Download Use of Optical Coherence Tomography to Assess Variations

Use of Optical Coherence Tomography to Assess
Variations in Macular Retinal Thickness in Myopia
Marcus C. C. Lim,1 Sek-Tien Hoh,1 Paul J. Foster,1,2 Tock-Han Lim,3,4 Sek-Jin Chew,1
Steve K. L. Seah,1,2 and Tin Aung1,3,5
PURPOSE. To investigate the variation in macular retinal thickness in otherwise normal young Asian myopic subjects by
using optical coherence tomography (OCT).
METHODS. One hundred thirty ophthalmically normal men 19
to 24 years of age with myopia (spherical equivalent, ⫺0.25
to ⫺14.25 D) underwent examination of one randomly selected eye. Visual acuity, refraction, slit lamp examination,
applanation tonometry, gonioscopy, A-scan ultrasound, fundus examination, visual field testing, and optic disc photography were performed. Exclusion criteria were visual acuity
worse than 20/30, previous intraocular surgery, intraocular
pressure ⬎21 mm Hg, or other ocular diseases. Three horizontal transfixation and three vertical transfixation OCT
scans (ver.4.1; Carl Zeiss Meditec, Dublin, CA) of 6 mm each
were conducted on each eye by a single operator. Neurosensory retinal thicknesses at 100 points along each scan
were measured, and the overall average, maximum, and
minimum retinal thicknesses were analyzed by simple linear
regression and analysis of variance.
RESULTS. The average macular retinal thickness (overall) was
230.9 ⫾ 10.5 ␮m and was not significantly related to the
degree of myopia. The mean maximum retinal thickness (at the
parafovea) was 278.4 ⫾ 13.0 ␮m, and correlated negatively
with axial length (P ⫽ 0.03). The mean minimum retinal
thickness (at the foveola) was 141.1 ⫾ 19.1 ␮m, and this was
positively correlated with axial length (P ⫽ 0.015) and spherical equivalent (P ⫽ 0.0002). The retina was thicker at the
superior and nasal parafovea compared to the inferior or temporal parafovea.
CONCLUSIONS. Average retinal thickness of the macula does not
vary with myopia. However, the parafovea was thinner and the
fovea thicker with myopia. (Invest Ophthalmol Vis Sci. 2005;
46:974 –978) DOI:10.1167/iovs.04-0828
From the 1Singapore National Eye Centre and Singapore Eye Research Institute, Singapore; the 2Department of Epidemiology, Institute
of Ophthalmology, London, United Kingdom; the 3Defence Medical
and Environmental Research Institute, Singapore; 4Tan Tock Seng
Hospital, Eye Institute, National Healthcare Group, Singapore; and the
5
Department of Ophthalmology, National University of Singapore,
Singapore.
Presented in part at the annual meeting of the Association for
Research in Vision and Ophthalmology, Fort Lauderdale, Florida, AprilMay, 2001.
Supported by a grant from the Defence Medical Research Institute, Singapore.
Submitted for publication July 14, 2004; revised August 30, 2004;
accepted August 30, 2004.
Disclosure: M.C.C. Lim, None; S.-T. Hoh, None; P.J. Foster,
None; T.-H. Lim, None; S.-J. Chew, None; S.K.L. Seah, None; T.
Aung, None
The publication costs of this article were defrayed in part by
page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate
this fact.
Corresponding author: Tin Aung, Singapore National Eye Centre,
11, Third Hospital Avenue, Singapore 168751; [email protected].
