ardson's stain joh. ln. ubels henr, fy. - arvo...

Volume 23Number 1 Reports 127

sure on retinal, prelaminar, laminar and retrolaminaroptic nerve blood flow in monkeys. INVEST OPHTWAL-MOL Vis SCI 18:1030, 1979.

6. Riva CE, Sinclair SH, and Grunwald JE: Autoregiila-tion of retinal circulation in response to decrease ofperfusion pressure. INVEST OPHTHALMOL VIS SCI

21:34, 1981.7. Bill A: Ocular circulation. In Physiology of the Eye,

Moses R, editor. St. Louis, 1975, The C. V. MosbyCo., p. 215.

8. Riva CE, Grunwald JE, and Sinclair SH: Laser Dopp-ler measurement of relative blood velocity in thehuman optic nerve head. INVEST OPHTHALMOL VISSCI 22:241, 1982.

A comparison of healing of corneal epithelialwounds stained with fluorescein or Rich-ardson's stain. JOHN L. UBELS, HENRY F.EDELHAUSER, AND KRISTINA H. AUSTIN.

The effects of fluorescein and Richardson's stain on cor-neal epithelial wound healing were compared in eyes ofrabbits whose corneas had the epithelium removed byscraping or by n-heptanol. One eye of each rabbit wasstained with fluorescein and the other eye was stainedwith Richardson's stain at intervals throughout thehealing process, and the wounds were photographedfor planimetry and determination of re-epithelializationrate. Corneal thickness was also measured throughoutthe re-epithelialization. These studies showed that Rich-ardson's stain, as compared with fluorescein, decreasesre-epithelialization rate, delays wound closure, andslows the return of the edematous cornea to normalthickness. Therefore fluorescein rather than Richardson'sstain should be used to stain epithelial defects in cornealwound healing studies and in the evaluation of the cor-neal toxicity of chemical agents. (INVEST OPHTHALMOL VISSCI 23:127-131, 1982.)

The clinical problems of corneal epithelial ero-sion and persistent epithelial defects have led toseveral studies of the mechanisms and biochem-istry of corneal wound healing.1"3 Evaluation of acorneal wound or infection requires the use of atopical staining technique. The most commonmethod used clinically has been fluorescein stain-ing. Fluorescein has also had wide usage in oph-thalmic research4 and has no known toxic effectson the corneas of humans or rabbits.5

A second stain, known as Richardson's stain, hasrecently been used in laboratory studies of cornealepithelial wound healing.3' 6> 7 This is a histologicstain containing methylene blue, azure II, andborax; when applied to the cornea, this stain yieldsa dark blue, well-delineated stain at the location of

an epithelial defect. It is visible under white lightand can be conveniently photographed withoutspecial filters.

Several characteristics of Richardson's stain,however, raise questions about its suitability as astain for corneal wounds. The pH is basic, 8.6, theosmolality is 15 mOsm and, most significantly,methylene blue is reported to be toxic to humanand rabbit corneas.5 Comparison of two reports inthe literature indicates that healing of chemicallywounded corneas may occur more rapidly in fluo-rescein-stained corneas8 than in corneas stainedwith Richardson's stain.6 A controlled study wastherefore undertaken to compare corneal healingrates with the use of fluorescein and Richardson'sstain. Rates of wound closure were determined bymeasurement of the area of the wound over time.Corneal thickness was also measured as an index ofthe functional integrity of the regenerating epi-thelium.

