Fluorescein angiography of the newborn rat. Implications in oxygen

Investigative Ophthalmology & Visual Science, Vol. 31, No. 5, May 1990Copyright © Association for Research in Vision and Ophthalmology

Fluorescein Angiography of the Newborn Rot

Implications in Oxygen-Induced Rerinoparhy

Luis I. Lorrozabol and John S. Penn

The current technique was developed to characterize the morphologic changes in the retinas of oxy-gen-reared rats, as an animal model of retinopathy of prematurity. Past studies have used ink perfu-sion to observe the retinal vasculature, but this method is static and requires the sacrifice of thesubject. Fluorescein angiography, however, is dynamic and relatively noninvasive, and allows thesurvival of the animal for further study. The fundus camera cannot be used because the source of lightthat is focused in an annulus is too large for the pupil size of a young (~14-day-old) rat. To overcomethis, a Nikon inverted microscope (Diaphot-TMD) was used. Using the proper exciting and barrierfilters for fluorescene, a photographic sequence was made by rapidly focusing to the plane of theretinal vessels. To our knowledge, similar photographs have not been previously published. Thistechnique was used in newborn pigmented ratlings that were 1) exposed to 80% oxygen for the first 14days of life; 2) exposed to 80% oxygen for the first 21 days of life; or 3) exposed for the first 14 daysfollowed by 7 days in room air. Age-matched controls were raised simultaneously in room air andevaluated with the same technique. Differences were observed between treatments in the amount ofretinal capillary loss, and in the tortuosity and diameter of the major retinal vessels. The hyaloidsystem also varied between treatment groups. Oxygen-exposed rats showed a persistence of thehyaloid vessels that was particularly prominent in the group returned to room air before analysis.Comparisons are made to past results obtained with other histologic techniques. Invest OphthalmolVis Sci 31:810-818, 1990

The recent increase in the survival of prematureinfants due to improving neonatal care has caused anincreased incidence of retinopathy of prematurity(ROP), but also a reduction in eye donations for basicresearch. This has led to renewed interest in the devel-opment of a suitable animal model for ROP. For anumber of reasons, much attention has been focusedon the rat in this role. First, the rat is an inexpensiveand easily maintained model in which the retinal vas-culature develops postnatally,12 facilitating its study.Over decades of research, newborn rats have shownconsistent susceptibility to oxygen-induced retinopa-thy. Further, the ontogeny of the rat retina parallelsthat of the human infant, particularly with respect tovascular development.3 Finally, the techniques pre-viously perfected to test retinal integrity in the ratwere believed to be appropriate and adequate.

The preferred method of studying retinal vascular

From the Department of Ophthalmology, University of Ar-kansas for Medical Sciences, Little Rock, Arkansas.

Supported by a grant to JSP from the National Eye Institute anda grant from Research to Prevent Blindness.

Submitted for publication: August 22, 1989; accepted October 9,1989.

Reprint requests: Dr. John S. Penn, Department of Ophthalmol-ogy, University of Arkansas for Medical Sciences, 4301 WestMarkham, Mail Slot 523, Little Rock, AR 72205.

morphology has been ink perfusion, a technique thatrequires the sacrifice of the animal. For decades, thismethod has been used extensively to describe alter-ations of the retinal blood vessels in response to hy-peroxic exposure of newborn animals.1'2'4'5 However,results obtained with this technique have been incon-sistent. This is particularly true with respect to thepossible occurrence of neovascularization. The argu-ment over the presence or absence of neovasculariza-tion in oxygen-exposed rats began with the work ofPatz and co-workers,6 who, using standard histologictechniques, found evidence of "capillary proliferationfrom the retina into the vitreous" and "intraocularhemorrhage" in newborn rats exposed to 80% oxy-gen. These findings were challenged by Ashton andBlach,1 who, using ink perfused and histologicallysectioned retinas, reported no evidence of abnormalneovascularization or hemorrhages in similarlytreated rats. Patz7 was later unable to reproduce hisoriginal results.

