, sterility, and antimi-
`crobial preservation. Although these added ingredients are
`listed as “inactive additives,” they can affect (i.e., enhance
`or reduce) the permeability of the drug across the ocular
`surface barriers and thus significantly alter the therapeutic
`effectiveness of a given drug (the active ingredient). Be—
`cause the effects of different additives are interrelated, the
`final product’s composition usually reflects a compromise
`between contradictory requirements rather than the opti-
`mized condition for each individual property.
`
`pH and Buffering
`
`The pH of an ophthalmic solution can play an important
`role in determining the therapeutic effectiveness of a drug.
`
`

`
`Most ophthalmic drugs, being weak acids or bases, are
`present in solutions as both the nonionized (norza.’issoci-
`area’) and the ionized (dissociated) species. It is generally
`accepted that the nonionized species (being more lipid sol-
`uble) diffuses across cellular barriers at a higher rate than
`the ionized (less lipid—soluble) species (6). The degree of
`ionization of a drug in solution is determined by its pK and
`the solution’s pH. Thus, a pH that favors a higher propor-
`tion of the nonionized species should result in a higher
`transcorneal permeability, as is well illustrated by the find-
`ings of Mitra and Mickelson (7), who showed that the per-
`meability of pilocarpine across isolated cornea, which was
`4.72 X 105 cm/s—‘ at pH 4.67 (99% of the pilocarpine in
`the ionized form),
`increased by almost twofold to 8.85
`X 105 cm/s*1 at pH 7.40 when 84% of the drug was in the
`nonionized form. Similar changes, that is, twofold to three
`fold increases in transcorneal permeability concomitant
`with 2 to 2.5 pH unit changes (5—7.5, 6.2-8.4, etc.), were
`reported for various carbonic anhydrase inhibitors (8).
`Similarly, alterations in the pH of solutions of the carbonic
`anhydrase inhibitor MK-927 from pH-7 (11% ionized) to
`pH-4.8 (91% ionized) or pH~9.1 (86% ionized), increased
`the lOP—lowering activity of the drug in rabbits by fivefold
`to sixfold, respectively (9).
`
`Tonicity
`
`To avoid irritation, ophthalmic formulations intended
`for topical instillation should be approximately isotonic
`with tears (10). Normal human tears are considered iso~
`tonic, or slightly hypertonic to plasma, that is, about 300
`mOsmol/kg or 0.90 to 1.01 NaCl equivalents. Significant
`interindividual and intraindividual variability has been re-
`ported for the tonicity of normal human tears, ranging be-
`tween 260 and 440 mOsmol (11-13).
`Various studies have shown that the eye can tolerate a
`considerable range of tonicity before any pain or discom-
`fort is detected (14,15). Also, increased tonicity of topi—
`cally applied solutions (above that of the body fluids) re-
`sulted in their immediate dilution by osmosis in the eye
`(16). Thus,
`the tonicity of an ophthalmic solution may
`range from 0.2% to 2.0% in NaCl equivalents or 220 to 640
`mOsM without exceeding the safety range. Most oph-
`thalmic drugs listed in the Physicians Desk Reference
`(PDR) 22 (17) do not exceed 5% of an active compound,
`and even with additional tonicity resulting from adjust~
`ments in pH, preservatives, and surfactants they are within
`this range of tonicity. Only a few ophthalmic solutions
`(i.e., pilocarpine 8% and 10%; phenylephrine 10%; sulfac~
`etamide 10%, 15%, and 30%) have an osmolarity of 700 to
`1,000 mOsM and would cause a strong buming-stinging
`sensation upon instillation.
`In one class of ophthalmic drugs, hypertonic ophthalmic
`solutions, which are used for temporary relief of corneal
`edema, the hypertonic factor (5% NaCl) is listed as the ac-
`
`OCULAR DRUG-DELIVERY SYSTEMS / 141
`
`tive ingredient. It is reasonable to believe that a similar ef-
`fect would occur during treatment with other hypertonic
`(high drug concentration) ophthalmic formulations.
