1776 INVESTIGATIVE OPHTHALMOLOGY G VISUAL SCIENCE / December 1986 Vol. 27

earlier study6 demonstrated that injections of a highlyhyperosmotic solution into the vitreous rapidly causedretinal detachment and disruption of RPE apical mor-phology in a sequence of changes similar to (but morerapid than) the ones described herein. The solutionsinjected into blebs during the present experiments wereessentially isoosmotic with tissue fluids, but other ab-normalities of the extracellular milieu might be present.

Our method for producing non-rhegmatogenousretinal detachments in rabbit is relatively gentle insofaras morphological criteria are concerned. Comparisonof our SEMs of the apical surface to published trans-mission electron micrographs of attached RPE-retina9'10" shows that delicate apical processes andcone sheaths survive our procedure intact, and are vis-ible immediately following detachment. Nonetheless,the halo region was obviously highly disturbed and wethink it represents the point at which the RPE was stilladherent to the inflating retina. Since the retina overthe bleb was necessarily stretched, there must have beenconsiderable force acting to stretch adherent RPE pro-cesses in the halo region, thus accounting for their ori-entation and appearance.

Key words: RPE, apical morphology, retinal detachment

Acknowledgments. The authors are indebted to SusanLauber and Suzanne Tharpe for valuable assistance in prep-arations for electron microscopy.

From the Division of Ophthalmology, Stanford University Schoolof Medicine, Stanford, California. This work was supported in partby National Eye Institute grant EY01678, National Research ServiceAward EY07053, and the Medical Research Service of the VeteransAdministration at the Ophthalmology Section, Veterans Adminis-

tration Medical Center, Palo Alto, California. *Present address: 13833Barton Court, Los Altos Hills, California 94022. fPresent address:Department of Ophthalmology, Kyoto University Faculty of Med-icine, Sakyo-Ku, Kyoto 606, Japan. Submitted for publication: March27, 1986. Reprint requests: Joe Immel, 13833 Barton Court, LosAltos Hills, CA 94022.

References

1. Kroll AJ and Machemer R: Experimental retinal detachment inthe owl monkey: III. Electron microscopy of retina and pigmentepithelium. Am J Ophthalmol 66:410, 1968.

2. Kroll AJ and Machemer R: Experimental retinal detachmentand reattachment in the rhesus monkey; electron microscopiccomparison of rods and cones. Am J Ophthalmol 68:58, 1969.

3. Inahara M: Studies on the fine structure of retinal detachment.Acta Soc Ophthalmol Jpn 77:1002, 1973.

4. Johnson NF and Foulds WS: Observations on the retinal pigmentepithelium and macrophages in experimental retinal detachment.Br J Ophthalmol 61:564, 1977.

5. Anderson DH, Stern WH, Fisher SK, Erickson PA, and BorgulaGA: Retinal detachment in the cat: The pigment epithelial-pho-toreceptor interface. Invest Ophthalmol Vis Sci 24:906, 1983.

6. Marmor MF, Martin LJ, and Tharpe S: Osmotically inducedretinal detachment in the rabbit and primate. Invest OphthalmolVis Sci 19:1016, 1980.

7. Marmor MF, Abdul-Rahim AS, and Cohen DS: The effect ofmetabolic inhibitors on retinal adhesion and subretinal fluid ab-sorption. Invest Ophthalmol Vis Sci 19:893, 1980.

8. Tso MOM and Friedman E: The retinal pigment epithelium: I.Comparative histology. Arch Ophthalmol 78:641, 1967.

9. Sjostrand FS and Nilsson SE: The structure of the rabbit retinaas revealed by electron microscopy. In The Rabbit in Eye Re-search, Prince JH, editor. Springfield, IL, Charles C. Thomas,1964, pp. 1-65.

10. Scullica L and Tangucci F: The ultrastructural relationship be-tween pigment cells and photoreceptors. J de Microscopie 7:1085, 1968.

