Indocyanine Green Videoangiography of Choroidal Neovascularization MARYANNA DESTRO, MD, CARMEN A. PULIAFITO, MD

Abstract: Choroidal neovascular membranes often are poorly defined on fluorescein angiography because of rapid or indistinct fluorescein leakage or because of blockage of hyperfluorescence by overlying hemorrhage, lipid, turbid fluid, or pigment. Indocyanine green (ICG) is a highly protein-bound dye with peak absorption (805 nm) and peak fluorescence (835 nm) in the near infrared portion of the spectrum. At these wavelengths, penetration through overlying pigments is increased. Using an infrared videoangiography system, the authors obtained ICG angiograms of 32 eyes with suspected choroidal neovasculari­zation. Compared with fluorescein angiography, ICG improved visualization of the choroidal circulation and enhanced visualization of some membranes that were poorly defined with fluorescein. In addition, after clearance of the dye from the retinal and choroidal circulations, ICG remained in and around the neovascular tissue. The authors conclude that ICG videoangiography may aid in the evaluation of selected patients with poorly defined membranes on fluorescein angiography. Ophthalmology 96:846-853, 1989

The current treatment of choroidal neovascularization is dependent on adequately visualizing the neovascular membranes via fluorescein angiography. I Unfortunately, studies have shown that up to 50% of membranes may not be seen clearly using this technique,2 because of flu­orescein blockage by overlying blood, lipid, melanin, or turbid fluid, or because of rapid fluorescein leakage ob­scuring neovascular detail. In contrast to fluorescein (peak absorption, 465 nm; peak fluorescence, 525 nm), indo-

Originally received: October 3, 1988. Revision accepted: December 5, 1988.

From the Laser Research Laboratory and Retina Service, Massachusetts Eye and Ear Infirmary, and Department of Ophthalmology, Harvard Medical School, Boston.

Presented at the American Academy of Ophthalmology Annual Meeting, Las Vegas, October 1988.

Supported by National Eye Institute grant R44-EY05532, National Eye In­stitute fellowship award F32-EY06020, Office of Naval Research contract N0014-86K-0117, National Institutes of Health Contract 1-R01-GM35459, and a Heed Ophthalmic Foundation fellowship.

The authors have no proprietary interest in the device described.

Reprint requests to Carmen A. Puliafito, MD, Laser Research Laboratory, Massachusetts Eye and Ear Infirmary, 243 Charles St, Boston, MA 02114.

846

cyanine green (leG) is a tricarbocyanine dye with its peak absorption (805 nm) and peak fluorescence (835 nm) in the near infrared spectrum. At these longer wavelengths, there is increased visibility through overlying pigments. 3

In addition, leG is highly protein-bound (98% compared with between 60 and 80% for fluorescein). Because of this, leG diffuses slowly out of fenestrated small choroidal vessels, in marked contrast to the rapid leakage of fluo­rescein.4

Indocyanine green is approved by the United States Food and Drug Administration for use in cardiac, hepatic, and ophthalmic studies. Ophthalmic angiography using leG was initially developed in the early 1970s.5-

33 How­ever, visualization of choroidal neovascularization proved disappointing and leG angiography failed to gain wide­spread use. Since then, however, significant advances in infrared imaging technology have occurred. These ad­vances include the use of an infrared video camera in addition to improved filters and illumination sys­tems.34

-39 Although the infrared leG videoangiography

system used in these studies has limited spatial resolution, its major advantages include significantly greater sensi­tivity than 35-mm infrared film and greater temporal res­olution than motorized shutter cameras. These properties are particularly useful for leG angiography because the fluorescence efficiency of leG is relatively poor (1/25th

DESTRO AND PUUAFITO • leG VIDEOANGIOGRAPHY

that of fluorescein) and because the initial transit of dye through the choroidal circulation is very rapid. We in­vestigated the possibility that ICG videoangiography could improve visualization of choroidal neovascular mem­branes that are not well seen with fluorescein angiography.

