anterior segment fluorescein angiography of the normal canine eye using a dslr camera adaptor

Anterior segment fluorescein angiography of the normal canine eyeusing a dSLR camera adaptor

Anthony F. Alario, Christopher G. Pirie and Stefano PizziraniDepartment of Clinical Sciences, Tufts Cummings School of Veterinary Medicine, 200 Westboro Road, North Grafton, MA 01536, USA

Address communications to:

Dr. C. G. Pirie

Tel.: (508) 839-5395

Fax: (508) 887-4363

e-mail: [email protected]

AbstractPurpose To describe anterior segment fluorescein angiography (ASFA) of the normalcanine eye using two different sedation/anesthetic protocols and a digital single lens-

reflex (dSLR) camera adaptor.Methods Dogs free of ocular and systemic disease were used for this study. Dogsreceived maropitant citrate (1.0 mg/kg SQ) and diphenhydramine (2.0 mg/kg SQ)

20 min prior to butorphanol [n = 6] (0.2 mg/kg IV) or propofol [n = 6] (4 mg/kg IVbolus, 0.2 mg/kg/min CRI). Standard color and red-free images were obtained prior

to administration of 10% sodium fluorescein (20 mg/kg IV). Image acquisition wasperformed using a dSLR camera (Canon 7D), dSLR camera adaptor, camera lens

(Canon EF-S 60 mm f/2.8 macro), and an accessory flash (Canon 580EXII). Imagingoccurred at a rate of 1/s immediately following bolus for a total of 30 s, then at 1, 2,

3, 4, 5, and 10 min.Results Twelve dogs with a combined mean age of 5.1 years and various iris colorswere imaged. Arterial, capillary, and venous phases were identified and time sequences

recorded. Visibility of the vascular pattern was inversely related to iris pigmentation.Complete masking of blood flow was noted with heavily pigmented irises. Vessel

leakage was noted in some eyes. Proper patient positioning and restricted ocularmovements were critical in acquiring quality images. No adverse events were noted.

Conclusion This study demonstrated that quality high resolution ASFA images wereobtainable using a novel dSLR camera adaptor. ASFA of the normal canine eye is

limited to irises, which are moderately to poorly pigmented. Use of general anesthesiaproduced higher quality images and is recommended for ASFA in the dog.

Key Words: anterior segment, anterior uvea, canine, fluorescein angiography, iris

INTRODUCTION

Fluorescein angiography is a well established diagnosticimaging modality, which has primarily focused on evaluat-ing the posterior segment vasculature of the eye, in bothphysician-based and veterinary medicine.1,2 However,shortly following its inception in 1961, fluorescein angiogra-phy expanded to incorporate the anterior segment of thehuman eye.1–4 Initial studies concentrated on angiographicfeatures of normal subjects.2–4 These studies established abaseline for future studies and provided significant and per-tinent information regarding the normal vascular patterns of

the human eye.2 Subsequent reports included angiographicfindings in various pathologic conditions, such as diabeticmicroangiopathy, glaucoma, uveitis, neoplastic conditions,trauma, and developmental disorders.1,2,5–7 Despite thevaluable information provided by this technique, its use hasremained limited.2 Within the veterinary literature, the onlypublished reports of anterior segment fluorescein angiogra-phy (ASFA) in the dog were by Schanz et al.,8,9 and thesewere limited to examination of the blood supply of conjunc-tival grafts. To the authors’ knowledge, there are currentlyno reports using this technique specifically to documentnormal iridal vasculature and flow patterns in canines.

While detailed descriptions of normal canine anteriorsegment vascular patterns have been reported, these studieswere conducted using corrosion casts.10,11 In these reports,the major blood supply of the iris is via the long posterior cil-iary arteries (nasal and temporal). These vessels penetrate

Presented as an abstract at ACVO 2011 Hilton Head, SC USA.

Grant/Financial support: None.

Financial disclosure: Anthony Alario (None) Chris G Pirie (P), Stefano

Pizzirani (None).

