Volume 74, No.5: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
284
Anuwat Jiravarnsirikul, M.D., Ngamkae Ruangvaravate, M.D., Chaovaporn Ubolviroj, M.D., Yuwared Chattagoon,
M.D., Sakaorat Petchyim, M.D.
Department of Ophthalmology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.
Ganglion Cell-inner Plexiform Layer Thickness
Measured by Cirrus High-denition Optical
Coherence Tomography Enhances Glaucoma
Diagnosis in Patients with Moderate or High Myopia
ABSTRACT
Objective: To assess the diagnostic ability of Cirrus high-denition optical coherence tomography (HD-OCT)
parameters in patients with moderate or high myopia for detecting glaucoma, and to compare the thickness of the
macular ganglion cell-inner plexiform layer (GC-IPL) in glaucomatous and normal eyes in both types of myopia.
Materials and Methods: is prospective study enrolled moderately (spherical equivalent -3.00 to -6.00 diopters)
and highly (spherical equivalent ≤ -6.00 diopters) myopic patients without (controls) and with (study) glaucoma.
Cirrus HD-OCT was used to determine the thickness of the peripapillary retinal nerve ber layer (RNFL) and the
GC-IPL. e area under the receiver operating characteristic curve was analyzed to evaluate the glaucoma detection
capability of each Cirrus HD-OCT parameter.
Results: Seventy eyes (31 moderate myopia, 39 high myopia) were included. e parameters with the best diagnostic
ability were minimum GC-IPL, inferior RNFL and average RNFL thickness in moderately myopic eyes, and average
RNFL, inferior RNFL and inferotemporal GC-IPL thickness in highly myopic eyes. All parameters were thinner in
glaucomatous than in normal eyes in both groups.
Conclusion: Although macular GC-IPL thickness demonstrated high ability to detect glaucoma in patients with
moderate or high myopia, it should be used in combination with other structural imaging and functional assessments
for diagnosing glaucoma.
Keywords: Ganglion cell-inner plexiform layer thickness; glaucoma; myopia; Cirrus high-denition optical coherence
tomography (Siriraj Med J 2022; 74: 284-293)
Corresponding author: Ngamkae Ruangvaravate
E-mail: ngamkae1@gmail.com
Received 15 November 2021 Revised 2 February 2022 Accepted 9 February 2022
ORCID ID: https://orcid.org/0000-0002-4342-0174
http://dx.doi.org/10.33192/Smj.2022.35
All material is licensed under terms of
the Creative Commons Attribution 4.0
International (CC-BY-NC-ND 4.0)
license unless otherwise stated.
INTRODUCTION
Glaucoma was reported to be the major cause of
irreversible blindness and the second most prevalent cause
of moderate and severe visual impairment.
1,2
Numerous
risk factors lead to the development of glaucoma, and
myopia is one of them.
3
Globally, myopia prevalence is
increasing, and approximately 5 billion people will be
aected by 2050, especially in East and Southeast Asia. In
addition to uncorrected myopia itself, pathologic ocular
complications from myopia or its comorbidities, such as
myopic macular degeneration, choroidal neovascularization,
and glaucoma, can lead to increased medical and economic
Jiravarnsirikul et al.
Volume 74, No.5: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
285
Original Article
SMJ
burden.
4,5
A previous study reported that myopic eyes
had a two to three times higher risk of having glaucoma
6
,
and high myopic eyes had a nearly six-fold increased
risk of having primary open-angle glaucoma.
7
us,
early detection and treatment of glaucoma in myopia
is crucial for preventing disease progression.
8
Previous study reported that detection of structural
abnormalities, i.e., retinal nerve ber layer (RNFL) defects,
may develop four to six years before the onset of visual
eld abnormalities.
9
As a consequence, recent advances
in ocular imaging, especially spectral-domain optical
coherence tomography (SD-OCT), which is a frequently
used structural imaging, may greatly contribute to the early
detection and treatment of glaucoma. Previous studies used
peripapillary RNFL thickness as a measurement of RGC
axon loss to detect glaucoma, and since approximately
50% of RGCs are located in the macular region, and
the inner retinal layers are preferentially aected by
glaucomatous damage
10,11
, further studies have assessed
macular parameters to aid in glaucoma diagnosis. Previous
studies demonstrated that the thickness of macular
ganglion cell complex (GCC) generated by an RTVue
SD-OCT (Optovue, Fremont, CA, USA), and the thickness
of macular ganglion cell-inner plexiform layer (GC-IPL)
by Cirrus high-denition optical coherence tomography
(HD-OCT) with ganglion cell analysis (GCA) algorithm
(Carl Zeiss Meditec, Jena, Germany) yielded glaucoma
detection ability similar to peripapillary RNFL thickness;
however, they found GC-IPL thickness measurement
not be confounded by the variation in RNFL.
10-14
Evaluation of RGC loss using various techniques of
structural imaging has been continuously improved to
achieve more accuracy in glaucoma detection, and with
better reproducibility. However, diagnosing glaucoma in
myopic eyes is still dicult since anatomical distortion
of the optic nerve head (ONH) (e.g., tilted disc, large
peripapillary atrophy) can mystify ophthalmologists to
dierentiate between glaucomatous damage and myopia-
related optic disc changes. erefore, objective structural
quantication may have a role in diagnosing glaucoma
in myopia.
