Effect of Benzalkonium Chloride/EDTA on the Ocular Bioavailability of
`Ketorolac Tromethamine following Ocular Instillation to Normal and
`De-epithelialized Corneas of Rabbits
`
`CHERUKURY MADHU, PETER J. RIX, MARTHA J. SHACKLETON, THAI G. NGUYEN, AND DIANE D.-S. TANG-LIUX
`Received October 5, 1995, from the Department of Pharmacokinetics, Allergan, Inc., 2525 Dupont Drive, P.O. Box 19534, Irvine, CA
`Accepted for publication January 9, 1996X.
`92713-9534.
`Final revised manuscript received December 8, 1995.
`
`Abstract 0 This study was designed to examine the effect of
`benzalkonium chloride/ethylenediaminetetraacetic acid (BAK/EDTA) on
`the ocular bioavailability (Focular) of ketorolac tromethamine after ocular
`instillation to normal and de-epithelialized corneas of rabbits both invitro
`and in vivo. The in vitro Focular of the formulations was measured in
`flow-through perfusion chambers. For invivo studies, a 35 (cid:237)L dose of
`0.5% ketorolac tromethamine with or without BAK/EDTA was instilled
`into rabbit eyes with intact or de-epithelialized corneas. At 0.5, 1, 2, 4,
`6, and 8 h postdose, rabbits were eu hanized, and the corneas and
`aqueous humor were collected from both eyes. The ketorolac concentra-
`tions from both invivoand invitrosamples were quantified by reversed-
`phase high-performance liquid chromatography. The invitrostudy results
`indicated that BAK/EDTA statistically significantly increased the Focular of
`ketorolac through de-epithelialized corneas but not through intact corneas.
`The in vivo study results showed that BAK/EDTA had no effect on he
`Focular of ketorolac in rabbits with intact corneas, based on the values of
`the area under the aqueous humor concentration versus time curves
`(AUC0-6h) of ketorolac. As expected, de-epithelialization of the corneas
`produced a faster and greater ocular absorption of ketorolac as evidenced
`by the smaller Tmax and larger AUC values compared to those for he
`intact corneas in vivo. However, BAK/EDTA decreased the ocular
`absorption of ketorolac in rabbits with de-epithelialized corneas. The
`half-lives (t1/2) of ketorolac in corneal
`issue and aqueous humor were
`longer in rabbits with intact corneas than hose in rabbits with de-
`epithelialized corneas.
`In conclusion, the invivoFocular of ketorolac was
`not altered by BAK/EDTA in rabbits with intact corneas, but
`it was
`decreased by BAK/EDTA in rabbits with de-epithelialized corneas.
`Therefore, the formulation with ketorolac alone may be better as a post-
`operative ocular analgesic.
`
`Introduction
`Steroids are used in the treatment of allergic ocular
`disorders, corneal burns, uveal tract inflammation, and other
`ocular inflammations, but their use is limited by their
`tendency to increase intraocular pressure and to cause
`cataracts upon chronic administration.1 The advantage of
`nonsteroidal anti-inflammatory drugs (NSAIDS) is that they
`do not increase intraocular pressure.2
`Ketorolac tromethamine is a potent NSAID, which is an
`effective treatment for postoperative inflammation in eyes.
`It is nonirritating when topically administered to eyes at
`concentrations of up to 0.5% and does not increase intraocular
`pressure.3 The corneal epithelium is often damaged during
`ocular surgery, and alterations of the corneal epithelium have
`been shown to influence the corneal permeability of various
`compounds. Also, preservatives such as benzalkonium chlo-
`ride (BAK) are known to enhance the corneal permeability of
`ketorolac in vitro.2 Therefore, the objective of this study was
`
`X Abstract published in Advance ACS Abstracts, March 1, 1996.
`
`to evaluate the effect of BAK/ethylenediaminetetraacetic acid
`(EDTA) (both known corneal penetration enhancers) on the
`ocular bioavailability of ketorolac following ocular instillation
`to both intact and de-epithelialized corneas of rabbit eyes.
