DTIC ADA512812: Trends in Retinal Damage Thresholds from 100-Millisecond Near-Infrared Laser Radiation Exposures: A Study at 1,110, 1,130, 1,150, and 1,319 nm

0.14 MB·English

by  Defense Technical Information Center

Form Approved
REPORT DOCUMENTATION PAGE
OMB No. 0704-0188
Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the
data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing
this burden to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-
4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently
valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.
1. REPORT DATE (DD-MM-YYYY) 2. REPORT TYPE 3. DATES COVERED (From - To)
Published Journal Article January 2003- October 2008
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
F41624-02-D-7003
TRENDS IN RETINAL DAMAGE THRESHOLDS FROM CONTINUOUS- 5b. GRANT NUMBER
WAVE NEAR-INFRARED LASER ADIATION: A STUDY AT 1110, 1130,
1150, 1139 NM 5c. PROGRAM ELEMENT NUMBER
63231F
6. AUTHOR(S) 5d. PROJECT NUMBER
Rebecca Vincelette, Benajmin Rocksell, Jeff Oliver, Semih Kumru, Robert 7757
Thomas, Kurt Schuster, Gary Noojin, Aurora Shingledecker, Dave Stolarski, 5e. TASK NUMBER
B2
Ashley Welch
5f. WORK UNIT NUMBER

26
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION REPORT
NUMBER
Air Force Research Laboratory Northrop Grumman-IT
Human Effectiveness Directorate 4241 Woodcock Dr., Ste B-100
Directed Energy Bioeffects Division San Antonio, TX 78228
Optical Radiation Branch
2624 Louis Bauer Dr.
Brooks City-Base, TX 78235-5128
9. SPONSORING / MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR’S/MONITOR’S ACRONYM(S)
Air Force Research Laboratory
Human Effectiveness Directorate AFRL/RHDO
Directed Energy Bioeffects Division
11. SPONSOR’S/MONITOR’S REPORT
Optical Radiation Branch
NUMBER(S)
2624 Louis Bauer Dr.
AFRL-HE-BR-JA-2008-0025
Brooks City-Base, TX 78235-5128
12. DISTRIBUTION / AVAILABILITY STATEMENT
APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED. Case file no. 09-002, 5 Jan. 09
Publisehd in Lasers in Surgery and Medicine 41:382-390 (2009)
13. SUPPLEMENTARY NOTES
14. ABSTRACT

Retinal damage thresholds from 100-millisecond exposures to laser radiation for wavelengths between 1,100 and
1,350nm have never previously been established. We sought to determine the retinal damage threshold for 100-
millisecond exposures of near-infrared (NIR) laser radiation wavelengths at 1,110, 1,130, 1,150, and 1,319 nm.
These data were then used to create trends for retinal damage thresholds over the 1,100–1,350nm NIR region based
upon linear absorption of laser radiation in ocular media and chromatic dispersion of the eye.

15. SUBJECT TERMS
eye safe; laser safety; laser tissue interaction; non-ionizing radiation; ocular hazard
16. SECURITY CLASSIFICATION OF: 17. LIMITATION 18. NUMBER 19a. NAME OF RESPONSIBLE PERSON
b. ABSTRACT c. THIS PAGE OF ABSTRACT OF PAGES 19b. TELEPHONE NUMBER (include area
a. REPORT UU 10 code)
Unclassified Unclassified Unclassified
Standard Form 298 (Rev. 8-98)

Prescribed by ANSI Std. Z39.18
LasersinSurgeryandMedicine41:382–390(2009)
Trends in Retinal Damage Thresholds From 100-millisecond
Near-Infrared Laser Radiation Exposures: A Study at
1,110, 1,130, 1,150, and 1,319nm
RebeccaL.Vincelette,MS,1*BenjaminA.Rockwell,PhD,2JeffW.Oliver,PhD,2SemihS.Kumru,PhD,2
RobertJ.Thomas,PhD,2KurtJ.Schuster,MS,3GaryD.Noojin,AAS,3AuroraD.Shingledecker,CVA,3
DaveJ.Stolarski,BA,3andAshleyJ.Welch,PhD1
1Department of Biomedical Engineering, Cockrell School of Engineering, University of Texas, 1 University Station,
C0800,Austin,Texas78712
2AirForceResearchLaboratory,711thHPW,OpticalRadiationBranch,2624LouisBauerDr.,BrooksCity-Base,Texas
78231
3NorthropGrumman,4241WoodcockDr.Ste.B-100,SanAntonio,Texas78228
Background and Objectives: Retinal damage thresh- INTRODUCTION
oldsfrom100-millisecondexposurestolaserradiationfor
Thenear-infrared(NIR)laserradiationregionbetween
wavelengths between 1,100 and 1,350nm have never
1,150and1,350nmisauniqueareaintheelectromagnetic
previously been established. We sought to determine the spectrumforlasersafety.Itisinthisregion,whereocular
retinaldamagethresholdfor100-millisecondexposuresof
media begins to absorb very strongly, but still transmits
near-infrared(NIR)laserradiationwavelengthsat1,110,
sufficient energy to pose a hazard to the retina. Retinal
1,130,1,150,and1,319nm.Thesedatawerethenusedto damage threshold studies are conducted at specific laser
createtrendsforretinaldamagethresholdsoverthe1,100–
radiation parameters to quantify these hazards. These
1,350nmNIRregionbaseduponlinearabsorptionoflaser
studiesarethenanalyzedbyinternationalsafetycommit-
radiationinocularmediaandchromaticdispersionofthe
tees such as the American National Standards Institute
eye. (ANSI)andtheInternationalCommissionofNon-Ionizing
Materials and Methods: The paramacula and macula Radiation Protection (ICNIRP) for making recommenda-
areas of the retina in Macaca mulatta (rhesus) subjects tionsforsafeexposurelevels[1,2].