15-17
Leung, et al. reported that, when using
Cirrus HD-OCT, temporally converged superior and
inferior RNFL bundles were detected in myopic eyes,
which resulted in abnormal RNFL measurement.
18
Additionally, a previous study concluded that optic disc
and cup margin detection errors were likely to be found
in myopic eyes.
15
us, particular care must be taken
when interpreting RNFL thickness map and neuroretinal
rim measurement obtained by SD-OCT in myopic eyes.
Analysis of macular parameters has been proposed as
an alternative method for diagnosing glaucoma to avoid
misdiagnosis of glaucoma via the inuence of optic disc
variation in myopia.
e thickness of the GC-IPL is a macular parameter
that has been studied in various research. Choi, et al.
demonstrated that in highly myopic eyes, GC-IPL has a
diagnostic potential for glaucoma that is comparable to
RNFL thickness.
19
In a previous study, the inferotemporal
GC-IPL thickness was found to have the largest area
under the receiver operating characteristic (AuROC)
curve, suggesting that it might be utilized as an eective
preperimetric glaucoma detection parameter in myopia.
20
Since the normative databases in SD-OCT do not
include data from highly myopic subjects, diagnosis of
glaucoma in these patients is still somewhat dicult.
21
Moreover, dierences in the severity of myopia can
aect the GC-IPL thickness measurement. Seo, et al.
concluded that GC-IPL in all sectors were thinner in
high myopes compared to low and moderate myopes.
22
Akashi, et al. reported that GC-IPL thickness had a high
AuROC curve value for distinguishing highly myopic
glaucomatous eyes from normal eyes in patients with
and without high myopia (spherical equivalent (SE)
≤ -6.00 diopters).
23
Accordingly, this study aimed to
determine the diagnostic capability of Cirrus HD-OCT
parameters for diagnosing glaucoma in patients with
moderate (SE -3.00 to -6.00 diopters) or high myopia
(SE ≤ -6.00 diopters), and to compare macular GC-IPL
thickness between glaucomatous and normal eyes in
both types of myopia.
MATERIALS AND METHODS
is prospective, comparative cross-sectional study
was conducted at Siriraj Hospital, was approved by the
Committee for the Protection of Human Participants
in Research of the Faculty of Medicine Siriraj Hospital,
Mahidol University, Bangkok, ailand [EC 103/2014],
and was registered in the ai Clinical Trials Registry
(identication number TCTR20210726001). is study
followed to the tenets of the Declaration of Helsinki.
Each subject provided a written informed consent before
enrollment.
is study was performed in the ophthalmology
outpatient clinic of Siriraj Hospital during February
2014 to December 2018. Patients aged older than 20
years with SE ≤ -3.00 diopters (D) and best spectacles-
corrected visual acuity of 20/40 or better were eligible
for inclusion. Subjects with retinal or macular pathology,
vision-associated systemic or neurologic diseases, receiving
drugs known to aect the macula (e.g., chloroquine,
ethambutol), previous intraocular surgery or laser within
1 month, inability to cooperate with the examination,
Volume 74, No.5: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
286
and/or having an allergic reaction to the eye drops used
in the study protocol were excluded. Participants visiting
the ophthalmology outpatient clinic who met these
criteria were enrolled in this study with written informed
consent. Only one eye was randomly selected in the nal
analysis if both eyes met all research eligibility criteria.
All recruited patients underwent a complete ophthalmic
examination, which consisted of assessment of visual
acuity (VA), refractive error measured by autorefractor
(ARK-530A; NIDEK, Aichi, Japan) and recorded as
spherical equivalents (SE), intraocular pressure (IOP)
measurement obtained with Goldmann applanation
tonometry, slit-lamp and dilated fundus examination,
gonioscopy, axial length obtained using a biometer (IOL
Master 500; Carl Zeiss Meditec), OCT RNFL and GC-IPL
parameter measurement by Cirrus HD-OCT soware
version 6.0.0.599 (Carl Zeiss Meditec), and visual eld test
by Humphrey visual eld analyzer (Carl Zeiss Meditec)
using the 24-2 Swedish interactive threshold algorithm
(SITA) strategy.
Enrolled subjects were classied into either the
moderately or highly myopic groups. Moderate and high
myopia were dened as SE -3.00 D to -6.00 D and ≤ -6.00
D, respectively. Within each of those two groups, patients
were further subdivided into either glaucoma (study) or
normal (control) groups. Using the reference standard
24
,
glaucomatous eyes were dened as eyes with glaucomatous
optic neuropathy with corresponding abnormal visual
eld tests. Glaucomatous optic neuropathy was dened
as enlarged cupping ≥ 0.5 vertical cup/disc (C/D) ratio
or an asymmetrical vertical C/D ratio greater than 0.2.
Glaucomatous visual eld defect was dened as any
one of the following: a cluster of ≥ 3 non-edge adjacent
points in one hemield of pattern deviation plot with
a probability < 5%, and including at least 1 point with
a probability < 1%; a pattern standard deviation (PSD)
showing a probability < 5%; or, a glaucoma hemield
test (GHT) showing outside normal limits. Normal
subjects were moderately and highly myopic patients
who visited ophthalmology outpatient clinic with mild
cataract, dry eyes or eye check up and without glaucoma,
which dened as those with an IOP ≤ 21 mmHg, no
glaucomatous optic neuropathy, and without detectable
visual eld defect. e visual eld of the normal group
must not meet any of the aforementioned criteria for
glaucomatous visual eld defect and a GHT showing
within normal limits, a mean deviation (MD) and PSD
within the 95% condence limit. OCT measurements
are not considered as part of the examinations to include
a patient in the normal control group.
