Download Measurement of Subfoveal Choroidal Thickness Using Spectral

i
■
m
a
g
c l i n i c a l
i
n
g
s c i e n c e
■
Measurement of Subfoveal Choroidal
Thickness Using Spectral Domain
Optical Coherence Tomography
Emily A. McCourt, MD; Brian C. Cadena, PhD; Cullen J. Barnett, CRA; Antonio P. Ciardella;
Naresh Mandava, MD; Malik Y. Kahook, MD
n BACKGROUND AND OBJECTIVE: To compare
subfoveal choroidal thickness (SFCT) in normal patients
and those with known ocular pathology using spectral
domain optical coherence tomography (SD-OCT).
n PATIENTS AND METHODS: This retrospective, observational case series was conducted at a tertiary care center where 194 consecutive eyes from 102
patients were imaged. Patients were not included or
excluded based on presence or absence of pathology.
One masked observer imaged the choroid and a second
masked observer measured SFCT. Multivariate analysis
was used and a statistical model created to analyze the
changes in SFCT induced by age, diabetic retinopathy,
glaucoma, wet and dry age-related macular degeneration, and other posterior pole pathology.
n RESULTS: The mean SFCT of the 194 eyes studied
was 246.59 ± 93.17 µm with a mean age of 55.50 ± 19.70
years. A strong negative relationship was found between
age and SFCT (R2 = 0.42), with an average 3.09-µm decrease in SFCT per additional year of age. Subgroup analysis demonstrated that patients with diabetic retinopathy,
wet or dry age-related macular degeneration, and glaucoma all had SFCT measurements that were statistically
significantly less than those of normal patients. However,
when regression analysis was used to control for age, this
difference was no longer significant.
n CONCLUSION: No differences were found in
SFCT in patients with glaucoma, macular degeneration, or diabetic retinopathy compared to eyes lacking
pathology when age was counted as a confounding variable. Age has a strong inverse relationship with SFCT,
independently confirming prior studies and creating a
foundation for more research on the relationship between ocular pathology and choroidal thickness.
[Ophthalmic Surg Lasers Imaging 2010;41:S28S33.]
From the Department of Ophthalmology (EAM, CJB, NM, MYK), Rocky Mountain Lions Eye Institute, University of Colorado School of Medicine, Aurora,
Colorado; the Department of Economics (BCC), University of Colorado at Boulder, Boulder, Colorado; and the Department of Ophthalmology (APC), University
of Bologna, Bologna, Italy.
Originally submitted November 19, 2009. Accepted for publication March 11, 2010.
Dr. Kahook has received consulting fees from Heidelberg. The other authors have no financial or proprietary interest in the materials presented herein.
Address correspondence to Malik Y. Kahook, MD, 1675 Aurora Court, Mail Stop F-731, Aurora, CO 80045. E-mail: [email protected]
doi: 10.3928/15428877-20101031-14
S28
Copyright © SLACK Incorporated
i
m
a
g
i
n
g
Figure 1. (A) Standard, upright
image obtained from Spectralis
spectral domain optical coherence tomography (Heidelberg
Engineering, Heidelberg, Germany). (B) The same patient with
choroid scan. The machine was
placed closer to the patient and
the image obtained is inverted.
(C) The choroid becomes easier
to see with black and white inverted. Software calipers were
used to measure thickness of the
choroid of this image.
INTRODUCTION
The choroid serves many essential functions for the
eye. It acts as a heat sink for the retina1,2 and delivers blood
to 85% of the eye, including the photoreceptors and the
prelaminar portion of the optic nerve head.3 Despite its
importance, we have few reliable techniques to study choroidal structure and function.4 Disturbance of choroidal
blood flow may play a key role in disease states such as
glaucoma, diabetic retinopathy, and age-related macular
degeneration (AMD).3,5,6 Resistance to flow is related to
the diameter of a vessel3; thus, choroidal thickness may be
proportional to the blood flow in the choroid and may be
an important metric for choroidal health. Therefore, it is
important to study choroidal thickness with more precision to better understand this vital structure.
A novel way for measuring the thickness of the choroid was recently described in patients without any underlying pathology.7 A decrease in subfoveal choroidal
thickness with increasing age was also recently reported.1
Subfoveal choroidal thickness in patients with AMD,
diabetic retinopathy, and glaucoma has not been reported in peer-reviewed literature to our knowledge. We
report the results of our retrospective review that details
the ability of spectral domain optical coherence tomography (SD-OCT) to measure subfoveal choroidal thickness in a tertiary care center in both normal patients and
patients with various ophthalmic diseases.