Materials and methodsExperimental procedures. Albino rabbits (2 to

2.5 kg) were sedated with ketamine HC1 (30mg/kg) and the corneas of the animals were anes-thetized with a drop of proparacaine HC1 (0.5%).A central epithelial wound was made on both cor-neas of each animal by means of either the n-hep-tanol method of Cintron et al.6 or the scrapingmethod described by Ho et al.2 A filter paperdisc 6.5 mm in diameter was used for hepta-nol wounds, and scraped corneas were debridedwithin a 6.5 mm trephine mark. The right eye ofeach animal was stained by applying a moistenedfluorescein sodium ophthalmic strip to the bulbarconjunctiva and rinsing the eye with an ophthal-mic irrigating solution (Dacriose; CooperVision,San German, Puerto Rico). The left eye wasstained with one or two drops of Richardson'sstain—1% methylene blue and 1% azure II in 1%borax 1:1 (this stain should actually be called Mal-lory's methylene blue-azure II stain9)—and rinsedwith irrigating solution. The corneal wounds werephotographed with 35 mm Ektachrome 160 filmand a Kowa fundus camera equipped with a cobaltfilter for fluorescein photography. Slides wereprojected onto paper and the corneal wound pe-rimeters were traced. The areas of these tracingswere determined by computerized planimetry (Ap-ple II computer and graphics tablet) calibrated formagnification. After the initial photographs, thewounds were stained and photographed at 4, 12,24, 30, 48, 54, 72, and 78 hr.

Healing was also monitored by measurement ofcorneal thickness with a Haag-Streit slit lamp andpachometer modified according to the method

0146-0404/82/070127+05$00.50/0 © 1982 Assoc. for Res. in Vis. and Ophthal., Inc.

Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933104/ on 06/02/2018

128 ReportsInvest. Ophthalmol. Vis. Sci.

July 1982

— Fluorescein—-Richardson's Stain

o

• oco

g

Init

100-

9 0

8 0

7 0

SO

5 0

4 0

3 0

2 0

10

B^ N • Fluorescein^ \ A A-—-Richardson's Stain

\ \

• \ \

\ \

\ v -

\ \

\ \

• \ \

V\ \\ \

\ *• . . . A . . .

10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80

Time (hr) Time(hr)

Fig. 1. Healing of corneal wounds stained with fluorescein or Richardson's stain. A, Scrapedwound. Fluorescein line fit by linear regression, slope = 0.97mm2/hr, r = 0.94. Richardson'sstain curve fit by hand. Line fit to 0, 12, 24, and 30 hr data, slope = 0.87 mm2/hr, r = 0.78. B,Heptanol wound. Fluorescein, slope = 0.89mm2/hr, r = 0.80. Richardson's stain curve fit byhand.

Table I. Corneal epithelial wound healing rates of eyes stained with fluorescein orRichardson's stain

Epithelial removal

Scraping

Rabbit No.

Fluorescein OD

Rate* r

Richardson's OS

Rate* r

n-Heptanol

Rabbit No.

Fluorescein OD

Rate* r

Richardson's OS

Rate* r

12345

MeanSE

1.000.751.000.921.30

0.99t0.089

0.940.910.990.970.94

0.440.410.800.860.56

0.610.092

0.880.880.800.920.84

6789

1011

0.920.921.001.000.910.89

0.9410.019

0.940.940.940.920.920.95

0.580.680.580.770.610.58

0.630.032

0.950.940.900.920.920.95

r = correlation coefficient.

* Expressed as mm2/hr.

t Significant difference between fluorescein and Richardson's (p ^ 0.05).

of Mishima and Hedbys.10 These readings weretaken before wounding and immediately prior tophotography. The central fixation lights were eas-ily visualized on the surface of wounded corneas.Although it is known that proparacaine can inhibitwound healing, it was necessary to anesthetize

wounded corneas so that the animal would openits eye to allow corneal thickness measurements tobe made.

The animals were euthanized by an overdose ofsodium pentobarbital at the conclusion of the ex-periments.



