Download Ultra-Wide–Field Green-Light (532-nm) Autofluorescence Imaging in

i
■
m
a
g
c l i n i c a l
i
n
g
s c i e n c e
■
Ultra-Wide–Field Green-Light (532-nm)
Autofluorescence Imaging in Chronic
Vogt-Koyanagi-Harada Disease
Florian M. Heussen, MD; Daniel V. Vasconcelos-Santos, MD, PhD; Rajeev R. Pappuru, MD;
Alexander C. Walsh, MD; Narsing A. Rao, MD; Srinivas R. Sadda, MD
n BACKGROUND AND OBJECTIVE: To assess
the prevalence of peripheral fundus autofluorescence
(FAF) abnormalities in chronic Vogt-Koyanagi-Harada
disease (VKH).
n PATIENTS AND METHODS: A retrospective
review of cases at the Doheny Eye Institute between
December 2009 and April 2010. Patients with chronic
VKH who had ultra-wide–field FAF and pseudo-color
imaging performed were included. All images were reviewed independently by two reading center certified
retina specialists.
ed in this analysis. Fourteen eyes of 7 patients (70%)
showed peripheral changes on FAF images outside the
posterior pole. Three different patterns were observed:
multifocal hypofluorescent spots (n = 11 eyes), hyperfluorescent spots (n = 8 eyes), and a unique lattice-like
pattern in both eyes of one patient. There were noticeable disparities between FAF and color images.
n CONCLUSION: Peripheral FAF abnormalities are frequent in chronic VKH and are readily revealed by widefield FAF imaging and manifesting with distinct patterns.
Further investigation in prospective studies is warranted.
n RESULTS: Twenty eyes of 10 patients were includ-
[Ophthalmic Surg Lasers Imaging 2011;42:272-277.]
INTRODUCTION
sive clinical stages.1-3 In the acute uveitic stage, leakage
at the level of the retinal pigment epithelium (RPE)
leads to bilateral exudative retinal detachments, which
usually resolve after adequate high-dose steroid thera-
Vogt-Koyanagi-Harada disease (VKH) is a bilateral granulomatous panuveitis characterized by succes-
From the Doheny Eye Institute and the Department of Ophthalmology, Keck School of Medicine, University of Southern California, Los Angeles, California.
Originally submitted January 11, 2011. Accepted for publication March 24, 2011. Posted online May 12, 2011.
Supported in part by NIH Grant EY03040 and NEI Grant R01 EY014375.
Drs. Sadda and Walsh share in royalties from intellectual property licensed to Topcon Medical Systems by the Doheny Eye Institute. Dr. Sadda serves on the scientific advisory board for Heidelberg Engineering and receives research support from Carl Zeiss Meditec and Optovue Inc. This is not related to the article’s subject
matter. The remaining authors have no financial or proprietary interest in the materials presented herein.
Address correspondence to Srinivas R. Sadda, MD, Doheny Eye Institute, 1450 San Pablo Street, Los Angeles, CA 90033. E-mail: [email protected]
doi: 10.3928/15428877-20110505-01
272
Copyright © SLACK Incorporated
i
m
a
g
py. Patients then enter a convalescent stage, but some
may present recurrences (chronic recurrent stage)1-4
and progressive chorioretinal damage mainly associated
with RPE changes.1,5 Such RPE changes can lead to vision-threatening complications, subretinal fibrosis, and
choroidal neovascularization.1,5
New advances in imaging of the outer retina and
RPE may aid in better characterizing and monitoring the
retinal changes in VKH, with a possible impact on the
optimal management of the disease.5 Fundus autofluorescence (FAF) is based on excitation of inherent fluorophores by light of a certain wavelength (typically between
488 and 585 nm) and recording the emitted light to create a brightness map. The nature of excited fluorophores
greatly depends on the excitation wavelength used. A
prevalent fluorophore is lipofuscin6 and its composites,
such as N-retinyl-N-retinylidene ethanolamine (A2E),
the accumulation of which is seen as an indicator for
metabolic stress on RPE cells. Therefore, an increase or
decrease in FAF has been shown to correlate with integrity
and functional status of the photoreceptor–RPE complex,
and FAF imaging is increasingly being used for qualitative
assessment of RPE pathology, also due in part to its noninvasive nature.7 Frequently, inflammatory diseases of the
posterior segment directly or indirectly involve the RPE
and underlying choriocapillaris, and as such many investigators have suggested that FAF imaging may be useful in
evaluating these diseases. Some studies have demonstrated
that FAF imaging may be superior to ophthalmoscopy for
the clinical assessment of RPE damage in these inflammatory entities,8 including in VKH.5
However, in VKH many of these changes manifest in
the periphery, which may be beyond the field of view of
traditional FAF imaging techniques that tend to focus on
the posterior pole. As a result, more peripheral FAF abnormalities in VKH have not been characterized or compared
with the ophthalmoscopic findings. Recently, a prototype
wide-field green-light (532-nm) autofluorescence imaging system (Optos P200CAF; Optos, Dunfermline, UK)
has been developed, allowing assessment of more peripheral FAF abnormalities. In this report, we identify the frequency and morphology of these peripheral FAF changes
in a series of patients with chronic VKH.