Recent work by Ricci and Calogero4 has rekindledthis controversy. These authors described both retinalneovascularization and hemorrhagic lesions in ink-perfused retinas from rats exposed to 80% oxygen for10 days followed by room air for 15 days. In somecases the neovascularization was described as "ex-traretinal."

810

Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933153/ on 03/26/2018

No. 5 FLUORE5CEIN ANGIOGRAPHY OF OXYGEN-REARED NEWBORN RATS / Lorrozobol ond Penn 811

Over the past 2 yr our laboratory has processedapproximately 200 ink-perfused retinas from new-born rats exposed for various durations to elevatedoxygen and room air. It has been our experience thatthe presence of "hemorrhagic lesions," as Ricci andCalogero describe them, is difficult to distinguishfrom pressure-induced leaks of ink. With careful ad-ministration of ink, we have eliminated the occur-rence of leaks, leading us to believe that they are mostoften, if not always, attributable to procedural arti-fact. Further, we see no evidence of neovasculariza-tion and little evidence of other forms of abnormalcapillary growth or capillary "tuft"8 formation withink. The true presence of hemorrhages or abnormalneovascutarization probably can be determined withextensive histologic scrutiny of retinal tissue. How-ever, we have chosen an alternative path for address-ing past contradictions resulting from ink perfusion.

Retinal fluorescein angiography has been usedlargely in clinical diagnosis. Its advantage is that truevessel patency under physiologic conditions can beevaluated. For studies using" animals, sacrifice of thesubject is not necessary, and it is possible to followprogression of, or recovery from, vascular deficitwithin the same retina. Also, vessel integrity can bedetermined in vivo, without the complication ofpressure artifacts, and thus, some understanding ofthe true integrity of the vessels can be gained from thepresence or absence of fluorescein leaks.

Fluorescein angiography has been achieved suc-cessfully in the adult rat with the clinical fundus cam-era,9 but in younger animals its use has been impossi-ble because of problems with illumination. Thesource of light in this instrument is focused in anannulus, and spreads out behind the pupil to illumi-nate the retina, allowing the image forming rays toreturn to the camera through the center of the an-nulus. This design eliminates light reflection by thecornea and lens. Unfortunately, the size of the new-born rat's entire eye is smaller than the center of theannulus, making it impossible to illuminate the ret-ina with the fundus camera.

Fluorescence is made possible by a process of ab-sorption and emittance of light in different wave-lengths. Using the proper exciting and barrier filters,fluorescein angiography can be achieved in verysmall eyes with a fluorescence microscope. The prob-lem of illumination is solved because, even though noannulus is used, the barrier filter renders the camerablind to the exiting wavelength, and scattered andreflected light is virtually eliminated.

This report describes an attempt to apply fluores-cein angiography to the study of oxygen-induced reti-nopathy in newborn rats. To our knowledge, this isthe first such attempt made. Further, we believe that

these experiments mark the first use of the fluores-cence microscope for angiography of the retina ineyes of this size. Comparisons are made to past resultsobtained using ink perfusion, with particular empha-sis on those facets of retinal vasculature that may beexplained by pressure artifact.

Materials and Methods

Eight litters of Long Evans pigmented rats wereused. Immediately after birth, some litters wereplaced with their mothers in 80% atmospheric oxy-gen. Other litters of control animals were simulta-neously placed in room air. Some rats received acombination of oxygen and room air maintenance.Surrogate mothers were alternated from room air tooxygen at regular intervals (3-4 days) to avoid respi-ratory distress. Mothers were fed rat chow ad libitum,and experiments were conducted under 12-hr lightlighting of 500 lux illuminance. All animals weremaintained in accordance with the ARVO Resolu-tion on the Use of Animals in Research.