`
`Experimental Section
`Ophthalmic solutions of 0.5% ketorolac tromethamine (pH 7.4) with
`or without 0.01% benzalkonium chloride/0.1% EDTA were provided
`by Allergan (Irvine, CA).
`Female New Zealand albino rabbits weighing between 2 and 3.5
`kg were obtained from Myrtle’s Rabbitry (Thomson Station, TN). The
`rabbits were quarantined for at least 1 week upon arrival and
`examined for clinically normal eyes. The rabbits were individually
`housed with food and water provided ad libitum.
`In Vitro StudiessCorneal DissectionsThe rabbits were eutha-
`nized with Eutha-6 (Western Medical Supply Co. Inc., Arcadia, CA).
`The corneal epithelium was removed by careful scraping of the
`cornea’s surface with a scalpel blade until the stroma was exposed.
`De-epithelialization was confirmed by microscopic examination of
`the corneas after scraping. The eyes were then enucleated, and the
`corneas were excised. The freshly-excised corneas were mounted in
`flow-through perfusion chambers as previously described.3
`Dosing and SamplingsGlutathione-enriched bicarbonate Ringer’s
`solution (GBR) was added to the receiver chamber4 and bubbled with
`O2/CO2 (95%/5%). One hundred fifty microliters of the 0.5% ketorolac
`tromethamine formulation was instilled into the donor chamber inflow
`line followed by 100 (cid:237)L of blank GBR. One minute after dosing, blank
`GBR buffer was infused into the donor chamber at a flow rate of (cid:24)28
`(cid:237)L/min. The donor effluent was collected over four 60 min intervals,
`and the volume and drug concentrations were measured. Samples
`of 100 (cid:237)L were collected with replacement from the receiver chamber
`at 15, 30, 45, 60, 90, 120, 150, 180, 210, and 240 min postdose. The
`ketorolac concentrations in all samples were quantified by HPLC. At
`the end of the experiment the corneas were weighed, soaked in 1 mL
`of methanol overnight, dried at 120 °C, and reweighed. The corneal
`hydration was calculated, and the methanol extracts were assayed
`for ketorolac content by HPLC.
`In Vivo StudiessDe-epithelialization of CorneassRabbits were
`anesthetized with ketamine (Ketaset; Fort Dodge Labs, Fort Dodge,
`IA) and xylazine (Xylazine; American Animal Health; Wisner, NE).
`One drop of proparacaine (Opthetic; Allergan Inc., Irvine, CA) was
`topically applied to the cornea of the left eye. The corneal epithelium
`of the left eye was removed by scraping of the cornea’s surface with
`a scalpel blade until the stroma was exposed. The right eye was used
`intact. For another set of animals, both eyes were scraped.
`Animal Dosing and Tissue CollectionsOne 35 (cid:237)L drop of 0.5%
`ketorolac tromethamine ophthalmic formulation with or without BAK/
`EDTA was instilled into the lower cul-de-sac of each eye after the
`animals recovered from the general anesthesia. The upper and lower
`eyelids were gently held closed for (cid:24)10 s to maximize drug-cornea
`contact. At 0.5, 1, 2, 4, 6, and 8 h postdose, six rabbits each were
`euthanized, after which the cornea and aqueous humor were collected
`and each stored in amber glass tubes containing 1 mL of methanol.
`All samples were stored at -20 °C until analysis. Additional rabbits
`(two rabbits per formulation) were treated with placebo formulations
`with or without BAK/EDTA, and tissues were taken at 2 h postdose.
