1176 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / July 1986 Vol. 27

10. Lowry OH, Rosenbrough NJ, Fair AL, and Randall RJ: Proteinmeasurement with the Folin phenol reagent. J Biol Chem 193:265, 1951.

11. Masterson E and Chader GJ: Characterization of glucose trans-port by cultured chick pigmented epithelium. Exp Eye Res 32:279, 1981.

12. Pascuzzo GJ, Johnson JE, and Pautler EL: Glucose transport inisolated mammalian pigment epithelium. Exp Eye Res 30:53,1980.

13. Zadunaisky JA and Degan KJ: Passage of sugars and urea acrossthe isolated retina pigment epithelium of the frog. Exp Eye Res23:191, 1976.

Retinal Hemorrhage Thresholds for Q-Switched

Neodymium-Yag Laser Exposures

Michael F. Blankenstein,* Joseph Zuclich,* Ralph G. Allen, f Harold Davis,^ Scoff J. Thomas,§ and Robert F. Harrison§

Thresholds for retinal vitreal and contained hemorrhages weredetermined for 1064 nm laser light at 30-nsec and 4-nsecpulsewidths. Rhesus monkeys received graded exposures froma neodymium-yag laser onto either the macular or extramac-ular region of the retina. Contained hemorrhages appearedas concentric ring structures with white punctate centers. Thevitreal hemorrhage was characterized by the presence of cho-roidal blood in the vitreal chamber at the exposure site. The30-nsec contained hemorrhage threshold (ED50) was 1.7 mJon the macula and 2.1 mJ for an extramacular exposure. The30-nsec vitreal hemorrhage macular threshold was 2.3 mJ,and the extramacular threshold was 6.6 mJ. The thresholdfor the 4-nsec pulsewidths to produce a hemorrhage (vitrealor contained) on the retina (macula or extramacular) was 340nJ. Invest Ophthalmol Vis Sci 27:1176-1179, 1986

Laser safety standards are a result of many studiesaimed at setting a maximum permissible exposure(MPE) for the various wavelengths and pulsewidthsobtainable with operational laser systems.1 The Amer-ican National Standards Institute (ANSI) MPE is basedon theshold data for the minimum visible lesion(MVL). Little has been done, however, investigatingthe effects of suprathreshold ocular exposures. Twosuch effects are the vitreal and contained hemorrhagein which choroidal blood either passes through theruptured retina into the vitreal chamber or is containedbeneath the retina. Both have been observed and re-ported for visible laser radiations. However, the hem-orrhagic lesion endpoints have not, in general, beenadequately denned with respect to parameters such asthreshold energies, wavelength and pulsewidth of thesource, or pathology of the damaged tissue; nor havepermanent and transient effects on vision been inves-tigated. These parameters are important in definingrisks, evaluating the transient or permanent nature ofthe injury, and in considering medical treatments withlasers.

Using a neodymium-yag (Nd:YAG) laser producing30-nsec and 4-nsec pulses at a wavelength of 1064 nm,we report thresholds for vitreal and contained hem-

orrhages at macular and extramacular positions, anddiscuss the experimental indications of mechanism.

Materials and Methods. Apparatus: The Nd:YAGflashlamp pumped laser used in this study was builtby the Los Alamos National Laboratory. An IsometModel 420 intracavity Q-switch was used to obtain 30-nsec pulsewidths. To generate 4-nsec pulses, the Q-switched 30-nsec laser pulses were directed into an ex-tracavity Pockels cell driven by a laser triggered sparkgap with a 4-nsec charge line.

To accurately position the beam on the retina, aninverted Zeiss fundus camera and animal translationstage were used. A helium-neon alignment beam, col-linear with the Nd:YAG laser beam, was used to po-sition the fundus camera line-of-site with the 1064 nmbeam path.

Spatial filters and lenses produced a collimated beamwith a 1/e2 diameter of 3 mm at the corneal plane.The beam divergence was less than 1 mrad. Additionaldetails of the experimental setup and beam diagnosticshave been published elsewhere.2

Animal models: The experimental subjects were 18rhesus monkeys (Macaca mulatto) weighing between2 and 6 kg with both sexes equally represented. Eachanimal had normal corneas, lenses, and retinae, with±0.5 diopter or better refraction and no astigmatism.