� 2012 American College of Veterinary Ophthalmologists

Veterinary Ophthalmology (2012) 1–10 DOI:10.1111/j.1463-5224.2012.01007.x

the sclera near the equator and bifurcate at the level of theiris base at the three and nine o’clock positions. From here,their branches travel superiorly and inferiorly to form themajor arterial circle (MAC) of the iris.10,11 In dogs, theMAC is considered to be incomplete. From the MAC, ves-sels branch off forming both radial ciliary arteries and radialiris arterioles.11 The radial ciliary arteries travel in a poster-ior direction toward the ciliary body and processes, whereasthe radial iris arterioles travel toward the pupillary border ina tortuous course branching along the way. Additionally, aproportion of radial iris arterioles have been reported tooriginate from radial ciliary arteries. The radial iris arteri-oles terminate as fine capillaries at the pupillary border.These capillaries subsequently drain into venules that travelposteriorly toward the iris base in a radial fashion, ultimatelydraining into the ciliary body vasculature.10,11

Anterior segment fluorescein angiography may be a clini-cally useful diagnostic imaging modality within veterinarymedicine; however, its current disuse may relate to equip-ment requirements and associated costs.2 Currently, ASFAis performed utilizing a photographic slit-lamp or modifiedfundus camera, appropriate lenses, light source, filters (exci-tation and barrier), and a photographic or video recordingsystem.1–4,12 Recently, a simple, low cost dSLR adaptor for astandard dSLR camera has been developed by one of theauthors (CGP). This adaptor has been shown to producehigh quality standard images of the anterior and posteriorsegment of many species.13 Additionally, it has demon-strated sodium fluorescein angiographic capabilities of theposterior segment of canine globes.14 The purpose of thisstudy was to determine whether this same adaptor can alsoperform ASFA of the canine globe, as well as to documentnormal parameters of ASFA of the canine iris. There are cur-rently no reports within the veterinary literature describingthis technique or providing normal parameters. Addition-ally, two different protocols were assessed to determinewhich would provide diagnostic quality images, yet be prac-tical in the clinical setting. A sedative (butorphanol) and ananesthetic agent (propofol) were chosen for this purpose.

MATERIALS AND METHODS

AnimalsTwelve client owned dogs from the Foster Hospital forSmall Animals at the Tufts Cummings School of VeterinaryMedicine were used for this study. This study was approvedby Tufts Cummings School of Veterinary Medicine ClinicalScience Review Committee.

Prior to enrollment, all dogs received a complete physicaland ophthalmologic examination by a boarded ophthalmol-ogist (CGP) and were deemed to be free from ocular and/orsystemic disease.

Study protocolHeart rate, body temperature, and Doppler blood pressure(Parks Medical Electronics, Inc, Aloha, OR) were recorded

prior to and following ASFA. At least 20 min prior toinjection of sodium fluorescein, all dogs received maropitantcitrate (1.0 mg/kg SQ, Pfizer, New York, NY) and diphen-hydramine (2.0 mg/kg SQ, Baxter healthcare, Deerfield,IL). A 20 gauge sterile IV catheter was then routinely placedin the right or left cephalic vein. Dogs were then eithersedated with butorphanol [n = 6] (0.2 mg/kg IV, Pfizer) oranesthetized using propofol without intubation [n = 6](4 mg/kg IV bolus, 0.2 mg/kg/min CRI, Baxter healthcare).Dogs sedated with butorphanol were gently manuallyrestrained during image acquisition. Dogs anesthetized withpropofol were positioned in sternal recumbency, and twostay sutures (5-0 nylon) were placed within the bulbar con-junctival tissue near the limbus to center the globe duringimage acquisition. One suture was placed at the ventronasalaspect of the globe and the other laterally. All globes beingimaged were regularly lubricated using a balanced saltsolution (BSS; Akorn Inc., Lake Forest, IL) to maintain thehealth of the cornea during imaging. In both groups, stan-dard color and red-free images were obtained prior toadministration of 10% sodium fluorescein (AK-FLUOR10%; Akorn Inc). Image acquisition was performed using anestablished dSLR adaptor system consisting of a digital sin-gle lens-reflex (dSLR) camera (Canon 7D), dSLR cameraadaptor, camera lens (Canon EF-S 60 mm f/2.8 macro), andan accessory flash (Canon 580EXII).13

Fluorescein angiographyFor angiographic purposes, an appropriate interference fil-ter combination (excitation and barrier filters) were insertedwithin the illumination and optical pathway of the adaptor,as previously described.14 Prior to the injection of sodiumfluorescein, images were obtained with the excitation andbarrier filters in place, serving as controls to determine thelevel of background pseudofluorescence.