The rats were divided into five groups: 1) room airfor 14 days; 2) 80% oxygen for 14 days; 3) room airfor 21 days; 4) 80% oxygen for 21 days; and 5) 80%oxygen for 14 days, followed by 7 days in room air(called the "recovery group").

Fluorescein angiography was performed at the endof each experimental period with a Nikon invertedfluorescence microscope (Diaphot-TMD; Nikon,Tokyo, Japan) equipped with a 100-W mercury lampand a 3.2X objective lens. Details of this procedure

Fig. I. A 14-day-old pigmented ratling is placed over the stage ofthe inverted microscope, with the coverslip creating a negative tens.


812 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / May 1990 Vol. 31

Table 1. Qualitative

Treatment

assessment*

Extent

of fluorescein angiograms of oxygen-

Hyaloid

TortuosityCapillary-free zones

and room air-reared

Retina

Major vesselnarrowing

newborn rats

Tortuosityof major vessels

14 days room air(n = 7)

14 days O2 80%(n = 6)

21 days room air(n = 6)

21 days O2 80%(n = 6)

14 days O2 80%+ 7 days roomair(n = 9)

* Range of observation from - (no occurrence) to ++++ (extensive occurrence); + / - (inconsistent presence).

are described elsewhere.10 The camera was a NikonF-3, set for automatic exposure time at ASA 3200.We used Kodak T-Max 3200 Professional Film, de-veloped according to manufacturer instructions atstandard film speed.

Typically, a ratling was anesthetized with sodiumpentobarbital (50 mg/kg) injected intraperitoneally.In the 14-day-old animals, a canthotomy was per-formed occasionally in order to retract the eyelidsfully. The pupil was dilated with 1% phenylephrineand 1% tropicamide, and the animal was placed overa heating pad to avoid hypothermia. At least 10 minwere allowed for complete dilation. The ratlings thenwere injected intracardially with 10 mg/kg of 1% so-

dium fluorescein. This method of administration wasnecessary because of the difficulty of canalizing veinsin animals of this age. A coverslip with a drop of 2.5%methylcellulose was placed on the warmed stage ofthe inverted microscope. Immediately after injection,the ratling was placed on the stage with its eye con-tacting the coverslip (Fig. 1). This created a negativelens and allowed for focusing in the plane of the reti-nal and hyaloid vessels. A photographic sequence wasmade, with the exposure times ranging between 0.5and 1.5 sec. The eye was reoriented manually bymoving the head of the animal, in order to photo-graph the different areas of the retina and hyaloidsystem.

Fig. 2. Retinal fluorescein angiograms. (a) 14-day-old rat reared in room air, showing major arteries and veins radiating from the optic discand a fully arborized capillary network, (b) 14-day-old rat reared in 80% oxygen, showing narrowed major vessels and missing capillarynetwork.


No. 5 FLUORESCEIN ANGIOGRAPHY OF OXYGEN-REARED NEWBORN RATS / Lorrozobol ond Penn 810

Fig. 3. Retinal fluorescein angiograms. (a) 21-day-old rat rearedin room air, showing major arteries and veins radiating from theoptic disc and a visible fully arborized capillary network. Detailsare clearer than in the 14-day-old room air-reared rat due to in-creased size of the eye and less persistent hyaloid system, (b) 21-day-old rat reared in 80% oxygen, showing narrowing of somemajor vessels and missing capillary network in the central retinaand portions of the periphery, (c) 21-day-old rat reared in 80%oxygen for 14 days and then returned to room air for 7 days,showing striking tortuosity and slight narrowing of the major ves-sels, and persistence of missing capillary network.

Results

Our qualitative interpretation of the fluoresceinangiograms is summarized in Table 1. Examples ofangiograms are found in composite Figures 2-7.

Figure 2a depicts the retina of a 14-day-old controlrat. The loss of retinal capillaries and the narrowingof the major vessels of the retina, both results of the14-day oxygen exposure, are evident in the exposedcounterpart (2b).