Cirrus HD-OCT measurement
Two scans were obtained through a dilated pupil, one
peripapillary RNFL scan and one macular scan (optic disc
cube 200x200 protocol and macular cube 512x128 protocol,
respectively). Macular scan measured GC-IPL thickness
within a 6x6x2 mm cube centered at the fovea using the
GCA algorithm, and the measurements were generated
into the minimum, average, and 6 sectorial parameters
(Inferotemporal, inferior, inferonasal, superonasal, superior
and superotemporal). e optic disc cube protocol measured
peripapillary RNFL thickness through a 6x6x2 mm
cube, aer which the measurements were analyzed into
the average and the 4 quadrant thicknesses (temporal,
superior, nasal, and inferior). Scans with a signal strength
lower than 6 on either macular GC-IPL or peripapillary
RNFL scan, visible eye movement, decentration, artifacts
from blinking, and/or image distortion from anatomical
abnormalities were discarded. Measurement values were
analyzed via comparison with the device’s normative
database, and the results were visually described as a
color-coded signicance map with the colors green, yellow,
and red indicating normal, borderline, and abnormal,
respectively. ere was no cuto point of OCT value to
dene glaucoma, but it was used in combination with
structural and functional tests to enhance glaucoma
diagnosis.
Statistical analysis
Patient characteristics are reported as mean plus/
minus standard deviation (SD) for normally distributed
continuous data, and as frequency and percentage for
categorical variables. Peripapillary RNFL and GC-IPL
thicknesses were compared between the glaucoma and
normal groups using unpaired t-test, and were compared
among 4 groups using 1-way analysis of variance (ANOVA)
by Tukey HSD or Games-Howell. e ability of a factor
to detect glaucoma was assessed by the AuROC curve.
All statistical analyses were performed using PASW
Statistics soware version 26. A p-value < 0.05 indicates
statistically signicant.
RESULTS
Subjects
Seventy eyes were included and categorized into
4 groups, as follows: moderately myopic normal group
(n=16), highly myopic normal group (n=16), moderately
myopic glaucomatous group (n=15), and highly myopic
glaucomatous group (n=23). Patient demographic and
baseline characteristics compared between glaucomatous
and normal eyes in the moderate and high myopia groups
Jiravarnsirikul et al.
Volume 74, No.5: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
287
Original Article
SMJ
are presented in Table 1. e mean age of the moderately
myopic group was 47.55±14.51 years (range from 22 to
68 years), and 19 were female (61.29%). e mean age of
the highly myopic group was 45.69±15.02 years (range
from 21 to 70 years), and 22 were female (56.41%). No
signicant dierences were found in age, VA, IOP, or
SE between normal and glaucomatous eyes in both the
moderate and high myopia groups.
Cirrus HD-OCT measurement
Macular GC-IPL and peripapillary RNFL thicknesses
as evaluated by Cirrus HD-OCT compared between
glaucomatous and normal eyes in the moderate and high
myopia groups are demonstrated in Table 2. No patients
were excluded for poor quality scans. In both groups, all
of the RNFL thickness parameters, except for the nasal
region in both groups (moderate myopia p=0.700, and
high myopia p=0.831) and the temporal region in the
highly myopic group (p=0.118), were signicantly thinner
in glaucomatous than in the normal eyes (all p<0.05).
Comparison of macular GC-IPL thickness between
glaucomatous and normal eyes in both the moderate
and high myopia groups revealed statistically signicant
dierences in minimum, average, and all sectors of
macular GC-IPL thickness (all p<0.05), except for the
superonasal GC-IPL sector in both the moderate and
high myopia groups (p=0.282 and 0.614, respectively),
and the superior GC-IPL sector in the high myopia
group (p=0.443).
Diagnostic ability
All of the AuROC values of macular GC-IPL and
peripapillary RNFL thickness in both the moderate
and high myopia groups were above 0.5 (Table 3). In
the moderately myopic group, the best parameters for
distinguishing glaucomatous eyes from normal eyes were
minimum GC-IPL, inferior RNFL, and average RNFL
thicknesses (AuROC: 0.963, 0.925 and 0.919, respectively)
(Fig 1). In the highly myopic eyes, average RNFL, inferior
RNFL, and inferotemporal GC-IPL thicknesses (AuROC:
0.980, 0.962 and 0.905, respectively) demonstrated the
best diagnostic ability for dierentiating glaucoma from
normal eyes (Fig 2).
TABLE 1. Patient demographic and baseline characteristics compared between normal and glaucomatous eyes in
the moderate and high myopia groups.