were exposed for 100milliseconds to NIR laser radiation For100-millisecondlaserradiationexposures,theretinal
wavelengthsusingaCoherentOPOlaserfor1,110,1,130,
damagethresholddataareafunctionofthreefundamental
and1,150nmandaLeelaserfor1,319nm.Probitanalysis
physical characteristics: (1) the transmission of laser
wasused to establish the estimated damage threshold in radiationtotheretina,(2)absorptionoflaserradiationby
the retina for 50% of exposures (ED50). Using trends of theretina,and(3)thediameterofthelaser-spotformedat
transmitted energy to the retina, refractive error of the theretina[3].Thediameterofthelaser-spotformedatthe
eye and linear absorption of the retina, a scaling factor retina is typically taken to be dependent on the natural
(SF) method was created to fit the experimental data, chromaticdispersionpropertiesoftheeye,independentof
predicting retinal damage thresholds over the 1,100– absorptionandthermalpropertiesofthetissue.
1,350nmregion. NIR1,300-nmlaserradiationretinaldamagethresholds
Results: The experimental retinal damage threshold, have been conducted by the United States Army and Air
ED50, for 100-millisecond exposures for laser radiation Forceoverthepastseveraldecades.Retinallesionstudies
wavelengths at 1,110, 1,130, and 1,319nm were deter- conducted on Macaca mulatta and rabbit subjects at
minedtobe193,270,and13,713mWofpowerdeliveredto 1,318nm discovered that lesions could be placed on a
thecornea,respectively.Theretinaldamagethresholdfor rabbit’s retinaregardless of exposure duration, butcould
the 1,150nm wavelength was statistically undetermined only be placed in a non-human primate’s retina for
duetolaser-powerlimitations,butwasachievedinoneout exposureslessthan1millisecond[4–8].Thisdiscrepancy
ofthreesubjectstested. in the occurrence of retinal lesions between species has
Conclusion: The SF predicts the experimental 100-
millisecond NIR ED50 value for wavelengths between Contract grant sponsor: Air Force Research Laboratory;
1,100and1,350nm.LasersSurg.Med.41:382–390,2009. Contractgrantnumber:FA8650-05-2-6501.
*Correspondenceto:RebeccaL.Vincelette,MS,109BlueSage
(cid:1)2009Wiley-Liss,Inc. Ln.,Cibolo,TX78108.E-mail:[email protected]
Accepted7April2009
PublishedonlineinWileyInterScience
Key words: eye safe; laser safety; laser tissue interac-
(www.interscience.wiley.com).
tion;non-ionizingradiation;ocularhazard DOI10.1002/lsm.20772
(cid:1)2009Wiley-Liss,Inc.
TRENDSINRETINALDAMAGETHRESHOLDS 383
never been observed in the previous 40 years of testing,
leaving retinal damage threshold studies on the conven-
tional M. mulatta animal models for 100-millisecond
exposures to NIR laser radiation inconclusive at
1,318nm.Additionally,retinaldamagestudiesinthelaser
radiationregionof1,100–1,300nmwavelengthshavenot
beenconductedfor100-millisecondexposures.
Thisarticlepresentsdatafromaretinaldamagethresh-
oldstudyusing100-millisecondexposuresto1,110,1,130,
1,150, and 1,319nm laser radiation wavelengths. The
paramacula was exposed for laser radiation wavelengths
of1,110,1,130,and1,150nm,whilethemaculawasusedfor
the1,319nmstudy.Thesedataarecomparedtotrendsin
thethreefundamentalcriteriapreviouslymentioned.
BACKGROUND Fig.1. Transmissionoflaserradiationinahumanandrhesus
In the 1,318nm NIR laser radiation retinal damage eye based on linear absorption data of Maher [15]. These
threshold studies conducted in rabbits by Zuclich et al. resultsassumedphysicalthicknessesofthecornea,aqueous
[4,5,9], lesions were never apparent after 1-hour post- humor,crystallinelens,andvitreoushumorfromFernandes
exposure, but were seen after 24hours. In other laser etal.[16],Lietal.[17],andWestheimer[18].
radiationretinaldamagestudies,below1,100nm,asmall
numberoflesionsappearedatthe24-hourendpoint,but respectively.Notethevariabilitybetweenthe1,110,1,130,
themajority(>90%)oflesionswerepresentat1hour[10]. and1,150nmwavelengthsislessthan1%,butthe1,319nm
Histology of the 1,318-nm radiation lesions also revealed wavelengthisaboutafactorof2.25timessmaller(Table1).
full-retinalthicknessdamageatthresholdlevelenergies,a Birngruberetal.[20]comparedthehistologicalsectionsofa
trait seen in visible lesions only for suprathreshold monkey’s retina for several subjects, reporting that the
exposures [5]. Zuclich et al. [5] also reported that lesions averagethicknessofthelayerofpigmentedgranuleswhich
which appeared at 24hours increased, even doubling, in are found in the cells of the RPE is approximately 5mm.