`HPLC AnalysissThe methanol extracts were centrifuged at 1500g
`for 15 min, and the supernatants were dried and reconstituted in
`mobile phase for HPLC analysis. A Beckman pump Model 126 (San
`Ramon, CA) was used to deliver the mobile phase at a flow rate of
`
`© 1996, American Chemical Society and
`American Pharmaceutical Association
`
`0022-3549/96/3185-0415$12.00/0
`
`Journal of Pharmaceutical Sciences / 415
`Vol. 85, No. 4, April 1996
`
`+
`
`+
`
`SENJU EXHIBIT 2090
`LUPIN v SENJU
`IPR2015-01105
`
`PAGE 1 OF 4
`
`

`
`Cumulativeketorolacinreceiver
`
`
`chamber(pg)
`
`A6
`
`—fi—lIlui couu - willrnrl BAK/EDTA
`—-—lnIuI tornn - will BAK/EDTA
`- -A- — D!-QpilIl|iIlil(d (one: without BAKIIZDTA
`- -E]- - Dc-cpitlelialiud corner with BAKIIIDTA
`
`3-NU)9OO
`
`0
`
`50
`
`150
`100
`Time (min)
`
`200
`
`250
`
`Figure 1—Eifect of BAK/EDTA on the penetraion profles of ketorolac through
`iliactand deepiihelialized rabbitcorneasin vitro_ Comeaswerepeflusedwilh
`0.5% ketrxolactrrmethaminesolrtionswiih orwithoulBAK.Vduesaremean:
`SEM, n=8.
`
`1.5 mllmin. A dry—packed precolumn was placed between the injector
`(Wisp 710B, Waters Assoc., Milford, MA) and the analytical column
`(Spherisorb ODS 511, 4.6 mm x 25 cm, Alltech, Deerfield, IL). The
`eflluent wm monitored at 254 nm with a UV detector (Beckman Model
`166, San Ramon, CA). The injection volume was 50 uL. The retention
`time and lowest limit of quantitation of ketomlac were ~l0.8 min
`and 15 ng/mL, respectively. Standards of ketorolac tromethamine
`ranging fmm 0.015 to 20.0 ug/mL were analyzed with samples Assay
`selectivity was verified by analysis of ocular tissues from animals
`treated with placebo formulations.
`
`Data Analysis
`In vitro StudierThe maximum cumulative total mass of
`
`drug in the receiver chamber (Q,..,,) was directly obtained from
`the cumulative amount of ketorolac versus time curve, which
`was corrected for the mass of drug removed during sampling
`at each time point. The in vitro ocular bioavailability was
`calculated as Fmh, = Q...”/dose.
`In vivo Studies—the maximum concentration (C...;.,,) of
`drug in the aqueous humor and the time required to reach
`C)“, (T...,,.) were obtained from the aqueous humor concentra-
`tion versus time curves. The area under the concentration
`
`versus time curve (AUC) was calculated as previously de-
`scribed.‘ The half-life (tug) of ketorolac was given by [1/2 =
`0.693Ik, where the rate constant (Ir) for ketorolac was obtained
`by log linear regression of the last three points (terminal
`portion) of the aqueous humor or corneal concentration versus
`time curve. Student's t—test was used to compare values
`between groups. The level of statistical significance was set
`at 0. = 0.05.
`
`Results
`
`In Vitro Studies—The effects of BAK/EDTA on the corneal
`
`absorption of ketorolac through intact and de—epithelialized
`corneas are shown in Figure 1 and Table 1. BAK/EDTA
`increased the corneal absorption of ketorolac in both intact
`and de—epithelialized corneas. As expected, de-epithelializa-
`tion of the corneas markedly increased the corneal absorption
`of ketorolac from both formulations, as evidenced by shorter
`T,.,,,,. and larger Funk, values compared to those for the intact
`corneas. BAK/EDTA increased the Funk, of ketorolac, but this
`increase was statistically significant only in the deLepithe—
`lialized corneas (Table 1). The ketorolac concentrations
`remaining in the corneal tissue 4 h after exposure to the
`ketorolac formulations with and without BAK/EDTA were 1 13
`
`:l: 17 and 142 :l: 15 pglg, respectively, in the intact corneas.
`
`416 / Journal of Phamraceulical Sciences
`Vol. 85. No. 4. Am‘! 1996
`
`PAGE 2 OF 4
`
`In the de—epithelialized corneas, the concentrations were 173
`i 12 and 123 :l: 15 fig/g with and without BAK/EDTA,
`respectively.
