Reports
`
`this generalization, however, at least one and prefera-
`bly several other units of this instrument should be
`tested to determine how much inter-unit variability
`exists. Presumably, there will be little. The exercise
`is worthwhile because the TP is such a convenient
`instrument to use.
`
`KeyWords
`
`cynomolgus monkey, glaucoma, intraocular pressure,
`nometry, Tono-Pen-
`
`to-
`
`Acknowledgments
`
`The authors thank Paul Dederichs of Mentor O 8: 0, Inc.
`(Nowell, MA) for loaning them the Tono-Pen.
`
`Rderenoes
`
`1. Minckler DS, Baeweldt G, Heuer DK, Qui|len—
`Thomas B, Walonker AF, Weinerj. Clinical evaluation
`of the Oculab Tono-Pen. Am j O/zhlhalmol. 1987;
`104:168-173.
`2. Boothe WA, Lee DA, Panek WC, Pettit TH. The Tono-
`Pen: A manometric and clinical study. Arch Ophthal-
`mol. 1988;106:1214-1217.
`3. Hessemer V, Rossler R, Jacobi KW. Comparison of
`intraocular pressure measurements with the Oculab
`
`10.
`
`1199
`
`Tono-Pen versus manometiy in humans shortly after
`death. Am] Ophthalmol. 1988; 105:678—682.
`Kao SF, Lichter PR, Bergstrom T], Rowe S, Musch DC.
`Clinical comparison of the Oculab Tono-Pen to the
`Goldmann applanation tonometer. Ophthalmology.
`1987;94:1541-1544.
`Frenkel REP, Hong 1(], Shin DH. Comparison of the
`Tono-Pen to the Goldmann applanation tonometer.
`Arch Ophthalmol. 1988; 1061750-753.
`Armstrong TA. Evaluation of the Tono-Pen and the
`Pulsair tonometers. Am j Ophlhalrrwl. 1990;109:716-
`720.
`
`Geyer O, Mayron Y, Loewenstein A, Neudorfer M,
`Rothkoff L, Lazar M. Tono-Pen tonometry in normal
`and post-keratoplasty eyes. Br ] Ophthalmol. 1992;
`76:538-540.
`
`Menage M], Kaufman PL, Croft MA, Landay SP. Intra-
`ocular pressure measurement after penetrating kera-
`toplasty: Minified Goldmann applanation tonometer,
`pneumatonometer, and Tono-Pen versus manometry.
`Br_] Ophlhalmol. 1994;78:671-676.
`Mendelsohn AD, Forster RK, Mendelsohn SL, et al.
`Comparative tonometric measurements of eye bank
`eyes. Cornea. 1987;6:219-225.
`Moore CG, Milne ST, Morrison jC. Noninvasive mea-
`surement of rat intraocular pressure with the Tono-
`Pen. Invest Ophlhalmol Vis Sci. 1993;34:363-369.
`
`Dexamethasone-Cyclodextrin—
`Polymer Co-complexes in Aqueous
`Eye Drops
`
`Aqueous Humor Pharmacokinetics in
`Humans
`
`jéhannes Ktiri Kristinss0n,* Hafnin Fridriksdéttinf
`Sigrtdur Th6risd6ttir,* Anna M. Sigurdardéttinf
`Einar Stefdnsson,* and Thorsteirm Loftsson1'
`
`Purpose. To test an aqueous eye drop solution con-
`taining a high concentration of dexamethasone in a
`cyclodextrin-based drug delivery system. This system in-
`creases both drug solubility in aqueous eye drops and
`
`From the ‘Department of ()phlhalmalog5, I,andakoLtrpilul.i, and lIIefI.)z[mr1mnIl of
`Phannruy, University q/‘Iceland. Reyhjavlk, Ireland.
`Pmmled in part at the annual meeting of the /lxxodalion for Research in Vision
`and Ophthalmology, Sarasota, Florida. May I994.
`Supported by Larulalwlsspflali Research Fund. (he Helga _[o'n.cdr5uir and Sigurlidz
`Krisljrirumn Memorial Fund, the University of Ireland Ruearch Fund, and the
`Icelandic Research Council
`Submiuedfor publication October 30, I 995,‘ reuicedjaruurry I6, I 996,‘ aaepled
`january 23, I996.
`Proprietary inlerzsl calegmy: P.
`Reprint requests: Einar Stejrinsson. Prafesmr of Ophthalmology, University of
`Iceland, LanIlaIwLr.vpi'lali, IS I0! Reykjavik, Iceland.