size, stabilizing after 48hours post-exposure. Again, Birngruber et al. reported finding equivalent pigmenta-
extensivegrowthinlesionsizeisnotacharacteristictrait tionsintheRPEandchoroidwithpigmentationdistributed
for lesions caused by visible laser radiation wavelengths. throughoutthechoroid(nopigmentationwasreportedin
Though histology is not provided for the study presented the choriocapillaris). From their histological findings,
here,theseobservationsfromZuclichetal.[4,5,9]providea Birngruber et al. [20] reported the absorption coefficient,
basisforunderstandingourdata. m ,oftheRPEtobethesameforthepigmentedregionsof
a
Lundetal.havecreatedanactionspectrumtofitretinal the choroid. Here, we are assuming that this absorption
damagethresholddatausingthreefundamentalcriteriaof coefficientfortheRPEandchoroidreportedbyBirngruber
transmission,absorption,andradius,subjecttochromatic et al. includes water absorption, and thus assume these
dispersion,ofthelaserradiationreachingtheretina[3,11– valuesaccountforanyabsorptionwhichwouldoccurinthe
14]. The transmission of NIR laser radiation from the neuralretina.Them forRPEoverwavelengthsfrom1,000
a
corneatotheretinainarhesusandhumanwaspredicted to1,400nmisshowninFigure2.Theexponentialfittothe
using one-dimensional Beer’s law and linear absorption absorptioncoefficients,m (cm(cid:2)1),forwavelength,l(nm),in
a
datafromMaher[15].Physicalthicknessesforthecornea, rhesusRPEis
aqueoushumor,crystallinelens,andvitreoushumorwere
m ðlÞ¼15874(cid:6)expð(cid:2)4:59(cid:6)10(cid:2)3lÞ ð1Þ
obtained from Fernandes et al. [16], Li et al. [17], and a
Westheimer [18]. The percent transmission decreases When a collimated beam is normal to the cornea, it is
sharply,asindicatedinFigure1,forthethreewavelengths focusedbythecorneaandlens.The1/e2laser-spotdiameter
of 1,110, 1,130, and 1,150nm from 57.3% to 42.2% and formedontheretina(Fig.3)bychromaticdispersionwas
finally12.4%,respectively,intherhesuseye.Thepercentof
laserradiationtransmittedat1,319nmforarhesuseyeis
(cid:1)3%.Thetotalabsorptionofthepre-retinalmediacanbe
TABLE1. TheLinearAbsorptionCoefficientofthe
estimatedfromFigure1bysubtractingthepercenttrans-
RPE,m ,fortheLaserRadiationWavelengthsinThis
missionfrom100%(thisneglectsreflection). a
Study
Lund et al. [14] determined the retinal pigment
epithelium (RPE) laser radiation absorption data which (nm) m (cm(cid:2)1)
a
providedthebest-fittoretinaldamagethresholddatawere
1,110 97.87
fromGabeletal.[19]andBirngruberetal.[20]wherethe
1,130 89.30
percentoflaserradiationreachingtheretinaisabsorbedby
1,150 81.48
the neural retina and RPE at 1,110, 1,130, 1,150, and
1,319 37.54
1,319nmwavelengthsis3.87%,3.55%,3.26%,and1.58%,
384 VINCELETTEETAL.
1,130,and1,150nmwavelengthswithrespecttoeachofthe
threefundamentalcriteria.
MATERIALSANDMETHODS
Thesubjectsinthisstudywerefouradult,malerhesus
(M. mulatta), approximately 4–5 years in age. Animals
usedinthisstudywereprocured,maintained,andusedin
accordancewiththeAnimalWelfareActandthe‘‘Guidefor
theCareandUseofLaboratoryAnimals’’preparedbythe
Institute of Laboratory Animal Resources, National
ResearchCouncil;andtheARVOResolutionontheUseof
AnimalsinResearch.Subjectswerescreenedforanyocular
defectspriortobeginningthestudy.Subjectswerechemi-
cally restrained with intramuscular injections of Telazol
Fig.2. Absorptionbyretinalpigmentepithelium(RPE)based (5–10mg/kgbodymass).Subjectsthenreceivedasubcuta-
onafittodatafromBirngruberetal.[20]. neousinjectionof0.16mgatropinesulfateandeyedrops;
two drops each of proparacaine hydrochloride 0.5%,
phenylephrine hydrochloride 2.5%, and tropicamide 1%,
determined by superimposing refractive error measure- todilatetheeye.Bilateralsaphenouscatheterswereused
mentsofhumaneyesontoasimplerhesuseyemodel[3]. bothforadministeringanesthetic(Propoflo11,500Mcg/kg
The predicted diameter formed at the rhesus retina for bolus,thencontinuousrateinfusionat300Mcg/kg/minute)
laserradiationwavelengthsof1,110,1,130,and1,150nmis and warm intravenous fluids (lactated ringers 10ml/kg/
146.5, 150.6, and 153.8mm respectively. The variability hour) throughout the procedure. Immediately after anes-
betweenthesethreewavelengths’laser-spotdiameterdue thesia was administered, the subject was intubated to
to chromatic dispersion is less than 5%. In Figure 3, the assuretheairwayremainedunobstructedthroughoutthe
laser-spotdiameterforthe1,319-nmwavelengthisapprox- experiment.Heartrate,respiration,andoxygensaturation
imately 184.6mm, or approximately 25% larger than the weremonitoredusingapulseoximeter.Thebodytemper-
otherthreelaserradiationwavelengthsinthisstudy. ature of the subject was checked periodically and main-
Of the three fundamental criteria leading to retinal tainedwithaBairHugger1(warmaircirculatingblanket).
damagefromlaserradiationpresentedinFigures1–3,the Tominimizeeyemovements,aperibulbarinjectionof1cc
transmitted energy is by far the distinguishing variable Xylocaine,4%,wasadministeredjustbeforebeginningthe
between1,110,1,130,and1,150nmwavelengths.Forthese laserradiationexposures.Theeyewasheldopenbyawire
threewavelengths,theabsorptionoftheRPEandthelaser- lidspeculumandirrigatedwith0.9%salinesolutionevery
spotdiameterdeliveredtotheretinavarybylessthan1% few seconds throughout the procedure. Subjects were
and 5%, respectively. However, the 1,319nm laser radia- securely restrained in the prone position on a custom
tion wavelength is considerably different from the 1,110, animalstagedesignedtoallowforcarefulmovementsofan
intubated subject in front of a Topcon TRC-50(cid:7) fundus
camera(Figure4).