Figure 3a clearly shows the presence of capillariesunder room-air conditions in the 21-day-old. Expo-sure to oxygen for this period (3b) resulted in theprolonged loss of most capillaries, with the exceptionof a limited number in the periphery of one or tworetinal quadrants. Also, some tortuosity of the majorvessels occurred during this extended exposure. Thislatter phenomenon was not observed in rats exposedfor 14 days. The recovery animals (3c) showed nosubstantial increase in physiologically patent capillar-ies during the room-air period, and also showed animpressive degree of tortuosity of all major vessels.

Figures 4 and 5 depict the hyaloid systems of 14-day- and 21-day-old rats, respectively. The 14-day-old reared in 80% oxygen (Fig. 4b) maintains aslightly more extensive hyaloid network than theroom-air counterpart (Fig. 4a), and the vessels of theformer are slightly more dilated. In the 21-day-oldcontrol (5a), only a small number of very narrowhyaloid vessels remains. The oxygen-exposed 21-day-old (5b) has retained a more extensive system withgreater arborization and vessel diameter. Rats raisedfor the first 14 days of life in 80% oxygen, followed by7 days in room air (5c), display an extensive hyaloidnetwork. Vessels are dilated and extremely tortuous,often displaying loops along their length from theoptic nerve herd to the rear of the lens.

Figure 6 reveals the possible presence of abnormalvessel growth and capillary formation in recovery an-imals. First, the formation of tuftlike capillary beds isdepicted in Figure 6a. This single recovery animal isour only evidence of "capillary tufts"8 obtained withfluorescein for any experimental treatment. We have


814 INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / May 1990 Vol. 31

Fig. 4. Hyaloid system fluorescein angiograms. (a) 14-day-old rat reared in room air, showing the normal extent found at this age.(b) 14-day-old rat reared in 80% oxygen, showing a more fully arborized system than the room-air control.

never witnessed unequivocal evidence of tuft forma-tion using ink perfusion. Second, vessel shunts can bedocumented with a battery of photographs taken atslightly different angles. Figure 6b is one of such aseries. The fluorescein is routed directly from a majorartery to a major vein with little capillary deviation.

Figure 7 compares one of our ink-perfused retinaswith a fluorescein angiogram from a rat that receivedan identical treatment. Each animal was in the recov-ery group. This comparison emphasizes two majordifferences in the results of the two techniques: 1) thefluorescein angiograms show no significant capillaryrecovery during the postexposure week in room air,indicating that those capillaries seen with ink in thisgroup are not physiologically patent; and 2) the tortu-osity of major vessels in the recovery group, which ismade obvious with fluorescein angiography, is notevident in ink-perfused retinas, except in rare cases inthe far periphery.

Discussion

This work adds a new tool to the study of retinalvascular degeneration in small animals. It is particu-larly well suited for the study of oxygen-induced reti-nopathy in rats. The conditions under which the datais recorded allow for a truer physiologic evaluation ofvascular integrity. This is especially important duringdevelopment of the eye when the newest vessels are

very fragile or are only patent under nonphysiologicpressures.

It was necessary to solve several technical problemsin order to visualize ocular blood vessels with thistechnique, and some problems still remain. We wereunable to adapt an electronic flash to our photo-graphic equipment, so the amount of light returningfrom the retina after filtering and reflection wassmall, and highly sensitive film was required. Thedisadvantage of this film is increased granularity, re-sulting in a loss of photographic quality, both aesthet-ically and in the resolution of small objects.

Because the lens of the rat is almost flat at birth andrapidly becomes more spherical, occupying most ofthe ocular space at the 16th day of life," the hyaloidsystem is very active during this time. It is highlyfluorescent during the angiography, and it is locatedbetween the retina and the camera. For these reasons,its scattered fluorescence partially obscures opticalresolution of the retinal plane. If the hyaloid system ismore developed or persistent, it becomes more diffi-cult to visualize retinal vessels, because of the in-creased scattered fluorescence in the optical pathway.