`
`drug permeability into the eye, through drug-cyclodex-
`trin-polymer ccrcomplexes.
`
`Meflwds. 2-hydroxypropyl-3-cyclodextrin is a water-so|u-
`ble oligosaccharide that can be used to dissolve lipo-
`philic drugs, such as dexamethasone, in aqueous solu-
`tions. Co-complexation with a polymer further in-
`creases the solubility and increases drug permeability
`through biologic membranes. Eye drops containing
`dexamethasone (0.32% and 0.67%). 2-hydroxypropyl-
`flqclodexuin. and polymer were given to patients be-
`fore cataract surgery, and the resultant dexamethasone
`concentration was measured from aqueous humor sam-
`ples.
`
`Results. The dexarnethasone—cyclodextrin drops give a
`significantly higher concentration of dexamethasone in
`aqueous humor than dexamethasone alcohol 0.1%
`(Maxidex). Heating of the dexamethasone-cyclode»
`m'n-polymer co-complexes appears to enhance the per-
`meability of the drug into the eye.
`
`Conclusions. The cyclodextrin-based drug delivery sys-
`tem enhances both the solubility of dexamethasone in
`aqueous eye drops and the permeability of the drug into
`the human eye. Dexamethasone concentration levels in
`the human aqueous humor exceed those reported with
`currently available steroid eye drops. Invest Ophthal-
`mol Vis Sci. 1996;37:1199—l203.
`
`SENJU—M|TSUB|SH| 2033
`
`

`
`I200
`
`Investigative Ophthalmology Sc Visual Science, May 1996, Vol. 37, No. 6
`
`Dexamediasone has been one of the most frequently
`used topical ocular corticosteroids.‘ Topical cortico-
`steroids are used for diseases of the outer eye and
`anterior segment of the eye. Inflammatory diseases
`in the posterior segment of the eye usually require
`systemic corticosteroids. Topical corticosteroids are li-
`pophilic water-insoluble compounds that are only sol-
`uble to a limited extent in aqueous eye drop formula-
`tions and, thus, frequently are formulated as suspen-
`sions, such as 0.1% dexamethasone alcohol, or as
`hydrophilic water-soluble prodrugs, such as 0.1% dex-
`amethasone sodium phosphate.
`To increase the bioavailability of topical ocular
`steroids, it is desirable to increase the concentration
`
`of the steroid drug in aqueous eye drops and the per-
`meability through the cornea or sclera. Our hypothe-
`sis is that both these aims can be accomplished with
`the use of the cyclodextrin-based drug delivery system
`for eye drops.
`2-hydroxypropyl-,6<:yclodextrin (I-IP/3CD) is a hy-
`droxypropyl substituted B-cyclodextrin.“ It is a cyclic
`oligosaccharide with a hydrophilic outer surface. Al-
`though soluble in water, it has a lipophilic cavity in
`the center. I-IPfiCD forms inclusion complexes with
`many lipophilic drugs by taking up a drug molecule,
`or part of it, into the cavity. In this way, it is possible to
`form aqueous drug-HPfiCD complexes of lipophilic
`drugs.‘ The complexes are readily dissociated because
`no covalent bonds are formed. As a rule, I-IPflCD mol-
`ecules do not penetrate biologic membranes but act
`as penetration enhancers by assuring constant high
`concentration of dissolved drug at the membrane sur-
`face. In topical drug formulations, HPHCD keeps wa-
`ter-insoluble drug molecules in solution, delivering
`them to the surface of the barrier where they partition
`into the barrier.” In doing so, I-IPflCD improves ocu-
`lar bioavailability of drugs by increasing their rate of
`absorption through the corneal barrier. Complex-
`ation of dexamethasone with HPBCD increases the
`absorption of dexamethasone 0.1% through the rab-
`bit comea by 40%.’
`We have discovered that in aqueous solutions, wa-
`ter-soluble polymers (hydroxypropyl methylcellulose
`[HPMC]) increase the solubilizing effect of cyclodex-
`trins on lipophilic drugs by increasing the stability
`constants of the drug—cyclodextrin complexes. The
`polymers form co-complexes with the drug and cyclo-
`dextrin molecules. At the cornea, the polymers may
`adhere to the surface. This promotes the release of
`drug molecules from the cyclodextrin inclusion com-
`plexes into the solution leading to a high concentra-
`tion of drug molecules at the corneal surface, resulting
`in permeability enhancement. However, the cyclodex-
`trin-based drug delivery system has not been found to
`alter the banier function of the comea.”