`
`Viscosity .
`
`Increasing the viscosity of topically applied ocular for—
`mulations is expected to reduce drainage, increase the resi-
`dence time in the conjunctiva] sac, and thus lead to an en-
`hanced intraocular penetration and this therapeutic effect.
`Substances like methylcellulose (MC), hydroxypropyl—
`methylcellulose (HPMC), hydroxypropylcellulose (HPC),
`hydroxyethylcellulose (HEC), and polyvinyl alcohol (PVA),
`listed in the U.S. Pharmacopeia (USP) 23 (5) as viscosity-
`increasing agents, are frequently added to ophthalmic liq-
`uid formulations.
`
`The use of MC for improving contact with the ocular
`surface was reported some 50 years ago by Swan (18).
`Linn and Jones (19) showed that an increase in the
`drainage time of topically applied fluorescein solution is
`directly related to the concentration of added HPMC in
`human subjects. A solution containing 0.25% HPMC in~
`creases drainage time by 15-fold and a 1% solution of
`HPMC by 3.5-fold compared with an aqueous solution. Al-
`though a 2.5% solution of HPMC increased drainage time
`by 4.5~fold, it caused irritation, which made it unsuitable
`for routine clinical use. Mueller and Deardorff (20) re-
`ported that although a 0.25% homatropine hydrobromide
`solution failed to produce a cycloplegic or mydriatic re-
`sponse in human subjects, the addition of 1% MC 4,000
`centipoise (cps) to this solution reduced accommodation
`by 80% and led to significant pupillary dilation. Later,
`Chrai and Robinson (21) showed that increasing the vis-
`cosity of a pilocarpine solution over the range of 1 to 100
`cps (achieved by increasing the added concentration of
`MC) resulted in a significant reduction in the drainage rate
`constant (by tenfold) and an increase in aqueous humor pi-
`locarpine concentration (by twofold). Patton and Robinson
`(22) reported that most of the improvement in ocular drug
`delivery was observed over the viscosity range of 1 to
`15 cps and suggested that the optimal viscosity should be
`12 to 15 cps. The use of formulations with higher viscosity
`causes ocular surface irritation, resulting in reflex blinking
`and lacrimation and increased drainage of the applied for-
`mulation.
`
`Saettone et al. (23) compared the effects of different
`polymers, including HPC and PVA, added to an aqueous
`solution of 0.2% tropicamide, on the mydriatic responses
`of albino rabbits and human subjects. All test solutions
`were isoviscous (73 cps). They found that the PVA—con—
`taining solution had the highest activity: a 3.7~fold increase
`compared with the aqueous solution and a twofold increase
`over the other polymers. The author suggested that the wt»
`face—spreading effect of PVA, which did not characterize
`the other polymer solutions tested, was responsible for the
`
`

`
`l4:2 / CHAPTER 10
`
`observed advantage of PVA. An important finding was that
`the advantage of the PVA—tropicamide formulation was ob-
`served in the human subjects but not in the albino rabbit.
`The lower blinking and tear production rates of the rabbit
`compared with humans could contribute to the discrepancy
`and highlights the importance of clinical testing in defining
`the actual advantages of modifications in ocular DDS.
`
`Preservatives
`
`Ophthalmic solutions packaged in multiple—dose contain-
`ers must contain a suitable substance, or mixture of sub-
`
`stances, to prevent the growth of or to destroy microorgan-
`isms accidentally introduced when the container is opened
`during use (USP 23) (5). Quaternary ammonium compounds
`(benzalkonium chloride), organic mercurials (thimerosal),
`parahydroxy benzoates, chlorobutanol, and aromatic alco-
`hols are used as preservatives in ophthalmic preparations.
`Banzalkonium chloride (BAk), the most commonly used
`preservative in ophthalmic preparations, was demonstrated
`repeatedly to enhance corneal permeability of various drugs.