11. Bunt AH: Fine structure and radiography of rabbit photoreceptorcells. Invest Ophthalmol Vis Sci 17:90, 1978.

Comparative Aqueous Outflow Facility Measurements byPneumatonography and Schiotz Tonography

Joseph G. Feghali,* Dimirri T. Azar,* ond Paul L. Kaufmanf

Tonography was performed on 36 eyes of 15 normal and 3primary open angle glaucoma patients using pneumatonog-raphy and classical Schiotz tonography. The average valuesof the coefficient of outflow facility (C) for the whole samplewere virtually identical with both methods. However, bothintersubject and interobserver variability were significantlyhigher with penumatonography. Although both methods pro-vide comparable aggregate estimates of aqueous outflow fa-cility, we think that Schiotz tonography is more reliable thanpneumatonography because of the greater mechanical stabilityof the Schiotz instrument on the eye. On the other hand,pneumatonography offers the advantage of a shorter test pe-

riod (2 min instead of 4). Invest Ophthalmol Vis Sci 27:1776-1780,1986

Measuring aqueous outflow facility (C) tonograph-ically has wide applications in clinical glaucoma re-search. However, because tonography involves someinherent sources of error,1'2 some technical limitations3

and subjective approximations in estimating the Cvalue,2 and because the test is usually not necessary forthe management of individual glaucoma patients,4'5 ithas been largely abandoned in clinical practice.

Downloaded From: https://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933357/ on 09/07/2018

No. 12 Reports 1777

Since their introduction by Langham and co-work-ers,6'7 pneumatonometry and especially pneumaton-ography have been subject to controversy in theophthalmic literature. While Langham8 considers thatapplication of the pneumatic tonometer to the meas-urement of outflow facility has theoretical and practicaladvantages over conventional Schiotz indentationtonography, Moses and Grodzki9 concluded that theinstrument is position and force sensitive, that itunderestimates the magnitude of a change in intra-ocular pressure, and that it is not well suited for ton-ography.

In this study, we compare measurement of the Cvalue and variability in interpreting tracings obtainedby each method.

Materials and Methods. Subjects: Fifteen normalvolunteers, aged 24 to 30 years, and three glaucomapatients, aged 42 to 67 years, were studied. All subjectswere white; 11 were male and 7 female. Subjects wereincluded in the study after written informed consentwas obtained and a complete eye examination per-formed. The following conditions disqualified patientsfrom entering the study: intraocular pressure higherthan 22 mmHg, poor fixation in either eye, nystagmus,tropias, phorias, contact lens wear, recent eye surfaceinfection, and corneal scarring or irregularities thatprevent reliable tonometry.

Tonography: All subjects underwent tonography us-ing an electronic Schiotz tonograph (V-Mueller andCo., Chicago, IL) on one eye, and a pneumatonograph(Alcon, Ft. Worth, TX) on the other eye. Three daysto 1 week later, at the same hour of day, tonographywas repeated with crossover use of the testing instru-ments. The initial method of testing on each eye wasrandomly chosen. In the meantime, no modificationwas made in the medical therapy of the glaucoma pa-tients, which consisted of timolol 0.5% eye drops ap-plied to each eye every 12 hours. All the test procedureswere performed by one of us (Dr. Feghali, a tonogra-pher having more than 2 years of intensive experiencewith both techniques), in a cool, quiet room, with thepatient in the supine position. A dim red light placedat 6 feet from the patient's eyes was used for fixation.Before each measurement, the instruments were cali-brated according to the manufacturer's instructions,the test procedure explained to the patient, and theimportance of fixation stressed. Baseline intraocularpressure was determined for each eye using initial pres-sure (Po) determination tables for Schiotz tonography,10

and by taking the mean level of the recorded pulsewave in the supine position for pneumatonography.11

Four min tracings were then obtained with the Schiotztonograph, and 2 min tracings with the pneumaton-ograph, to which probe a 10 g weight was added.