PATIENTS AND METHODS
A retrospective review was conducted of all patients with VKH consecutively seen at Doheny Eye
Ophthalmic Surgery, Lasers & Imaging · Vol. 42, No. 4, 2011
i
n
g
Institute from December 2009 to April 2010. The
diagnosis of VKH established by one of the authors
(NAR) according to the most recently revised criteria4,9 and chronic disease was defined based on a
duration of intraocular inflammation longer than 3
months. Additionally, for inclusion in the study these
patients had to have been imaged with the P200CAF
Wide-field-Autofluorescence scanning laser ophthalmoscope prototype (532-nm; Optos). All patients
underwent both FAF imaging and pseudo-color (red
and green only) imaging by the same machine, in addition to color fundus photographs in a regular fundus camera. Ten cases met the aforementioned criteria
and were included in this analysis. For all 10 subjects,
a detailed medical chart review was performed and
anonymous demographic and clinical data were recorded in a separate database, including gender, age,
ethnicity, disease duration, visual acuity, and findings
for slit-lamp biomicroscopy and mydriatic ophthalmoscopy. The research protocol was approved by
the Institutional Review Board of the University of
Southern California and was in accordance with tenets set forth in the Declaration of Helsinki. All subjects signed a written informed consent form.
All FAF images were reviewed and graded by two
retina specialists (SRS and ACW), who are certified
graders and investigators at the Doheny Image Reading Center. Regions of increased, decreased, or isointense FAF relative to background fluorescence were
noted and compared with corresponding locations
on the pseudo-color images. Similarly, abnormalities
noted on the pseudo-color images were compared and
correlated with corresponding locations on the FAF
images. Particular care was taken to compare areas of
sunset glow fundus, evident on the color images, with
the FAF images. Common patterns of peripheral FAF
abnormalities were identified. Pseudo-color images
were also compared and validated against regular color
photographs.
RESULTS
Of the 10 patients included in this study, 9 were
women and 1 was a man. The ages ranged from 25
to 57 years (mean: 40.0 ± 8.7 years; median: 40.0
years). Three patients were in the convalescent stage
of VKH and were not receiving any anti-inflammatory therapy at the time of evaluation. Three patients
273
i
m
a
g
i
n
g
Table
Clinical Features of a Series of Patients With Chronica
Vogt-Koyanagi-Harada Disease and Peripheral FAF Changes
Case
No.
BCVA
Sex
Age
(Y)
Disease
Duration (Mo)
Disease
Status
Current
Treatment
OD
OS
1
F
43
137
Inactive
Prednisonec
20/20
20/40
OD/OS multifocal hypofluorescent spots (< 20),
solitary hyperfluorescent
spots (< 20)
2
F
25
34
Inactive
MFM, prednisonec
20/30
20/30
OD/OS multifocal hypofluorescent spots (> 100),
solitary hyperfluorescent
spots (< 20)
3
F
32
36
Inactive
None
20/20
20/20
None
4
F
42
7
Inactive
MFM, prednisone
20/40
20/25
None
5
M
37
240
Inactive
20/30
20/300
OD/OS multifocal hypofluorescent spots (> 20 and < 100)
6
F
36
218
Inactive
None
20/20
20/25
OD/OS multifocal hypofluorescent spots (> 20 and < 100), solitary hyperfluorescent spots (< 20)
7
F
47
24
Active
Prednisonec
20/25
20/200
OD hyperfluorescent spots
(+), OS multifocal hypofluorescent spots (< 20),
solitary hyperfluorescent
spots (< 20)
8
F
57
4
Inactive
CSA, prednisone
20/50
20/25
OD/OS lattice-like pattern
9
F
43
90
Inactive
Prednisonec
20/20
20/25
OD/OS multifocal hypofluorescent spots (> 20 and < 100)
10
F
38
51
Inactive
None
20/20
20/20
Topical prednisone
Peripheral FAF
Abnormalitiesb
None
FAF = fundus autofluorescence; BCVA = best-corrected visual acuity; OD = right eye; OS = left eye; MFM = mycophenolate mofetil; CSA = cyclosporine A.