`A 50- to 80-fold increase in the permeability of carb-
`aminochline chloride caused by 0.02% BAk was reported
`by O’Brien and Swan (24). A 10- to 18-fold increase in in-
`ulin permeability was attained by 0.01% and 0.02% BAk,
`respectively (25). This preservative was also reported to
`increase the permeability of pilocarpine (26) and fluores-
`cein (27,28).
`
`In rabbits, preservatives used in ophthalmic solutions
`(BAk and others) are toxic to the ocular surface following
`topical application as well as to retinal functions following
`intravitreal administration (29—34). The clinical signifi-
`cance of the toxic effects and the enhanced ocular penetra-
`tion attributed to preservatives are not completely clear,
`however. Most of these studies used rabbits as experimental
`subjects, and rabbits are more susceptible to ocular surface
`drainage (28) and permeability changes (35) than humans.
`
`OPHTHALMIC SUSPENSIONS
`
`in
`Ophthalmic suspensions are sterile preparations
`which relatively water-insoluble drugs are delivered in the
`form of solid particles dispersed in a liquid vehicle. They
`are sometimes commercially identified by the “forte” suf-
`fix. Because the small drug particles tend to remain in the
`cul—de-sac longer than an aqueous solution, drug delivery
`from a suspension is characterized by two consecutive
`phases: the first, rapid delivery of the dissolved drug; the
`second, slower, more prolonged delivery resulting from
`dissolution of the retained particles. Thus, in order for a
`suspension to attain a higher bioavailability than that of a
`saturated solution, it must have a rapid, significant rate of
`dissolution in the tear film (36).
`
`Ocular bioavailability from topically applied suspension
`is correlated with particle size (37): (a) the area accessible
`
`for drug dissolution, and hence the dissolution rate, is re-
`lated to particle size; (b) larger particles can lead to in-
`creased ocular irritation with enhanced tearing and drug
`loss by drainage. Use of the drug in a micronized form with
`a particle size <10 um should minimize ocular discomfort
`and irritation (38). Formulation factors that affect the ocu-
`
`lar bioavailability of topically instilled ophthalmic solution
`(i.e., pH and buffering, tonicity, viscosity) would also in-
`fluence the bioavailability from ophthalmic suspensions.
`Because precipitation is the major problem with oph-
`thalmic suspensions, they must be resuspended before use
`to obtain an accurate dose. The degree of resuspension of
`commercially available formulations varies considerably
`among patients and formulations (39).
`
`OPHTHALMIC OINTMENTS
`
`Ointments are a popular, frequently prescribed ocular
`DDS. Fifty—eight ophthalmic ointments are listed in the
`USP 23 (5) compared with 59 ophthalmic solutions and 29
`ophthalmic suspensions. The ointment bases commonly
`used in ophthalmic preparations are usually those that per-
`mit the incorporation of aqueous solutions and the forma-
`tion of a water-in—oil emulsion (e. g., white hydrophilic
`petrolatum and anhydrous liquid lanolin). The addition of
`mineral oil to the petrolatum base lowers the melting point
`of the base, allowing the ointment to become a liquid at eye
`temperature, with improved spreading and mixing with the
`tear film and reduced effect on vision. Hydrophilic drugs
`are dispersed in the ointment as fine, solid particles, simi-
`lar to aqueous suspensions, whereas lipid—soluble drugs are
`dissolved in the ointment base. Like ophthalmic solutions
`and suspensions, ophthalmic ointments must contain
`preservatives to prevent the growth of microorganisms in-
`troduced accidentally during routine use. The effects re-
`lated to inactive ingredients (e.g., preservatives, pH, to-
`nicity), described earlier for ophthalmic solutions, are
`pertinent to ophthalmic ointments and should be controlled
`accordingly.
`
`Similar to ophthalmic solutions, some ophthalmic oint-
`ments contain a high drug concentration, exceeding tears’
`tonicity, and thus would cause stinging, burning, and reflex
`tearing upon instillation. A hypertonic ointment (5% NaCl
`formulated in white petrolatum, mineral oil, and lanolin) is
`available for temporary relief of corneal edema.