ACoherentChameleonlasersetto834-nmwavelength
pumped the Coherent Mira-OPO tuneable laser for the
1,110,1,130,and1,150nmwavelengthsusedinthisstudy.
SeeTable2foradata-matrixwithparametersofexposures.
The bandwidth of the OPO laser ranged between 12 and
15nmfull-widthathalfmax.Awaveplateandanelectronic
shutter were used to control the power and exposure
durationoftheNIRlaseratthecornealplane.Athinpiece
ofglassplacedslightlyoffaxisjustbeforetheshutterwas
usedtoreflectasmallportionoftheNIRlaserradiationtoa
Coherentpowermeter,PM3,readbyaCoherentPowerMax
5000;thisisreferredtoasthereferencepowermeter.The
OPOlasedinTEM modewithitsGaussianbeamhavinga
00
full-angledivergenceand1/e2diameteratthecornealplane
of1.5mradand4.8mm,respectively,withinthespecifica-
Fig. 3. The 1/e2 laser-spot diameter for NIR laser radiation tions by the manufacturer. A Melles Griot 633-nm HeNe
wavelengths delivered to the rhesus retina with chromatic laserwasexpandedandattenuatedforuseasapointerto
dispersion. Results shown here are based upon a collimated allow for careful placement of retinal lesions. A Spectra
beamwith a4.24-mm 1/e2 diameter delivered tothe corneal Physics532-nmMillennialaserwasusedtoplacemarker
plane. lesions to map a grid on the retina. A second electronic
TRENDSINRETINALDAMAGETHRESHOLDS 385
thecornealplaneandtoensurestabilityandrepeatability
oftheexperiments.Toverifytheproperexposureduration,
a photodiode connected to an oscilloscope was placed
perpendicular to the corneal plane. In addition to the
photodiode,anElectroPhysicsNIRcamerawasplacedoff
axistozoominonthecornealplaneandrecordedonaDVD.
TheNIRcamerahelpedtoconfirmthatthelaserbeamwas
not clipped by the subject’s iris when delivering the NIR
laser radiation. The fundus camera was equipped with a
digitalcameramountedtoviewthefundusinadditiontoa
35-mm film camera mounted in another position to take
traditional photos. Fundus photographs were taken pre,
1hour,and24hourspost-NIRlaserradiationexposure.
A set of 532-nm marker lesions using 48mW at the
cornealplaneexposedfor20millisecondsweremadealong
Fig.4. Theopticalset-up.L1andL2,lensesformeda5(cid:7)beam
the macula–paramacula boundary to map out a grid for
expander for the 633-nm pointing laser; M1-5, broad-band
referencing laser exposures in our data set. NIR laser
visible mirrors; M6-8, silver mirrors. M7 and M8 formed a
radiation,all100-millisecondsexposureswerethenplaced
periscopetobringthebeamupverticallytotheviewinglevelof
along thegrid mapinthe paramacula. Subject’smaculas
thefunduscamera.BS1,abeamsplitterwhichcombinedthe
werenotavailabletouseintheOPOstudy.Theparamacula
twovisiblelasers;BS2andBS3,identicalbeamsplittersused
grid location and reference power meter reading were
tocombineallthreewavelengths;WP,waveplate.S1andS2
recorded.Threeobserversinspectedthesubject’sretinafor
were two independently controllable electronic shutters. P1,
lesions at 1- and 24-hour end points. These ‘‘lesion/no
reference power meter; P2, photodiode connected to an
lesion’’datawereprocessedinProbitanalysissoftwareto
oscilloscope;NIRCam.,near-infraredcamera.
determinetheestimateddosewhichcausesdamage50%of
the time, called an ED . The Probit analysis takes the
50
shutter controlled the delivery of the 532-nm laser binomial lesion/nolesiondataformingthesigmoiddose–
radiationtotheocularplane.Allthreelaser-wavelengths responsecurveandfitsthedatatoalinearregressionusing
(NIR, 633, and 532nm wavelengths) were co-aligned to a maximum likelihood algorithm. The Probit reveals
deliverlaserradiationtotheocularplaneofthesubject.The informationontheconfidenceintervals(CI)includingthe
subject was placed such that its fundus was in focus and upper and lower fiducial limits (CI¼95%) and the ED
50
observable by a fundus camera by the scientific team (CI¼50%).ProbitanalysishasbeenreportedbyCainand
(Fig.4)witha1-in.beamsplitterplacedjustbetweenthe Noojintoprovidethemostsuitablemethodfordetermining
fundus camera’s lens and the subject’s eye. This beam ED thresholdsandisawell-knownmethodfordetermin-
50
splitterwashighlytransmissiveinthevisibleallowingfora ing dose–response curves [21,22]. Slit lamp photographs
clearfundusimageevenduringlaserexposure.Thetotal were taken of subjects post-exposure. To check for any
lengthoftheopticaltrainfromtheOPOlaserheadtothe leakagefromtheplacedlesions,fundusfluoresceinangiog-
cornealplanewasmeasuredtobe261cm. raphy was performed and recorded using 35-mm film
With the OPO tuned to the desired wavelength, the immediately following the 1- and 24-hour retinal lesion
alignment of the system was checked 24hours and reads.Thedoselevelsforfundusfluoresceinangiography
30minutes before the arrival of the subject. At were 0.5–1mg/kg IV (acepromazine) given 20minutes
thesetimes,anotherCoherentpowermeterwasplacedat priortoadministeringtheFluorescein(0.4ml/subject)dye.