Statistical analysis: The tonography tracings wererandomly and separately presented to one of us (Dr.Azar) and to four other assistants, each of whom wasasked to draw a best fitting line along each curve andto determine the initial and final readings on the scale.Tracings which two or more observers found difficultto fit with a line because of multiple irregularities, andthose which failed to display ocular pulsations overmore than one quarter of their total span, were con-sidered technically unsatisfactory and the correspond-ing eyes excluded from the analysis. Thirteen tono-grams out of 72 (18%) were considered technically un-satisfactory (8 pneumatonograms and 5 Schiotztonograms), causing exclusion of 11 eyes and 2 patients(both with normal eyes), as well as 9 satisfactory ton-ograms in the 11 excluded eyes. If two satisfactorytracings were obtained from only one eye of an indi-vidual, only data from that eye were analyzed; if bothtracings from both eyes were satisfactory, data fromthe two eyes were averaged for each tonographicmethod, and the averages used in the analysis. Thecomparison of the techniques is thus based on 25 eyesof 16 patients, with each patient contributing one valuefor each technique and constituting a single statisticalunit. Determination of outflow facility from the trac-ings was done using Schiotz10 and pneumatonographytables,12 with no additional correction for scleral rigid-ity. Analysis of the first 2 min of the Schiotz tonographytracings, for comparison with the 2 min pneumaton-ography, data was not attempted. Analyses of such ab-breviated Schiotz tracings are fraught with a variety ofuncertainties and were abandoned by investigatorsyears ago (personal communication with Dr. RobertMoses), so that 4 min Schiotz tonography remains theaccepted standard.

For each individual subject, the C values determinedfrom the pneumatonography tracing by each of thefive observers (CiP) were averaged to give a mean Cand standard deviation for that individual subject withthat technique (CiP ± SD1P). The individual means(Cip) were averaged to give a mean C and standarddeviation for the entire group of subjects (n = 16) withthat technique (C2p ± SD2p). The individual standarddeviations (SD1P) were averaged to give a mean stan-dard deviation for the entire group of subjects withthat technique (SD1P). The same calculations were donefor the C values obtained from the Schiotz tonographytracings, generating Cis, C1S ± SD1S, C2S ± SD2S, andSDis. These parameters permitted statistical compar-ison of mean facility, intersubject variability and in-terobserver variability with the two techniques, usingt- or F-tests, as described below.

Results. Mean outflow facility for the entire groupof 16 subjects was similar with the two techniques. For

Downloaded From: https://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933357/ on 09/07/2018

1778 INVESTIGATIVE OPHTHALMOLOGY & VI5UAL SCIENCE / December 1986 Vol. 27

O.5O

C1S (fj\ x min"1 x mm Hg"1)

Fig. 1. Mean Schiotz and pneumatonographic outflow facility foreach individual subject (C|S, C,P) plotted against one another; 45°line = identity. Large symbol represents mean facility + SD obtainedby each technique for the entire group of 16 subjects (C2s + SD^,C2p + SD2p); r = correlation coefficient; p = probability that thiscorrelation occurred by chance {i.e., that r = 0).

0.10

0.05SD1S (jj l x min" x mm Hg")

Fig. 2. Standard deviation of mean Schiotz and pneumatonographicoutflow facility for each individual subject (SD,S, SD1P) plotted againstone another; 45° line = identity. Large symbol represents mean stan-dard deviation obtained by each technique for the entire group of 16subjects (SDlS, SD|P);r = correlation coefficient; NS = not significant.

Fig. 3. Pneumatonograph (left) and Schiot tonograph (right) probetips. Note concavity and size of Schiotz instrument. Arrowhead in-dicates guide ring.

pneumatonography, C2P ± SD 2 P = 0.242 ± 0.114 jitlX min"1 X mmHg"1; for Schiotz tonography Q s± SD2S = 0.220 ± 0.070 /d X min"1 X mmHg."1 Pairedt-test comparison confirmed the absence of any statis-tically significant difference between these values.However, intersubject variability was greater withpneumatonography, as evidenced by the larger stan-dard deviation. This difference in intersubject vari-ability was statistically significant by the F-test of thevariance ratio [(SD2P)2/(SD2S)2 = 2.652; P < 0.05].

In Figure 1, the mean facilities for each individualsubject obtained by pneumatonography and Schiotztonography (ClP, Cis) are plotted against one another.While the correlation is statistically highly significant(r = 0.635; P < 0.001), it is far from complete. Indeed,only 40% (r2 = 0.403) of the intersubject variationabout the mean pneumatonographic facility (C2P) isaccounted for by the intersubject variation about themean Schiotz tonographic facility (C2S). Note also thegreater scatter of the data along the y (pneumatonog-raphy) axis, despite the similarity of the mean facilities.