PATIENTS AND METHODS

PATIENT POPULATION

All patients were examined by us at the Massachusetts Eye and Ear Infirmary between December 1987 and July 1988. Patients were enrolled in this study after approval from our human studies committee and after informed written consent. Because ICG contains a small amount of iodine «5%), patients were excluded from this study if they had a known sensitivity to iodine or shellfish. Thirty-two eyes of 16 patients were studied. Seven patients were men and nine were women. Patients ranged in age from 24 to 79 years. All patients had symptoms of visual loss and clinical findings suggestive of choroidal neovas­cularization in at least one eye on initial presentation. No patient had a history of diabetic retinopathy or other ret­inal or choroidal disorder.

INDOCYANINE GREEN FUNDUS CAMERA AND VIDEO RECORDING SYSTEM

For ICG videoangiography, a Topcon fundus camera (TRC-W) (Paramus, NJ) was modified in the following manner: the xenon flash lamp was replaced with a con­tinuous 300 W halogen bulb with a variable setting con­trol. The barrier and exciter filters were replaced to trans­mit the peak absorption and fluorescence bands of ICG (peak absorption, 805 nm; peak fluorescence, 835 nm; filter transmittance overlap, 0.5%). The camera lenses were treated with near infrared antireflective coating, and the 35-mm camera back was replaced with a near infrared video camera (Topcon Corporation; spatial resolution, 350 X 450 lines; threshold sensitivity, 2 lux). Retinal ex­posures were limited to comply with the maximal per­missible exposure guidelines for near infrared light as rec­ommended by the American National Standards Institute (ANSI 1980).34 The video signal was interfaced with a video timer (timing interval, 0.01 second) and simulta­neously displayed on a high-resolution black and white monitor (Ikegami Corporation; spatial resolution, 800 lines). Images were recorded onto three-quarter-inch magnetic tape (Sony V-Matic 5800 Video Recording Sys­tem and Sony Broadcast Quality tape, recording speed: 30 frames/second), and hard copy slides and prints were obtained by direct photography from the monitor (Pan­atomic-X, 35-mm film, 32 ASA).

INDOCYANINE GREEN

Indocyanine green (Cardiogreen, Hynson, Wescott and Dunning, Inc, Baltimore, MD) was reconstituted with the manufacturer-supplied aqueous solvent to a concentration of 50 mg/ml. Angiography was performed using 3.0 mg/

kg of ICG injected into a peripheral arm vein and was followed immediately by a 5.0-ml flush of sterile saline via a three-way stopcock.

COLOR PHOTOGRAPHY AND FLUORESCEIN ANGIOGRAPHY

Color fundus photographs (Kodak Ektachrome 64 film) and fluorescein angiograms (Kodak Tri-X film) were ob­tained on the same day as the ICG videoangiograms. So­dium fluorescein 10% (3.0 to 5.0 ml) was injected into a peripheral arm vein in the standard fashion.

RESULTS

Thirty of 32 eyes studied had evidence of chorioretinal disease on color photographs, fluorescein angiograms, and ICG videoangiograms. In all, 25 eyes had findings con­sistent with age-related maculopathy, including drusen, atrophy or hypertrophy of the retinal pigment epithelium, and subretinal or intraretinal macular hemorrhage, exu­date, or serous fluid. Two eyes had findings consistent with presumed ocular histoplasmosis syndrome, including peripapillary atrophy, macular lesions, and punched-out lesions in the posterior pole and midperiphery. Two eyes had isolated choroidal neovascularization, and neovas­cularization had developed in one eye after traumatic choroidal rupture. Five eyes had been previously treated with laser photocoagulation. In all, 26 of 32 eyes displayed evidence of choroidal neovascularization on fluorescein and/or ICG angiography, including three eyes with recurrent or persistent neovascularization after laser treatment.