the corneal plane and the ratio of the reference power The exposures for the 1,319-nm laser radiation used a
meter to the delivered corneal power was recorded. The similarexperimentaltechnique,saveforthelasersourcewas
ratioofthe532-nmlaserradiationdeliveredtothecorneal aLEE-laserandtheshutterwasmanufacturedbynmLaser
planecomparedtothelaserheadsettingwasalsoobtained. Products,Inc(SanJose;CA).Exposuresat1,319nmwerefor
Theseratioswereusedtocomputethepowerdeliveredto 80milliseconds to the macula of the same subjects used in
the OPO experiment. The 1,319-nm beam characteristics
TABLE2. TheParameterSetsforNIRExposures atthecornealplanewerea1/e2diameterof(cid:1)5mmwitha
InvestigatedinThisStudy full-anglebeamdivergenceof13.3mradandM2of(cid:1)50.
An estimated fit of the damage threshold data was
Exposure 1/e2diameterat Exposure performed using a scaling algorithm developed with the
l(nm) duration(ms) cornealplane(mm) location three fundamental criteria previously described. The
scalingfactor(SF)wascreatedbybreakingthefundamen-
1,110 100 4.8 P
talcriteriaintoasimpleheatsourceterm,S(W/cm3),fora
1,130 100 4.8 P
wavelength,l,intheNIRbasedon
1,150 100 4.8 P
1,319 80 5.0 M
P
SðlÞ¼ ðm Þ ð2Þ
PandMareforparamaculaandmacula,respectively. pr2 a
386 VINCELETTEETAL.
wherePisthepowerreachingtheretina(W),risthe1/e2 exposureonthefundus.Combiningallyes/nodataforboth
radiusofthebeamattheretina(seeFig.3),andm isthe exposuresessions(datafrombothFigure6aandb)gavethe
a
linear absorption coefficients of the RPE (assumed to resultingED of439mWreportedinTable3.
50
includewater)incm(cid:2)1.Ifaninputisassumedtobeunity
at the corneal plane, then the power reaching the retina DISCUSSION
can be taken as the fraction of transmitted power, P , to Based on fundus camera observation, all of the NIR
T
theretinabasedontheresultsfromFigure1. retinallesionsinthisstudyhadthetypicalappearanceof
Thesimpleheatsourceterm,S,canbecalculatedforthe lookinglargerthanthemarkerlesionsproducedbya532-
NIR region using Equation (2), then divided into nm wavelength laser. At higher NIR power exposures,
S(1,110nm) to create the SF relative to the 1,110-nm lesionscommonlyappearedlargerthanlesionsfromlower
damagethreshold.Thus,theSFfortheNIRiscalculatedby power exposures (Fig. 6a). Lesions always continued to
! grow over the 24-hour period with many lesions only
SF¼ðPT;1110nm=r21110nmÞ ma;1110nm ð3Þ becoming apparent at the 24-hour end point (Table 3).
ðPT;l=r2lÞ ma;l These observations are similar to those made by Zuclich
et al. [5,9,24,25] who reported that the delayed onset
A fit to ED data in the NIR can then be made by
50 inflammation is a characteristic of NIR laser–tissue
multiplying the SF by the experimental 1,110-nm ED
50 interactionwiththeretina.Webelievetheretina’sdelayed
value.
responsetoNIRlaserradiationthresholdenergylevelsis
related to the large beam diameter owing to chromatic
RESULTS
dispersionandthestrongincreaseinwaterabsorptionat
The results of the Probit analysis of the yes/no data at
these laser radiation wavelengths, resulting in a more
1and24hoursforeachwavelengthexposureforindividual
uniformly distributed heating effect in the retina as
and grouped subjects are in Table 3. Probit results for
comparedtoalocalizedeffectduetostrongRPEabsorption
combinedsubjectsinTable3representthecompleteProbit
at visible laser radiation wavelengths. In the study
analysisperformedonalltheyes/nodataobtainedfromthe
presented here, power levels that produced lesions never
corresponding data set at the reported laser radiation
resultedinaretinalhemorrhage.
