Abstract
purpose. To develop a new method with which to visualize leukocytes moving
through the choroidal vessels of pigmented animals and enable the
evaluation of leukocyte dynamics in the choroidal microcirculation.
methods. Pigmented rabbits and monkeys were used in this study. Leukocytes,
collected by centrifugal separation of autologous blood, were stained
with indocyanine green (ICG) dye. The ICG-stained leukocyte fluid was
injected into the vein, and the fundus image was obtained with a
scanning laser ophthalmoscope. The image was recorded on videotapes and
analyzed with a personal computer-based image analysis system.
results. In pigmented rabbits, fluorescent leukocytes moving in the choroidal
circulation were clearly visible for more than 1 hour. In monkeys,
distinct fluorescent dots were seen moving approximately 50 to 200 μm
in the foveal avascular zone for more than 30 minutes after the
injection of the ICG-stained leukocyte fluid. Dim fluorescent dots were
seen moving in the fundus. Although the movement of these dim dots was
difficult to trace, they seemed to be moving in the choroidal vessels.
In the rabbits, the mean flow velocity of leukocytes moving without
plugging was 0.48 ± 0.14 mm/sec in the peripheral
choriocapillaris. In the monkeys, the mean flow velocity of distinct
fluorescent leukocytes without plugging was 2.45 ± 0.48 mm/sec in
the posterior choroid.
conclusions. In pigmented rabbits and monkeys, this method allows visualization of
leukocytes passing through the choroidal vessels and provides a new way
to investigate, noninvasively and in vivo, leukocyte dynamics in the
choroidal microcirculation.
Leukocyte dynamics in the retinal and choroidal circulation have
been investigated in vivo, because leukocytes may play a key role in
the microcirculation.
1 2 3 4 Acridine orange digital
fluorography enables evaluation of leukocyte dynamics in the retinal
circulation under physiological and pathologic
conditions.
1 Because acridine orange is spectrally similar
to fluorescein and stains all nuclear material including the retinal
pigment cell nuclei, leukocytes circulating in the choroid cannot be
seen. The use of this method is limited to animal studies, because
acridine orange is carcinogenic and may be phototoxic to cell
lysosomes. Matsuda et al.
2 reported that leukocytes,
stained with indocyanine green (ICG) dye, were observed as
hyperfluorescent dots in the choroidal microcirculation of nonpigmented
rats by using regular ICG angiography with a scanning laser
ophthalmoscope (SLO). However, it is well known that regular ICG
angiography does not enable the visualization of leukocytes in the
choroid of other animals, such as rabbits and monkeys, or in humans. In
the report by Matsuda et al., visualization of leukocytes in the
choroidal circulation of pigmented rats was not possible because of
significant pigment in the retinal pigment epithelium (RPE) and
choroid. Yang et al.
3 4 developed a new method, called
fluorescein leukocyte angiography and studied the leukocyte dynamics in
the retinal and choroidal circulation. However, it is also impossible
to investigate the choroidal circulation of pigmented animals with this
method, because of staining with fluorescein sodium and the
pigmentation of the RPE and choroid. To our knowledge, no method has
yet been available to study in vivo leukocyte dynamics in the choroid
of pigmented animals, such as pigmented rabbits and monkeys, or in
humans.
We developed a new method with which to directly visualize leukocytes
in the choroid of pigmented animals and investigated leukocyte dynamics
in the choroidal circulation of pigmented rabbits and monkeys through
autologous leukocytes stained with ICG dye. Our method, called
indocyanine green leukocyte angiography (ILA), can be applied to the
evaluation of leukocyte dynamics in the choroidal circulation and may
be feasible for use in humans.
Ten to 20 ml blood was withdrawn from a vein (in rabbits from the
ear vein; in monkeys from the cubital vein) into a sterile test tube.
The blood was mixed with a mixture of Ficoll and metrizoate (Mono-Poly
Resolving Medium; Dainippon Pharmaceutical, Osaka, Japan) and separated
using a centrifuge at 1800 rpm for 30 minutes. Most of the plasma and
the bulk of erythrocytes were removed and the white-coat layer of
leukocytes was mixed with 0.01 ml (in rabbits) or 0.05 ml (in monkeys)
ICG solution (Diagnogreen injection; Daiich Pharmaceutical, Tokyo,
Japan). Additional centrifugal separation of the leukocyte fluid at
1500 rpm for 5 minutes was performed to collect ICG-stained leukocytes
more densely. Three milliliters of phosphate-buffered saline (PBS) was
added to the ICG-stained leukocytes. The whole leukocyte fluid was
estimated to contain 10 to 60 million leukocytes by counting the
leukocytes in 0.1 μl of the leukocyte fluid.