Moderately myopic group (n = 31) Highly myopic group (n = 39)
Normal Glaucoma P value* Normal Glaucoma P value
†
(n =16) (n =15) (n =16) (n =23)
Female, n (%) 11 (68.8) 8 (53.3) 0.38 12 (75) 10 (43.5) 0.10
Age, years 43.56 (13.47) 51.80 (14.80) 0.53 42.50 (15.19) 47.91 (14.83) 0.82
VA, logMAR 0.07 (0.09) 0.11 (0.11) 0.85 0.09 (0.10) 0.10 (0.10) 1.00
IOP, mmHg 15.69 (1.85) 14.33 (2.38) 0.55 14.56 (2.94) 13.65 (2.48) 0.82
SE, D -4.39 (0.87) -4.57 (0.99) 1.00 -7.61 (1.14) -8.40 (4.79) 0.95
AL, mm 25.51 (0.89) 25.14 (1.25) 0.95 26.42 (0.81) 27.69 (1.49) <0.01
MD, dB -0.87 (1.49) -5.10 (4.20) <0.01 -1.35 (0.99) -6.27 (3.56) <0.01
PSD, dB 1.47 (0.25) 5.25 (3.65) <0.01 1.67 (0.56) 5.60 (3.64) <0.01
Data are given as mean (SD).
Abbreviations: VA; visual acuity, IOP; intraocular pressure, SE; spherical equivalent, AL; axial length, MD; mean deviation, PSD; pattern
standard deviation
* Value for comparison of normal and glaucomatous eyes in moderately myopic group
†
Value for comparison of normal and glaucomatous eyes in highly myopic group
Volume 74, No.5: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
288
TABLE 2. Peripapillary RNFL and macular GCIPL thickness as measured by Cirrus HD-OCT compared between
normal and glaucomatous eyes in the moderate and high myopia groups.
Moderately myopic group (n = 31) Highly myopic group (n = 39)
Normal Glaucoma P value* Normal Glaucoma P value
†
(n =16) (n =15) (n =16) (n =23)
RNFL thickness parameters, μm
Average 97.31 (8.13) 72.73 (14.00) <0.001 89.81 (7.07) 70.96 (7.38) <0.001
Superior 119.75 (16.18) 89.13 (18.50) <0.001 103.50 (16.74) 81.39 (14.03) <0.001
Nasal 68.63 (9.17) 63.20 (11.01) 0.700 65.63 (12.38) 61.48 (12.01) 0.831
Inferior 120.19 (20.87) 77.67 (21.23) <0.001 114.13 (15.86) 74.13 (15.02) <0.001
Temporal 80.50 (11.17) 61.27 (17.26) 0.001 75.69 (8.06) 65.74 (13.65) 0.118
Macular GC-IPL parameters, μm
Average 80.88 (6.64) 71.33 (8.02) 0.004 75.88 (4.26) 67.17 (9.13) 0.004
Minimum 79.44 (6.73) 61.87 (8.73) <0.001 73.19 (7.48) 58.52 (12.06) <0.001
ST 80.50 (6.28) 71.33 (8.73) 0.019 76.25 (4.12) 67.78 (10.99) 0.015
S 81.69 (7.29) 70.40 (9.83) 0.047 77.44 (4.79) 71.13 (16.82) 0.443
SN 83.00 (7.04) 75.60 (10.31) 0.282 78.19 (5.91) 73.22 (14.42) 0.614
IN 82.00 (6.19) 71.93 (8.56) 0.001 74.56 (6.06) 67.04 (6.53) 0.007
I 78.50 (6.97) 66.47 (8.08) 0.001 73.50 (4.18) 61.04 (10.37) <0.001
IT 79.88 (6.87) 66.00 (9.51) <0.001 75.69 (5.00) 62.96 (8.49) <0.001
Data are given as mean (SD).
Abbreviations: ST; superotemporal, S; superior, SN; superonasal, IN; inferonasal, I; inferior, IT; inferotemporal
* Value for comparison of normal and glaucomatous eyes in moderately myopic group
†
Value for comparison of normal and glaucomatous eyes in highly myopic group
DISCUSSION
is study aimed to demonstrate the diagnostic
performance of OCT parameters, including peripapillary
RNFL and macular GC-IPL thickness obtained by Cirrus
HD-OCT, to detect glaucoma in moderately and highly
myopic patients. We found the capability of macular
GC-IPL thickness parameters for discriminating between
glaucomatous and normal eyes to be high and comparable
to peripapillary RNFL thickness in both myopia severity
groups.
Diagnosing glaucoma in myopic patients can be a
challenge due to anatomical variations in the optic disc
and retina, such as tilted disc, nasal elevation, temporal
attening, rotation, posterior staphyloma, and large
peripapillary atrophy, can make it dicult for clinicians
to distinguish between structural damage from glaucoma
and anatomical changes due to myopia.
15,16
In addition
to the diculties in making a clinical diagnosis by direct
visualization of ONH or optic disc photographs, structural
evaluation via peripapillary RNFL parameters obtained by
SD-OCT can also demonstrate inaccuracies of measurement.
Hwang, et al. reported that neuroretinal rim measurement
errors were observed in myopic patients when using
Cirrus HD-OCT.
15
Leung, et al. concluded that increasing
Jiravarnsirikul et al.
Volume 74, No.5: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
289
Original Article
SMJ
TABLE 3. e AuROC value (SE) for each Cirrus HD-OCT parameter, which indicates the strength of that parameter
for distinguishing between normal and glaucomatous eyes, in the moderate and high myopia groups.