wavelength.Probitanalysisshouldnotbeconfusedasan
Inregardstothe100-millisecondparamacularexposures
average over all subjects. Slit lamp photography always
tothe1,150-nmlaserradiation,onlyonesubjectoutofthree
appearednegativeforlesionsinthecornea,aqueous,and
testedyieldedanyfunduscopicallyobservablelesions.This
lens.TheGaussian-beamprofileshad1/e2diametersof4.8
single subject’s 1-hour ED data were not statistically
and 5mm for the OPO and LEE lasers, at the cornea, 50
significantsincetheupperfiduciallimitwasgreaterthan
respectively. The OPO and LEE lasers’ full-angle beam
50% of the corresponding ED . However, the 24-hour
divergencewere1.5and13.3mrad,respectively. 50
Probit of subject 1’s 1,150-nm wavelength data set main-
The80-millisecondthresholdexposureforthe1,319-nm
tainedtheupperandlowerfiduciallimitswithin50%ofthe
laser radiation wavelength can be scaled to 100milli-
ED . Subject 1’s 24-hour ED 1,150-nm laser radiation
secondsbasedon 50 50
thresholdvalueof439mW(deliveredtothecornea)wasa
(cid:3)100(cid:4)ð3=4Þ power levelof approximately2.27 timesgreater thanthe
ED50(cid:2)100msðJÞ¼ED50(cid:2)80msðJÞ(cid:6) 80 ð4Þ 193-mW24-hourED50from1,110nm.BaseduponFigure5,
the 100-millisecond ED for 1,150-nm laser radiation is
50
where J are units of Joules, as reported in studies 1,181mW,oraSFof(cid:1)6timeshigherthanthe1,110-nm24-
investigating the time-dependence relationship to max- hourED (Table4),buttheOPOlaserwasonlycapableof
50
imum permissible exposure levels [23]. Using Equation deliveringupto(cid:1)870mW,tothecornealplane.Subjects2
(2),theSFfromEquation(3),overseveralwavelengthsis and 3 were tested at the 1,150-nm laser radiation wave-
giveninTable4. lengthandyieldednofunduscopicallyobservablelesionson
The24-hourED datainTable3areplottedinFigure5 theretina.The24-hourED valuesfollowedtheSFtrend
50 50
(all 1,319-nm data were scaled to 100milliseconds using wellforsubjects2and3at1,110,1,130,and1,319nmlaser
Eq. 4). A fit to the ED data in Figure 5 was made by radiation wavelengths and for subject 1 at 1,130 and
50
multiplyingtheSF(Table4)bythe24-hourED valuefor 1,319nm.
50
combined subjects at 1,110-nm laser radiation (193-mW TheSFmethodpreviouslypresentedbasedonthethree
deliveredtocornea). fundamentalcriteriacreatesausefultechniqueforfitting
Subject1wasexposedto1,150-nmlaserradiationinthe ED data to 100-millisecond exposures from NIR laser
50
paramaculaontwoseparateoccasions;3weeksapart.The radiation. The SF method predicted the 1,319-nm laser
firstsetofexposuresresultedinverylargelesions(largeat radiationED forcombinedsubjectstobe15.3Wdelivered
50
1hour, and even larger at 24hours) with an ED of atthecornea.Thispredictedvalueisonly(cid:1)12%largerthan
50
406mW,seeFigure6a,whilethesecondsetofexposures the24-hourED experimentalvalueof13.7WinTable3
50
resultedinmuchsmallerlesions(remainedsmallat1and (experimental value was scaled to the 100-millisecond
24hours)withanED of567mW,seeFigure6b.Lesions exposure using Eq. 4). This error is within the upper
50
labeled without numbers in Figure 6 are marker lesions and lower fiducial limits of the 24-hour ED , 15.3
50
used to reference the grid location of each NIR radiation and 11.8W, respectively (fiducial limits here were scaled
TRENDSINRETINALDAMAGETHRESHOLDS 387
Lowerfiducialmit(mW) 14813221516031823115.6241255N/AN/A8729,5808,05015,80012,500 onds.N/A
li ec
s
W) milli
Upperfiducialmit(m 172132373237457314241301524N/AN/A87214,20019,10023,30016,200 or100
urs li m)f
o n
4h W) 50
2 0000 1
s (m50 161132299193364266210270439N/AN/A8722,504,509,804,50 or1,
d D 1111 f
hol E able
Thres No 1823135424196491010525871025 ravail
ge we
ama Yes 17111442111526522100211891340 mpo
D u
m
rMacula) Lowerfiducialmit(mW) 171153310207370201248295489N/AN/AN/A13,600N/A20,20017,000 mW(maxi
fo li 30
M 8
(cid:1)
al(PforParamacula, 1hour UpperfiducialED(mW)limit(mW)50 1701971531533761472634054121,1503042,610,000275306346501765287N/AN/AN/AN/AN/AN/A15,90018,20016,500N/A23,70028,20018,90021,800 houghexposureswereatfortherelevantdataset.
n te
eti enabl
mtheR No 2125196529221768201053515101540 erved,evunattain
Fro Yes 1498316121533110011116825 obswas
e
d erL
ne 4 3 3 3 wF
aObtai Subj. 2432,3,and3121,2,and1321,2,and1321,2,and olesionscatesthe
SummaryoftheDat Exp.dur.milliseconds)Loc. 100P 100P 100P 80M indicatesndforanED50afiduciallimit(FL)indi
3. ( ren
E tedi
TABL (nm)l 1,110 1,130 1,150 1,319 N/Aenentere
388 VINCELETTEETAL.
TABLE4. TheScalingFactor,SF,FromEquation(2),
forSeveralWavelengthsintheNIR
l(nm) SF
1,100 0.872
1,110 1.000
1,130 1.563
1,150 6.119
1,200 12.299
1,250 12.343
1,300 27.899
1,315 60.267
1,319 79.696
1,320 85.898
The SF is intended to be multiplied by the 1,110-nm laser Fig. 5. The 24-hour, 100-millisecond, ED retinal damage
50
radiationwavelength’s24-hourED50. thresholds from Table 3 (1,319-nm data were scaled to
100milliseconds by Eq. 4) and the SF fit determined by
mulitplying the SF values (Table 4) by the 1,110-nm ED
50
thresholdforcombinedsubjects.
to100-millisecondexposuresusingEq.4),fromtheexperi-
ment (Table 3). The action spectrum produced by Lund
et al. [14] shows that the expected ED for 100milli- retinaanddoesnotrequireabsolutevalues.Moreadvanced
50
seconds,1,110-nmlaserradiationisapproximatelyalittle modeling methods are required to determine and under-
less than 200-mW power delivered to the cornea. Our stand absolute values and their relationship on retinal
experimentaldatafromthreesubjectsfor100milliseconds, damagethresholds.