Pupils were dilated using 0.5% tropicamide ophthalmic solution. The
leukocyte fluid was injected into the ear vein (in rabbits) or the
antecubital vein (in monkeys), and the fundus images were obtained with
infrared laser and an SLO (model 101; Rodenstock Instrument, Munich,
Germany). The SLO was operated using a 40° field size in rabbits or a
20° field size in monkeys. The images, obtained at 30 frames/sec,
were recorded on S-VHS videotapes. Twenty minutes after the injection,
simultaneous fluorescein and ICG angiography,20 showing
retinal vascular landmarks, was performed to locate ICG-stained
leukocytes in the monkeys.
A few seconds after the injection of the leukocyte fluid,
relatively weak ICG fluorescence of plasma was first visible. For the
initial 3 minutes in rabbits (5 minutes in monkeys), the transit of
phase similar to regular ICG angiography was seen, and the
choroidal arteries and veins were identified by their diffusion time.
After that, ICG fluorescence of plasma significantly decreased, and
choroidal arteries and veins in rabbits (retinal and choroidal in
monkeys) were outlined as negative features against the background.
Composite photographs of early-phase images and late-phase images
were used for the differentiation between arteries and veins. Many
fluorescent dots became distinctly visible approximately 5 minutes
after injection. For more than 1 hour, each ICG-stained leukocyte
passing through the choroidal circulation was distinctly observed as a
single fluorescent dot moving in the choroidal vessels. Because
choroidal arteries and veins were outlined as negative figures against
the background, it was possible to observe these leukocytes moving
rapidly in the choroidal arteries, passing very slowly through the
choroidal capillaries and draining into the choroidal veins at
increasing velocity
(Fig. 1) .
Transient plugging, defined as absence of flow over two intervals
between frames, was frequently seen during passage of leukocytes
through the capillaries. Most of the plugged leukocytes stagnated for
less than 1 second.
The flow velocities of the fluorescent leukocytes that did not plug
decreased rapidly at the entry point of the choroidal capillaries and
were constant during passage through the capillaries. The flow
velocities of leukocytes that did not plug, which passed through the
peripheral capillaries, ranged from 0.26 to 0.93 mm/sec (mean velocity±
SD, 0.48 ± 0.14 mm/sec).
In regular ICG angiography, fluorescence of plasma masks leukocyte
fluorescence in the early to middle phase of the angiography, and in
the late phase, in which the fluorescence of plasma disappears,
leukocytes with decreased fluorescence cannot be recognized because of
blockade by much pigment of the RPE and choroid. ILA is based on the
differential fluorescent staining among blood contents, in that
leukocytes are intensely stained with ICG dye. In the middle-
to late-phase of ILA (in rabbits, 5 minutes to more than 1 hour after
the injection of ICG-stained leukocyte fluid; in monkeys, 10–40
minutes after that), leukocytes can be recognized as hyperfluorescent
dots in the choroid. ILA can be also applied to nonpigmented animals.
In our preliminary study, clear visualization of leukocytes was
confirmed in the choroidal circulation of albino rabbits.
The flow velocity of leukocytes moving in the choroidal vessels was
measured in the present study. The fastest velocity that could be
measured by using the viewing system in the present study was estimated
to be 30 mm/sec. In rabbits, fluorescent dots in the choroidal arteries
were clearly observed. However, the movement in the arteries was too
fast to be traced on a monitor, and it was difficult to evaluate the
velocity of them accurately. Koyama et al.
7 reported that
the flow velocity of erythrocytes moving in the choriocapillaris of
albino rabbits, which was measured by using high-speed videography
taken through a scleral window, ranged from 0.28 to 2.11 mm/sec and
that leukocytes moved more slowly than erythrocytes. Matsuda et
al.
2 reported that leukocyte velocity in the choroidal
capillaries was 0.74 ± 0.06 mm/sec in the nonpigmented rats. In
the present study, the mean velocity was 0.48 ± 0.14 mm/sec. The
discrepancy may be due to the difference in species or measurement
sites.
In primates, little is known about leukocyte rheology in the choroidal
circulation, and the flow velocity of leukocytes in the choroid is
unknown, because there is no available method. In contrast, the mean
flow velocity of leukocytes passing through the retinal perifoveal
capillaries was reported to be 0.92 ± 0.32 mm/sec in
monkeys
1 and 1.41 ± 0.29 or 0.54 ± 0.19 mm/sec
in humans.
4 8 These data are in close agreement with our
data (1.10 ± 0.37 mm/sec), despite the different methods used. In
contrast, the mean velocity of leukocytes moving in the posterior
choroid of the monkeys was 2.45 ± 0.48 mm/sec in the present
study, approximately two times higher than the velocity in the retinal
capillaries. The large choroidal vessels are at a depth different from
that of the choriocapillaris. We observed leukocyte flow at a constant
focal plane during a sequence of data collection. Thus, we may be
biasing our observations to leukocytes that do not deviate
significantly from the plane.