All angiograms were performed following a rapid intrave-nous bolus of 10% sodium fluorescein at a dose of 20 mg/kg. All injections were performed manually during whichtime the photographic sequence and timer were initiatedsimultaneously. Imaging occurred at a rate of 1/s for a totalof 30 s, then at 1, 2, 3, 4, 5, and 10 min. After image acquisi-tion was complete, stay sutures were removed from thosedogs in the propofol group, and animals were recoveredfrom anesthesia/sedation. Reexamination of the animals wasperformed upon recovery from anesthesia/sedation toensure no sustained ill effects had occurred during the pro-cedure. Animals were then returned to their owners.

Fluorescein angiography evaluationMeasurements were performed in accordance with Kottow1

and included the onset of arterial, capillary, and venousphases. Briefly, the arterial phase was defined as the periodduring which sodium fluorescein entered the MAC andended upon reaching the pupillary capillaries. The capillaryphase encompassed the time from initial capillary filling andended upon filling of the iris venules. Finally, the venous

2 a l a r i o E T A L .

� 2012 American College of Veterinary Ophthalmologists, Veterinary Ophthalmology, 1–10

phase was defined by the onset of iris venule filling at thepupillary border.1 All time measurements marking the onsetof each phase were performed by one author (CGP) in dupli-cate and subsequently averaged. Qualitative presence of ves-sel leakage and/or leakage within the aqueous humor wasalso recorded.

Statistical analysisHeart rate, body temperature, and blood pressure measure-ments obtained prior to and following ASFA using the afore-mentioned protocols were compared via a two-tailed pairedt-test (Microsoft� Excel 2010). A P value <0.05 was consid-ered statistically significant.

RESULTS:

Animal population and protocol assessmentTwelve dogs with a combined mean age of 5.1 ± 2.9 yearswere used in this study. The breed distribution for dogs usedwas as follows: four Australian Shepherds, three SiberianHuskies, two mixed-breed dogs, and one German Shepherd,American Bulldog, and Golden Retriever. Seven castratedmales and five spayed females were used. Iris pigmentationvaried including six poorly (blue), three variably (hetercho-mic), one moderately (light brown), and two heavily pig-mented (dark brown).

Use of butorphanol for the purpose of this study resultedin satisfactory sedation for approximately 1 h, allowing nec-essary handling and manual positioning of the dogs duringthe imaging sequence. However, ocular movement wasunrestricted. These movements, while minor, periodicallyresulted in poor focus throughout the final image(s) as dem-onstrated in Fig. 1. Use of propofol resulted in an adequateplane of anesthesia for necessary handling and patient posi-tioning. While ventromedial globe rotation was notedrequiring placement of two stay sutures, once in place, globeposition remained fixed and central. This translated intoimproved quality and focus throughout the final image. Noadverse events were noted with either protocol.

There was no statistical difference between pre- and post-study values regarding body temperature (Butorphanol/

propofol; P = 0.35/0.17), HR (P = 0.078/0.66), or BP(P = 0.52/0.91) for either group of animals undergoingASFA.

Preangiographic imagingCamera settings employed based on preliminary data (notshown) included a shutter speed (1/30), effective aperture (f/8), and ISO of 800. Representative standard and red-freeimages of blue-eyed and heterochromic dogs are depicted inFigs 2–5(a,b). Control photographs obtained prior to theinjection of sodium fluorescein failed to demonstrate evi-dence of pseudofluorescence in the dogs evaluated.