`In Vivo StudierThe in viva studies showed that the
`
`aqueous humor concentrations of ketorolac correlate well with
`the corneal tissue concentrations (Figure 2). BAK/EDTA had
`no effect on the corneal tissue or aqueous humor concentra-
`tions of ketorolac in rabbits with intact corneas (Figure 2,
`panels A and B, respectively). Deepithelialization of the
`corneas initially increased the corneal tissue and aqueous
`humor concentrations of ketorolac, but the concentrations
`decreased more rapidly than in rabbits with intact corneas
`(Figure 2, panels A and B). In rabbits with de—epithelialized
`corneas, BAK/EDTA decreased the corneal and aqueous
`humor concentrations of ketorolac compared to those of
`ketorolac alone.
`
`On the basis of the pharmacokinetic parameters, BAKI
`EDTA had no effect on the ocular absorption of ketorolac in
`rabbits with intact corneas (Table 2). De-epithelialization of
`the corneas produced a faster and greater ocular absorption
`of ketorolac as evidenced by shorter Tm, and larger C,,,,,, and
`AUC values than for the intact corneas (Table 2). However,
`BAK/EDTA decreased the ocular absorption of ketorolac in
`the de—epithelialized corneas as evidenced by lower Cm, (2.98
`:l: 0.36 _ug/mL) and AUC (403 :l: 47 ug-min/rnL) values than
`those of ketorolac alone (7.39 :l: 1.15 lug/mL and 854 :h 96 pg-—
`min/mL, respectively). BAK/EDTA had no efiect on Tm,
`values in either intact or deepithelialized corneas.
`In rabbits with intact corneas, the apparent half-lives (tm)
`of ketorolac in corneal tissue were longer (3.36 h without BAKI
`EDTA and 5.34 h with BAK/EDTA) than those in rabbits with
`de—epithelialized corneas (1.60 and 2.41 h, respectively) for
`both formulations. The half-lives of ketorolac in aqueous
`humor correlate well with corneal tissue half-lives. In aque-
`ous humor, the half-life of ketorolac was shorter for rabbits
`with deepithelialized corneas compared to that of rabbits with
`intact corneas (Table 2). BAK/EDTA had no elfect on the half-
`lives of ketorolac (Table 2).
`
`Discussion
`
`Our In vitm results are in agreement with previous reports
`in which BAK increased the in vitm corneal penetration of
`various compounds such as fluorescein,7 horseradish peroxi-
`dase,3 prednisolone phosphate,” dexamethasone and pilo-
`carpine,‘° and ketorolac} In the present study, BAK in-
`creased the corneal penetration of ketorolac in vitro ('I'able
`1). Two mechanisms have been suggested to explain the BAK-
`enhanced corneal penetration of ketorolac:
`(a) BAK may
`disrupt the integrity of the epithelial membrane; and (b) BAK
`and ketorolac may form a more lipid-soluble ion pair, which
`may enhance corneal penetration.3 However, the exact mech-
`anism by which BAK enhances corneal penetration of ketrolac
`is not known.
`
`Additionally, the ophthalmic solution with BAK also con-
`tains a low concentration of EDTA (0.1%). EDTA, a known
`calcium-chelating agent, has been shown to act on cell
`junctions by interfering with calcium ions and altering inter-
`cellular integrity.“ EDTA also disrupts the plasma mem-
`brane and consequently increases intercellular permeability."
`EDTA has been shown to increase the absorption of various
`compounds through intact comeas.”-13 Therefore, it is likely
`that EDTA in conjunction with BAK mayjointly increase the
`extent of absorption through de—epithelialized corneas. Our
`results cannot exclude this possibility in that BAK/EDTA
`markedly enhanced the In vitm penetration of ketorolac
`through deepithelialized corneas compared to that of ketoro-
`lac solution without BAK/EDTA (Figure 1).