Prior to exposure, subjects were preanesthetized with10.0 mg/kg ketamine (IM), and anesthetized with asingle injection of 18.0 mg/kg sodium pentobarbitalinto the saphenous vein. Eyes were dilated with 1%tropicimide applied topically. The eyelids were re-tracted with an ophthalmic speculum and the corneairrigated with lactated Ringer's solution to prevent de-siccation. The subject was then placed on a stage (hav-ing 5° of freedom) in front of the fundus camera sothe retina could be viewed and positioned such thatany retinal area could be accurately exposed. Each eyereceived one macular and one or more extramacularexposures, none of which were made over any observ-able venation. The exposed retina was viewed imme-diately after exposure to observe lesion development.

Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933359/ on 05/16/2018

No. 7 Reporrs 1177

Fig. 1. Funduscopic photo-graph of rhesus monkey retinawhich has been exposed to aneodymium-yag laser, dem-onstrating three lesion types:minimum visible lesion (ver-tical column of four reflectivecircular areas at center), vitrealhemorrhage (far left and farright), and contained hemor-rhage (upper center). The right-hand vitreal hemorrhage iscentered on the macula.

Eyes used for histopathology were perfused in vivowith Karnovsky's fixative. They were enucleated post-mortem approximately 2 hr after exposure. Eyes werepost-fixed in Osmium tetroxide, dehydrated in a gradedethanol solution and embedded in LDX-112. The spe-cific techniques used for eye preparation have beendescribed previously.3'4 Thin sections were cut with anultramicrotome and stained with Toluidine Blue.

The animals involved in this study were procured,maintained, and used in accordance with the AnimalWelfare Act of 1970 and the "Guide for the Care andUse of Laboratory Animals," prepared by the Instituteof Laboratory Animal Resources-National ResearchCouncil. This investigation also conformed to theARVO Resolution on the Use of Animals in Research.

Lesion definition and data analysis: Visible lesionswere placed in one of three categories: minimal visiblelesion (MVL) without hemorrhage, contained hem-orrhage (CHL), or vitreal hemorrhage (VHL). TheMVL appeared as a small spot on the target site within1 hr after exposure. CHLs were large structures, ap-proximating the shape and size of the macula. Theyexhibited concentric ring structures with a white punc-tate center, and varied somewhat in ring diameter andcolor. A VHL was differentiated from the other lesiontypes by the leakage of blood into the vitreal chamber.Examples of each lesion type may be seen in Figure 1.

The 4 nsec hemorrhage threshold was taken fromboth macular and extramacular exposures and com-bined VHL and CHL data. Thirty nsec thresholds were

Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933359/ on 05/16/2018

1178 INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / July 1986 Vol. 27

Table 1. Vitreal and contained hemorrhage thresholds (ED50) for 30-nsec Nd:Yag (1064 nm) laser radiation

Typehemorrhage

Retinalposition ED,

Confidence limits Lowest energyw/hemorrhage*

Highest energyw/o hemorrhage*

Contained

Vitreal

MaculaExtra-macularCombined

MaculaExtra-macularCombined

1.72.11.9

2.36.64.2

0.04-2.2f0.6-2.71.4-2.3

0.6-3.91.6-27.3$3.1-280

1.614

2.1l.S

3.3

3.33.9

• Total intraocular energy in millijoules.t 90% confidence limits. 95% Fiducial limits not valid.

$ ED05 and ED9J. No valid fiducial limits could be calculated.

resolved into VHL and CHL data for both macularand extramacular exposures.