PATIENTS AND METHODS
This is a retrospective consecutive case series of
patients who underwent subfoveal choroidal thickness
measurements at the Department of Ophthalmology
at the University of Colorado Denver between January
and June 2009. One hundred ninety-four eyes of 102
consecutive patients were scanned using the Heidelberg
Spectralis SD-OCT machine (Heidelberg Engineering,
Heidelberg, Germany). The study included eyes with a
variety of pathology and patients were not excluded or
included in the study based on any known eye disease
or lack thereof. The eyes classified as normal controls
did not have any pathology other than mild to moderate cataracts.
The choroid was imaged using the same method as
previously described7 by positioning the SD-OCT device closer to the eye to create an inverted image (Figs.
1A and 1B). The image color was changed from black
on white to white on black to increase the contrast between the choroid and the sclera to measure the choroidal thickness more accurately (Fig. 1C). The subfoveal
choroidal thickness was measured by a masked observer using the calipers within the software of the OCT
machine and positioning them from the outer aspect of
Bruch’s membrane to the border of the sclera.
The data were analyzed using STATA software
(StataCorp, College Station, TX). Each eye was treated
as a separate observation in all analyses. Standard errors
from the regression analysis were adjusted for potential
within-patient correlation. The resulting standard errors provide valid statistical inference even if choroidal
thickness is more similar in two eyes from the same
patient than in two randomly sampled eyes. This was
done to prevent doubling our data because, for example, diabetic retinopathy tends to be a bilateral disease
process.
Ophthalmic Surgery, Lasers & Imaging · Vol. 41, No. 6 (Suppl), 2010
S29
i
m
a
g
i
n
g
Table 1
Overall Subfoveal Choroidal Thickness and Age Statistics
Variable
Mean
Median
Standard Deviation
Min
Max
Subfoveal choroidal thickness (µm)
246.59
240.5
93.17
52
539
55.5
60
19.7
21
91
Age (y)
Table 2
Percentage of Eyes in Each Category
Disease Process
Percent
Glaucoma
9.3
Diabetic retinopathy
15.5
Wet AMD
7.7
Dry AMD
7.7
Other posterior pathology
20.1
Total – any posterior pathology
56.2
AMD = age-related macular degeneration.
RESULTS
Figure 2. Histogram demonstrating subfoveal choroidal thickness
range for all patients.
Tables 1 and 2 provide an overview of our patient
population. The overall mean subfoveal choroidal thickness was 246.59 ± 93.2 µm and the overall mean age was
55.5 ± 19.7 years. Figure 2 shows a histogram of subfoveal choroidal thickness in our patient population.
Mean subfoveal choroidal thickness was 198.3 µm
for eyes with glaucoma (n = 18) and 221.7 µm for eyes
with diabetic retinopathy (n = 30). Average subfoveal
choroidal thickness was 209.9 µm for eyes with wet
AMD (n = 15) and 162.4 µm for eyes with dry AMD
(n = 15). The 39 eyes with posterior pole pathology
that did not fit into one of the above categories had
a mean subfoveal choroidal thickness of 252.3 µm.
Normal eyes (n = 63) had a mean subfoveal choroidal
thickness of 305.7 µm. As demonstrated in Table 3,
when compared to the normal groups, all pathologies
evaluated had a statistically significant decrease in subfoveal choroidal thickness. However, when regression
analysis was used to control for patient age, none of the
differences in subfoveal choroidal thickness remained
statistically significant (Table 4).
With regression analysis, several models for choroidal thickness as a function of age and adjustment for
pathology were fitted (Table 4). The simplest regression
model includes only age as an explanatory variable,
and the results suggest that choroidal thickness falls by
approximately 3.085 µm per year. The R2 value demonstrates the strength of the relationship of subfoveal
choroidal thickness to age. In our study, age accounted
for 42% of the variation in subfoveal choroidal thickness. Figure 3 plots patient age and subfoveal choroidal
thickness for the entire sample and the fitted line from
the regression model using only age as a covariate. The
same analysis with normal eyes can be seen in Figure 4.
In addition, Figure 5 displays each point according to
the presence and type of pathology.
The remaining question is whether eyes with pathology deviate from their age-adjusted expected thicknesses in any systematic way. The figure does not provide any obvious support for that hypothesis, and the
regression results confirm that we cannot reject the null
hypothesis that eyes with varying pathologies and normal eyes have equal subfoveal choroidal thickness after
accounting for age. In this model, glaucoma would add
approximately 5.6 µm to a patient’s age-adjusted subfoveal choroidal thickness and diabetes mellitus would
add 8.89 µm. Wet AMD would add 47.78 µm and
dry AMD would subtract 14.1 µm. However, none of
these differences are statistically significantly different
from zero.
S30
Copyright © SLACK Incorporated
i
m
a
g
i
n
g
Table 3
t Test of Equal Means Comparing Each Category of Eye With Both Normal and Overall Patient Population
Group of Interest
No.
SFCT (µm)
Reference Group
No.