COMPARISON OF FLUORESCEIN ANGIOGRAPHY AND INDOCYANINE GREEN VIDEOANGIOGRAPHY

In all eyes, visualization of the choroid was greater with ICG angiography than with fluorescein angiography. This was due to (1) greater penetration of ICG fluorescence through macular xanthophyll and retinal pigment epi­thelium melanin and (2) decreased leakage ofICG out of the choriocapillaris (Fig 1). In addition, the high-speed videoangiography system allowed imaging of rapid cho­roidal filling, which was not easily captured with standard fluorescein angiography.

In general, spatial resolution of small retinal vessels and of well-defined neovascular membranes was greater with fluorescein angiography than with ICG angiography. In addition to higher spatial resolution, with fluorescein angiography, the hypofluorescent retinal pigment epithe­lium provided a high-contrast background against which fine capillaries could be seen (Fig 1). However, for poorly defined membranes, several advantages were noted with ICG compared with fluorescein.

Neovascular membranes for which ICG videoangiog­raphy was particularly helpful were those with overlying hemorrhage, or those that had recurred on the edge of previous treatment areas. For membranes with overlying

847

OPHTHALMOLOGY • JUNE 1989 • VOLUME 96 • NUMBER 6

Fig 1. Bilateral "idiopathic" choroidal neovascularization. Results of slit-lamp biomi­croscopy showed subretinal fluid overlying macular (top left) and peripapillary (top right) lesions. Fluorescein (second row) and leG (third row) angiograms of the right eye showed progressive hy­perfluorescence of the lesions consistent with choroidal neovascularization. Notice increased visualization of the choroidal circulation with leG compared with fluores­cein, with late pooling of dye in the neovascular mem­brane. Visualization of the choriocapillaris detail was suboptimal due to resolution limitations and masking by the underlying choroidal vessels. Notice also that the optic disc and the retinal pig­ment epithelium window de­fects temporal and inferior to the macula and adjacent to the disc appeared hyperflu­orescent on fluorescein an­giography but hypofluores­cent with leG. Fluorescein (bottom left) and leG (bot­tom right) angiograms of the left eye showed hyperflu­orescence of the peripapillary membrane. Notice the adja­cent retinal pigment epithe­lium window defect appeared hyperfluorescent with fluo­rescein but hypofluorescent with leG (arrow).

blood (Fig 2), there was greater penetration and therefore greater visibility of the neovascular tissue with ICG than with fluorescein. For those membranes adjacent to cho­rioretinal scars (Figs 3, 4), there was greater contrast be­tween the neovascularization and the adjacent scar with ICG angiography than with fluorescein. With ICG, cho­rioretinal scars were hypofluorescent in contrast to the hyperfluorescence of the neovascular tissue. This was due to the lack of extravasation of ICG and the relative lack

848

of choroidal vessels in those areas. In contrast, with flu­orescein, both the neovascular tissue and the adjacent chorioretinal scars displayed hyperfluorescent staining.

In addition, after clearing ofICG from the surrounding retinal and choroidal circulation, ICG selectively re­mained in and around the choroidal neovascular mem­branes (Figs 1-4).

In general, patients found the near-infrared illumina­tion ofICG videoangiography more comfortable than that

DESTRO AND PULlAFITO • leG VIDEOANGIOGRAPHY

Fig 2. Suspected choroidal neovascularization in a pa­tient with sudden visual loss in the right eye and a history of drusen and disciform scar in the left eye. Notice large subretinal hemorrhage in­volving the fovea (top left). Fluorescein angiography (second row) showed a "poorly defined" neovascular membrane with significant blockage of fluorescence by overlying blood. Indocyanine greenllngiography (third row) showed increased visualiza­tion of the Choroidal vessels and decreased blockage by blood. Although visualiza­tion through the blood was not complete, the area of neovascularization appeared to primarily involve the in­ferior macula (arrow). Three months later (top right), the blood had resorbed; however, the area of suspected neovas­cularization had enlarged, with subretinal fluid extend­ing throughout the macula. Fluorescein angiography (fourth row) and leG an­giography (bottom) showed a large area of subfoveal hy­perfluorescence.