Angiographic imaging and phasesAngiograms of the canine iris vasculature were noted to beconsiderably variable in their appearance(s). The detailobserved was dependent on the degree of iridal pigmenta-tion, demonstrating an inverse relationship. Poorly pig-mented irises (e.g. blue) allowed adequate visibility ofvasculature, while moderate-to-heavily pigmented irises(e.g. golden to brown) failed to clearly demonstrate thepresence of iris vessels. Variably pigmented irises (e.g.heterochromic) demonstrated focal blockade of fluorescencein the region(s) of increased pigmentation. Representativeangiograms demonstrating these effects are shown inFigs 2–5(c–f). Additionally, videos S1 and S2 are provided asa better means to assess/demonstrate vascular flow patterns.Angiographic phases were divided into arterial, capillary,and venous phases. A summary of the mean time, along withthe standard deviation (SD), at which each phase was notedis presented in Table 1.

Arterial phaseThe onset of the arterial phase was noted to occur on aver-age 11.0 (butorphanol) and 7.2 (propofol) s following injec-tion. This phase was characterized by the initial filling ofterminal branches from the long posterior ciliary arterieslocated at the medial and lateral iris base. These vesselscoursed both superiorly and inferiorly, forming the MAC(Figs 2c and 3c). Their location within the iris base was vari-able and in some, despite lack of significant iris pigmenta-

Figure 1. Anterior segment fluorescein angiography (ASFA) image obtained during the venous phase [16 s] using butorphanol from a blue-eyed 7-

year-old SF Siberian Husky dog. Color and black and white* images are represented. Note, poor focus and detail of the ciliary region resulting from

slight axial globe movement during imaging. *Black and white images were generated from color images obtained using Adobe Photoshop CS4 black

and white tool function.

a n t e r i o r s e g m e n t f l u o r e s c e i n a n g i o g r a p h y i n d o g s 3


tion, was not visualized (Fig. 4c). Complete filling of theMAC was rapid, often occurring within 1 s of initial detect-able vascular fluorescence. Communication between theneighboring superior and/or inferior branches was incom-plete in most; however, in one animal, anastomoses werenoted forming a complete circle (Fig. 2c). Shortly uponentering the MAC, fluorescence of both radial ciliary arter-ies and radial iris arterioles was noted. Radial ciliary arteriesbranched directly from the posterior aspect of the MAC andwere of a larger caliber as compared to radial iris arterioles.These arteries demonstrated a straight course traveling pos-teriorly and often bifurcated immediately upon leaving theMAC (Figs 2c and 3c). Of note, however, radial ciliary arter-ies were not always visible and, if present, were variable intheir number and location. Radial iris arterioles were notedto originate from the anterior aspect of the MAC and/orradial ciliary arteries. These vessels demonstrated a straight(Fig. 4c) to slightly tortuous course (Fig. 3c) with their cali-ber tapering as they progressed toward the pupillary border.Numerous collateral vessels often occurring at right angleswere noted, connecting neighboring arterioles to oneanother (Fig. 6). In some animals, at the level of the collar-ette, several of these vessels were noted to merge with neigh-boring vessels, forming a rudimentary/incomplete minorarterial circle (Fig. 3c–e). Numerous bifurcations were alsoobserved, forming terminal dichotomous branches at thepupillary border (Fig. 6a). Upon reaching the pupillary bor-der, radial iris arterioles terminated into fine capillaries.

Capillary phaseThe onset of the capillary phase was noted approximately14.2 (butorphanol) and 13.0 (propofol) s following injectionand was short lived. This phase was characterized by rapidfilling of the capillaries located at the pupillary border(Figs 2–5d). In most animals, capillaries were noted to forman abrupt 180� bend at the pupillary edge continuing intothe deeper iridal veins (Fig. 6b). In one animal, however, alattice of fine vessels was noted forming a peri-pupillary net-work (Fig. 4d).

Venous phaseThe onset of the venous phase was noted to occur on average15.2 (butorphanol) and 14.0 (propofol) s following injec-tion. This phase was characterized by progressive and com-plete filling of iridal veins that demonstrated radial dyemovement toward the iris base (Figs 2–5e). This phase

occurred shortly after and, in some cases, almost simulta-neously with the onset of the capillary phase. The size, pat-tern, and distribution of the iridal veins closely mimickedthose of the arterial system, making a clear distinction some-times difficult. As such, tracing of flow patterns was oftenrequired to differentiate between venous and arterial sys-tem(s) (Fig. 6 and Videos S1 and S2). Iridal veins oftenappeared deeper and parallel to their neighboring arteries;however, some veins appeared to twist around their neigh-boring artery, often becoming superficial. No lamination ofthe vessel wall was noted during this phase.