`
`aqueous dexamethasone—HPfiCD-HPMC co-com-
`plex with a dexamethasone concentration up to
`1.28%. In a previous study,” we compared the admin-
`istration of 1.28% dexamethasone eye drop solution
`in rabbits to 0.1% dexamethasone alcohol suspension
`(Maxidex; Alcon Laboratories (Fort Worth, TX).
`
`Dexamethasone concentration in aqueous humor
`measured, on average, 4.7 times greater with 1.28%
`dexamethasone than with 0.1% dexamethasone. The
`
`purpose of this study was to determine the ocular bio-
`availability of 0.32% and 0.67% dexamethasone-
`HPHCD-I-IPMC by administrating the eye drops to
`patients undergoing cataract surgery and comparing
`it to the bioavailability of the commercially available
`dexamethasone 0.1% suspension (Maxidex). Further-
`more, we tested the efiect of heating (i.e., forming
`the co-complex) on the ocular bioavailability of 0.67%
`dexamethasone.
`
`MATERIALS AND METHODS. Patients sched-
`
`uled to undergo cataract surgery were recruited to the
`study. Infonned consent was obtained. Patients with
`corneal disease,
`inflammatory ocular disease, glau-
`coma, or with the only potentially seeing eye undergo-
`ing surgery or receiving systemic or topical steroid
`treatment were excluded.
`
`Dexamethasone was obtained from Sigma Chemi-
`cal (St. Louis, MO), HP[3CD of molar substitution 0.6
`was obtained from Wacker—Chemie (Munich, Ger-
`many), hydroxypropyl methylcellulose (HPMC) 4000
`was obtained from Mecobenzon (Copenhagen, Den-
`mark), and Maxidex eye drops were obtained from
`Alcon Laboratories. The dexamethasone-I-IPfiCD—
`HPMC co-complex was produced by heating the dexa-
`methasone—I-IPBCD with 0.10% hydroxypropyl meth-
`ylcellulose in an autoclave (l20°) for 20 minutes.
`
`Four types of preparations were tested:
`
`Preparation 1
`0.32% dexamethasone
`0.1% HPMC
`
`5% HPHCD
`0.05% sodium edetate
`0.7% sodium chloride
`0.01% benzalconium chloride
`
`Preparation 2
`0.67% dexamethasone
`0.1% HPMC
`
`10% I-IP,8CD
`0.05% sodium edetate
`0.55% sodium chloride
`0.01% benzalconium chloride
`
`We have formed eye drops containing a novel
`
`Heated in an autoclave (l20°) for 20 minutes
`
`

`
`1201
`
`
`
`—O'— 0.32% dexamethasone
`
`*°— 0.1% dexamethasone (MaxidexO)
`
`§
`
`§
`
`(ng/ml)
`
`
`
`
`
`Dexamethssoneconcentrationinanteriorchamber
`
`Time from application (hours)
`
`FIGURE 1. Dexamethasone concentration in aqueous humor
`after the administration of 0.32% dexamethasone—2-hy-
`droxypropyl-3-qclodextrin-hydroxypropyl methylcellulose
`and 0.1% dexamethasone alcohol (Maxidex). Mean concen-
`tration 1' SEM is shown at appropriate time points after
`administration of the eye drops.
`
`This research followed the tenets of the Declara-
`
`tion of Helsinki. Approval was obtained from the eth-
`ics committee of Landakotsspitali and the Icelandic
`Drug Administration (Ministry of Health).
`
`RESULTS. One hundred twenty-five patients par-
`ticipated in the study. Of those, 47 received 0.32%
`dexamethasone-HPBCD-I-IPMC, 26 patients
`re-
`ceived heated 0.67% dexamethasone-—HP/3CD-
`HPMC, 18 patients received unheated dexametha-
`sone 0.67%—1-1P/3CD, and 34 patients received Maxi-
`dex. Two additional patients were excluded because
`of abnormally high values of dexamethasone in the
`anterior chamber, one after receiving heated 0.67%
`dexamethasone-HPfiCD-HPMC (655 ng/ml after 4
`hours, mean concentration. ,,.,,,,, 1- SEM = 50.5 I 33.5
`
`ng/ml) and the other after receiving unheated 0.67%
`dexamethasone—HPBCD—HPMC (116 ng/rnl after
`1.9 hours, mean concent.ration._g .,.,..., 1 SEM = 6.4 1
`
`6.4 ng/ml).