849

OPHTHALMOLOGY • JUNE 1989 • VOLUME 96 • NUMBER 6

Fig 3. Suspected recurrence of neovascularization after laser photocoagulation. Top left. an acute posttreatment photograph of a choroidal neovascular membrane tem­poral to the fovea. On foUow­up examination (top right), there was evidence of subret­inal fluid nasal to the treat­ment area and involving the center of the fovea. Fluores­cein angiography (center) showed hyperfluorescent staining of the chorioretinal scar with the suggestion of a "poorly defined" neovascular membrane involving the fovea. In contrast, the early phase leG angiogram (bot­tom left) showed a suspicious tuft of choroidal vessels (ar­rows) nasal to the hypoflu­orescent chorioretinal scar with progressive hyperflu­orescence in the late phase angiogram (bottom right).

with fluorescein angiography and no patients experienced nausea or other adverse or allergic effects from ICG. In addition, because ICG does not rapidly extravasate from the circulation and is rapidly cleared by the liver, there was no discloration of the urine, skin, or mucous mem­branes in contrast with fluorescein.

DISCUSSION

Multicentered prospective randomized trials such as the Macular Photocoagulation Study have shown that la­ser photocoagulation may reduce the risk of severe visual loss in some eyes with choroidal neovascular mem­branes.35-37 The benefits of such treatment, however, have been clearly documented only for membranes that are extrafoveal and well defined on fluorescein angiography. Clear definition implies that the entire extent of the neo­vascular membrane is well visualized.I .35-37

Unfortunately, many membranes are poorly defined on fluorescein angiography, and attempted photocoagu-

850

lation of such occult or poorly defined membranes may be incomplete.

Limitations in visualizing choroidal vessels with fluo­rescein led to the development of ICG choroidal angiog­raphy in the early 1970s.5-33 Kogure et al30.31 in 1969 and 1970 used ICG absorption angiography to study the pial circulation in dogs. After this, ICG was injected into the carotid circulation of both animals and patients under­going carotid surgical procedures and for the first time, the larger choroidal veins were able to be routinely vi­sualized. Several technical advances, including the use of improved filters and the use of black and white infrared film instead of color film, resulted in successful angiog­raphy after intravenous injection, and with this technique, the early clinical studies began through the work of Flower and HochheimerlO.13.15.16 and Orth et al. 11 There were a number of technical difficulties with this absorption tech­nique, however, and the same group subsequently altered the technique from ICG absorption angiography to ICG fluorescence angiography. This resulted in improved vi­sualization of medium-sized choroidal veins and arteries, although it still was not possible to routinely visualize the

DESTRO AND PULIAFITO • leG VIDEOANGIOGRAPHY

choriocapillaris in humans. This technique involved the use of a modified Zeiss camera (Carl Zeiss, Inc, Thorn­wood, NY), a motorized shutter and movie camera and infrared-sensitive 35-mm black-and-white film. Using this technique, several hundred angiograms were performed in both normal volunteers and in patients with chorio­retinal disorders, and the preliminary results of these studies were reported by Patz and Orth in 1976.11

•25 Un­

fortunately, this technique never gained widespread clin­ical acceptance, either for the evaluation of choroidal neovascularization or for other choroidal disorders. In re­view of those angiograms, Patz found that leG angiog­raphy improved the visualization of choroidal neovas­cularization over fluorescein angiography in only 2 of 25 patients with suspected membranes. In addition, it was not possible to view the fundus during the time of the angiogram, and despite the use of a movie camera (up to 20 frames/second), it was not always possible to capture the very rapid transit of dye through the choroidal cir­culation.