In the monkeys, the fluorescent leukocyte appeared abruptly on a video
monitor, moved approximately 50 to 200 μm corresponding to the
dimension of the lobules of the choriocapillaris,
9 and
disappeared abruptly. These movements seem to reflect the choroidal
vasculature, in that feeding arteries and drainage veins are at a right
angle to the choriocapillaris. Leukocytes in the choroidal arteries and
veins move much faster than those in the capillaries. Indeed, we
observed dim fluorescent dots moving too fast to be recognized in the
still image, the quality of which was much lower than the real-time
image. Fluorescence of leukocytes in the artery or vein of the deeper
choroid may be more strongly blocked by much pigment of the choroid
than that in the capillaries. Thus, we supposed that distinct
fluorescent dots represent leukocytes passing through the choroidal
capillaries. Japanese monkeys (
Macaca fuscata) used in this
study have much pigment in the choroid compared with humans. Chang et
al.
10 described that they used a higher dose of ICG (2
mg/kg) in the rhesus monkeys (
Macaca mulatta) to gain
high-contrast ICG angiographic images, compared with a dose of ICG (1
mg/kg) in human eyes. More distinct images of fluorescent leukocytes in
the choroidal circulation might be gained in human eyes or other
monkeys with less pigment. Further studies are necessary to prove the
assumption that the distinct fluorescent spots seen in the present
study are leukocytes passing through the choroidal capillary.
Transient plugging can be seen in the capillary under physiological and
pathologic conditions.
1 2 3 4 As described, transiently
plugged leukocytes in the choriocapillaris were frequently seen in
rabbits. The majority of them remained stagnant for less than 1 second.
Plugging was sometimes seen in the monkeys. However, it is sometimes
difficult to distinguish whether true plugging occurs or whether it is
artifactual, due to movement of the leukocyte perpendicular to the
choriocapillaris.
To our knowledge, this study is the first to visualize leukocytes
moving in the choroidal microcirculation of pigmented animals including
monkeys. ILA is available to evaluate leukocyte dynamics in vivo and
noninvasively and would be useful for the experimental study of
choroidal hemodynamics in pigmented subjects under pathologic
conditions. Additionally, this method may be applicable in humans
because of the minimal toxicity of ICG and holds a possibility of
disclosing a role of leukocytes in the pathogenesis of various ocular
diseases.
Supported in part by the Health Sciences Research Grants from the Ministry of Health and Welfare, Tokyo.
Submitted for publication June 28, 1999; revised December 17, 1999 and April 14, 2000; accepted April 26, 2000.
Commercial relationships policy: N.
Corresponding author: Fumio Shiraga, Department of Ophthalmology, Okayama University Medical School, 2-5-1 Shikata-cho, Okayama 700-8558, Japan.
[email protected]
Nishiwaki H, Ogura Y, Kimura H, Kiryu J, Honda Y. Quantitative evaluation of leukocyte dynamics in retinal microcirculation. Invest Ophthalmol Vis Sci
. 1995;36:123–130.
[PubMed]Matsuda N, Ogura Y, Nishiwaki H, et al. Visualization of leukocyte dynamics in the choroid with indocyanine green. Invest Ophthalmol Vis Sci
. 1996;37:2228–2233.
[PubMed]Yang Y, Kim S, Kim J. Visualization of retinal and choroidal blood flow with fluorescein leukocyte angiography in rabbits. Graefes Arch Clin Exp Ophthalmol
. 1997;235:27–31.
[CrossRef] [PubMed]Yang Y, Kim S, Kim J. Fluorescein dots in Fluorescein angiography and fluorescein leukocyte angiography using a scanning laser ophthalmoscope in humans. Ophthalmology. 1997;104:1671–1675.
McNeil DR, Trussell TJ, Turner JC. Spline interpolation of demographic data. Demography
. 1977;14:245–252.
[CrossRef] [PubMed]Zeimer RC, Mori MT. An interactive model eye for use with ophthalmic instruments. Arch Ophthalmol
. 1988;106:126–127.
[CrossRef] [PubMed]Koyama T, Matsuo N, Nakayama T, Koyama M. High speed cinematography of microcirculation in the photocoagulated choroid. Folia Ophthalmol Jpn. 1990;15:107–111.
Fallon TJ, Chowiencyzk P, Kohner EM. Measurement of retinal blood flow in diabetics by the blue-light entoptic phenomenon. Br J Ophthalmol
. 1986;70:43–46.
[CrossRef] [PubMed]Torczynski E, Tso MOM. The architecture of the choriocapillaris at the posterior pole. Am J Ophthalmol
. 1976;81:428–440.
[CrossRef] [PubMed]Chang AA, Morse LS, Handa J, et al. Histologic localization of indocyanine green dye in aging primate and human ocular tissues with clinical angiographic correlation. Ophthalmology
. 1998;105:1060–1068.
[CrossRef] [PubMed]