Vessel leakage was noted in several animals (4/12) duringthe venous phase of the angiogram, which often originatedfrom the peripheral iridal venules (Figs 4f and 7). Leakageprogressed slowly throughout the remainder of the imagingsequences. No late (recirculation) phase was noted in thecurrent study. Progressive dye leakage within the iris stromaoccurred several minutes postinjection, becoming completeat the time of final image acquisition, which resulted in anegative contrast image (Figs 2f, 3f, 5f). Dye leakage into theaqueous humor occurred in the latter time periods in someanimals (3/12) and appeared to originate from within thepupillary opening (Figs 2f and 5f).

DISCUSSION

The results of this study demonstrate that high quality, diag-nostic ASFA images of the canine globe can be obtained usingthe equipment and procedures described in this paper. To theauthors’ knowledge, there are no previous reports describingASFA involving the iris vasculature of the canine eye.

In this study, two sedation/anesthesia protocols were per-formed and evaluated to determine which would consistentlyprovide the highest quality angiograms, while being practi-cal in the clinical setting. While both protocols enabledASFA images to be collected, image sequences obtained withpropofol were more consistent and subjectively superior inimage quality. This was simply because of complete controlof all ocular movements, thereby allowing proper centrationand alignment of the iris plane to that of the camera sensor.This superior control is of great importance while imagingat higher magnifications (e.g. 1:1 or greater) as in this currentreport. While canines receiving butorphanol were consid-ered sedate enough for proper positioning, minor rotationalmovements of the globe were present and problematic.These movements translated in displacement of the iris

Table 1. Age and number of animals enrolled in the study, followed by timing of the arterial, capillary, and venous phases. Timing data listed include

time of onset after intravenous fluorescein injection for each phase and duration of time from the beginning of each phase to the beginning of the next

phase (phase interval). Values are represented by their mean ± SD

ProtocolNumber ofanimals

Meanage (years)

Arterialphase onset (s)

Phaseinterval (s)

Capillaryphase onset (s)

Phaseinterval (s)

Venous phaseonset (s)

Propofol 6 4.8 ± 1.8 7.2 ± 1.1 5.6 ± 2.3 13.0 ± 2.9 1.0 ± 0 14.0 ± 2.9Butorphanol 6 4.8 ± 3.0 11.0 ± 6.7 3.2 ± 0.9 14.2 ± 5.9 1.0 ± 0 15.2 ± 6.0



plane from the plane of the camera sensor, often resulting inseveral frames within the image sequence being partially outof focus. Additionally, based on a previous report document-ing emesis in 15% of canines following intravenous sodiumfluorescein, maropitant citrate was administered prior toinjection using both protocols.15 However, no adverseevents were noted in the current report.

Vascular patterns observed in this study were consistentwith previous reports examining the canine anterior segmentvia corrosion cast techniques.10,11 Some minor differences

were noted, however. In one dog, the MAC was observed tobe complete with ASFA (Fig. 2c). Whether this finding rep-resents a truly complete MAC or is evidence of smaller con-necting vessels as previously described cannot be definitivelydetermined.11

Mean time values of angiographic phases and their flowpatterns, using either protocol, were comparable with thosedescribed in humans.1,2 Despite these similarities regardingthe temporal phases, several differences including vesselleakage were noted (Figs 4f and 7).

(a)

(d)

(e)

(f)

(c)

(b)

Figure 2. Standard (a), red-free (b), and anterior

segment fluorescein angiography (ASFA) (c–f)

images obtained using propofol from a blue-eyed 3-

year-old SF mixed-breed dog. For ASFA images,

phases include: (c) arterial phase [12 s], (d) capillary

phase [17 s], (e) venous phase [20 s], and (f) late time

period [10 min]. Color and black and white* ASFA

images are represented. Bifurcations of the radial cil-

iary arteries forming a V-like pattern are readily

apparent in (c), in addition to, the presence of a com-

plete major arterial circle (MAC). In (f) leakage of

fluorescein within the iris stroma is apparent and

produces a negative contrast image (MAC is now

black). Furthermore, aqueous humor leakage is

demonstrated and noted to originate from within

the pupillary opening. *Black and white images were

generated from color images obtained using Adobe

Photoshop CS4 black and white tool function.