`
`

`
`Table 1—Phannacokinetic Parameters after an Acute Administration of 0.5% Ketorolac Tromelhamite Solutions with and without BAKIEDTA to intact
`and De-epithelialized Rabbit Comeas in Vil‘ro'
`
`ComealTrealrnenl
`Intact
`
`Deepiihelialized
`
`BAKIEDTA
`—
`+
`—
`+
`
`G,.....’(ugIg)
`113i 17
`142i15
`123145
`173i12' '
`
`Gill: (#9)
`3.44 i 0.77
`6_07i1.00
`13.3 i 1.8
`25.8 i 3.6‘
`
`nil‘
`236 i 4
`23614
`135 i 21
`105 1 10
`
`% Foufle
`0.636 i 0.122
`1.07i0.18
`2.48 i 0.28
`4.64 i 0.57‘
`
`'ValuesaremeaniSEM, n= 8. "Kelorolacconoenlralion in corneal tissue, measuredatlheendollhe4 hpetfusion. ‘Max'Inalcumulativemassolkelorolac
`iireceivercharber. ‘Tlne remiiredioreachq... °Per0enioctlarl)ioavaiwiity. 'Slahsfir:dy§gnifiwrmydifiaemfiunsdufimwmnflBAWEDTAhdeepiflieiaIized
`cornea (p < 0.05).
`
`BAK/EDTA is dependent on various conditions such as the
`concentration of BAK used in the dosing solution, number of
`ophthalmic doses administered, coadrninistration with other
`compounds, and the species tested. It has been reported that
`0.01% BAK“ and 0.02% BAK‘-" increased the corneal perme
`ability of fluorescein and inulin, respectively, in rabbits in
`vivo. In contrast, 0.01% BAK failed to increase the corneal
`permeability of inulin. 15 Similarly 0.01% BAK did not alter
`the corneal permeability of fluorescein in humans in viva.
`Additionally, three repeated administrations (50 [IL each) of
`0.01% BAKIO.1% EDTA with 2 min intervals failed to increase
`
`the corneal permeability, but five repeated administrations
`of this solution did increase the corneal permeability in
`humans in v1'vo.“‘ Additionally, it has been reported that
`neither 0.34 nor 1.0% EDTA eye drops had any influence on
`the anterior chamber fluorescein concentration in humans. ‘3
`Therefore, it is not surprising that, in the present study, a
`single ophthalmic dose of 0.01% BAK/0.1% EDTA had no eifect
`on the corneal penetration of ketorolac in rabbits with intact
`corneas.
`
`In vivo ocular bioavailability is known to be altered by
`changes in lacrimation;‘7 that is, increased lacrirnation causes
`increased washout of drug, thereby decreasing the ocular
`absorption of drug from the precomeal region. BAK is known
`to cause ocular irritation. ‘3 Therefore, one can speculate that
`BAK may enhance ocular absorption of ketorolac in rabbits
`with intact comeas as it does for other oompourids3-7‘9-14-19 but,
`at the same time, BAK in combination with EDTA (another
`potential ocular irritant) might have produced increased
`irritation, and thus increased lacrimation, thereby reducing
`drug absorption. This effect could be expected to be exacer-
`bated in the d¢,Lepithelialized cornea in viva. Our results are
`consistent with this hypothesis.
`In the de-epithelialized
`corneas, BAK/EDTA decreased ocular absorption of ketorolac
`as evidenced by lower C...“ and AUC values than those
`observed after topical administration of ophthalmic solution
`with ketorolac alone, which was shown to be nonirritating
`when applied to the corneal surfaces of rats, dogs, and rhesus
`monkeys at concentrations up to 0.5%.?”
`We need to explain why the corneal penetration enhance-
`ment of BAK/EDTA was observed in vitro, but not in viva.
`This is likely due to the fact that the in vitro model is
`completely devoid of complication by variability in precomeal
`factors such as blinking, lacrimation, tear turnover, and drug
`washout. Therefore, the corneal penetration enhancement of
`BAK/EDTA was not diminished in vitro.
`
`It has been reported that deepithelialization of the cornea
`increased the penetration of pilocarpine, dexamethasone, and
`sorbitol."-Z‘ In agreement with these reports, our In vitro and
`in viva results showed that deepithelialization of the cornea
`produced faster and greater ocular absorption of ketomlac as
`evidenced by shorter Tm. and larger AUC values (Tables 1
`and 2). These results suggest that the corneal epithelium is
`rate limiting in the ocular absorption of ketorolac.