1,110-nmlaserradiationyieldedanED of193-mWpower Forthe1,150-nmwavelength,subject1wasexposedto
50
atthecornealplane,veryclosetothepredictionmadeby thislaser radiationearly oninthestudy, andbrought in
Lundetal.[14].TheSFvaluespresentedherearebasedon asecondtime3weekslatertodetermineifthe1,150-nm
the1,110-nmreferenceof193mWdeliveredtothecornea. lesionswererepeatable.Inthefirst1,150-nmexposure,at
It should be stressed that the SF algorithm depends on 1hour,subject1hadlargelesionsevenat500mWofpower
relative values between wavelengths for percent trans- delivered to the cornea and an 1,150-nm 24-hour ED
50
mission,absorption,andbeam-waistradiideliveredtothe (fromthisfirstexposuresetonly)of407mW(seeFig.6a).In
Fig. 6. Funduscameraimagescapturedofsubject1at24-hourpost-exposureto1,150-nm
laserradiation(funduscamerawassetto358fieldofviewineachimage).TheletterMdenotes
amarkerlesion,withM actingasareferencemarkerbetweenaandb.Thenumbersmarked
0
below the grid locations represent the power (mW) delivered to the corneal plane for
100milliseconds.a:Displaysthelargelesionsobservedforthefirst1,150-nmexposurewitha
resulting ED of 406mW. b: Displays lesionsobserved forthe second 1,150-nm exposure,
50
conducted3weekslater,onthesameeyeresultinginanED of567mW.Thearrowspointto
50
exposureareasoffthemarkergridinattemptstocheckexperimentalrepeatability.
TRENDSINRETINALDAMAGETHRESHOLDS 389
thesecond1,150-nmexposure,theonly1-hourlesionswere REFERENCES
the 830-mW power levels delivered to the cornea giving 1. ANSI.AmericanNationalStandardforSafeUseofLasers.
much smaller visible lesions than the first data set with Orlando,FL:LaserInstituteofAmerica;2000.
2. ICNIRP.Revisionofguidelinesonlimitsofexposuretolaser
a resulting 1,150-nm 24-hour ED (from this second
50 radiationofwavelengthsbetween400nmand1.4microns.
exposuresetonly)of567mW(seeFig.6b).Thepigmenta- HealthPhys2000;79(4):431–440.
tion of the two regions in the paramacula which were 3. VinceletteRL,RockwellBA,ThomasRJ,LundDJ,WelchAJ.
exposed to the 1150-nm laser radiation were noted to be Thermallensinginocularmediaexposedtocontinuous-wave
near-infraredradiation:1150-1350nmregion.JBiomedOpt
fairlyuniformanddidnotappearanylighterordarkerthan
2008;13(5):054005.
anyothersubject’sparamacula. 4. ZuclichJA,LundDJ,StuckBE,EdsallPR.Oculareffectsand
The SF method predicts the ED of 1,150-nm laser safety standard implications for high-power lasers in the
50
1.3–1.4 mm wavelength range. San Antonio, TX: Air Force
radiation (100-millisecond exposure) should be (cid:1)6 times
Research Laboratory; 2004. Report no. AFRL-HE-BR-TR-
greater(Table4)thanthe1,110-nmED50(Fig.5).Subject1 2004–0187.pp.1–16.
yielded1130and1319nmED resultsconsistentwiththe 5. ZuclichJA,SchuscherebaST,ZwickH,BoppartSA,Fujimoto
50
SF method (Table 3 and Fig. 5), suggesting absorptive JG, Cheney FE, Stuck BE. A comparison of laser-induced
retinal damage from infrared wavelengths to that from
properties of this subject’s retina are not largely deviant
visible wavelengths. Lasers Light Ophthalmol 1997;8(1):
fromothersubjects.Giventhesedatatrendsandthefact 15–29.
thatsubject1hadtwoverydifferentED ’sovera3-week 6. ZuclichJA,LundDJ,StuckBE.Wavelengthdependenceof
50
oculardamagethresholdsinthenear-IRtofar-IRtransition
intervalforthe1150-nmexposurewithremarkablysmaller
region:ProposedrevisionstoMPEs.HealthPhys2007;92(1):
lesions on the second exposure, it is possible that the 15–23.
refractivestateofsubject1’seyevariedbetweenexposures. 7. Lund DJ, Stuck BE, Beatrice ES. Biological research in
support of project MILES. San Francisco, CA: Letterman
Possiblesourcesforcausingthistypeofvariationinclude
Army Institute of Research; 1981. Report no. 96. pp. 1–
the anesthesia to induce the relaxed state of the eye and 54.
oculargrowth[16].Allsubjectswerecheckedforrefractive 8. LundDJ,EdsallP,StuckBE.OcularhazardsofQ-switched
error several months prior to beginning the damage near-infraredlasers.In:StuckBE,BelkinM,eds.Laserand
NoncoherentLightOcularEffects:Epidemiology,Prevention,
threshold studies, but are not routinely checked. In
andTreatment,III.SanJose,CA:SPIE;2003;85–90.
addition,pathlengthwasnevermeasuredatanypointin 9. ZuclichJA,ZwickHT,SchuschurebaST,StuckBE,Cheney
the study. Data for this subject suggest that it may be FE. Ophthalmoscopic and pathologic description of ocular
damage induced by infrared laser radiation. J Laser Appl
beneficialtoroutinelyscreensubjectsbeforethestartofan
1998;10(3):114–120.
exposure,afterthesubject’seyeisbelievedtobeinafully 10. CainCP,NoojinGD,StolarskiDJ,ThomasRJ,RockwellBA.
relaxed state, to check for consistency in refractive error Near-infrared ultrashort pulse laser bioeffects studies. In:
and/orocularpathlength. Defense Do, eds. Volume AFRL-HE-BR-TR-2003-0029. Air
ForceResearchLaboratory;2003.