While considered to be a normal finding in older humanpatients (>50 years of age), vessel leakage is considered tobe pathologic in younger patients.1,2 When present, leak-age is most commonly seen within the peri-pupillary areaas small tufts, which begin to appear in the capillary and/or early venous phase of the angiogram. Under pathologi-cal conditions, this finding may be associated with systemicand/or local disease(s) processes. Hypoxia is reported to bea common cause and may be associated with arteriosclero-sis, chronic respiratory disease, microangiopathies, and, in

some cases, deep planes of anesthesia.1 As animals imagedin the current study were young and considered to be freeof ocular and/or systemic disease, vessel leakage may rep-resent a normal finding of ASFA in dogs. Alternatively,this may be a result of anesthesia; however, we do notbelieve this to be a cause as vessel leakage was noted inboth sedated and anesthetized dogs. In this current report,fluorescein leakage was also noted within the aqueoushumor during the late time periods of the angiogram; afinding noted in humans as well.1,2 This leakage is believed

(a)

(d)

(e)

(f)

(c)

(b)


segment fluorescein angiography (ASFA) (c–f) images

obtained using propofol from a blue-eyed 3-year-old

MN Siberian Husky dog. For ASFA images, phases

include: (c) arterial phase [8 s], (d) capillary phase

[10 s], (e) venous phase [13 s], and (f) late time period

[10 min]. Color and black and white* ASFA images

are represented. Presence of an incomplete minor

arterial circle is first noted in (c), becoming more

apparent in (d) and (e). Fluorescein leakage within

the iris stroma is apparent in (f) creating a negative

contrast image. *Black and white images were gener-

ated from color images obtained using Adobe Photo-

shop CS4 black and white tool function.



to originate from the ciliary body and was supported byobserving dye migration through the pupillary opening,prior to complete filling of the anterior chamber1 (Fig. 2f).Leakage occurred during the late stages of the angiogram,after the 5 min time point, which is consistent with reportswithin the physician-based literature.1 Alternatively, aque-ous humor leakage may be the result of changes involvingthe posterior pigmented iris epithelium, a finding noted indogs less than 2 years of age. These findings ranged fromstripping to hyperplasia of the posterior iris epithelium

and may be linked to disruption of the blood aqueousbarrier.16

In this current report, diagnostic quality images couldonly be obtained from regions within the canine iris lackingsignificant pigmentation. This was supported by failure toadequately identify iridal vessels within moderately pig-mented (light brown) and regionally pigmented irises (het-erochromic). In the case of the latter, vessels could only bevisualized in regions lacking significant pigmentation(Figs 4c–f and 5c–f). These results are comparable with

(a)

(d)

(e)

(f)

(c)

(b)



images obtained using butorphanol from a hetero-

chromic 6-year-old MN Australian Shepherd dog.

For ASFA images, phases include: (c) arterial phase

[10 s], (d) capillary phase [12 s], (e) venous phase

[14 s], and (f) late venous phase [22 s]. Color and

black and white* ASFA images are represented.

Regional pigmentation located within the dorsolat-

eral iris is apparent in (a) and (b), resulting in focal

blockage of fluorescence (hypofluorescence)

throughout (c–f). Note, lack of the MAC and pres-

ence of straight radial iris arterioles in (c). A fine

peri-pupillary capillary network is noted in (d). Ves-

sel leakage within the ciliary region is apparent in (f).

*Black and white images were generated from color

images obtained using Adobe Photoshop CS4 black




those observed in humans.1–3,12 As such, ASFA appears to beonly possible in poorly pigmented (blue irises) or hetero-chromic irises, where pigment is lacking. Despite this limita-tion, however, ASFA may be of value when evaluating anumber of disease states regardless of iridal pigmentation.For example, in diseases that result in neovascularizationand pre-iridal fibrovascular membrane (PIFM) formation,the ability to directly visualize abnormal vessel formationand/or vessel leakage would be of great significance. Such anobservation may allow the clinician to diagnose either of theaforementioned processes earlier and initiate appropriate

treatment options. Further studies are required to documentthese hypotheses.