974
O
ptical coherence tomography (OCT) is a new noninvasive
technique that can be used to measure retinal thickness
using time delays of reflected or backscattered light and interferometry.1,2 It is increasingly being used to image lesions at
the macula, such as macular thickening in diabetic macular
edema.3 OCT may also have a role in the clinical assessment of
glaucoma, as recent studies have shown macular and peripapillary thinning in glaucomatous eyes.4 – 6
Myopia is increasing in prevalence in East Asia. This trend
has been documented in Taiwan7 and Singapore.8 Estimates of
the proportion of myopia in the young population of Singapore
range from 30% to 65%.8 –10 According to the histopathologists
Yanoff and Fine,11 in high myopia (generally greater than
⫺6.00 D), the retina thins and degenerates, especially at the
posterior pole. These changes in the myopic retina with increase in myopia and axial length may confound retinal thickness measurements by OCT. The purpose of our study was to
investigate the variations in macular retinal thickness with
myopia in a demographically homogenous group of young
healthy male subjects.
METHODS
One hundred thirty myopic Asian men with various degrees of myopia
were randomly recruited from a cohort of 15,000 Singaporean National
Service enlistees between July 1996 and July 1997. Informed consent
was obtained from all study subjects, and the project had the approval
of the ethics committees of the Singapore Eye Research Institute and
the Defence Medical Research Institute, Singapore. This work was
performed in accordance with The World Medical Association’s Declaration of Helsinki.
Study subjects had participated in an earlier study that assessed the
influence of myopia and its method of correction on automated static
perimetry.12 The methodology of the study has been described previously, a summary of which is provided herein. All the subjects were
ophthalmically normal. In one randomly selected eye, a complete
ocular examination was conducted that included measurement of best
corrected visual acuity, subjective refraction, slit lamp examination,
Goldmann applanation tonometry, gonioscopy, A-scan ultrasound biometry, dilated fundus examination with a 78-D fundus lens, visual
field testing in the form of static automated white-on-white threshold
perimetry (program 24-2, model 750; Carl Zeiss Meditec, Dublin, CA),
and OCT. Exclusion criteria were best corrected visual acuity worse
than 6/9; previous intraocular or refractive surgery; IOP greater than
21 mm Hg; gonioscopic findings of angle closure; evidence of pseudoexfoliation, uveitis, or pigment dispersion syndrome; corneal or media
opacities, retinal disease; or neurologic conditions that could affect
visual fields. In addition, a history of glaucoma in a first-degree relative
and a history of glaucoma or any other optic neuropathy were exclusion criteria. Apart from optic disc and peripapillary changes associated with myopia, subjects did not have any other ocular abnormalities.
Optical Coherence Tomography
OCT (Carl Zeiss Meditec, Dublin, CA) was performed by the same
operator (STH) and analyzed with the version 4.1 software. The OCT
instrumentation (first generation) and imaging technique have been
Investigative Ophthalmology & Visual Science, March 2005, Vol. 46, No. 3
Copyright © Association for Research in Vision and Ophthalmology
Downloaded From: http://tvst.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/932929/ on 05/08/2017
IOVS, March 2005, Vol. 46, No. 3
OCT Measurement of Retinal Thickness in Myopia
975
simply read off the vertical thicknesses from the individual A-scan
measurements. Horizontal and vertical 6-mm scans centered against
the point of fixation of each eye were taken by the same operator
(STH). Three good-quality horizontal and three good-quality vertical
scans were retained for analysis. Each scan was composed of 100
A-mode scans providing 100 data points of retinal thickness. The
version 4.1 software did not have the capability of providing a mean
retinal thickness of all 100 data points for each scan. The operator
recorded the retinal thickness measurements manually at 10 evenly
spaced points along each scan, starting with point 10, 20, 30, and so
on, until point 100 was reached (Fig. 1).
Two points of maximum retinal thickness (RT-Max) measured on
either side of the fovea were recorded from each scan (Fig. 2). These
readings were presumed to correspond to the parafoveal crests and
generally lay around points 25 and 75 (of 100). From the direction of
the scan, it was also possible to correspond the point to the area of the
macula—that is, superior, inferior, temporal, or nasal to the fovea and
these were termed RT-Max (sup), RT-Max (inf), RT-Max (temp), and
RT-Max (nas), respectively. One point of minimum retinal thickness
measurement, which is presumed to correspond to the fovea (RT-Min)
and that lay around point 50, was also recorded (Fig. 2).