`Dexamethasone concentration in the anterior
`
`chamber was higher after the administration of 0.32%
`dexamethasone-HPflCD—HPMC solution compared
`with Maxidex suspension (Fig. 1) or 2.6 times higher
`levels comparing AUC curves of the preparations (Ta-
`ble 1), which was a statistically significant difference
`(P < 0.001).
`
`Administering dexamethasone solution contain-
`ing an approximately 2-fold higher concentration of
`dexamethasone, or 0.67% compared to 0.32%, did
`not result in a much higher dexamethasone concen-
`tration in the aqueous humor, with a 0.67% dexameth-
`asone-0.32% dexamethasone concentration ratio of
`
`1.13. Conversely, the duration of activity seemed to be
`longer with 0.67% dexamethasone because dexameth-
`
`Reports
`
`Preparation 3
`0.67% dexamethasone
`0.1% HPMC
`
`11.5% HPfiCD
`0.05% sodium edetate
`0.55% sodium chloride
`0.01% benzalconium chloride
`Unheated
`
`Preparation 4
`0.1% dexamethasone alcohol (Maxidex)
`
`Other preoperative procedures were routine and
`included the administration of 1 drop of eye drop
`solution containing 0.1% indomethacine 60 and 30
`minutes before surgery and 1% cyclopentolate and
`10% phenylephrine, each 60, 45, and 30 minutes be-
`fore surgery.
`One drop (50 pl) of 0.32% dexamethasone-
`HP/3CD-HPMC or 0.67% dexamethasone-HP[3CD—
`HPMC (heated or unheated) or 0.1 % dexamethasone
`alcohol (Maxidex) was administered into the lower
`conjunctiva] fornix of the eye prepared for cataract
`surgery at predetermined time points before surgery.
`Twelve patients acted as controls and received no dex-
`arnethasone preparation. The selection of prepara-
`tions to patients was randomized.
`Before opening the anterior chamber, 0.1 ml of
`aqueous humor was withdrawn with a small needle
`and syringe. The time interval between the topical
`administration of the dexamethasone eye drops and
`aspiration of the aqueous humor was recorded. Quan-
`titative determination of dexamethasone concentra-
`
`tion was performed by liquid chromatography on a
`high-performance liquid chromatography component
`system (ConstaMet.ric [Bic 8c Bemsten a/s, Denmark]
`3200 solvent delivery system, Rheodyne [Cotati, CA]
`7125 injector, a Beckman Ultrasphere [Caltech, Pasa-
`dena, CA] ODS 5 pm [4.6 X 150 mm] column and a
`Spectro Monitor [Bic 8c Bemsten] 3200 UV, Milton
`Roy [Spectronic Instruments, Rochester, NY], vari-
`able-wavelength detector operated at 242 nm). The
`mobile phase consisted of acetonitrite, tetrahydrofur-
`ane, and water (30:1:69). The flow rate was 1.8 m1/
`
`minute, and the retention time was 4.4 minutes. Aque-
`ous humor samples were injected directly into the col-
`umn without pretreatment. The drug recovery from
`aqueous humor samples that had been spiked with
`dexamethasone was estimated to be approximately
`100%.
`
`The area under the curve (AUC) for each prepa-
`ration was generated as a linear combination of means
`from 0 through the last observation time. Statistical
`methods for demonstrating bioequivalence by com-
`paring the drug preparations were adopted from
`Schoenwald et al." These included calculating the
`variance estimates for each AUC and using them to
`construct 95% confidence limits on each area and
`
`calculating a t-test comparing two areas.
`
`

`
`1202
`
`Investigative Ophthalmology & Visual Science, May 1996, Vol. 37, No. 6
`
`TABLE 1. Statistical Comparison of Dexamethasone Concentration
`Curves After Administration of 0.1% Dexamethasone Alcohol
`Suspension (Maxidex) and 0.32% Dexamethasone-HPBCD-HPMC
`(0.32%) Solution
`
`Maxidex
`0.32%
`
`AUG
`
`124.11
`327.57
`
`SE
`
`A
`
`26.42
`34.53
`
`95% a
`
`97.69-150.52
`293.034.52.10
`
`«yr
`
`9.13
`26.64
`
`AUC = area under the curve; SE = standard error of the estimate of AUC; 95% C1 = 95 percent
`confidence interval for AUG; :1] = approximate degrees of freedom for the standard error of the
`estimate of the AUC.
`
`0.32% vs. Maxidex (nest): AU(3umm,.¢, = 203.46; SEM,,m-,,,,..,, = 43.38; dj{,m.,.,,.,, = 32.86; tvalue =
`4.68; P < 0.001.