Subsequently, Hayashi et a138-

42 developed a technique of leG videoangiography and found that this not only

Fig 4. Suspected recurrence of choroidal neovasculariza­tion after laser photocoagu­lation. Initially (top left), there was no evidence of residual neovascularization on slit­lamp biomicroscopy or flu­orescein angiography. How­ever, I year later (top right) there was evidence of new subretinal blood and fluid nasal to the chorioretinal scar. Fluorescein angiogra­phy (center) showed progres­sive staining of the chorio­retinal scar without clear evidence of neovascular re­currence. Indocyanine green angiography (bottom), how­ever, showed a hypofluores­cent scar with hyperfluoresc­ence nasally in the region of suspected neovasculariza­tion.

improved the sensitivity of the angiograms but also im­proved temporal resolution. In addition, Hayashi et al improved the illumination system and separation between the excitation and barrier filters, cutting the overlap trans­mission to 0.5%.

In our study, we report the first clinical investigation of choroidal neovascularization using leG videoangiog­raphy. In addition, we further modified the leG videoan­giography technique by using a higher dose of dye (3 mg! kg compared with the 0.5 mg/kg used in previous studies). This dose was found to be optimal for visualizing neo­vascular membranes in our earlier animal studies, is less than the dose used in some clinical hepatic studies, and appears to be well tolerated in humans.43

•44 Using this

technique, we found that leG videoangiography improved visualization of some neovascular membranes over con­ventional fluorescein angiography because of (1) higher intravascular retention and greater transmission of leG fluorescence through overlying blood, melanin, and ex­udate and (2) chorioretinal scars appeared hypofluorescent with leG in contrast with the hyperfluorescent staining seen with fluorescein, improving the ability to distinguish

851

OPHTHALMOLOGY • JUNE 1989 • VOLUME 96 • NUMBER 6

neovascular recurrence on the edge of treatment scars. In addition, after clearance from the surrounding circulation, leG selectively remained in and around the choroidal neovascular tissue.

Although the resolution limitations of this technique still do not equal that of standard fluorescein angiography and although leG will not replace the use of fluorescein, the information that leG can add to fluorescein angiog­raphy may possibly improve the detection and subsequent treatment of some poorly defined neovascular membranes that are currently considered untreatable. In addition, ICG may possibly provide a dye-enhanced target for infrared laser photocoagulation, a technique that may possibly improve selective neovascular closure.45 Furthermore, leG videoangiography may provide useful quantitative data concerning normal and abnormal choroidal hemo­dynamics, as shown in a recent study by Priinte and Nie­sel. 46 Further refinements in video technology, including the use of newer high-resolution video cameras, laser disc recording devices, real time digital acquisition systems, infrared scanning laser ophthalmoscopes, and comput­erized image analysis, may continue to improve the qual­ity and utility of this technique.47

REFERENCES

1. Bressler NM, Bressler SB, Fine SL. Age-related macular degeneration. Surv Ophthalmol 1988; 32:375-413.

2. Bressler NM, Bressler SB, Gragoudas ES. Clinical characteristics of choroidal neovascular membranes. Arch Ophthalmol1987; 105:209-13.

3. Geeraets WJ, Berry ER. Ocular spectral characteristics as related to hazards from lasers and other light sources. Am J Ophthalmoll968; 66:15-20.

4. Bischoff PM, Flower RW. Ten years experience with choroidal an­giography using indocyanine green dye: a new routine examination or an epilogue? Doc Ophthalmol1985; 60:235-91.

5. Hyviirinen L, Flower RW. lndocyanine green fluorescein angiography. Acta Ophthalmoll980; 58:528-38.

6. Forsius H, Hyviirinen L, Nieminen H, Flower R. Fluorescein and in­docyanine green fluorescence angiography in study of affected males and in female carriers with choroideremia: a preliminary report. Acta OphthalmoI1977; 55:459-70.

7. Chopdar A, Turk AM, Hill OW. Fluorescent infra-red angiography of the fundus oculi using indocyanine green dye. Trans Ophthaimol Soc UK 1978; 98:142-6.