• Fluorescein*•—Richardson's Stain

oo

0 10 20 30 40 50 60 70 80

Time (hr)10 20 30 40 50 60 70 80

Time (h r )

Fig. 2. Corneal thickness during healing of an epithelial wound. A, Scraped wound. Thick-nesses were compared by paired t test at 12, 24, 48, and 75 hr. Dagger, No significantdifference, asterisk, significant difference (p < 0.05) mean ± S.E., n = 5. B, Heptanolwound. Thicknesses were compared at 24, 30, 54, and 72 hr. Dagger, No significant difference;asterisk, Significant difference (p < 0.05) mean ± S.E., n = 6.

Data analysis. Corneal re-epithelialization ratesfor each eye of each animal were determined bylinear regression. These healing rates were thencompared for each experimental group by thepaired t test (p < 0.05). This is the most accu-rate method of comparing healing rates, since ithas been reported that healing characteristics aremore similar within animals than among animals.8

Corneal thicknesses were also compared by t test(p ^ 0.05) at selected times during the healingprocess.

Results. Initial wound areas varied from 40 to 60mm2. In the case of heptanol wounds, this is dueto diffusion of the heptanol away from the filterpaper applicator. In scraping, the knife blade oc-casionally crossed the trephine mark, resulting insome variation of wound size. There was, how-ever, no correlation between initial wound sizeand healing rate. This has also been shown byHo et al.2

The average healing rate of corneas wounded byscraping and stained with fluorescein was 0.99mmVhr. Staining with Richardson's stain de-creased the average healing rate to 0.61 mnWhr(Table I). Staining of a heptanol wound withfluorescein or Richardson's stain resulted inhealing rates of 0.94 and 0.63 mm2/hr, respec-tively.

Graphs of pooled data expressed as percent ini-

tial wound area (Fig. 1) show that fluorescein-stained corneas wounded by scraping or by hep-tanol healed at a constant rate and that these ratesare similar to those listed in Table I. The use ofRichardson's stain resulted in a decrease in healingrate after 40 hr, yielding a sigmoid healing curve.As illustrated in Fig. 1, A, a line can be fitted onlyto the early part of this curve and results in thecalculation of a healing rate of 0.87 mm2/hr.

By comparison with fluorescein, Richardson'sstain caused a delay in corneal wound closure. Infour out of five animals with scraped cornealwounds, the fluorescein-stained wound closed ear-lier than the wound stained with Richardson'sstain. In five out of six animals with heptanolwounds, the fluorescein-stained wound closed ear-lier than the wound stained with Richardson'sstain. The epithelium of the cornea stained withRichardson's stain sloughed at 78 hr in three of theanimals with heptanol wounds.

Immediately after epithelial wounding the cor-nea swells rapidly (Fig. 2). Scraped corneas stainedwith fluorescein continued to swell for 12 hr andthen returned to control thickness by 72 hr afterwounding. Corneas wounded with heptanol con-tinued to swell for 24 hr and reached a significantlygreater thickness than scraped corneas (0.78 vs.0.68 mm). When stained with fluorescein, theheptanol-wounded corneas did not reach control


130 ReportsInvest. Ophthalmol. Vis. Sci.

July 1982

thickness by 72 hr; however, subsequent experi-ments indicated that control thickness was reachedby 90 hr after wounding.

Whether corneas are scraped or wounded withheptanol, there is no significant difference in themaximum thickness reached by corneas stainedwith Richardson's stain as compared with thosestained with fluorescein. The thickness of corneasstained with Richardson's stain, however, reacheda plateau and the deswelling phase of corneal heal-ing was significantly delayed (Fig. 2). Measure-ments on several animals indicated that the thick-ness of corneas stained with Richardson's stainmay not return to normal for as long as 1 weekafter wounding.

Discussion. The results of this study show that,as compared with fluorescein, the use of Richard-son's stain results in a decrease in wound healingrate, a delay of wound closure, and slowing of thereturn of the edematous cornea to normal thick-ness. Although the cellular mechanisms by whichthe components of Richardson's stain inhibit thenormal corneal healing process are not preciselyknown at this time, the delay in wound closureindicates possible interference with the sliding ofepithelial cells. The thickness data indicate thatRichardson's stain may also delay the establishmentof a functional epithelial barrier.