SFCT (µm)
P
Glaucoma
18
198.3
Non-glaucoma
176
251.5
.01
Glaucoma
18
198.3
Normal
63
305.7
< .0001
Diabetic retinopathy
30
221.7
Non-diabetic
164
251.1
.08
Diabetic retinopathy
30
221.7
Normal
63
395.7
< .001
Wet AMD
15
209.9
Non-wet AMD
179
249.7
.12
Wet AMD
15
209.9
Normal
63
305.7
.001
Dry AMD
15
162.4
Non-dry AMD
179
253.65
.001
Dry AMD
15
162.4
Normal
63
305.7
< .0001
Other posterior pathology
39
253.3
Non-other path
155
244.9
.587
Other posterior pathology
39
253.3
Normal
63
305.7
.003
Any pathology
131
218.2
Normal
63
305.7
< .0001
SFCT = subfoveal choroidal thickness; AMD = age-related macular degeneration.
Table 4
Subfoveal Choroidal Thickness Subgroup Regression Analysis to Account for
the Variations in Subfoveal Choroidal Thickness With Agea
Variable
(1)
(2)
(3)
Age (y)
-3.085*** (0.319)
-3.318*** (0.369)
-3.087*** (0.347)
Glaucoma
5.586 (20.85)
Diabetic retinopathy
8.893 (21.08)
Wet AMD
47.78 (29.05)
Dry AMD
-14.07 (31.16)
Other posterior pathology
22.21 (16.25)
Any posterior pathology
Constant
Observations
2
R
0.143 (13.99)
417.8*** (18.53)
421.8*** (19.09)
417.8*** (18.38)
194
194
194
0.424
0.447
0.424
AMD = age-related macular degeneration.
a
Three statistical models demonstrating the changes of subfoveal choroidal thickness with age and pathology. Robust standard errors in parentheses.
***P < .01; ** P < .05; * P < .10.
DISCUSSION
This study details the measurement of subfoveal
choroidal thickness in a variety of ophthalmic diseases
and compares these findings to eyes lacking pathology.
The major contribution of this study is the finding of
a strong inverse relationship between subfoveal choroidal thickness and age. This finding independently
confirms and further extends the work done by Spaide
et al. and provides information to build on for future
prospective studies.1,2
Patients with diabetic retinopathy were found to
have increased subfoveal choroidal thickness compared
with the overall study population, but this difference
was not statistically significant when accounting for
differences in age. This finding was somewhat surpris-
Ophthalmic Surgery, Lasers & Imaging · Vol. 41, No. 6 (Suppl), 2010
S31
i
m
a
Figure 3. Subfoveal choroidal thickness for all patients.
Figure 5. Subfoveal choroidal thickness for all patients categorized by disease process.
ing in this group of patients with microvascular disease,
where one might predict a decrease in choroidal thickness. Decreased subfoveal choroidal blood flow has
been previously documented in patients with diabetic
retinopathy and was proved to be even lower in eyes
with diabetic macular edema.6 Of note, the one eye in
our study with diabetic macular edema demonstrated a
very thin choroid (105 µm). Changes in subfoveal choroidal thickness, or the lack thereof, in patients with
diabetes mellitus may help predict the patients who
would be most at risk to acquire diabetic macular edema, the most common cause of impaired vision in patients with type 2 diabetes mellitus.6 Further research is
needed to explore a possible relationship here, but from
our pilot study, no relationship seems to exist. In fact,
it is possible that there is no correlation between subfoveal choroidal thickness and quantitative choroidal
blood flow, and further studies are needed to directly
S32
g
i
n
g
Figure 4. Subfoveal choroidal thickness for all normal patients.
correlate choroidal blood flow to subfoveal choroidal
thickness. The prospect for such studies is currently
limited by the lack of precise tools for measuring blood
flow in choroidal vessels.
The choroidal thickness in eyes with dry AMD was
14.07 µm thinner than was expected when controlled
for age. However, this result was not statistically significant due to the high standard error. This is probably
best explained by the very strong relationship that both
AMD and choroidal thickness have with age. Thus, a
higher powered study may prove AMD and age to be
additive in the future. Eyes with wet AMD were found
to have a subfoveal choroidal thickness that was 47.78
µm thicker than expected for adjusted age. This result
was also not statistically significant due to high standard error. This was an unexpected result because multiple previous studies have shown choroidal thickness
and blood flow to be decreased in patients with wet
AMD. Additionally, both choroidal thickness and choroidal blood flow have been proven to be decreased in
patients with wet AMD compared with those with dry
AMD.5 A possible explanation for this finding is the
increased vasculature and vascular endothelial growth
factor release that is associated with wet AMD.8
It is noteworthy that decreased choroidal blood
flow and volume as measured by Doppler laser has
been demonstrated in patients with AMD4 and has
been shown prospectively to indicate higher risk of disease progression.5 Metelista et al. found a thinner choroid in patients with the wet variety of AMD,5 which is
the opposite of what we found in our patients. Again,
due to the extremely strong relationship between both
subfoveal choroidal thickness and AMD with age, we
Copyright © SLACK Incorporated
i
m
a
would need a larger number of patients to achieve the
power required to demonstrate a statistically significant
difference and further studies are underway to explore
these relationships.