`The mean apparent half-life of ketorolac in aqueous humor
`was longer in rabbits with intact corneas than in rabbits with
`de-epithelialized corneas (Table 2). These results correlate
`
`JoumdofPharmaceulicaISa'ences/417
`Vol.85, No.4.Apn‘I1996
`
`——A—l-out cornea - without BAKIEDTA
`—fl—lntacI cornea - with BAK/EDTA
`- -A- - De-eplttnclialiud ¢ornu- without BAK/EDTA
`- -13- - De-tpilhelilliud cnrneo . -in BAKIEDTA
`
`I —A—lIt.|ct cornea - without BAK/EDTA
`—|— Intact cornea - will BAK/F.l)’l'A
`- -A- V De-tpithelialixnl cornea without BAK/EDTA
`
`| - —El- - De-epithelialiud corner with BAK/EDTA
`
`0
`
`2
`
`6
`4
`Time (hrs)
`
`8
`
`to
`
`Figue2—EfiedofBAKIE)TAmketmdacuxicermafimshuxrIealfissue(parieI
`A) and aqueous hum0r(pmeI B), alter topical application ofa 35;4Leye drop of
`0.5% of ketorolac tromehamine wih or without BAKIEDTA to intact and de
`epilheliaizedrabbitoomeasinvivo.ValuesaremeaniSEM,n=6f0rinlaci
`oomea, n = 12 for deepithelialized cornea.
`
`Table 2—Phannacokinetic Parameters of Ketorolac in Aqueous Hllllof
`following Topical Application of a 35 pL Eye Drop of 0.5% Ketorolac
`Tromethamine either with or without BAKIEDTA to Intact and
`De-epith'elialized Rabbit corneas in vivo
`
`CornealTrealrneni BAKIEDTA G....‘((4gIrr||_) T....."(h)
`Intact‘
`—
`0229:0071
`4
`+
`024510020
`1
`—
`7.39:1.15
`1
`+
`2.98i0.36'9
`0.5
`
`Deepilhel|ized'
`
`AUC0—a."
`(pg-miIImL)
`rmqn)
`57.1:0.9
`222
`57.2:7_3
`2.00
`0.746 854i96
`0.594 403147-
`
`'Max'lntln oonoenlra ion of ketorolac in ameous humor. "Tine required to
`reach C... °MemappaenthdHifeofketoro|acitaqtteoushunx)r. ‘Areamder
`aqueoushumorconcenlrationvslinecuvelromolofih. ‘Valuesaremeani
`SEM, n = 6. ‘Values are mean 4; SEM, n = 12. 9 (') Statistically significan ly
`dilfereni from solution with BAKIEDTA in deepilhelidized cornea (p < 0.05).
`When ketorolac solutions with and without BAKIEDTA
`
`were instilled into rabbit eyes In viva, BAKIEDTA failed to
`increase the corneal penetration of ketorolac. This may be
`due to the fact that the corneal penetration enhancement of
`
`PAGE 3 OF 4
`
`l00_0 A
`
`‘E
`H A
`,2 g
`E3 10 0
`E 9
`'
`3 5
`g E0 .-
`
`fl
`3 E
`'- o
`S ta0
`I
`
`1.0
`
`0.]
`
`100.0 B
`
`Ketorolacconcentrationinaqueous
`
`
`humor(pg/ml)
`
`lo 0
`
`1.0
`
`0.1
`
`
`
`

`
`well with the corneal tissue half-lives of ketorolac, suggesting
`that the corneal epithelium may be acting as a reservoir for
`drug accumulation, similar to the situation reported for
`ketorolac22 and pilocarpine.23 Thus the longer half-lives
`observed in rabbits with intact corneas may be due to a
`continued flux of drug into the aqueous humor from the
`corneal reservoir as previously reported.22,23
`Our results indicate that the corneal epithelium is impor-
`tant in the elimination/loss of drug from the anterior chamber.
`It has been reported that the mean half-life of [14C]ketorolac
`in the anterior chamber after intracameral injection to rabbits
`with intact corneas was 2.1 h.22 In this study, the mean half-
`lives of ketorolac in aqueous humor after ophthalmic admin-
`istration to rabbits with de-epithelialized corneas were much
`shorter (0.594-0.746 h).