In a recent study of corneal epithelial woundhealing,7 data from the first 24 hr of healing wereused to determine a re-epithelialization rate forRichardson's stained corneas. Our data (Fig. 1, A)show that because of the sigmoid shape of the heal-ing curve when Richardson's stain is used, the useof data from the first 30 hr of healing will result inan overestimation of healing rate and an underes-timation of wound closure time.

The results of the present study also show thatvaluable data about the toxicity of a chemical sub-stance can be obtained by measurement of cornealthickness during the healing of a corneal epithelialwound. Thus it is recommended that pachometrybe used in studies of corneal epithelial woundhealing and in the evaluation of topical drugs,chemicals, and household substances. In additionto the comparison of fluorescein and Richardson'sstain by measurement of corneal thickness, wehave also shown by this method a possible toxiceffect of heptanol on the cornea, since cornealthickness is greater after a heptanol wound thanafter scraping. Two other groups have also re-ported damage to the cornea caused by heptanol.Cintron et al.6 showed that the anterior stro-mal keratocytes are damaged by heptanol treat-

ment. Hirst et al.,1 who compared several meth-ods of epithelial debridement, found that althoughepithelial hemidesmosomes begin to regenerateby day 3 of healing in a scraped cornea, no epithe-lial hemidesmosomes are present at day 7 in theheptanol-wounded cornea. This lack of junctionalcomplexes may, in part, account for the greaterswelling of the heptanol-wounded cornea. The ab-sence of hemidesmosomes may also explain thegreater tendency of the epithelium of the heptanol-wounded corneas stained with Richardson's stainto slough.

In conclusion, it is clear that Richardson's staininhibits corneal wound healing and that this stainis not acceptable for use in laboratory studies ofcorneal re-epithelialization. Fluorescein is knownto be safe for many clinical applications, and it canbe easily photographed with black and white film,using a Wratten 47B filter on an electronic flashand a Wratten 12 filter on the camera lens,8 orwith color film, using a cobalt filter on the elec-tronic flash. Fluorescein, rather than Richardson'sstain, should therefore be used to stain epithelialdefects in future studies.

From the Departments of Physiology and Ophthal-mology, Medical College of Wisconsin, Milwaukee, Wis.Supported by NIH grant EY-05450 and by a grant fromAlcon Laboratories, Inc. Submitted for publication Nov.30, 1981. Reprint requests: Henry F. Edelhauser, De-partment of Physiology, Box 26509, Medical College ofWisconsin, Milwaukee, Wis. 53226.

Key words: corneal epithelium, fluorescein, Richard-son's stain, wound healing, pachometry

REFERENCES1. Hirst LW, Kenyon KR, Fogle JA, Hanninen L, and

Stark WJ: Comparative studies of corneal surfaceinjury in the monkey and rabbit. Arch Ophthalmol99:1066, 1981.

2. Ho PC, Davis WH, Elliot JH, and Cohen S: Kinet-ics of corneal epithelial regeneration and epidermalgrowth factors. INVEST OPHTHALMOL 13:804, 1974.

3. Jumblatt MM, Fogle JA, and Neufeld AH: Choleratoxin stimulates adenosine 3',5'-monophosphatesynthesis and epithelial wound closure in the rabbitcornea. INVEST OPHTHALMOL VIS SCI 19:1321, 1980.

4. Maurice DM: The use of fluorescein in ophthalmo-logical research. INVEST OPHTHALMOL 6:464, 1967.

5. Grant WM: Toxicology of the Eye. Springfield,1974, Charles C Thomas, Publisher.

6. Cintron C, Hassinger L, Kublin CL, and Friend J: Asimple method for removal of rabbit corneal epithe-lium utilizing n-heptanol. Ophthalmic Res 11:90,1979.