a
Defined as during > 3 months.
b
Number of observed lesions in parentheses.
c
Dose < 5 mg/day.
were receiving a low oral dose (< 5 mg/day) of prednisone alone, two patients were receiving oral prednisone and mycophenolate mofetil, and one patient
was receiving a combination of oral prednisone and
cyclosporine-A. Disease duration ranged widely from
4 to 240 months (median: 43.5 months) and active
intraocular inflammation was noted in only 1 patient
at the time of wide-field FAF imaging. Best-corrected
visual acuity (BCVA) ranged from 20/300 to 20/20
(Table). On ophthalmoscopy, all eyes were noted to
have varying degrees of sunset glow fundus.
Peripheral FAF changes were seen in 14 eyes of 7
274
patients, whereas 3 patients had no FAF abnormalities in either eye. The predominant FAF abnormality was the presence of multifocal, nummular foci
of decreased autofluorescence signal in 11 eyes of 6
patients, which could be correlated to chorioretinal
scars on the pseudo-color images in most instances
(Fig. 1). Eight of the 14 eyes with FAF abnormalities also showed focal spots of increased autofluorescence signal, which were generally far fewer in number than the hypofluorescent foci and did not appear
to correlate with abnormalities on the pseudo-color
images (Figs. 1 and 2). One patient presented distinc-
Copyright © SLACK Incorporated
i
m
a
g
i
n
g
Figure 2. Right eye of case 1. Few sharply demarcated hypofluorescent spots are visible on the fundus autofluorescence image
(A), which correlate to scars on the pseudo-color image (B). In this
case, a hyperfluorescent spot marked with an arrow on A shows
no correlate on B. Similarly, a small group of hypofluorescent spots
is clearly visible on A (arrow with asterisk), yet more difficult to
spot on image B.
Figure 1. Fundus autofluorescence (FAF) (A and B), pseudo-color
(C and D), and color images (E and F) of case 2 showing both
eyes. Multifocal hypofluorescent foci are clearly visible on both
FAF images, which correlate well with the chorioretinal scars in C
and D. In addition, few hyperfluorescent spots are seen. Interestingly, hyperpigmentation does not reliably show up on FAF images
(arrows) and other disparities are clearly seen (arrows with asterisk). The sunset glow fundus changes seen on the pseudo-color
and the color images show no correlate on the FAF images.
tive changes on FAF, namely a striking lattice-like
pattern of increased autofluorescence signal (Fig. 3).
On closer inspection of the pseudo-color images (and
regular color photographs), many of these streaks correlated with linear bands of hyperpigmentation. No
correlation was observed between the pattern or extent of FAF abnormalities and current disease activity
or treatment regimen. A summary of these findings is
shown in the table. The sunset glow fundus changes
were apparent on regular color fundus photographs
and less so on the wide-field pseudo-color images.
Interestingly no direct correlate for the sunset glow
changes could be seen on the FAF images, independent of the degree of changes. Overall, there was
good agreement between the chorioretinal lesions on
regular color fundus photographs and their aspect on
wide-field pseudo-color imaging, although the latter
was able to show more peripheral lesions.
Ophthalmic Surgery, Lasers & Imaging · Vol. 42, No. 4, 2011
Figure 3. Both eyes of case 8. The fundus autofluorescence images (A and B) demonstrate a lattice-like pattern of hyperfluorescence along the vascular arcades. Some streaks can be correlated to pigmented lines seen on the color images (C and D) as
indicated by the arrows (B and D).
DISCUSSION
In the current study, we report the frequency and
pattern of peripheral FAF abnormalities in a series of
patients with chronic VKH in southern California.
The functional assessment of the RPE layer through
FAF imaging may be valuable in these patients5 because RPE changes have been associated with severe
vision-threatening complications in chronic VKH.1
The peripheral changes reported in the current study
were documented by a wide-field FAF scanning laser
ophthalmoscope prototype, which allows for a more
thorough assessment of the retinal periphery than
conventional FAF instruments. Not unexpectedly in
this disease population, 7 of 10 cases (14 of 20 eyes)
showed peripheral FAF changes and common patterns
of FAF abnormalities could be identified.