The effective doses for 50% probability of damage(ED50) with confidence limits were calculated by probitanalysis.5 Insufficient lesion data at high energy levelsprevented fiducial limit calculation for the extramacu-lar VHL ED50s. Therefore, the ED05, ED50, and ED95

Fig. 2. Photomicrographs of contained hemorrhage (CHL) andvitreal hemorrhage lesions (VHL) in the rhesus retina. A, CHL pro-duced with 2.7 mJ, 30 nsec exposure of 1064 nm laser light (X42).B, VHL produced with 2.2 mJ, 30 nsec exposure of 1064 nm laserlight (X50). Both sections were stained with Toluidine Blue.

are reported for this data. Limited lesion data also low-ered the macular CHL ED50 fiducial limits to 90%.

Results. The ED5Os and their confidence limits forthe 30-nsec pulsewidth are summarized in Table 1.Listed also is the lowest energy observed to produce ahemorrhage and the highest energy that did not pro-duce a hemorrhage.

Figure 2 demonstrates a CHL produced by an ex-posure of 2.7 mJ and a VHL from an exposure of 2.2mJ. Both exposures were of 30 nsec duration. Damageto the retina due to a CHL was characterized by retinaldetachment. The retinal pigment epithelium (RPE) andBruch's membrane had both been penetrated withblood filling the space between RPE and neural retina.The outer nuclear layer (ONL) had pyknotic cells andthe outer plexiform layer (OPL) was swollen with serousexudate (Fig. 2A). The most prominent features as-sociated with VHL damage were a portal through theneural retina and choroidal hemorrhage which throughthat portal entered the vitreal chamber (Fig. 2B).

The retinal hemorrhage ED50 for the 4-nsec pulse-width was calculated by probit analysis to be 340 MJwith confidence limits at the 95% level of 260 j J and440 / J . The highest energy exposure producing nohemorrhage was 330 /xJ. The lowest energy exposureto produce a hemorrhage was 270 /xJ.

Discussion. The 30 nsec ED5Os were consistent withpreviously reported data6 in that the macular thresholdsare lower than the extramacular thresholds for bothcontained and vitreal hemorrhages. The macula's lowerthreshold may be related to the decreased thickness ofthe retina and internal limiting membrane (ILM)thickness in this area.7 From observations on the for-mation of numerous VHLs, it was noted that devel-opment frequently proceeds through an intermediateCHL stage. The CHL increased in diameter until itruptured, forming a VHL. The elapsed time from ex-posure until the time of rupture never exceeded 10 sec.On two occasions the foci of rupture occurred some100 Mm from the exposure site. This shift between ex-posure and rupture sites could be due to local variationsin thickness and shear strength of the retina and ILM.7

In Figure 2A, a typical CHL, the laser radiation wasabsorbed principally by the RPE with damage pene-

Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933359/ on 05/16/2018

No. 7 Reports 1179

trating into the choroid. Choroidal blood has passedthrough the damaged Bruch's membrane and the RPEand caused the neural retina to separate from the RPEand be pushed upward. Serous exudate in the OPL isthought to be plasma filtering through the outer nuclearlayer. ONL cell damage is thought to result from RPEdamage and not directly from interaction with the 1064nm laser light.8 With the VHL, Figure 2B, the higherenergy density of the exposure has caused greater dam-age to the choroidal vaculature than in the CHL. Theincreased free blood pressure coupled with anteriorlydirected shock waves from the RPE have created aportal for blood passage from the choroid to the vit-reous compartment. Neural retinal detachment andOPL enlargement are also present.

Within the nanosecond pulsewidth range, both ther-mal and mechanical damage mechanisms may be op-erative with the dominant mechanism for lesion for-mation shifting from thermal to mechanical as the peakpower increases. With suprathreshold exposure energiessuch as those required to produce hemorrhages, theeffects are generally cataclysmic and can be attributedto both mechanisms.910 The hemorrhage threshold de-termined using a 4 nsec pulsewidth (340 iiJ) was foundto be a factor of six below the contained hemorrhagethreshold for 30 nsec pulses (1.9 mJ), suggesting thatat the shorter pulsewidth the mechanical damagemechanism was more dominant.Key words: hemorrhage, neodymium-yag, laser, retina,monkey

Acknowledgment. The authors thank W. Butcher, of theComparative Pathology Branch, School of Aerospace Med-icine, Brooks Air Force, Texas, for tissue preparation andmounting.