We were unable to use our preferred model, thealbino rat, because its retinal epithelium is not pig-mented. The lack of pigment in the albino wouldhave allowed the fluorescence from the choroidalvessels to mask that of the retinal vessels. Also, be-cause of the immaturity of the ratlings, they had to be


No. 5 FLUORESCEIN ANGIOGRAPHY OF OXYGEN-REARED NEWBORN RATS / Lorrozobol ond Penn 815

Fig. 5. Hyaloid system fluorescein angiograms. (a) 21-day-old ratreared in room air, showing the normal extent found at this age. (b)21-day-old rat reared in 80% oxygen, showing a more persistentsystem with increased vessel caliber, (c) 21-day-old rat reared in80% oxygen for 14 days and then returned to room air for 7 days,showing striking tortuosity and dilation of the hyaloid vessels withincreased arborization.

closely monitored during anesthesia, with special at-tention paid to the dose of anesthetic and the bodytemperature. Several animals were lost during recov-ery from anesthesia, after the angiography.

The sum of the above difficulties was manifested inpoor capillary resolution in comparison with that ofink perfusion. However, we were able to record datawith fluorescein angiography which reflects the truephysiology of the circulatory system of the eye. This isimpossible in ink-perfused retinas for a number ofreasons, the most notable of which is the likelihood ofpressure artifact.

In an attempt to determine if the ink leaks ob-served by Ricci and Calogero in perfused retinal flatmounts and reported as "hemorrhagic lesions"4

could have been the result of pressure artifact, wescrutinized our fluorescein angiograms of similarlytreated animals for fluorescein leaks. We were unableto demonstrate any unequivocal leakiness of retinalvessels in angiograms. Also, in our ink-perfused ret-inas, we were unable to observe any leakiness of reti-nal vessels that could not be the possible result ofabnormal pressure. Moreover, we never observedcapillary tufts with the ink procedure; however, thepresence of tufts, albeit rare, was confirmed by theangiography (Fig. 6a).

Based on ink perfusions, this laboratory has sug-gested previously that complete recovery of the capil-lary net occurs within a week after return to room air,after a 14-day exposure to 80% oxygen.12 This resultwas not supported by the angiograms of animals withidentical experimental treatment (Fig. 7b). The strik-ing tortuosity of major vessels and the presence oflarge capillary-free zones were not evidenced in ink-perfused retinas. This vast discrepancy of vascularmorphology serves to emphasize the advantage ofobservation under physiologic conditions, such as isachieved with fluorescein angiography. It should benoted that this tortuosity of the major vessels greatlyresembles the so-called "plus disease" used in theclassification of ROP. This phenomenon, as Flynn13

describes it, "is predictable and, in fact, inevitable" incases where large capillary-free zones and vascularshunting are evident in the infant retina.


INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / May 1990 Vol. 31

-4

Fig. 6. Retinal fluorescein angiogram of a 21-day-old rat reared in 80% oxygen for 14 days and then returned to room air for 7 days,(a) Tuftlike formation in the periphery, (b) Shuntlike formation in the periphery.

There has been some question in the past about thetiming of the normal disappearance of the hyaloidsystem. Cairns,2 using ink perfusion, reported that

the hyaloid system was vestigial or absent by the 11 thday of life. In other, more recent work employingscanning electron microscopy, Hollenberg and Dick-

•r

Fig. 7. Comparison between ink perfusion (a) and retinal fluorescein angiography (b) of 21-day-old rats reared in 80% oxygen for 14 daysand then returned to room air for 7 days. Note two clear differences: the extent of tortuosity with narrowing and dilation of different majorvessels, and the degree of capillary presence.


No. 5 FLUORESCEIN ANGIOGRAPHY OF OXYGEN-REARED NEWBORN RATS / Larrazabal ond Penn 817

son14 demonstrated the patency of the hyaloid systemon the 14th day after birth. Its extent at this agewas similar to that which we see in our angiograms(Fig. 4a).