This study provides preliminary evidence demonstratingthat ASFA in dogs is possible using the equipment setup andprocedure described herein. These results may serve as abaseline of normal ASFA findings in the canine eye, particu-larly in blue-eyed dogs. Currently, ASFA is an important,yet limited diagnostic imaging modality utilized in physi-cian-based ophthalmology and may be utilized to diagnose anumber of conditions including diabetic microangiopathy,glaucoma, uveitis, neoplastic conditions, trauma, and devel-

(a)

(d)

(e)

(f)

(c)

(b)



images obtained using propofol from a heterochro-

mic 1-year-old SF Australian Shepherd dog. For

ASFA images, phases include: (c) arterial phase

[11 s], (d) capillary phase [15 s], (e) venous phase

[20 s], and (f) late stage [10 min]. Color and black

and white* ASFA images are represented. Sectoral

hypopigmentation located within the ventromedial

iris is apparent in (a) and (b). Fluorescence and the

presence of iris vasculature are apparent only within

this hypopigmented region (c–f). Fluorescein leak-

age within the iris stroma is apparent in (f) creating a

negative contrast image. Additionally, aqueous

humor leakage creating a vertical line is noted.

*Black and white images were generated from color

images obtained using Adobe Photoshop CS4 black




opmental disorders.1,2,5–7 Likewise, ASFA may prove to beuseful in veterinary medicine. Specifically, this techniquemay allow for early detection and/or characterization of uve-itis, PIFM formation, glaucoma, and various neoplasias indogs. Additionally, the protocol and equipment used in thisstudy (excluding the novel adaptor) are widely available andcan easily be performed in a clinical setting. This may allowfor greater availability of a currently underutilized diagnos-tic technique.

Limitations of the current study primarily include thesmall population of normal dogs evaluated. However, as thisstudy was conducted to first evaluate the ease and feasibilityof performing ASFA in canine eyes and to determine an

appropriate sedative/anesthetic protocol, the number of ani-mals used was limited. Future studies and a larger samplesize are likely required to more accurately determine normaltimes frames of the temporal phases; however, this has notproven useful within the physician-based literature.1

In summary, diagnostic quality ASFA images were obtain-able using the equipment setup described herein. Anteriorsegment fluorescein angiography appears to be a useful diag-nostic imaging technique in the canine eye. The detailobserved in a normal angiogram of the canine iris is, how-ever, dependent on the degree of iridal pigmentation. Properpatient positioning and restricted ocular movement wereconsidered imperative for obtaining quality images.

(a)

(b)

(c)

Figure 6. Anterior segment fluorescein angiogra-

phy images obtained using propofol from a blue-

eyed 11-year-old MN Australian Shepherd dog.

Images are magnified by a factor of 4 (400%) to

demonstrate the finer vascular detail around the

pupillary border. Phases include: (a) arterial phase

[11 s], (b) capillary phase [12 s], and (c) venous phase

[13 s]. Color and black and white* ASFA images are

represented. Bifurcation of the radial iris arterioles

in a Y-like pattern are noted in (a). Fine capillaries

are noted in (b), which form an abrupt 180� bend at

the pupillary edge and continue into deeper iridal

venules in (c). Numerous collaterals between neigh-

boring arterioles and venules are also noted. *Black

and white images were generated from color images

obtained using Adobe Photoshop CS4 black and

white tool function.

Figure 7. Anterior segment fluorescein angiography image obtained using propofol during the late phase(s) [1 min] from a blue-eyed 4-year-old SF

mixed-breed dog. Color and black and white* images are represented. Note, the presence of a rudimentary vascular circle at the collarette, in addition

to, vessel leakage primarily within the ciliary region and to a lesser degree the collarette and pupillary border. *Black and white images were generated

from color images obtained using Adobe Photoshop CS4 black and white tool function.



SUPPORTING INFORMATION

Additional Supporting Information may be found in theonline version of this article:

VIDEO S1. Representative time-elapsed video demon-strating vascular flow patterns of a blue eyed 3 yr old SFmixed breed dog. (Same eye depicted in figure 2) Progressiveradial dye movement towards (arterial phase) and away(venous phase) from the pupillary border is clearly demon-strated. Additionally, note dye leakage within the anteriorchamber which emanates from the pupillary opening. Videowas created using still photographs obtained during ASFAimaging and Windows Movie Maker.