275
i
m
a
The most prevalent FAF changes were multifocal
sharply demarcated foci of decreased autofluorescent
signal (in 11 eyes of 6 patients). These hypofluorescent foci appeared to have a stereotypic appearance
across cases and were easy to identify due to the high
contrast of wide-field FAF images. On pseudo-color
(and regular color) images, these foci corresponded to
nummular atrophic scars with various degrees of pigmentation, which have been shown in both opticalcoherence-tomography–based5 and histopathological
studies10,11 of patients with chronic VKH. In these areas, the decreased autofluorescence signal is associated
with disruption in the outer retina, RPE atrophy, and
subsequent loss of fluorophores.5,11
Eight eyes also demonstrated focal spots of increased autofluorescence signal, but the morphology of
these lesions was more variable, with spots varying in
both size and brightness. Some foci of increased autofluorescence signal did not correlate to any fundus
findings seen on pseudo-color imaging (and regular
color fundus photographs). It is possible that these
lesions represent areas of subclinical disease that may
eventually develop into zones of RPE atrophy, similar
to the progression of hyperfluorescent abnormalities in
macular diseases such as atrophic age-related macular
degeneration.12
Interestingly, one patient also presented with a
lattice-like pattern of hyperfluorescent streaks in both
eyes (Fig. 3). Only after careful comparison were some
of the streaks found to match faint pigmented lines in
the corresponding color image. Even within our database of more than 400 imaged patients with more than
20 different posterior uveitides, we did not see another
case with this lattice-like FAF pattern. The clinical significance of this unique finding and whether it is only
associated with chronic VKH remains unclear. Based
on our previous fundus autofluorescence and spectraldomain optical coherence tomography (SD-OCT)
studies in VKH, these areas likely correspond to foci of
RPE proliferation with a possibly spared outer retinal
architecture.5 However, the precise histologic correlate
is still speculative, because the far peripheral location of
some of these abnormalities precluded obtaining SDOCT scans through these lesions.
One should bear in mind that hyperpigmentation
does not necessarily contribute to an increased autofluorescence signal, which is rather due to an increased
level of fluorophores.13 Although there are pigmented
276
g
i
n
g
fluorophore compounds such as melanolipofuscin, the
dominant fluorophores do not seem to have a close relation to the level of pigmentation. However, in areas
where the RPE has proliferated, an increase in the total
amount of fluorophores or their accumulation in the
outer retina might still be associated with an increased
autofluorescence signal.5 Particularly in inflammatory diseases, a high fluorophore content might also
be associated with a local pro-oxidative environment,
although the nature and exact localization of these specific fluorophores are largely unknown.14
Of note, some spots of decreased autofluorescence
signal could not be clearly correlated to respective sites
on the pseudo-color images, and similarly some findings on the pseudo-color images were not associated
with any abnormality on FAF imaging. Underlying
reasons for this phenomenon are manifold. Heavily
depigmented spots on the pseudo-color images might
mimic RPE atrophy and even atrophic areas might still
show a level of autofluorescence due to choroidal and
scleral autofluorescent properties. In addition, the autofluorescent signal can be blocked by structures/pigment anterior to the source of emission.7
All of our patients exhibited varying degrees of sunset glow fundus on clinical examination. This fundus
change is associated with depigmentation of the choroid10,11 and has been regarded as a distinctive clinical
feature of chronic VKH.3 One aim of our study was to
seek a correlation between FAF changes and the sunset
glow fundus. However, in concordance with previous
reports,5 the autofluorescence signal was preserved in
these areas and undistinguishable from normal background autofluorescence. This is presumably because
choroidal changes do not contribute significantly to the
green-light autofluorescence signal, which largely originates from the lipofuscin content of the RPE.6 Thus,
even a heavily depigmented choroid is not detectable
via fundus autofluorescence alone in the presence of vital RPE. This shows that although the autofluorescence
signal carries additional clinical information, it should
be regarded as adjunctive imaging in combination with
color fundus photographs. Also of note, regular color
fundus photographs more easily demonstrated the sunset glow fundus when compared with pseudo-color images in this study.