The standard deviation of the five readings obtained

Downloaded From: https://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933357/ on 09/07/2018

No. 12 Reports 1779

Fig. 4. Position of pneumatonograph (left) and Schiotz tonograph (right) probe tips on cornea.

for each subject by each technique (SDiP, SDIS) gavean average standard deviation (SDiP, SDiS), reflectinginterobserver variability in reading the tracings gen-erated by each technique. For pneumatonography,SDIP = 0.048 p\ X min"1 X mmHg"1; for Schiotz ton-ography, SD,S = 0.022 fi\ X min"1 X mmHg"1. Anal-ysis of variance confirmed significantly greaterinterobserver variability with pneumatonography(F = 2.182, P < 0.01). In Figure 2, the standard devia-tions for each individual subject (SD1P, SDjS) are plot-ted against one another; there is virtually no correlation(r = 0.026), and the greater scatter along the y (pneu-matonography) axis is apparent.

Discussion. The 13 tonograms which were consid-ered technically unsatisfactory could not be fitted witha proper line because of multiple irregularities in thetracings. These disorderly recordings were often due toinvoluntary squeezing of the patient's eyelids, suddenblinking movements, or temporary shift of the contra-lateral eye from the fixation target. Less frequently,they were caused by fine movements of the examiner'shand or arm. The relatively high frequency (18%) ofunsatisfactory tracings could be partly attributed to the

young age of the study population. Young patients areusually more tense during an eye examination, andhave more brisk blinking reflexes than older individ-uals.

The average C values obtained by both methods werenot statistically different. However, both intersubjectan interobserver variability were significantly higherfor pneumatonography. We attribute this wider scatterto a greater mobility of the pneumatonography probetip on the corneal surface, and more transmission ofeye and hand movements to the transducer. Unlikethe tip of the Schiotz instrument which is concave, hasan external diameter of 10.0 mm and rests steadily onthe cornea, that of the pneumatonograph is flat, has a5.3 mm. external diameter, and can slide on the cornealsurface (Figs. 3-4). Furthermore, the presence of a guidering (Fig. 3) with an internal diameter of 11.5 mmaround the Schiotz tonograph probe leaves a free spaceon the sides preventing transmission to the probe offine horizontal movements of the patient's eye or ex-aminer's hand. In the case of the pneumatonograph,which lacks such a ring and has greater mobility onthe cornea, more hand and eye movements are trans-

Downloaded From: https://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933357/ on 09/07/2018

1780 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / December 1986 Vol. 27

mitted to the transducer, causing more irregularities inthe tracing.

Thus, although the average C values obtained bySchiotz and pneumatonography are comparable andcorrelated, the greater mechanical stability of theSchiotz instrument on the eye results in less variabilitythan with pneumatonography when interpreting thetracings. On the other hand, pneumatonography maybe performed in 2 min instead of 4, which saves onexamination time and may render the test more ac-ceptable to the patient. Although we did not specificallyask individual subjects about their preference betweenSchiotz and pneumatonography, it was our impressionthat patients preferred the shorter procedure.

The results of this study were obtained within thenormal intraocular pressure range (<22 mmHg); theirapplicability at higher pressures remains to be deter-mined.

Key words: aqueous humor, tonography, pneumatonography,Schiotz tonography, aqueous outflow facility

Acknowledgment. The authors wish to thank David L.DeMets, PhD, for his expert statistical advice.

From the Departments of Ophthalmology, ^American UniversityMedical Center, Beirut, Lebanon, and the fUniversity of WisconsinMedical School, Madison, Wisconsin. The authors have no proprie-tary interests in the tonography units or companies mentioned. Thisstudy was supported by grant 18-5261 from the Faculty of MedicineResearch Grants, American University of Beirut. Submitted for pub-lication: May 15, 1986. Reprint requests: Joseph G. Feghali, MD,

Department of Ophthalmology, West Virginia University MedicalCenter, Morgantown, WV 26506-6302.