8. Bonnet M, Habozit F, Magnard G. Valeur de I'angiographie en infra­rouge au vert d'indocyanine <;Ians Ie diagnostic clinique des angiomes de la chordide (observation anatomo-clinique). Bull Soc Ophthalmol Fr 1976; 76:713-6.

9. Craandijk A, Van Seek CA. Indocyanine green fluorescence angiog­raphy of the choroid. Br J Ophthalmol1976; 60:377-85.

10. Flower RW, Hochheimer BF. Indocyanine green dye fluorescence and infrared absorption choroidal angiography performed simulta­neously with fluorescein angiography. Johns Hopkins Med J 1976; 138:33-42.

11 . Orth OH, Patz A, Flower RW. Potential clinical applications of indo­cyanine green choroidal angiography-preliminary report. Eye Ear Nose Throat Men 1976; 55:15-28, 58.

12. Hill OW, Young S. Infrared angiography of the cat fundus oculi. Arch Ophthalmol1975; 93:131-3.

852

13. Flower RW, Hochheimer BF. Infrared fundus angiography [Letter]. Br J Ophthalmol 1974; 58:635.

14. Brown N, Straney R. Infrared fundus angiography. Br J Ophthalmol 1973; 57:797-802.

15. Flower RW. Injection technique for indocyanine green and sodium fluorescein dye angiography of the eye. Invest Ophthalmol1973; 12: 881-95.

16. Hochheimer BF. Angiography of the retina with indocyanine green. Arch Ophthalmol1971; 86:564-5.

17. Kulvin S, Stauffer L, Kogure K, David NJ. Fundus angiography in man by intracarotid administration of dye. South Med J 1970; 63:998-1000.

18. Burchell HB. Symposium on indocyanine green and its clinical appli­cations: introduction. Proc Mayo Clin 1960; 35:729-32.

19. Fox IJ, Wood EH. Indocyanine green: physical and physiologic prop­erties. Proc Mayo Clin 1960; 35:732-44.

20. Edwards AWT, Bassingthwaite JB, Sutterer WF, Wood EH. Blood level of indocyanine green in the dog during multiple dye curves and its effect on instrument calibration_ Proc Mayo Clin 1960; 35:745-51_

21 . Hunton DB, Bollman JL, Hoffman HN N. Hepatic removal of indocyanine green. Proc Mayo Clin 1960; 35:752-55.

22. Amorim OS, Weidman WH, Wood EH. Use of indicator-dilution curves for selection of site for injection of contrast medium for selective an­giography. Proc Mayo Clin 1960; 35:756-63.

23. Sinclair JO, Sutterer WF, Wolford JL, et aI . Problems in comparison of dye-dilution curves with densitometric variations at the same site on the circulation measured from simultaneous cineangiograms. Proc Mayo Clin 1960; 35:765-73.

24. Marshall HW, Fox JJ, Rodich FS, Wood EH. Method for simultaneous measurement of total and lower-body blood flow using indicator-dilution technics. Proc Mayo Clin 1960; 35:774-82.

25. Patz A, Rower RW, Klein ML, et aI. Clinical application of indocyanine green angiography. In: deLaey JJ. International Symposium on Flu­orescein Angiography: Ghent, 28 March-1 April, 1976. The Hague: Dr. W. Junk, 1976; 245-51 (Doc Ophthalmol Proc Ser; 9).

26. Flower RW. High speed human choroidal angiography using indo­cyanine green dye and a continuous light source. In: deLaey JJ. In­ternational Symposium on Ruorescein Angiography: Ghent, 28 March-1 April, 1976. The Hague: Dr. W. Junk, 1976; 59-66. (DocOphthaimol

Proc Ser; 9). 27. Flower RW, Hochheimer BF. Quantification of indicator dye concen­

tration in ocular blood vessels. Exp Eye Res 1977; 25:103-11. 28. Hochheimer BF, O'Anna SA. Angiography with new dyes. Exp Eye

Res 1978; 27:1-16. 29. Flower RW. Choroidal fluorescent dye filling patterns: a comparison

of high speed indocyanine green and fluorescein angiograms. Int Ophthalmol1980; 2:143-9.