Moderately myopic group Highly myopic group
(n = 31) (n = 39)
RNFL thickness parameters, μm
Average 0.919 (0.057) 0.980 (0.018)
Superior 0.892 (0.058) 0.855 (0.064)
Nasal 0.648 (0.108) 0.548 (0.093)
Inferior 0.925 (0.045) 0.962 (0.027)
Temporal 0.850 (0.077) 0.795 (0.076)
Macular GC-IPL parameters, μm
Average 0.842 (0.073) 0.887 (0.058)
Minimum 0.963 (0.031) 0.897 (0.065)
ST 0.810 (0.086) 0.891 (0.057)
S 0.838 (0.075) 0.760 (0.079)
SN 0.756 (0.091) 0.758 (0.082)
IN 0.823 (0.077) 0.832 (0.074)
I 0.881 (0.059) 0.904 (0.054)
IT 0.894 (0.055) 0.905 (0.053)
Data are given as AUROC (SE).
Abbreviations: ST; superotemporal, S; superior, SN; superonasal, IN; inferonasal, I; inferior, IT; inferotemporal
Fig 1. e area under the receiver operating characteristic
(AuROC) curve for the 3 best parameters by Cirrus
HD-OCT for distinguishing between normal and
glaucomatous eyes in the moderate myopia group.
Abbreviations: RNFL; peripapillary retinal nerve ber
layer, GCIPL; macular ganglion cell-inner plexiform
layer.
Volume 74, No.5: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
290
severity of myopia associated with temporally converged
superotemporal and inferotemporal bundles of RNFL,
which led to abnormal measurement of RNFL.
18
Kim,
et al. reported that in healthy eyes, false-positive OCT
RNFL ndings were inuenced by smaller disc areas
and longer axial lengths, and these factors are usually
found in myopic eyes.
25
In contrast to the limitations of RNFL measurement
for diagnosing glaucoma in myopic patients, macular
parameters, which are less aected by the optic nerve
head variations found in myopia, have good diagnostic
ability and their application has been reported in several
studies.
14,19,20,23,26-28
Kim, et al. reported the eectiveness of
using GCC thickness, composed of inner plexiform layer,
ganglion cell layer, and RNFL, measured by RTVue SD-
OCT for diagnosing glaucoma in myopic eyes to be similar
to the diagnostic results obtained from using peripapillary
RNFL thickness.
27
Previous study found that global loss
volume from GCC algorithm was the best parameter for
dierentiating between highly myopic glaucoma and
nonglaucoma subjects.
28
Despite its advantages, recent
study showed the RNFL in myopic eyes to be thinner
than the RFNL in normal eyes
23
, so the results of macular
GCC thickness may be confounded by these variations
of RNFL in myopic subjects.
When excluding RNFL measurement, another macular
parameter, GC-IPL thickness, was introduced. Previous
study reported that in highly myopic subjects, macular
GC-IPL thickness demonstrated a similar capability to
identify glaucoma as the RNFL thickness. In the highly
myopic group (SE -6.00 to -20.00 D), they found the SD-
OCT parameters with the highest AuROC values to be
inferior RNFL and inferotemporal GC-IPL thicknesses
(AuROC: 0.906, 0.852, respectively). In the non-highly
myopic eyes (SE -0.25 to -6.00 D), the best AuROC values
were average RNFL (AuROC: 0.920) and minimum
GC-IPL thicknesses (AuROC: 0.908).
19
Seol, et al. found
GC-IPL thickness at inferotemporal sector to be the
best parameter for detecting preperimetric glaucoma
in both the non-highly myopic (SE -0.50 to -6.00 D;
AuROC: 0.747) and highly myopic groups (SE ≤ -6.00
D; AuROC: 0.737).
20
ese results are consistent with
our ndings. In our study, minimum GC-IPL, inferior
RNFL, and average RNFL thicknesses (AuROC: 0.963,
0.925, and 0.919, respectively) were the parameters
with the best diagnostic performance in the moderately
myopic group. In the highly myopic group, average RNFL,
inferior RNFL, and inferotemporal GC-IPL thicknesses
(AuROC: 0.980, 0.962, and 0.905, respectively) were the
best parameters for detecting glaucoma.
e parameters that demonstrated the greatest
AuROC in the moderately myopic eyes (mean SE: -4.48
± 0.92 D) in our study were similar to those identied
in the non-highly myopic group (mean SE: -2.46 ± 1.69
D) in previous study
19
; however, greater AuROC values
were found in our study. Since increasing severity of
myopia temporally converges the inferior and superior
RNFL bundles and brings them closer to the macula
18
,
using macular parameters may be an appropriate tool
for detecting glaucomatous damage in eyes with a higher
severity of myopia. Moreover, previous study reported
the glaucoma diagnostic ability of macular GCA to be
Fig 2. e area under the receiver operating characteristic
(AuROC) curve for the 3 best parameters by Cirrus
HD-OCT for distinguishing between normal and
glaucomatous eyes in the high myopia group.
Abbreviations: RNFL; peripapillary retinal nerve ber
layer, GCIPL; macular ganglion cell-inner plexiform
layer.
Jiravarnsirikul et al.
Volume 74, No.5: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
291
Original Article
SMJ
inuenced by the angular distance between the RNFL
defect and the fovea, which means that defects located
far from the fovea may be dicult to detect with GCA
maps.