`The major advantage of ointments as an ocular DDS is
`their tendency to serve as a drug depot, made possible by
`their extended retention in the conjunctival sac (i.e., in-
`creased ocular contact time), resulting in enhanced and
`sustained corneal absorption (40,41). This enhanced effect
`is related to the lipid—aqueous differential solubility of the
`drug. The bioavailability of a water~soluble drug like pilo-
`carpine resulted in a fourfold increase when a 0.01 M oint-
`ment was compared with a 0.01 M solution. On the other
`
`hand, the bioavailability of the lipid—soluble drug, fluo-
`
`

`
`rometholone, formulated as an ointment, increased by more
`than eightfold compared with that observed following treat-
`ment with a saturated solution of fluorometholone (42).
`On the other hand, Riegelman (43) pointed out that the
`solubility of most drugs in commercial petrolatum oint-
`ments is very low. Therefore, most of the drug is present
`within the ointment as solid rnicrocrystals and must diffuse
`through the petrolatum to reach the surface. Thus, under
`conditions of low tear turnover (i.e., during sleep), the drug
`may reach the tear film at a rate lower than necessary to
`achieve therapeutic levels, as was the case for various
`preparations of 5 mg/g of neomycin ointment. Thus, oint-
`ment may not be always advantageous to solutions and
`suspensions.
`The major disadvantages of ophthalmic ointments are an
`annoying blurring of vision following instillation; difficul-
`ties in properly applying an exact dose of the drug com-
`pared with applying a solution or a suspension; an initial
`delay in drug delivery; and sensitivity to ambient tempera-
`tures
`(under cold—weather conditions petrolatum-based
`ointments are difficult to extrude from the ointment tube,
`and the formulation exhibits a poor ocular drug release
`rate). Water—containing bases tend to separate into two
`phases. Elevated temperatures also can enhance melting
`and nonhomogeneity within the ointment tube.
`
`STRIPS
`
`Paper strips impregnated with fluorescein are used for
`diagnostic purposes by staining the anterior segment of the
`eye. The sterile paper strips are impregnated with a suffi-
`cient amount of the drug and then released by the tears on
`contact with the bulbar conjunctiva. Direct contact of the
`paper with the eye can be avoided by leaching the drug
`from the paper strip with the aid of sterile water or sodium
`chloride solution. Like many other ocular DDS, these pa-
`per strips contain, in addition to the active ingredient, a
`preservative, surface—acting agent, and buffering agents.
`
`OCULAR INSERTS
`
`Hydrogel Contact Lenses
`
`Hydrogel contact lenses can absorb water up to 80% of
`their weight;
`thus, when soaked in a drug solution and
`placed over the cornea, they can greatly extend the contact
`time of the drug solution with the ocular surface and hence
`increase drug penetration. Waltman and Kaufman (44)
`showed that placing a contact lens presoaked with fluores-
`cein over the cornea resulted in increased fluorescein levels
`
`in the anterior chamber of rabbits (by fourfold) and humans
`(by eightfold) compared with levels observed following fre-
`quent topical instillation of fluorescein solution.
`Similar findings were later reported for various ophthal-
`mic drugs, including idoxuridine, polymyxine B, phenyl-
`
`OCULAR DRUG-DELIVERY SYSTEMS / 143
`
`ephrine (45), chloramphenicol and tetracycline (46), pilo-
`carpine (47,48), prednisolone (49), and carbonic anhydrase
`inhibitors (50). Despite the demonstrated improved ocular
`drug delivery by this system, it did not gain widespread use
`and is restricted to a few clinical examples.