`
`TABLE 2. Statistical Comparison of Dexamethasone Concentration
`Curves After Administration of Unheated 0.67% Dexatnethasone-
`HPfiCD-HPMC (Unheated) and Heated 0.67% Dexamethasone-
`HPfiCD-HPMC (0.32%) Solution (Heated)
`
`Group AUC
`
`Group SEM
`
`95 % (1
`
`Unheated
`Heated
`
`90.70
`260.08
`
`19.88
`59.84
`
`7083-1106
`200.24- 319.91
`
`df
`
`1.61
`3.84
`
`AUC = area under the curve; SE. = standard entor of the estimate of AUC; 95% C1 = 95 percent
`confidence interval for AUG; :1] = approximate degrees of freedom for the standard error of the
`estimate of the AUC
`Heated vs. unheated (btest): AUC,g.m.,,,e = 169.38; SEM..,m..,.,.¢, = 63.05; dfi.m.,...., = 4.67; (value =
`2.69; P < 0.05.
`
`ability of dexamethasone compared with the commer-
`cially available 0.1% dexamethasone, Maxidex. No
`toxic effects were observed.
`
`§
`
`
`
`
`
`Dexamethasoneconcentrationinanteriorchamber nglml §
`
`0
`
`2
`
`4
`
`
`
`*0— Heated (co-complex)
`—°— Unheated (no co-complex)
`
`6
`
`8
`
`10
`
`Time front application (hours)
`
`FIGURE 2. Dexamethasone concentration in aqueous humor
`after the administration of heated 0.67% dexarnethasonc—
`
`2-hydroxypropyl-fl-cyclodextr-in—hydroxypropyl methylcel-
`lulose and unheated 0.67% dexamethasone-2-hydroxypro-
`pyl-fl-cyclodextrin. Mean concentration 2 SEM is shown at
`appropriate time points after adminisuation of the eye
`drops.
`
`asone concentration was measurable in the aqueous
`humor9 hours after instillation with 0.67% dexameth-
`asone, but not with 0.32% dexamethasone or Maxi-
`dex.
`
`By heating the 0.67% dexamethasone—HP,6CD—
`HPMC (i.e., formation of the cyclodextrin-polymer
`co-complex), a higher concentration was obtained in
`the anterior chamber compared to unheated 0.67%
`dexamethasone—HP;3CD-HPMC (Fig. 2) or 2.9 times
`higher levels comparing AUC curves of the prepara-
`tions (Table 2), which was statistically significant (P
`< 0.05).
`
`DISCUSSION. This study shows that it is possible
`to achieve efiective trans-ocular delivery of lipophilic
`corticosteroids, such as dexamethasone, by producing
`a three-way co-complex consisting of the drug, 2-hy-
`droxypropyl-‘5—cyclodextrin, and hydroxypropyl meth-
`ylcellulose. The drug delivery system produces a
`higher concentration of dexamethasone in aqueous
`eye drops than would otherwise be possible. Heating
`the dexamethasone—cyclodextrin-polyrner co-com-
`plexes seems to enhance permeability into the eye
`regardless of the drug concentration.
`The cyclodextrin-based drug delivery system is as-
`sociated with a large increase in the intraocular avail-
`
`

`
`Reports
`
`TABLE 3. Adjusted Mean Peak Concentrations
`(: SEM) of Corticosteroids in Aqueous
`Humor of Human Patients After Topical
`Administration
`
`Cor-tieostemids
`
`Mean Conoentmtion (ng/ml)
`
`0.32% dexamethasone-
`HPflCD-HPMC
`Dexamethasone alcohol
`0.1%, Maxidex
`Prednisolone acetate 1%"’
`
`140.5 I 36.4
`
`59.5 1- 20.8
`95.7 : 19.0
`
`"Prednisolone is a sevenfold weaker steroid than
`dexamethasone.‘ "" The intraocular concentrations reported by
`McGhee et al'" are adjusted accordingly.
`
`Watson and associates” measured the dexametha-
`
`sone in human aqueous humor after the administra-
`tion of Maxidex with results that were similar to ours.
`Prednisolone concentration in the anterior chamber
`
`after the administration of prednisolone acetate 1%
`is 20-fold higher than dexamethasone concentration
`after the administration of Maxidex 0.1%." However,
`
`prednisolone is a 7-fold less potent steroid than dexa-
`methasone,""5 calling for adjustment of the concen-
`tration in anterior chamber when compared with dex-
`amethasone preparations (Table 3). Results also show
`that heating the dexamethasone-HP[3CD complex
`with 0.10% hydroxypropyl methylcellulose increases
`the delivery of dexamethasone into the human eye.