We did not find any association between the number and type of FAF changes and disease duration or
inflammatory activity. However, this is a small series
Copyright © SLACK Incorporated
i
m
a
g
and FAF imaging was only reported at a single time
point. Therefore, we cannot rule out the influence of
therapy and recurrences in the prevalence of these FAF
changes.
There are limitations to our study, mainly the small
sample size with a variable duration of disease and the
lack of longitudinal data. It is not yet known how these
FAF patterns evolve over time and whether changes in
FAF patterns may be used as a reliable marker of disease
progression. In addition, the abnormalities identified
in this study are based on scanning laser ophthalmoscope green-light autofluorescence. As such, the patterns may not correspond to findings from flash-based
FAF approaches or other wavelengths, such as blue or
near-infrared. Finally, the use of pseudo-color widefield images (which are based on a reconstructed color
map of only two color channels: red and green but not
blue) might not provide images as accurate as regular
color fundus photography. Although it is known that
little information is carried in the blue channel, we also
compared the pseudo-color images with regular color
fundus photographs to attenuate this possible bias.
We observed that peripheral FAF abnormalities
are frequent in patients with chronic VKH and may be
readily detected using wide-field FAF imaging. These
FAF abnormalities do not always correlate with abnormalities visible on ophthalmoscopy or color photography. At the same time, ophthalmoscopic findings do
not necessarily have a correlate on FAF images, as is the
case with sunset glow fundus changes that are not apparent on FAF images. Still, we believe that wide-field
FAF imaging might be useful to assess peripheral RPE
Ophthalmic Surgery, Lasers & Imaging · Vol. 42, No. 4, 2011
i
n
g
changes in patients with chronic VKH. However, the
exact clinical relevance of these FAF changes needs to
be investigated in future prospective studies.
REFERENCES
1. Moorthy RS, Inomata H, Rao NA. Vogt-Koyanagi-Harada syndrome.
Surv Ophthalmol. 1995;39:265-292.
2. Sugiura S. Vogt–Koyanagi–Harada disease. Jpn J Ophthalmol.
1978;22:9-35.
3. Rao NA, Gupta A, Dustin L, et al. Frequency of distinguishing
clinical features in Vogt-Koyanagi-Harada disease. Ophthalmology.
2010;117:591-599.
4. Read RW, Holland GN, Rao NA, et al. Revised diagnostic criteria for
Vogt-Koyanagi-Harada disease: report of an international committee
on nomenclature. Am J Ophthalmol. 2001;131:647-652.
5. Vasconcelos-Santos DV, Sohn EH, Sadda S, Rao NA. Retinal pigment
epithelial changes in chronic Vogt-Koyanagi-Harada disease: fundus
autofluorescence and spectral domain-optical coherence tomography
findings. Retina. 2010;30:33-41.
6. Delori FC, Dorey CK, Staurenghi G, et al. In vivo fluorescence of the
ocular fundus exhibits retinal pigment epithelium lipofuscin characteristics. Invest Ophthalmol Vis Sci. 1995;36:718-729.
7. Schmitz-Valckenberg S, Holz FG, Bird AC, Spaide RF. Fundus autofluorescence imaging: review and perspectives. Retina. 2008;28:385409.
8. Haen SP, Spaide RF. Fundus autofluorescence in multifocal choroiditis and panuveitis. Am J Ophthalmol. 2008;145:847-853.
9. Rao NA, Sukavatcharin S, Tsai JH. Vogt-Koyanagi-Harada disease
diagnostic criteria. Int Ophthalmol. 2007;27:195-199.
10. Inomata H, Rao NA. Depigmented atrophic lesions in sunset
glow fundi of Vogt-Koyanagi-Harada disease. Am J Ophthalmol.
2001;131:607-614.
11. Rao NA. Pathology of Vogt-Koyanagi-Harada disease. Int Ophthalmol. 2007;27:81-85.
12. Schmitz-Valckenberg S, Fleckenstein M, Scholl HP, Holz FG. Fundus
autofluorescence and progression of age-related macular degeneration.
Surv Ophthalmol. 2009;54:96-117.
13. Bindewald A, Bird AC, Dandekar SS, et al. Classification of fundus
autofluorescence patterns in early age-related macular disease. Invest
Ophthalmol Vis Sci. 2005;46:3309-3314.
14. Spaide RF. Chorioretinal inflammatory disorders. In: Holz FG,
Schmitz-Valckenberg S, Spaide RF, Bird AC, eds. Atlas of Fundus Autofluorescence Imaging. Berlin: Springer; 2007:207-239.
277