Data Analysis
FIGURE 1. The points of retinal thicknesses measured along the length
of a single OCT scan.
described in detail elsewhere.1–5 Pupils were dilated to at least 5-mm
diameter during the OCT examination. All eyes’ axial lengths and
refractive errors were entered into the machine. Correction for magnification errors were provided by means of the Littman formula–
based correction factors implemented in the software. Scan length was
adjusted to 6 mm before scanning, to allow for correction of scan size
for magnification error. The scanner was calibrated with a Gullstrand
model eye before each session. An automated computer algorithm
developed by the manufacturer analyzed optical scattering properties
of the retinal layers based on the threshold values calculated from the
overall signal quality of the longitudinal (z-axis) OCT scan. Retinal
thickness was measured automatically with the retinal thickness algorithm built into version 4.1 of the OCT software, which automatically
determines the anterior and posterior borders of the internal limiting
membrane and retinal pigment epithelium respectively. The operator
All data were analyzed on computer (Statistical Package for Social
Sciences, ver. 11.0; SPSS Inc, Chicago, IL), and statistical significance
was assumed at the P ⬍ 0.05 level. Descriptive statistics, simple linear
regression, and analysis of variance (ANOVA) were performed on the
data.
RESULTS
Of the 130 study subjects, there were 111 Chinese (85.4%), 12
Malays (9.2%), and 7 Indians (5.4%), and the average age was
21.2 ⫾ 1.1 years (range, 19 –24). The mean spherical equivalent was ⫺5.9 ⫾ 3.5 D (range, ⫺0.25 to ⫺14.25 D). Forty-six
(55%) subjects ranged from ⫺0.25 to ⫺4.00 D, 49 (38%) from
greater than ⫺4.00 to ⫺8.00 D and 35 (27%) greater than
⫺8.00 D. The mean axial length was 26.0 ⫾ 1.43 mm (range,
23.22–29.23). The longest axial length of 29.23 mm was within
the focusing range of the machine. The axial length correlated
significantly with the spherical equivalent (Pearson correlation
coefficient ⫽ ⫺0.84, P ⬍ 0.0001). Simple regression of RT-Min
FIGURE 2. The points of RT-Max
and RT-Min on the OCT scan.
Downloaded From: http://tvst.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/932929/ on 05/08/2017
976
Lim et al.
IOVS, March 2005, Vol. 46, No. 3
TABLE 1. Summary of OCT Measurements and Their Variation with Myopia
Macular Thickness
Mean
(95% CI)/␮m
SD
(␮m)
Axial Length*
R2 (95% CI)/␮m
P
Spherical Equivalent*
R2 (95% CI)/␮m
P
Average thickness (overall)
RT-max
RT-max(sup)
RT-max(inf)
RT-max(temp)
RT-max(nas)
RT-min
230.9 (229.1 to 232.7)
278.4 (276.1 to 280.7)
288.3 (284.8 to 291.8)
278.4 (275.5 to 281.3)
262.2 (259.6 to 264.8)
284.2 (281.2 to 287.2)
141.1 (137.8 to 144.5)
10.5
13.0
20.2
16.3
14.6
17.1
19.1
⫺0.58 (⫺1.86 to 0.70)
⫺1.77 (⫺3.33 to ⫺0.21)
⫺3.27 (⫺1.14 to ⫺5.39)
⫺2.47 (⫺4.27 to ⫺0.67)
⫺1.31 (⫺2.83 to 0.21)
⫺1.76 (⫺3.66 to 0.15)
2.85 (0.56 to 5.14)
0.37
0.03†
0.003†
0.008†
0.09
0.07
0.015†
⫺0.11 (⫺0.63 to 0.42)
0.23 (⫺0.42 to 0.88)
0.74 (⫺0.15 to 1.63)
0.54 (⫺0.21 to 1.29)
0.26 (⫺0.37 to 0.89)
0.21 (⫺0.58 to 0.99)
⫺1.78 (⫺2.68 to ⫺0.87)
0.69
0.48
0.10
0.16
0.42
0.60
0.0002‡
* Unstandardized univariate linear regression.