`
`Membrane-bound Devices
`
`The first marketed device to achieve the goal of a zero-
`order delivery kinetics was the Ocusert. The drug, pilo-
`carpine, bound to olginic acid and present as a free base, is
`contained in a reservoir formed by two thin, transparent
`ethylene—vinyl—acetate (EVA) membranes. An annular ring
`of EVA, made opaque by impregnation with titanium diox-
`ide, aids in visualization and handling of the insert. The hy-
`drophobic polymer impedes the permeation of water into
`the device, and the drug delivery rate is determined by the
`coefficient of diffusion and the concentration gradient. The
`elliptical device, measuring 13.4 by 5.7 mm, 0.3 mm thick,
`delivers 20 pig/h (Pilo—20) or 40 tog/h (Pilo-40); the higher
`delivery rate is achieved by the addition of a flux enhancer
`(di(2-ethylhexyl)phtalate) to the reservoir. The initial clini-
`cal studies with Ocusert (51,512) showed that the 20 /ig/h
`device, delivering 500 ptg/day, was effective in maintain-
`ing IOP as a standard drop treatment delivering 4 mg/day
`(an eightfold reduction in the daily dose). The device could
`be used in combination with other antiglaucoma drugs, for
`example, epinephrine (53,54). The major advantage of this
`ocular DDS is the maintenance of therapeutic effectiveness
`using a smaller amount of drug concomitant with a lower
`incidence of induced miosis and myopia and reduction in
`visual acuity (53).
`Ocusert therapy is more expensive than standard oph-
`thalmic solution therapy. Patients must check that the de-
`vice has not been lost and that it is in position in the lower
`cul-de—sac, that is, has not migrated around the eye. The
`device must be removed and replaced once a week. Exces-
`sive foreign body sensation can sometimes preclude its
`use. Cases of sudden leakage of the drug have been re-
`ported (51).
`
`COLLAGEN SHIELDS
`
`Collagen shields were originally developed by Fyodorov
`(55) for use as a corneal bandage after radial keratotomy,
`keratorefractive procedures, and corneal abrasion. The
`commercially available collagen shields (Bio—Cor, Bausch
`and Lomb, Pharmaceuticals, Tampa, FL, U.S.A.) are
`biodegradable contact—lens—shaped clear films made of
`porcine scleral collagen. They dissolve over 12 to 72
`hours, depending on the degree of collagen cross-linking
`induced by ultraviolet irradiation during production of the
`device. Collagen shields are indicated for relief of discom-
`fort and to promote corneal epithelial wound healing.
`
`

`
`144 / CHAPTER 10
`
`Because corneal shields have a water content >60%,
`they can be used as a depot for ophthalmic drug delivery.
`Drugs can be coalesced into the collagen matrix during
`production, absorbed during rehydration of the device be-
`fore ocular application, or added topically over a shield al-
`ready installed in the eye. Animal studies have demon-
`strated repeatedly that corneal shields can deliver various
`drugs to the cornea, aqueous humor, or vitreous better than
`multiple—drop treatments,
`including tobramycin (56,57),
`gentamycin, vancomycin (58-60), dexamethasone (61),
`prednisolone acetate (62), cyclosporine A (63), and ampho-
`tericin B (64).
`Studies in animal disease models (Pseudomonas kerati-
`tis, anterior chamber fibrin, and corneal allograft rejection)
`also showed that the efficacy of collagen shields is superior
`to that of drop treatments of tobramycin (57), cyclosporine
`A (65), heparin (66), and tissue plasminogen activator (67).
`Some studies, however, reported comparable or even lower
`efficacy for the collagen shield compared with drops of to-
`bramycin (68), amphotericin B (69), and gentamycin
`(70,71).
`
`Only a limited number of clinical studies using collagen
`shields as ocular DDS have been published. Reidy et al.
`(72) found better delivery of fluorescein to aqueous humor
`by collagen shields in human subjects compared with de-
`livery by a soft Contact lens or frequent drop application.
`Poland and Kaufman (73) tested collagen shields rehy—
`drated in 4% tobramycin in an uncontrolled series of 60 pa-
`tients with epithelial defects.