VIDEO S2. Representative time-elapsed video demon-strating vascular flow patterns of a 1 yr old SF AustralianShepherd dog with heterochromia. (Same eye depicted infigure 5) Note significant impairment of fluorescence is pre-sent as a result of iridal pigmentation. Progressive radial dyemovement towards (arterial phase) and away (venous phase)from the pupillary border is clearly demonstrated within theregion of hypo-pigmentation. Video was created using stillphotographs obtained during ASFA imaging and WindowsMovie Maker.

Please note: Wiley-Blackwell Publishing are not respon-sible for the content or functionality of any supporting mate-rials supplied by the authors. Any queries (other thanmissing material) should be directed to the correspondingauthor for the article.

REFERENCES

1. Kottow M. Anterior Segment Fluorescein Angiography. Williams and

Wilkins, Baltimore, 1978.

2. Brancato R, Bandello F, Lattanzio R. Iris fluorescein angiography

in clinical practice. Survey of Ophthalmology 1997; 42: 41–

70.

3. Mitsui Y, Matsubara M, Kanagawa M. Fluorescence irido-corneal

photography. The British Journal of Ophthalmology 1969; 53:

505–512.

4. Mapstone R. Fluorescein iridography. The British Journal of Oph-

thalmology 1971; 55: 400–406.

5. Dart JK, Marsh RJ, Garner A et al. Fluorescein angiography of

anterior uveal melanocytic tumours. The British Journal of Oph-thalmology 1988; 72: 326–337.

6. Marigo FA, Finger PT. Anterior segment tumors: current con-

cepts and innovations. Survey of Ophthalmology 2003; 48: 569–593.

7. Sanborn GE, Symes DJ, Magargal LE. Fundus-iris fluorescein

angiography: evaluation of its use in the diagnosis of rubeosis iri-

dis. Annals of Ophthalmology 1986; 18: 52–58.

8. Schantz B. Fluorescein angiography of canine conjunctival auto-

and allografts. Bibliotheca Anatomica 1977; 16 (Pt 2): 269–273.

9. Schantz B. Fluorescein angiography of canine conjunctival auto-

and allografts. Analysis by sequential, still photographic recording

of the spreading of fluorescence over the graft surfaces following

intravenously administered fluorescein. Acta Chirurgica Scandinav-ica 1975; 141: 575–581.

10. Anderson BG, Anderson WD. Vasculature of the equine and canine

iris. American Journal of Veterinary Research 1977; 38: 1791–1799.

11. Sharpnack DD, Wyman M, Anderson BG et al. Vascular path-

ways of the anterior segment of the canine eye. American Journal

of Veterinary Research 1984; 45: 7, 1287–1294. 19 ref.12. Fetkenhour CL, Chromokos E. Anterior segment fluorescein

angiography with a retinal fundus camera. Archives of Ophthalmol-

ogy 1978; 96: 711–713.

13. Pirie CG, Pizzirani S. Anterior and posterior segment photo-

graphy. An alternative approach using a dSLR camera adaptor.

Veterinary Ophthalmology 2011; doi:10.1111/j.14635224.2011.

00963.x, Early View.

14. Pirie CG, Pizzirani S. Fluorescein angiography of the canine pos-

terior segment using a dSLR camera adaptor. Veterinary Ophthal-mology 2011; doi:10.1111/j.1463-5224.2012.00995.x, In Press.

15. Gelatt KN, Henderson JD, Steffen XX. Fluorescein angiography

of the normal and diseased ocular fundi of the laboratory dog.

Journal of the American Veterinary Medical Association 1976; 169:

980–984.

16. Pizzirani S, Desai SJ, Pirie CG et al. Age related changes in the

anterior segment of the eye in normal dogs (abstract). 41st Annual

Meeting of the American College of Veterinary Ophthalmologists 2010;

13: 407–423.



anterior segment fluorescein angiography of the normal canine eye using a dslr camera adaptor

Documents