† Significant correlation between macular thickness and axial length (P ⬍ 0.05).
‡ Significant correlation between macular thickness and spherical equivalent (P ⬍ 0.05).
measurement variance as the dependent variable and axial
length as the independent variable yielded P ⫽ 0.26.
The OCT macular thickness measurements are summarized
in Table 1. The average macular thickness (overall) was
230.9 ⫾ 10.5 ␮m. ANOVA showed no difference between
myopia and mean macular thickness (P ⫽ 0.73). Using simple
linear regression, mean macular thickness did not relate to
axial length (P ⫽ 0.37) or to spherical equivalent (P ⫽ 0.69).
The mean RT-Max was 278.4 ⫾ 13.0 ␮m. RT-Max showed a
negative correlation with axial length (P ⫽ 0.03) with a slope
of ⫺1.77 ␮m/mm (Fig. 3), but was not correlated with spherical equivalent (P ⫽ 0.48). The mean RT-Max (sup), RT-Max
(inf), RT-Max (temp), and RT-Max (nas) were 288.3 ⫾ 18.0,
278.4 ⫾ 15.1, 262.2 ⫾ 14.6, and 284.2 ⫾ 17.1 ␮m, respectively. RT-Max (sup) decreased with increasing axial length
(P ⫽ 0.003) at a rate of ⫺3.27 ␮m/mm. RT-Max (inf) decreased
with increasing axial length (P ⫽ 0.008) at a rate of ⫺2.47
␮m/mm. RT-Max (temp) and RT-Max (nas) were not related to
axial length. The mean RT-Min was 141.1 ⫾ 19.1 ␮m. RT-Min
increased with axial length (P ⫽ 0.015; Fig. 4) and increasing
spherical equivalent (P ⫽ 0.0002).
Figure 5 shows the lack of relationship between the variance of RT-Min measurements and increasing axial length (P ⫽
0.26).
This study has shown that in a young healthy cohort of myopic
subjects, overall average macular retinal thickness does not
vary with increasing myopia or axial length. This is in agreement with recent reports in which OCT was used to investigate
retinal thickness variations in myopia.13 Our results contrast
with histologic descriptions of increasing scleral and retinal
thinning with myopia.11 In another study, chorioretinal atrophy in the posterior pole was found to be more frequent in
eyes of longer axial length.14 We believe that the disparity in
our findings may be related to the different approach used to
measure retinal thickness, with OCT retinal measurements
having been shown to be highly reproducible,15 being more
sensitive in detecting subtle and regional variations in retinal
thickness compared to histologic mounts and their susceptibility to shrinkage and processing artifacts. Another study16
showed increasing chorioretinal degeneration in the peripheral fundus with myopia. Unfortunately, it is difficult to measure peripheral retinal thickness at present with the OCT. One
report using ultrasound and Fourier analysis showed that the
midperipheral retina in myopic eyes was thinner than in emmetropic eyes.17 If this is the case, then retinal thinning in
myopia may be more common in the peripheral retina.
The overall average macular thickness in this study was
230.9 ⫾ 10.5 ␮m, which compares favorably with the thickness in Wakitani et al.13 of 231 ⫾ 15 ␮m and in Hee et al.1 of
230 ⫾ 15 ␮m. Subjects in the former study were a mixture of
male and female with mean age 46.2 ⫾ 15.9 years (range,
12–74). In a similarly heterogeneous group, Asrani et al.18
found an average posterior pole retinal thickness of 229 ␮m.
The similar values of mean macular thickness found in these
studies suggests that there may not be great variation in aver-
FIGURE 3. Relationship between maximum macular thickness and
increasing axial length.
FIGURE 4. Relationship between minimum macular thickness and increasing axial length.