The average subfoveal choroidal thickness in patients with glaucoma in our study was 5.59 µm thicker
than would have been expected when controlled for age.
However, this result was not statistically significant. Decreased ocular blood flow in glaucoma has been reported, although a direct causal relationship has not been
firmly established.3 The disruption of oxygenated blood
flow to the retina and optic nerve head might lead to
ischemia and cell apoptosis with eventual loss of ganglion cells and thinning of the optic nerve head rim.
Still, it is unclear how this would relate to the fact that
the choroid serves 85% of the outer retina, and most of
these structures are spared while the inner portion of the
retina (the ganglion cells) is injured. Additionally, our
study did not find any difference in subfoveal choroidal
thickness when the data were controlled for age.
There are important limitations in this study. This
was a retrospective study of a limited number of consecutive patients. The group of patients used was heterogeneous with a variety of ophthalmic pathology and
a wide range of ages. Additionally, we did not account
for the sub-population differences such as the type of
glaucoma or diabetic retinopathy. Another limitation
was the younger age and smaller number of normal
patients used as controls in this study. Because most
of our pathology was found in patients with advanced
ages, and the negative relationship between subfoveal
choroidal thickness and age is so strong (R2 = 0.42),
we will need a more highly powered study to tease out
differences if they exist.
There are no related studies that we can use to compare our data because our study is, to the best of our
knowledge, the first to compare the subfoveal choroidal
thickness of different pathologies using the Spectralis
OCT. We did find a much greater decrease in subfoveal choroidal thickness with age in normal eyes when
compared with previous data published by Spaide et
al., who found a 1.56-µm decrease in subfoveal choroidal thickness with each year.2 The difference between
our results can be explained by the different populations studied. We had a larger population of normal
g
i
n
g
patients who were younger than those previously studied. This could cause the steeper slope in our regression
line. Additionally, we measured the subfoveal choroidal
thickness from the outer border of Bruch’s membrane,
whereas Spaide et al. used the outer border of the
retinal pigment epithelium as their landmark.1,2 This
would make our measurements thinner and would not
explain the increase in subfoveal choroidal thickness
that we found in our patient population when compared with previously reported data.
Future prospective studies of patients with glaucoma, AMD, and diabetic retinopathy will help us to
better understand the role that the choroid plays in the
pathophysiology of these diseases and perhaps help us
better predict morbidity in these patients. In addition,
although the subfoveal choroidal thickness does reflect
the anatomic thickness of the choroid, further studies
will be needed to establish a correlation between subfoveal choroidal thickness and the volume or rate of
choroidal blood flow. Measurement of subfoveal choroidal thickness is a noninvasive method to image the
choroid and may prove to be a reliable metric to diagnose disease, monitor response to therapy, or monitor
progression of disease. Further studies are needed to
understand the full potential of this mode of imaging.
REFERENCES
1. Spaide R. Age-related choroidal atrophy. Am J Ophthalmol.
2009;147:801-810.
2. Margolis R, Spaide R. A pilot study of enhanced depth imaging optical coherence tomography of the choroid in normal eyes. Am J Ophthalmol. 2009;147:811-815.
3. Flammer J, Orgul S, Costa VP, et al. The impact of ocular blood flow
in glaucoma. Progr Retin Eye Res. 2002;21:359-393.
4. Harris A, Chung HS, Ciulla TA, Kagemann L. Progress in measurement of ocular blood flow and relevance to our understanding of
glaucoma and age-related macular degeneration. Prog Retin Eye Res.
1999;18:669-687.
5. Metelitsina TI, Grunwald JE, DuPont JC, Ying GS, Brucker AJ, Dunaief JL. Foveolar choroidal circulation and choroidal neovascularization in age-related macular degeneration. Invest Ophthalmol Vis Sci.
2008;49:358-363.
6. Nagaoka T, Kitaya N, Sugawara R, et al. Alteration of choroidal circulation in the foveal region in patients with type 2 diabetes. Br J
Ophthalmol. 2004;88:1060-1063.
7. Spaide RF, Koizumi H, Pozonni MC. Enhanced depth imaging
spectral-domain optical coherence tomography. Am J Ophthalmol.
2008;146:496-500. Erratum in: Am J Ophthalmol. 2009;148:325.
8. Xu H, Chen M, Forrester JV. Para-inflammation in the aging retina.
Prog Retin Eye Res. 2009;28:348-368.
Ophthalmic Surgery, Lasers & Imaging · Vol. 41, No. 6 (Suppl), 2010
S33