From Technology Inc., Life Sciences Division, San Antonio,Texas, the t Vulnerability Assessment Branch and ^Comparative Pa-thology Branch, the School of Aerospace Medicine, Brooks AFB,Texas, and the §Los Alamos National Laboratory, Los Alamos, NewMexico. Supported under Contract F33615-8O-C-O61O let by theVulnerability Assessment Branch, USAF School of Aerospace Med-icine, Brooks Air Force Base. Submitted for publication: October 1,1985. Reprint requests: Michael F. Blankenstein, Technology Incor-porated, Life Sciences Division, 300 Breesport, San Antonio, TX78216.

References

1: ANSI: American National Standard for the Safe Use of Lasers.Standard Z 136.1. New York, American National Standards In-stitute, 1980.

2. Allen R, Thomas S, Harrison R, Zuclich J, and BlankensteinM: Ocular effects of pulsed neodymium laser radiation: variationof threshold with pulsewidth. Health Physics 49:685, 1985.

3. Butcher W, Schmidt R, Elias F, and Hammond M: A rapidmethod for resectioning of semithin large epoxy sections forelectron microscopy. Micron 10:141, 1979.

4. Schmidt R and Zuclich J: Retinal lesions due to ultraviolet laserexposure. Invest Ophthalmol Vis Sci 19:1166, 1980.

5. Finney D: Probit Analysis, 3rd Ed., New York, Cambridge Univ.Press, 1971.

6. Gibbons W and Allen R: Retinal damage from suprathresholdQ-switch laser exposure. Health Physics 35:461, 1978.

7. Fine B and Yanoff M: Ocular Histology: A Text and Atlas. NewYork, Harper & Row, 1972.

8. Goldman A, Ham W Jr, and Mueller H: Mechanisms of retinaldamage resulting from the exposure of rhesus monkeys toultrashort laser pulses. Exp Eye Res 21:457, 1975.

9. Allen R: Retinal thermal injury. In Non-ionizing Radiation.Proceedings of a Topical Symposium. Amer Conf Govt IndustrialHygenists, Inc. Cincinnati, OH, 1980, pp. 161-168.

10. Ham W Jr, Ruffolo J Jr, Mueller H, and Guerry D III: Thenature of retinal radiation damage: dependence on wavelength,power level and exposure time. Vision Res 20:1105, 1980.

Full-Field Electroretinogroms in Miniature PoodlesWith Progressive Rod-Cone Degeneration

Michael A. Sondberg, Basil S. Pawlyk, and Elior L. Berson

Full-field electroretinograms (ERGs) were recorded betweenthe ages of 4 and 34 months from 8 miniature French poodleswith inherited progressive rod-cone degeneration (PRCD) andfrom 11 normal miniature poodles. Three stages of retinalfunction were observed in the affected dogs, although the agesof transition from one stage to the next were variable amongdogs. Rod and cone ERGs were normal in amplitude in thefirst stage, rod ERGs were reduced in amplitude and coneERGs were normal in the second, and rod ERGs were ex-tremely reduced in amplitude or nondetectable and cone ERGswere reduced in amplitude in the third. On average, thesepoodles with PRCD lost 7.2% of remaining rod amplitudeper month and 2% of remaining cone amplitude per month.Affected poodles had normal rod and cone b-wave implicittimes over all three stages, in contrast to humans with the

early stages of progressive forms of retinitis pigmentosa whohave delayed rod and/or cone b-wave implicit times. InvestOphthalmol Vis Sci 27:1179-1184, 1986

The miniature poodle with autosomal recessive pro-gressive rod-cone degeneration (PRCD) initially showssymptoms of night blindness, followed by impaired dayvision, eventually resulting in blindness.1 A similar se-quence of events is a well-recognized feature of thetypical forms of human retinitis pigmentosa.2 Electro-retinograms (ERGs) in response to Ganzfeld (full-field)stimulation in the early stages of human retinitis pig-mentosa have been characterized not only by reduc-

Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933359/ on 05/16/2018