29
In the moderately myopic group, the AuROC
for minimum GC-IPL thickness was the highest in our
study, and higher than that reported from previous study,
and this may be due to a greater degree of myopia, as
discussed above.
In the high myopia group, we found average RNFL,
inferior RNFL, and inferotemporal GC-IPL thickness
to have the greatest AuROC values, which is consistent
with previous studies.
19,20
However, among all of the
SD-OCT parameters, the RNFL parameters showed
the best AuROC values. Despite the higher degree of
myopia, it remains unclear why the RNFL parameters
have superior diagnostic ability. Of interest, recent studies
reported several factors specic to highly myopic eyes
that may reduce the reliability of GC-IPL measurements.
Kim, et al. reported that in eyes with myopic tilted disc,
increasing degree of optic disc torsion corresponded
with distortion of the posterior pole contour
30
, which
may imply that GC-IPL measurement in high myopes
could be inuenced by anatomical misalignment in the
macular area. Furthermore, although GC-IPL thickness
assessments in highly myopic eyes demonstrated satisfying
long-term reproducibility, various factors, such as retinal
thinning caused by chorioretinal atrophy and posterior
staphyloma from myopia-related changes, may signicantly
inuence reproducibility.
31
In addition to thinning of the
macular region from stretching of the posterior globe,
abnormal macular thickening or macular retinoschisis
may also develop in highly myopic eyes.
32
ese factors
will eventually confound GC-IPL evaluation, so caution
must be exercised when interpreting GC-IPL thickness
in high myopia.
The current study demonstrated that all of the
parameters obtained by SD-OCT were thinner in
glaucomatous than in normal eyes in both moderately
and highly myopic patients. Among the normal eyes,
RNFL and GC-IPL thickness parameters were both lower
in high myopes than in moderate myopes. In contrast
to our ndings, Seo, et al. reported that temporal RNFL
thickness analyzed by Cirrus HD-OCT was signicantly
greater in highly myopic normal eyes than moderately
myopic normal eyes, which could be attributed to the
temporalization of RNFL when the degree of myopia
and axial length increase.
22
Further studies that include
more normal subjects in both groups may conrm these
hypotheses. inning of RNFL and GC-IPL thicknesses in
higher severity myopia could be artifactual ndings due
to the ocular magnication eect. A recent study showed
that the signicant negative correlation found between
OCT parameters (GC-IPL and RNFL thicknesses) and
axial length was the result of ocular magnication eect
since the area scanned by OCT was greater in elongated
globes.
33,34
erefore, further study using magnication
correction factors is needed to remedy this limitation,
and to avoid misdiagnosis of glaucoma in patients with
myopia.
Since the prevalence of myopia is increasing globally,
and especially in Asia
17
, data collection from various
population groups is needed to construct normative
databases for myopia, which will enhance our ability to
diagnose glaucoma in myopic patients. Previous studies
were mainly conducted in Korean, Japanese, and Chinese
participants.
14,18-20,22, 23,26
To the best of our knowledge, this
is the rst study to investigate the common structural
imaging parameters and their diagnostic abilities in
dierent severities of myopia in ai population. ese
results from dierent ethnic groups provide clinically
useful data that can be used to develop myopia-specic
normative databases for general population. One of the
strengths of this study is that we enrolled normal subjects
in both myopia severity groups so that AuROC values
could be generated to satisfactorily detect glaucoma among
myopic patients. However, according to the Hodapp,
Parish, and Anderson criteria
35
, the glaucomatous eyes
included in this study were classied as having early to
moderate defects, which means that our ndings may
not be applicable to eyes with greater or lesser severity of
glaucoma damage. Future study evaluating the diagnostic
performance of these parameters for dierentiating
dierent severities of glaucoma is warranted.
Limitations
There are certain limitations to this study that
should be mentioned. First, the study population size
was relatively small, and this may have given our study
insucient statistical power to identify all statistically
signicant dierences and associations between groups.
Additional studies in a larger number of participants
may be needed conrm and/or improve the validity
of our results. Second, GC-IPL thinning in myopic
glaucoma patients was not necessarily detected by
HD-OCT measurement. Because the Cirrus HD-OCT
equipment utilized in this research assessed macular
GC-IPL thickness inside a 6x6x2 mm cube centered at
the fovea, the presence of any glaucomatous defect away
from this area might not be detected by this algorithm,
as shown in Fig 3. Measurement of a larger eld of the
macula may optimize the glaucoma diagnosis, even in
patients with myopia.
Volume 74, No.5: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
292
Fig 3. An example of a moderate myopia patient, and a high myopia patient.
Moderately myopic patient (A-D). Right eye; SE -5.00 D, AL 26.68 mm, HVF MD -2.00 dB.
(A) Red-free fundus photograph showed multiple RNFL defects (arrow heads). Generalized RNFL thinning in the inferior retina was
observed as the choroidal vessels became clearly visible, so the RNFL defect was more easily detected in the superior hemield than in the
inferior hemield.
(B) Detection of inferior defect on GCIPL thickness map indicated that the superior defect was located away from the macular area, thus
the scanned area of the GCIPL thickness map was unable to include the superior defect into the analysis.
(C) e GCIPL deviation map was consistent with the thickness map.