`It is likely that, once the drug
`reaches Cmax in the aqueous humor, it may diffuse back
`through the cornea. This would lead to a more rapid elimina-
`tion of ketorolac from the aqueous humor of rabbits with de-
`epithelialized corneas than that observed after intracameral
`injection where the corneal epithelium of rabbits was still
`intact.22
`In conclusion, BAK had no effect on the ocular absorption
`of ketorolac in intact corneas in vivo. The ocular absorption
`of ketorolac was increased by de-epithelialization of the
`corneas in vivo, but it was decreased by BAK. Therefore, the
`formulation with ketorolac alone may be better as a postop-
`erative ocular analgesic. This result is unexpected and should
`be of interest to ophthalmic formulators.
`
`References and Notes
`1. Haynes, R. C., Jr. In The pharmacological Basis of Therapeutics;
`Gilman, A. G., Rall, T. W., Nies, A. S., Taylor, P., Eds.; Pergamon
`Press: New York, 1990; pp 1456-1458.
`
`2. Flach, A. J.; Jaffe, N. S.; Akers, W. A. Ann. Ophthalmol. 1989,
`21, 407-411.
`3. Fu, R. C.-C.; Lidgate, D. M. Drug Dev. Ind. Pharm. 1986, 12,
`2403-2430.
`4. Richman, J. B.; Tang-Liu, D. D.-S. J. Pharm. Sci. 1990, 79, 153-
`157.
`5. O’Brien, W. J.; Edelhauser, H. F. Invest. Ophthalmol. 1977, 16,
`1093-1103.
`6. Tang-Liu, D.; Burke, P. J. Pharm. Res. 1988, 5, 238-241.
`7. Green, K.; Tonjum, A. Am. J. Ophthalmol. 1971, 72, 897-905.
`8. Tonjum, A. M. Acta Ophthalmol. 1975, 53, 335-347.
`9. Green, K.; Downs, S. J. Invest. Ophthalmol. 1974, 13, 316-319.
`10. Camber, O.; Edman, P. Int. J. Pharm. 1987, 39, 229-234.
`11. Grass, G. M.; Wood, R. W.; Robinson, J. R. Invest. Ophthalmol.
`Visual Sci. 1985, 26, 110-113.
`12. Ashton, P.; Diepold, R.; Platzer, A.; Lee, V. H. L. J. Ocular
`Pharmacol. 1990, 6, 37-42.
`13. Rojanasakul, Y.; Liaw, J.; Robinson, J. R. Int. J. Pharm. 1990,
`66, 131-142.
`14. Burstein, N. L. Invest. Ophthalmol. 1984, 25, 1453-1457.
`15. Keller, N.; Moore, D.; Carper, D.; Longwell, A. Exp. Eye Res.
`1980, 30, 203-210.
`16. Ramselaar, J. A. M.; Boot, J. P.; van Haeringen, N. J.; van Best,
`J. A.; Oosterhuis, J. A. Curr. Eye Res. 1988, 9, 947-950.
`17. Conrad, J. M.; Reay, W. A.; Polcyn, R. E.; Robinson, J. R. J.
`Parenter. Drug Assoc. 1978, 32, 149 -161.
`18. Kennah, H. E.; Higney, S.; Laux, P. E.; Dorko, J. D.; Barrow, C.
`S. Fundam. Appl. Toxicol. 1989, 12, 258-268.
`19. Smolen, V. F.; Clevenger, J. M.; Williams, E. J.; Bergdolt, M.
`W. J. Pharm. Sci. 1973, 62, 958-961.
`20. Mohoney, J. M.; Waterbury, L. D. Invest. Ophthalmol. Visual
`Sci. (Suppl.) 1983, 24, 151.
`21. Ashton, P.; Diepold, R.; Platzer, A.; Lee, V. H. L. J. Ocular
`Pharmacol. 1990, 6, 37-42.
`22. Ling, T. L.; Combs, D. L. J. Pharm. Sci. 1987, 76, 289-294.
`23. Sieg, J. W.; Robinson, J. R. J. Pharm. Sci. 1976, 65, 1816-1822.
`JS9504189
`
`418 / Journal of Pharmaceutical Sciences
`Vol. 85, No. 4, April 1996
`
`+
`
`+
`
`PAGE 4 OF 4