There also has been controversy over the persis-tence of the hyaloid system in oxygen-treated ani-mals. Ashton and Blach found no evidence of abnor-mal hyaloid persistence in oxygen-exposed rats withair survival after hyperoxia,1 but it should be notedthat in their experiments the rats were not exposedimmediately after birth in most cases. On the otherhand, Patz et al found histologic evidence of persis-tence of the hyaloid system in oxygen-treated ani-mals, occasionally in 14-day-olds and consistently in21 -day-olds.6 Bischoff et al found the same results inmice reared in oxygen, using scanning electron mi-croscopy.15 The fluorescein angiograms in the currentstudy corroborate the latter two studies. The hyaloidappears markedly more persistent in oxygen-treatedrats than in controls, at both 14 and 21 days of age(Figs. 4, 5). According to Bischoff et al, since theretinal and hyaloid arteries share a common bloodsupply, retinal vasoconstriction resulting from oxy-gen exposure causes an elevation of pressure in thehyaloid system which inhibits the normal process ofhyaloid vascular atrophy.15

The persistence of the hyaloid in recovery animalswas even more striking, with dilation and extremetortuosity of the vessels. When similarly treated re-covery rats (14 days, 60 or 80% oxygen; 7 days roomair) are perfused with ink, the retina appears com-pletely rearborized.12 Since, in this technique, thehyaloid system is removed with the vitreous in orderto visualize the retina, its persistence is difficult todetermine. Fluorescein angiography shows us apoorly arborized retinal capillary network and an ex-tensive hyaloid system in recovery animals. It appearsas though when the marker (ink or fluorescein) ispressure-fed through vessels, its preferred target is theretina. When administered under physiologic bloodpressure, the marker prefers the hyaloid. Yet the twosystems are hydrostatically coupled at their originwithin the optic nerve head.

The apparent persistence of the hyaloid evidencedby angiography in recovery rats is not easily ex-plained. Bischoff et al theorized that

regression of the hyaloid vasculature is inhibited by the combi-nation of sustained application of an abnormally high pressuregradient through it and the presence of an abnormally highintravascular [oxygen] concentration within it. When that ox-ygen concentration suddenly drops to a normal level, the hya-loid vessel walls may suddenly undergo rapid atrophic changeswhich lead to even greater distention while the pressure gra-dient [still] remains high.15

This condition, the above authors argue, would re-main until the retinal vessels reopen after further re-covery in normoxia. The current study does not ad-dress this question and sheds no new light on thepersistence of the hyaloid, other than to document itin yet another manner.

Originally, retrolental fibroplasia was believed tobe due to "persistence and thickening of the posteriorfibrovascular sheath of the lens."16 Later experimen-tation pointed toward the retinal vessels as the solemediator, and our attention has been focused onthem since. Clearly, in both rats and humans, thehyaloid and retinal vessels, sharing a common source,are intimately connected. The effect of hyperoxia onthe hyaloid system appears at least as drastic as itseffect on the retina. The suggestions of Bischoff sgroup15 warrant further study, and fluorescein angi-ography is an effective tool with which to do so. Withthis technique, both retinal and hyaloid vessels can bestudied under physiologic conditions, and the possi-ble role of the hyaloid system in the development ofretinopathy of prematurity can be reexamined.

Summary

We make the following conclusions with respect tofluorescein angiography and its role in the study ofoxygen-induced retinopathy:

1. Satisfactory fluorescein angiography can be ac-complished using the fluorescence microscopeon eyes too small for the fundus camera.

2. Fluorescein angiography is a powerful tool forstudying oxygen-induced alterations of retinalvasculature, eliminating many of the possibleartifacts inherent in other methods.