DISCUSSION
Downloaded From: http://tvst.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/932929/ on 05/08/2017
IOVS, March 2005, Vol. 46, No. 3
FIGURE 5. Relationship between variance of RT-Min measurements
and increasing axial length.
age retinal thickness with age,18,19 though one Japanese paper
on OCT reported macular retinal thinning with age.20
We observed regional differences in parafoveal retinal thickness in different sectors of the macula. The maximum parafoveal thickness at the superior, inferior, temporal, and nasal
regions were 288.3 ⫾ 20.2, 278.4 ⫾ 16.3, 262.2 ⫾ 14.6, and
284.2 ⫾ 17.1 ␮m, respectively, indicating that the retina is
thickest in the superior and nasal parafovea. This is consistent
from results in other studies showing maximum thickness at
the superior and nasal parafoveal quadrants.18
Our study showed that the point of maximum retinal thickness on either side of the fovea, which we assume to correspond to the parafovea, decreased with increasing axial length
(Fig. 2), whereas minimum thickness corresponding to foveolar thickness increased. From the scatterplot (Fig. 3), it is
evident that this is a linear relationship, and we used unstandardized univariate linear regression for this analysis. In contrast, Wakitani et al.13 did not note any change in foveolar or
parafoveolar thicknesses with increasing myopia. They calculated average thicknesses for three concentric zones: 350,
1850, and 2850 ␮m in diameter and compared the groups with
an ANOVA. For their data, perhaps a simple regression would
yield results similar to ours. Subjects in that study were also
more heterogenous, whereas our subjects were all men and of
a similar age.
Looked on in isolation, the finding of decreasing parafoveolar thickness with myopia may be consistent with retinal thinning in this region. However, when we found simultaneous
increasing foveolar thickness, the possibility of poorer fixation
with high myopia arose, as fixation not through the fovea
would produce overestimated foveal thicknesses, perhaps because in our version of OCT scanning, using version 4.1 software, the landmark point is used as a fixation target. The
subjects were told to fixate this target, and the scans were
positioned over it, giving rise to transfixation scans. Therefore,
the scans were obtained over the points of fixation but this
does not guarantee that they correspond to the foveola of
subjects. Off-foveola fixation may have resulted in larger foveolar thickness measurements.
However, our subjects fixated quite consistently. Evidence
of this is shown in the scatterplot (Fig. 5) showing the relationship between variance of RT-Min measurements and increasing axial length, which does not show any evidence of
OCT Measurement of Retinal Thickness in Myopia
977
increasing variance with increasing axial length (P ⫽ 0.26, R2
⫽ 0.0097). If fixation was inconsistent with increasing myopia,
then the variance of measurements through the thinnest point
of the retina should be significantly higher.
The mean thickness of the thinnest retinal thickness point
was 141.1 ⫾ 19.1 ␮m, remarkably similar to the mean thickness of the foveola of 142 ⫾ 18 ␮m, as measured by Gobel et
al.21 in subjects ranging from 13 to 92 years of age, and by
Kanai et al.20 (142 ⫾ 15 ␮m), and it suggests that the thinnest
point we measured was indeed the foveola. From our results,
it can be concluded that the variation in thickness between the
foveal pit and parafoveolar crest decreases in increasing myopia. In effect, the macular area might be said to become flatter.
Our results may be affected by enrollment bias, as the study
subjects recruited were all young and had good visual acuity
with no evidence of retinal degeneration. It is likely that older
myopic subjects may have greater ocular morbidity related to
myopia. In summary, we have found unique variations in macular thickness with myopia. Although overall mean macular
thickness is not affected by degree of myopia, there is evidence
that the thickest point at the parafoveal region decreases with
myopia, whereas the fovea thickness increases with myopia.
These variations in macular thickness should be considered in
the evaluation of retinal diseases and glaucoma using the OCT.
References
1. Hee MR, Izatt JA, Swanson EA, et al. Optical coherence tomography of the human retina. Arch Ophthalmol. 1995;113:325–332.