References

1. Moses RA: Errors in tonography. Trans Am Acad OphthalmolOtolaryngol 65:145, 1951.

2. Kolker AE and Hetherington J Jr: Tables for tonography andtonometry. In Becker-Shaffer's Diagnosis and Therapy of theGlaucomas. St Louis, CV Mosby, 1983, pp. 519-535.

3. Spencer RW, Helmick ED, and Scheie HG: Tonography. Tech-nical difficulties and control studies. Arch Ophthalmol 54:515,1956.

4. Podos SM and Becker B: Tonography-current thoughts, editorial.Am J Ophthalmol 75:733, 1973.

5. Kolker AE and Hetherington J Jr: Tonometry and tonography—clinical applications. In Becker-Shaffer's Diagnosis and Therapyof the Glaucomas. St Louis, CV Mosby, 1983, pp. 107-118.

6. Langham ME: Discussion on pneumatic applanation tonometer.Trans Am Ophthalmol Otolaryngol 6:1041, 1965.

7. Langham ME and McCarthy E: A rapid pneumatic applanationtonometer: comparative findings and evaluation. Arch Ophthal-mol 79:389, 1968.

8. Langham ME: Reliability and application of the pneumatono-graph. Arch Ophthalmol 98:384, 1980.

9. Moses RA and Grodzki WJ: The pneumatonograph, a laboratorystudy. Arch Ophthalmol 97:547, 1979.

10. Moses RA and Becker B: Clinical tonography: the scleral rigiditycorrection. Am J Ophthalmol 45:196, 1958.

11. Quigley HA and Langham ME: Comparative intraocular pressuremeasurements with the pneumatonograph and Goldmann to-nometer. Am J Ophthalmol 80:266, 1975.

12. Langham ME, Leydhecker W, Krieglstein G, and Waller W:Pneumatonographic studies on normal and glaucomatous eyes.Adv Ophthalmol 32:108, 1976.

Inhibition of Lens Opocificotion in X-lrrodioted Rots Treoted With WR-77913

Thomas D. Osgood,* Thomas W. Menard,f John I. Clark,:}: and Kenneth A. Krohn§

Radiation induced cataracts are models for studying mech-anisms of lens opacification. WR-77913, S-3-(amino-2-hy-droxypropyl) phosphorothioate (NCS-318809), has beenidentified as a radioprotective agent. Injection of WR-77913(1160 mg/kg, i.p.) 15 to 30 min before exposure to 15.3 grayof x-irradiation inhibited rat lenses from developing radiationcataracts. Irradiated rats which did not receive the drug de-veloped dense cataracts. Lenses from control rats which re-ceived no radiation remained transparent. Individual lenseswere weighed, homogenized, and assayed for protein contentusing the Lowry method. The molecular weight distributionof soluble protein was determined by HPLC. Mean lensweights were: controls 48.2 mg; irradiated, drug-treated 45.9mg; and irradiated, nontreated 45.5 mg. Protein accountedfor over 40% of the lens weight in control and drug-treatedrats and less than 20% for the nontreated cataractous lenses.Water was less than 60% of the lens weight in control anddrug-treated rats and over 80% in cataractous lenses. Insolubleprotein ranged from 12 to 17% of the total lens weight for

each group. The ratio of insoluble to soluble lens protein was0.40 for control, 0.65 for drug-treated, and 11.28 for cata-ractous rat lenses. HPLC confirmed a dramatic loss of solubleprotein and a complete absence of protein below 25K daltonsin cataractous lenses. Proteins below 25K daltons accountedfor over 25% of the soluble protein in control and drug-treatedrat lenses. WR-77913 stabilizes protein composition and ap-pears to be an effective inhibitor of radiation cataractogenesis.Invest Ophthalmol Vis Sci 27:1780-1784, 1986.

Lens opacification occurs in animals exposed to x-ray doses as low as 0.25 gray, which is well below levelsused for clinical radiotherapy of the head and neck.'Development of opacity following x-irradiation ischaracterized by an increase in the proportion of in-soluble protein, increased light scattering, and hydra-tion of the lens.2 These changes are common to othercataracts and thus the opacification produced by x-

Downloaded From: https://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933357/ on 09/07/2018