7. Jumblatt MM and Neufeld AH: Characterization of



cyclic AMP-mediated wound closure of rabbit cor-neal epithelium. Curr Eye Res 1:189, 1981.

8. Moses RA, Parkinson G, and Schuchardt R; A stan-dard large wound of the corneal epithelium in therabbit. INVEST OPHTHALMOL VIS SCI 18:103, 1979.

9. Putt FA: Manual of Histopathological Technique.New York, 1972, John Wiley & Sons, Inc.

10. Mishima S and Hedbys BO: Measurement of cor-neal thickness with the Haag-Streit pachometer.Arch Ophthalmol 80:710, 1968.

Episcleral venous pressure in normoten-sive and glaucomatous beagles. KIRK N.GELATT, GLENWOOD G. GUM, REUBEN E.

MERIDETH, AND NANCY BROMBERC.

Episcleral venous pressure was measured by a noninva-sive method with a modified force-displacement trans-ducer in six laboratory-quality normotensive and 12glaucomatous beagles. The dogs were anesthetized byketamine-xylazine, acepromazine-ketamine, and halo-thane. Simultaneous intraocular pressure (10?) andblood pressure were recorded. The mean episcleral ve-nous pressures in normotensive beagles were 11.4 to 11.6mm Hg with the three methods of anesthesia; in the glau-comatous beagles the mean episcleral pressures were 10.6to 12.5 mm Hg. There were no significant differences inepiscleral venous pressure (p < 0.19 and greater) andblood pressure (p < 0.53 and greater) between the nor-motensive and glaucomatous beagles. IOP was signifi-cantly different between the normotensive and glauco-matous beagles anesthetized with acepromazine-ketamine(meanIOP,23.4and34.2mmHg,respectively;p < 0.02)and halothane (mean IOP, 19.9 and 27.4 mm Hg, respec-tively; p < 0.001) but not significant with anesthesia withketamine-xylazine (mean IOP, 26.0 and 37.8 mm Hg,respectively; p < 0.12). Episcleral venous pressure is un-changed as the disease progresses in the glaucomatousbeagle. (INVEST OPHTHALMOL VIS SCI 23:131-135, 1982.)

Noninvasive measurements of episcleral venouspressure in man and animals have used directpressure to temporarily blanch and collapse thevessel by various devices and a stream of air.1"4

These methods have employed a torsion balance,pressure chamber with transparent membranes,and force-displacement transducer. Measurementof episcleral venous pressures in rabbits and catsby direct cannulation under general anesthesiayielded highly reproducible results and had a dis-tinct end point but was an invasive procedure.5"8

Pressure-chamber methods using various mem-branes provided values similar to direct1 cannula-tion.3 The torsion-balance system, evaluated in'

Fig. 1A. Isometric force-displacement transducerwith applanating Incite cone and 20 diopter lensfor the measurement of episcleral venous pressure.

Fig. IB. Position of the episcleral venous pressuremeasuring system on the dog's eye.

man and rabbits, resulted in greater difficulty inobtaining an end point and provided higher val-ues.5 With an isometric force-displacement sys-tem, episcleral venous pressures of normotensiveand glaucomatous humans were not significantlydifferent.2 However, in one family with glaucoma,episcleral venous pressure was elevated.9 Theepiscleral venous pressure in persons with ocularhypertension was significantly lower than that insubjects with normal levels of intraocular pressure(IOP) and glaucoma.4

Normal episcleral venous pressures in rabbits,cats, and man range from 5 to 15 mm Hg. Pressurechamber measurements of the episcleral venouspressure in the dog yielded a range of 10 to 18 mmHg.8 The values in these species are remarkablysimilar in spite of the different venous pathways ofthe anterior segment.

0146-O404/82/070131+05$00.50/0 © 1982 Assoc. for Res. in Vis, and Ophthal., Inc.


ardson's stain joh. ln. ubels henr, fy. - arvo...

Documents