2. Huang D, Swanson EA, Lin CP, et al. Optical coherence tomography. Science. 1991;254:1178 –1181.
3. Hee MR, Puliafito CA, Duker JS, et al. Topography of diabetic
macular edema with optical coherence tomography. Ophthalmology. 1998;105:360 –370.
4. Guedes V, Schuman JS, Hertzmark E, et al. Optical coherence
tomography measurement of macular and nerve fiber layer thickness in normal and glaucomatous human eyes. Ophthalmology.
2003;110:177–189.
5. Hoh ST, Greenfield DS, Mistlberger A, et al. Optical coherence
tomography and scanning laser polarimetry in normal, ocular hypertensive, and glaucomatous eyes. Am J Ophthalmol. 2000;129:
129 –135.
6. Tanito M, Itai N, Ohira A, Chihara E. Reduction of posterior pole
retinal thickness in glaucoma detected using the retinal thickness
analyzer. Ophthalmology. 2004;111:265–275.
7. Lin LL, Shih YF, Hsiao CK, et al. Epidemiologic study of the
prevalence and severity of myopia among schoolchildren in Taiwan in 2000. J Formos Med Assoc. 2001;100:684 – 691.
8. Wong TY, Foster PJ, Hee J, et al. Prevalence and risk factors for
refractive errors in adult Chinese in Singapore. Invest Ophthalmol
Vis Sci. 2000;41:2486 –2494.
9. Au Eong KG, Tay TH, Lim MK. Race, culture and Myopia in
110,236 young Singaporean males. Singapore Med J. 1993;34:
29 –32.
10. Tay MT, Au Eong KG, Ng CY, Lim MK. Myopia and educational
attainment in 421,116 young Singaporean males. Ann Acad Med
Singapore. 1992;21:785–791.
11. Yanoff M, Fine BS. Ocular Pathology: A Text and Atlas. 3rd ed. JB
Lippincott; 1989;408.
12. Aung T, Foster PJ, Seah SK, et al. Automated static perimetry: the
influence of myopia and its method of correction. Ophthalmology.
2001;108:290 –295.
13. Wakitani Y, Sasoh M, Sugimoto M, et al. Macular thickness measurements in healthy subjects with different axial lengths using
optical coherence tomography. Retina. 2003;23:177–182.
14. Curtin BJ, Karlin DB. Axial length measurements and fundus
changes of the myopic eye. Am J Ophthalmol. 1971;1:42–53.
15. Massin P, Vicaut E, Haouchine B, et al. Reproducibility of retinal
mapping using optical coherence tomography. Arch Ophthalmol.
2001;119:1135–1142.
Downloaded From: http://tvst.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/932929/ on 05/08/2017
978
Lim et al.
16. Karlin DB, Curtin BJ. Peripheral chorioretinal lesions and axial
length of the myopic eye. Am J Ophthalmol. 1976;81:625– 635.
17. Chujo S, Kobayashi Y, Emi K, et al. The biometry of each thickness
of the human retina, choroid and sclera by using ultrasound and
Fourier analysis: high myopia. (in Japanese). Nippon Ganka Gakkai Zasshi. 1983;87:70 –73.
18. Asrani S, Zou S, d’Anna S, et al. Noninvasive mapping of the normal
retinal thickness at the posterior pole. Ophthalmology. 1999;106:
269 –273.
IOVS, March 2005, Vol. 46, No. 3
19. Landau D, Schneidman EM, Jacobovitz T, Rozenman Y. Quantitative in vivo retinal thickness measurements in healthy subjects.
Ophthalmology. 1997;104:639 – 642.
20. Kanai K, Abe T, Murayama K, Yoneya S. Retinal thickness and
changes with age (in Japanese). Nippon Ganka Gakkai Zasshi.
2002;106:162–165.
21. Gobel W, Hartmann F, Haigis W. Determination of retinal thickness in relation to the age and axial length using optical coherence
tomography (in German). Ophthalmologe. 2001;98:157–162.
Downloaded From: http://tvst.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/932929/ on 05/08/2017