(D) HVF pattern deviation map showed inferior arcuate scotoma and early superior nasal step both of which correspond with the RNFL
defects. Of note, GCA could not detect abnormality in the superior hemield despite mild inferior eld loss.
Highly myopic patient (E-H). Right eye; SE -6.00 D, AL 27.23 mm, HVF MD -7.20 dB.
(E) Red-free fundus photograph showed multiple RNFL defects (arrow heads) in both the superior and inferior hemields.
(F) e GCIPL thickness map was able to detect both superior and inferior hemield defect, and the inferior defect was much more severe
than superior defect.
(G) e ndings of the GCIPL deviation map were consistent with those of the GCIPL thickness map.
(H) HVF pattern deviation map showed double arcuate scotomas, and severe defect was found in the superior visual eld, which corresponds
with the observed RNFL defects and GCA.
CONCLUSION
Macular GC-IPL thickness demonstrated high
ability to detect glaucoma in patients with moderate or
high myopia and thus can enhance glaucoma diagnosis
in both groups of myopic patients. However, it should
be utilized in conjunction with a comprehensive clinical
examination, additional structural imaging modalities,
and functional assessments to elucidate a more precise
diagnosis of glaucoma in myopia.
ACKNOWLEDGEMENTS
e authors gratefully acknowledge the patients
that generously agreed to participate in this study, and
Assistant Professor Chulaluk Komoltri, DrPH (Biostatistics)
from the Oce for Research and Development, Faculty
of Medicine Siriraj Hospital, Mahidol University, for
assistance with statistical analysis.
Conict of interest declaration
All authors declare no personal or professional
conicts of interest, and no nancial support from the
companies that produce and/or distribute the drugs,
devices, or materials described in this report.
Funding disclosure
is was an unfunded study.
REFERENCES
1. Adelson JD, Bourne RRA, Briant PS, Flaxman SR, Taylor
HRB, Jonas JB, et al. Causes of blindness and vision impairment
in 2020 and trends over 30 years, and prevalence of avoidable
blindness in relation to VISION 2020: the Right to Sight: an
analysis for the Global Burden of Disease Study. Lancet Glob
Health. 2021;9(2):e144-e60.
2. Foster A, Resniko S. e impact of Vision 2020 on global
blindness. Eye. 2005;19(10):1133-5.
Jiravarnsirikul et al.
Volume 74, No.5: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
293
Original Article
SMJ
3. McMonnies CW. Glaucoma history and risk factors. J Optom.
2017;10(2):71-8.
4. Holden BA, Fricke TR, Wilson DA, Jong M, Naidoo KS,
Sankaridurg P, et al. Global Prevalence of Myopia and High
Myopia and Temporal Trends from 2000 through 2050.
Ophthalmology. 2016;123(5):1036-42.
5. Pan CW, Ramamurthy D, Saw SM. Worldwide prevalence and
risk factors for myopia. Ophthalmic Physiol Opt. 2012;32(1):
3-16.
6. Mitchell P, Hourihan F, Sandbach J, Wang JJ. e relationship
between glaucoma and myopia: the Blue Mountains Eye Study.
Ophthalmology. 1999;106(10):2010-5.
7. Pan CW, Cheung CY, Aung T, Cheung CM, Zheng YF, Wu RY,
et al. Dierential associations of myopia with major age-related
eye diseases: the Singapore Indian Eye Study. Ophthalmology.
2013;120(2):284-91.
8. Weinreb RN, Khaw PT. Primary open-angle glaucoma. e
Lancet. 2004;363(9422):1711-20.
9. Sommer A, Katz J, Quigley HA, Miller NR, Robin AL, Richter
RC, et al. Clinically detectable nerve ber atrophy precedes the
onset of glaucomatous eld loss. Arch Ophthalmol. 1991;109(1):
77-83.
10. Curcio CA, Allen KA. Topography of ganglion cells in human
retina. J Comp Neurol. 1990;300(1):5-25.
11. Tan O, Chopra V, Lu AT, Schuman JS, Ishikawa H, Wollstein
G, et al. Detection of macular ganglion cell loss in glaucoma
by Fourier-domain optical coherence tomography. Ophthalmology.
2009;116(12):2305-14.e1-2.
12. Guedes V, Schuman JS, Hertzmark E, Wollstein G, Correnti A,
Mancini R, et al. Optical coherence tomography measurement of
macular and nerve ber layer thickness in normal and glaucomatous
human eyes. Ophthalmology. 2003;110(1):177-89.
13. Mwanza JC, Oakley JD, Budenz DL, Chang RT, Knight OJ,
Feuer WJ. Macular ganglion cell-inner plexiform layer: automated
detection and thickness reproducibility with spectral domain-
optical coherence tomography in glaucoma. Invest Ophthalmol
Vis Sci. 2011;52(11):8323-9.
14. Kim KE, Park KH. Macular imaging by optical coherence
tomography in the diagnosis and management of glaucoma.
Br J Ophthalmol. 2018;102(6):718-24.
15. Hwang YH, Kim YY, Jin S, Na JH, Kim HK, Sohn YH. Errors in
neuroretinal rim measurement by Cirrus high-denition optical
coherence tomography in myopic eyes. Br J Ophthalmol. 2012;
96(11):1386-90.