3. Fluorescein angiography yields a new view ofthe effect of oxygen on the newborn rat retinawith respect to each of the following:A. capillary loss and recoveryB. major vessel diameterC. tortuosity of major vessels

4. Fluorescein angiography also can be useful inthe study of the hyaloid. Inasmuch as the atro-phy of hyaloid vessels seems directly coupled tothe development of retinal vessels, the role ofthis system in the development of ROP war-rants further investigation.

Key words: fluorescein angiography, retinal vascular de-generation, hyaloid system, newborn rat, oxygen-inducedretinopathy

AcknowledgmentsThe authors wish to thank J. Johnson for preparation of

the manuscript and G. Miranda and A. Kogan for produc-


818 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / May 1990 Vol. 31

tion of the figures. S. Thompson provided technical assis-tance.

References1. Ashton N and Blach R: Studies on developing retinal vessels:

VIII. Effect of oxygen on the retinal vessels of the rattling.Communications. Br J Ophthalmol 45:321, 1961.

2. Cairns JE: Normal development of the hyaloid and retinalvessels in the rat. Br J Ophthalmol 43:385, 1959.

3. ICretzer FL, Mehta RS, Goad D, and Hittner HM: Animalmodels in research on retinopathy of prematurity. In Retinop-athy of Prematurity: Current Concepts and Controversies,McPherson AR, Hittner HM, and Kretzer FL, editors. BCDecker, 1986, pp. 79-88.

4. Ricci B and Calogero G: Oxygen-induced retinopathy in new-born rats: Effects of prolonged normobaric and hyperbaricsupplementation. Pediatrics 82(2): 193, 1988.

5. Penn JS and Thum LA: The rat as an animal model for reti-nopathy of prematurity. In Inherited and Environmentally In-duced Retinal Degenerations, LaVail MM, Anderson RE, andHollyfield JG, editors. AR Liss Press, 1989, pp 623-642.

6. Patz A, Eastham A, Higgenbotham DH, and KJeh T: Oxygenstudies in retrolental fibroplasia: II. The production of the mi-croscopic changes of retrolental fibroplasia in experimentalanimals. Am J Ophthalmol 36:1511, 1953.

7. Patz A: The role of oxygen in retrolental fibroplasia. Trans AmOphthalmol Soc 66:940, 1968.

8. Ricci B: Effects of hyperbaric, normobaric, and hypobaric ox-ygen supplementation on retinal vessels in newborn rats: Apreliminary study. Exp Eye Res 44:459, 1987.

9. Spertus AD, Slakter JS, Weissman SS, and Henkind P: Experi-mental carotid occlusion: Fundoscopic and fluorescein angio-graphic findings. Br J Ophthalmol 68:47, 1984.

10. Larrazabal LI, Penn JS, and Merin LM: Study of ocular vascu-lature in the newborn rat: Retinal fluorescein angiographyusing an inverted microscope. Journal of Ophthalmic Photog-raphy 11(2):49, 1989.

11. Sivak JG and Dovrat A: Early postnatal development of the ratlens. Exp Biol 43(1 ):57, 1984.

12. Penn JS, Thum LA, Rhem MN, and Dell SJ: Effects of oxygenrearing on the electroretinogram and GFA-protein in the rat.Invest Ophthalmol Vis Sci 29(11): 1623, 1988.

13. Flynn JT: An international classification of retinopathy ofprematurity. Ophthalmology 92:987, 1985.

14. Hollenberg MJ and Dickson DH: Scanning electron micros-copy of the tunica vasculosa lentis of the rat. Can J Ophthalmol6:301, 1971.

15. Bischoff PM, Wajer SD, and Flower RW: Scanning electronmicroscopic studies of the hyaloid vascular system in newbornmice exposed to O2 and CO2. Graefes Arch Clin Exp Ophthal-mol 220:257, 1983.

16. Terry TL: Extreme prematurity and fibroplastic overgrowth ofpersistent vascular sheath behind each crystalline lens: I. Pre-liminary report. Am J Ophthalmol 25:203, 1942.


Documents

Fluorescein angiography of the newborn rat. Implications in oxygen