`Our results indicate that by using a cyclodextrin-
`based drug delivery system, it is possible to raise the
`intraocular corticosteroid concentration from what is
`
`possible with currently available steroid eye drops, and
`presumably to exert a greater anti-inflammatory effect
`inside the eye. Further study is needed to establish
`the clinical efficacy of these drug formulations.
`
`KeyWords
`
`aqueous humor, corticosteroids, cyclodextrin, dexametha-
`sone, drug penetration
`
`Admowledgmam
`
`The authors thank Drs. Fridbertjonasson, Oli Bjorn Han-
`nesson, Ingimundur Gislason, and Gudmundur Viggosson
`for their invaluable assistance in acquiring samples for analy-
`sis. They also thank Sigurborg Sigurjonsdottir, Sveinbjorg
`Gunnarsdottir, and other nurses at the Department of Oph-
`thalmology, Landakot Hospital, for their critical contribu-
`tion to the study.
`
`1203
`
`Rderences
`
`1. Gordon DM. Dexamethasone in ophthalmology. Am
`_[ Ophlhalmol. 1959;48:656-660.
`2. Muller BW, Brauns U. Solubilization of drugs by modi-
`fied B-cyclodextrins. IntjPha1macoL 1985;26:77-88.
`3. Pitha J, Milecki j, Fales H, Pannell L, Uekama K. Hy-
`droxypropyl-B-cyclodextrin: Preparation and charac-
`terization: Effects on solubility of drugs. Int] Pharma-
`col. 1986;29:73-82.
`4. Loftsson T, Brewster ME, Derendorf H, Bodor N. 2-
`hydroxypropyl-B-cyclodextrinz Properties and usage
`in pharmaceutical
`formulations. Pharm Ztg Wirw.
`l991;136:5-10.
`
`5. Szejtlij. Cydodactrin Technology. Dordrecht: Kluwer Ac-
`ademic; 1988:39-43, 188- 192.
`6. Loftsson T. Olafsdéttir B], Bodor N. The effects of
`cyclodextrin on transdermal delivery of drugs. Eur j
`Pharmaool Biopharmacol. 1991;37:30--33.
`7. Usayapant A, Karara AH, Narurkar MM. Effect of 2-
`hydroxypropyl-B-cyclodextrin on the ocular absorp-
`tion of dexamethasone and dexamethasone acetate.
`
`Pharmaool Res. l99l;12:l-195-1499.
`8. van Doome H. Interactions between cyclodextrins
`and ophthalmic drugs. Eur] Pharmacol Biophamtacol.
`1993;39:133-139.
`9. jansen T, Xhonneux B, Mesens j, Borgers M. Beta-
`cyclodextrins as vehicles in eye-drop formulations: An
`evaluation of their effects on rabbit corneal epithe-
`lium. Lens Eye Toxicity Res. 1990; 7:459-468.
`10. Loftsson T, Fridriksdéttir H, 'I’h6n'sd6ttir S, Stefans-
`son E. The efiect of hydroxypropyl methylcellulose
`on the release of dexamethasone from aqueous 2-
`hydroxypropyl-[3-cyclodextrin formulations. IntjPhar-
`maool. 1994;l04:181—l84.
`ll. Schoenwald RD, Harris RG, Turner D, Knowles W,
`Chien D—S. Ophthalmic bioequivalence of steroid/
`antibiotic combination formulations. Biophann Drug
`Dispos. l987;8:527-548.
`12. Watson D, Noble M], Dutton GN, Midgley]M, Healey
`TM. Penetration of topically applied dexamethasone
`alcohol into human aqueous humor. Arch Ophthalmol.
`1988; l06:686-687.
`
`13. McGhee CN], Watson DG, Midgley_[M, Noble M], Dut-
`ton GN, Fern Al. Penetration of synthetic corticoste-
`roids into human aqueous humour. Eye. 1990;4:526-
`530.
`
`14. Havener WH. Corticosteroid therapy. In: Havener
`WH, ed. Ocular Pharmacology. 5th ed. St. Louis: CV
`Mosby; 1983:4133-500.
`15. Haynes RC, Murad F. Goodman and Gilmanit The Phar-
`macological Basis ofTherapeutics. 7th ed. New York: Mac-
`Millan; l985:1475—1-177.