11. Lund DJ, Edsall P. Action Spectrum for Retinal Thermal
Injury. Laser and Noncoherent Light Ocular Effects: Epi-
CONCLUSION
demiology Prevention and Treatment, III. San Jose, CA:
TheexperimentalED retinalthresholdsfor100-milli- SPIE;1999.pp324–334.
50
secondexposurestolaserradiationwavelengthsat1,110, 12. Lund DJ, Edsall P, Stuck BE. Spectral dependence of
retinal thermal injury. In: Exarhos GJ, Guenther AH,
1,130,and1,319nmweredeterminedtobe193,270,and
Kozlowski MR, Lewis KL, Soileau MJ, eds. Laser-Induced
13,713mW of power delivered to the corneal plane, Damage in Optical Materials. San Jose, CA: SPIE; 2000.
respectively.TheexperimentalED retinalthresholdfor pp22–34.
50
13. Lund DJ, Edsall P, Stuck BE. Wavelength dependence of
100-millisecondexposurestolaserradiationatthe1,150-
laser-inducedretinalinjury.In:MannsF,So¨derbergPG,Ho
nmwavelengthwasnotachievedduetolimitationsonthe A, Stuck BE, Belkin M, eds. Ophthalmic Technologies XV.
availabilityofpowerfromthelasersource.TheSFmethod Bellingham,WA:SPIE;2005.pp383–392.
14. LundDJ,EdsallP,StuckBE.Spectraldependenceofretinal
allowsfortrendsinthefundamentalcriteriaoftransmitted
thermalinjury.JLaserAppl2008;20:2.
energy,retinalabsorption,andrefractiveerrortoshapethe 15. MaherEF.Transmissionandabsorptioncoefficientsforthe
fit,asopposedtoexactvaluesofthesecriteria,whichare ocular media of the Rhesus monkey: USAF Schoolof Aero-
difficult to determine. Given that the ED for 1,150-nm spaceMedicine;1978,September15,1974toSeptember15,
50 1976.Reportno.SAM-TR-78-32.pp1–104.
laser radiation was achieved on one subject, it may be
16. FernandesA,BradleyDV,TiggesM,TiggesJ,HerndonJG.
worthwhile to routinely screen for refractive error and Ocular measurements throughout the adult life span of
measure the ocular path length of subjects prior to rhesus monkeys. Invest Ophthalmol Vis Sci 2003;44(6):
2373–2380.
beginningaNIRretinalthresholdexperiment.Thiswould
17. LiG,ZwickH,StuckB,LundDJ.Ontheuseofschematiceye
eliminate the uncertainty in the subject’s path length or models to estimate retinal image quality. J Biomed Opt
refractive errorvariability whether from growth or anes- 2000;5(3):307–314.
thesia. 18. Westheimer G. The eye. In: Mountcastle VB, ed. Including
Central Nervous System Control of Eye Movements. St.
Louis,MO:C.V.MosbyCompany;1980.pp482–503.
ACKNOWLEDGMENTS 19. Gabel VP, Birngruber R, Hillencamp F. Visible and near
infraredlightabsorptioninpigmentepitheliumandchoroid.
Support was provided by the Consortium Research In: Shimizu K, ed. XXIII Concilium Ophthalmologicum
(ReprintedfromtheInternationalCongressSeriesNo.450).
FellowshipProgramattheAirForceResearchLaboratory
Kyoto:ExcerptaMedica,Amsterdam-Oxford;1978.pp658–
undercontractnumberFA8650-05-2-6501. 662.
390 VINCELETTEETAL.
20. Birngruber R, Hillenkamp F, Gabel VP. Theoretical inves- 24. Zuclich JA, Lund DJ, Edsall PR, Stuck RE, Hengst G.
tigations of laser thermal retinal injury. Health Phys Highpower lasers in the 1.3- to 1.4-mm wavelength range:
1985;48(6):781–796. Ocular effects and safety standard implications. In: Stuck
21. CainCP,NoojinGD.Acomparisonofvariousprobitmethods BE, Belkin M, eds. Laser and Noncoherent Light Ocular
foranalyzingyes/nodataonalogscale.In:DefenseDo,eds. Effects:Epidemiology,Prevention,andTreatment.SanJose,
VolumeAL/OE-TR-1996-0102.BrooksCity-Base,TX:USAF CA:SPIE;2001.pp78–88.
ArmstrongLaboratory;1996. 25. ZuclichJA,SchuscherebaS,ZwickH,CheneyF,StuckBE.
22. FinneyDJ.Probitanalysis.Cambridge,England:University Comparing laser-induced retinal damage from IR wave-
Press;1971. lengthstothatfromvisiblewavelengths.In:StuckBE,Belkin
23. Henderson R, Schulmeister K. Laser Safety. Philadelphia, M,eds.Laser-InflictedEyeInjuries:Epidemiology,Prevention,
PA:InstituteofPhysicsPublishing;2004.pp132–220. andTreatment.SanJose,CA:SPIE;1996.pp66–79.