16. Chang RT, Singh K. Myopia and glaucoma: diagnostic and
therapeutic challenges. Curr Opin Ophthalmol. 2013;24(2):96-
101.
17. Tan NYQ, Sng CCA, Jonas JB, Wong TY, Jansonius NM, Ang
M. Glaucoma in myopia: diagnostic dilemmas. Br J Ophthalmol.
2019;103(10):1347-55.
18. Leung CK-S, Yu M, Weinreb RN, Mak HK, Lai G, Ye C, et al.
Retinal Nerve Fiber Layer Imaging with Spectral-Domain
Optical Coherence Tomography: Interpreting the RNFL Maps
in Healthy Myopic Eyes. Invest Ophthalmol Vis Sci. 2012;
53(11):7194-200.
19. Choi YJ, Jeoung JW, Park KH, Kim DM. Glaucoma Detection
Ability of Ganglion Cell-Inner Plexiform Layer ickness by
Spectral-Domain Optical Coherence Tomography in High
Myopia. Invest Ophthalmol Vis Sci. 2013;54(3):2296-304.
20. Seol BR, Jeoung JW, Park KH. Glaucoma Detection Ability of
Macular Ganglion Cell-Inner Plexiform Layer Thickness
in Myopic Preperimetric Glaucoma. Invest Ophthalmol Vis
Sci. 2015;56(13):8306-13.
21. Knight ORJ, Girkin CA, Budenz DL, Durbin MK, Feuer WJ,
Cirrus OCT Normative Database Study Group . Eect of
Race, Age, and Axial Length on Optic Nerve Head Parameters
and Retinal Nerve Fiber Layer ickness Measured by Cirrus
HD-OCT. Arch Ophthalmol. 2012;130(3):312-8.
22. Seo S, Lee CE, Jeong JH, Park KH, Kim DM, Jeoung JW.
Ganglion cell-inner plexiform layer and retinal nerve ber layer
thickness according to myopia and optic disc area: a quantitative
and three-dimensional analysis. BMC Ophthalmol. 2017;17(1):22.
23. Akashi A, Kanamori A, Ueda K, Inoue Y, Yamada Y, Nakamura
M. e Ability of SD-OCT to Dierentiate Early Glaucoma
With High Myopia From Highly Myopic Controls and Nonhighly
Myopic Controls. Invest Ophthalmol Vis Sci. 2015;56(11):6573-
80.
24. European Glaucoma Society Terminology and Guidelines for
Glaucoma, 5th Edition. Br J Ophthalmol. 2021;105(Suppl 1):
1-169.
25. Kim KE, Jeoung JW, Park KH, Kim DM, Kim SH. Diagnostic
classication of macular ganglion cell and retinal nerve ber
layer analysis: dierentiation of false-positives from glaucoma.
Ophthalmology. 2015;122(3):502-10.
26. Kim YK, Yoo BW, Jeoung JW, Kim HC, Kim HJ, Park KH.
Glaucoma-Diagnostic Ability of Ganglion Cell-Inner Plexiform
Layer ickness Dierence Across Temporal Raphe in Highly
Myopic Eyes. Invest Ophthalmol Vis Sci. 2016;57(14):5856-63.
27. Kim NR, Lee ES, Seong GJ, Kang SY, Kim JH, Hong S, et al.
Comparing the ganglion cell complex and retinal nerve bre
layer measurements by Fourier domain OCT to detect glaucoma
in high myopia. Br J Ophthalmol. 2011;95(8):1115-21.
28. Shoji T, Sato H, Ishida M, Takeuchi M, Chihara E. Assessment of
Glaucomatous Changes in Subjects with High Myopia Using
Spectral Domain Optical Coherence Tomography. Invest
Ophthalmol Vis Sci. 2011;52(2):1098-102.
29. Hwang YH, Jeong YC, Kim HK, Sohn YH. Macular ganglion
cell analysis for early detection of glaucoma. Ophthalmology.
2014;121(8):1508-15.
30. Kim YC, Moon J-S, Park H-YL, Park CK. ree Dimensional
Evaluation of Posterior Pole and Optic Nerve Head in Tilted
Disc. Sci Rep. 2018;8(1):1121.
31. Lee M-W, Park K-S, Lim H-B, Jo Y-J, Kim J-Y. Long-term
reproducibility of GC-IPL thickness measurements using
spectral domain optical coherence tomography in eyes with
high myopia. Sci Rep. 2018;8(1):11037.
32. Benhamou N, Massin P, Haouchine B, Erginay A, Gaudric A.
Macular retinoschisis in highly myopic eyes. Am J Ophthalmol.
2002;133(6):794-800.
33. Francisconi CLM, Freitas AM, Wagner MB, Ribeiro RVP. Eects
of axial length on retinal nerve ber layer and macular ganglion
cell-inner plexiform layer measured by spectral-domain OCT.
Arquivos Brasileiros de Oalmologia (Online). 2020;83(4):269-76.
34. Huang D, Chopra V, Lu AT-H, Tan O, Francis B, Varma R, et al.
Does optic nerve head size variation aect circumpapillary
retinal nerve ber layer thickness measurement by optical
coherence tomography? Invest Ophthalmol Vis Sci. 2012;53(8):
4990-7.
35. Hodapp E, Parrish RK, Anderson DR. Clinical decisions in
glaucoma: Mosby Incorporated; 1993.