30. Kogure K, Choromokos E. Infrared absorption angiography. J Appl PhysioI1969; 26:154-7.

31 . Kogure K, David NJ, Yamanouchi U, Choromokos E. Infrared ab­sorption angiography of the fundus circulation. Arch Ophthalmol1970; 83:209-14.

32. Patz A. Clinical and experimental studies on retinal neovascularization. XXXIX Edward Jackson Memorial Lecture. Am J Ophthalmol 1982; 94:715-43.

33. Quentel G, Coscas G. Angiographie en fluorescene infrarouge au vert d 'indocyanine. Bull Soc Ophthalmol Fr 1984; 84:559-63.

34. The Laser Institute of America: American National Standard for the safe use of lasers: ANSI Z136.1-1986. Toledo, OH: The Institute, 1986. (cartridge 1007, frame 1894-943).

35. Macular Photocoagulation Study Group. Argon laser photocoagulation for neovascular maculopathy: three-year results from randomized clinical trials. Arch Ophthalmoll986; 104:694-701 .

36. Macular Photocoagulation Study Group. Argon laser photocoagulation for senile macular degeneration: results of a randomized clinical trial. Arch Ophthalmol1982; 100:912-8.

DESTRO AND PUUAFITO • ICG VIDEOANGIOGRAPHY

37. Macular Photocoagulation Study Group. Recurrent choroidal neovas­

cularization after argon laser photocoagulation for neovascular mac­

ulopathy. Arch Ophthalmol1986; 104:503-12.

38. Hayashi K, Okuyama H, Nakase Y, et aI. Indocyanine green fluorescein

angiography. Report 1. Fundamental studies. Nippon Ganka Gakkai

Zasshi. 1981; 85: 1 028-35.

39. Hayashi K, Ii M, Nakase y, et al. Indocyanine green fluorescein an­

giography. Report 2. Studies on new interference filters. Nippon Ganka

Gakkai Zasshi 1982; 86:1532-9.

40. Hayashi K, delaey JJ. Indocyanine green angiography of choroidal

neovascular membranes. Ophthalmologica 1985; 190:30-9.

41. Hayashi K, delaey JJ. Indocyanine green angiography of submacular

choroidal vessels in the human eye. Ophthalmologica 1985; 190:20-9.

42. Hayashi K, Hasegawa Y, Tokoro T. Indocyanine green angiography

of central serous chorioretinopathy. Int Ophthalmol 1986; 9:37-41.

43. Destro M, Puliafito CA, Dobi E. Indocyanine green videoangiography

of choroidal neovascularization. ARVO Abstracts. Invest Ophthalmol

Vis Sci 1988; 29(Suppl):339.

44. Leevy CM, Smith F, Longueville J, et al. Indocyanine green clearance

as a test for hepatic function: evaluation by dichromatic ear densi­

tometry. JAMA 1967; 200:236-40.

45. Puliafito CA, Destro M, To K, Dobi E. Dye enhanced photocoagulation

of choroidal neovascularization. ARVO Abstracts. Invest Ophthalmol

Vis Sci 1988; 29(Suppl):414.

46. Prilnte C, Niesel P. Quantification of choroidal blood-flow parameters

using indocyanine green Video-fluorescence angiography and statis­

tical picture analysis. Graefes Arch Clin Exp Ophthalmol 1986; 226:

55-8.

47. Bursel! S, Mainster MA. Methods of vitreoretinal evaluation. In: New­

some DA, ed. Retinal Dystrophies and Degenerations. New York:

Raven Press; 1986: 5-20.

853