Clinical science
Br J Ophthalmol: first published as 10.1136/bjophthalmol-2019-314696 on 15 November 2019. Downloaded from http://bjo.bmj.com/ on November 25, 2019 at Medizinische Lesehalle
1
Tajimi Iwase Eye Clinic, Tajimi,
Gifu, Japan
2
Ophthalmology, University of
the Ryukyus, Nakagami-­gun,
Okinawa, Japan
3
Ophthalmology, The University
of Tokyo Graduate School of
Medicine Faculty of Medicine,
Bunkyo-­ku, Tokyo, Japan
4
Kanto Central Hospital of the
Mutual Aid Association of Public
School Teachers, Setagaya-­ku,
Japan
Correspondence to
Aiko Iwase, Tajimi Iwase Eye
Clinic, Tajimi, Gifu, Japan; ​aiko-​
gif@​umin.​net
SS since deceased
Received 6 June 2019
Revised 11 September 2019
Accepted 2 November 2019
© Author(s) (or their
employer(s)) 2019. No
commercial re-­use. See rights
and permissions. Published
by BMJ.
To cite: Iwase A,
Sawaguchi S, Tanaka K, et al.
Br J Ophthalmol Epub ahead
of print: [please include Day
Month Year]. doi:10.1136/
bjophthalmol-2019-314696
Relationship between ocular risk factors for glaucoma
and optic disc rim in normal eyes
Aiko Iwase  ‍ ‍,1 Shoichi Sawaguchi,2 Kenji Tanaka,1 Tae Tsutsumi,3 Makoto Araie  ‍ ‍3,4
ABSTRACT
Aim  To study relationships between reported risk
factors for glaucoma and neuroretinal rim area in normal
eyes.
Methods  The Kumejima study participants, 3762 of
the 4632 eligible Kumejima residents 40 years and older,
underwent a detailed ocular examination including
sequential disc stereo photography. In a randomly chosen
eye of a subject whose both eyes met the inclusion
criteria, fundus photographs of 2474 ophthalmologically
normal eyes of the 2474 subjects were analysed by
computer-­assisted planimetry to measure the disc, rim
and β-peripapillary atrophy (PPA) areas. The rim was
divided into the superior and inferior halves by a line
connecting the fovea and disc centre.
Results  The disc, superior and inferior halves rim and
β-PPA areas averaged 2.53±0.50 (SD), 0.82±0.15,
0.84±0.16 and 0.45±0.66 mm2. After adjustment for
other systemic and ocular factors including age, disc
and β-PPA areas, disc–fovea distance (p=0.013, 0.016)
correlated positively and intraocular pressure (IOP)
(p=0.004, 0.006) and axial length (AL) (p<0.000, 0.004)
negatively with the superior and inferior halves rim area,
respectively; central corneal thickness (CCT) (p=0.008)
and mean blood pressure (mBP) (p=0.020) correlated
positively and male gender (p=0.012) negatively only
with the superior half rims.
Conclusions  Besides previously reported risk factors
for glaucoma such as age or IOP, thinner CCT, lower
mBP and male gender were newly found to significantly
correlate with smaller rim area only in the superior half
disc, and a greater disc–fovea distance with greater
superior and inferior half rim areas in normal adult eyes.
Introduction
Primary open-­angle glaucoma (POAG) is a progres-
sive morbidity of the optic nerve head (ONH) with
characteristic abnormalities of the neuroretinal rim
tissue. Correlations between peripapillary chorio-
retinal atrophy (PPA), especially the β-PPA area and
glaucomatous damage also have been reported.1
For example, the area or location of the β-PPA is
correlated with visual field (VF) and/or neuroretinal
rim loss,2–4 progression of glaucomatous damage5
and disc haemorrhage.6 Besides, β-PPA, higher age
and intraocular pressure (IOP),7 myopic refrac-
tion,8 9 thinner central corneal thickness (CCT),10
lower ocular perfusion pressure (OPP)11 and male
gender9 12 have been reported as factors related
to development and/or progression of glaucoma.
Further, recent studies have suggested a correla-
tion between the disc configuration, disc ovality or
tilt (maximal disc diameter/minimal disc diameter)
Iwase A, et al. Br J Ophthalmol 2019;0:1–5. doi:10.1136/bjophthalmol-2019-314696
and disc torsion, and progression of glaucomatous
damage.13 14
Previous population-­based studies have reported
significant correlations between the neuroretinal
rim area and age,15 16 myopic refraction or axial
length (AL),16 17 disc area,18IOP16 or body mass
index (BMI)19 in normal eyes, but correlations with
several of the above listed factors, such as CCT,
OPP, gender, β-PPA, disc ovality and disc torsion
have not been identified or estimated.
The Kumejima study is a population-­ based
epidemiologic study of 3762 participants who
focused on ocular diseases in Kumejima in south-
west Japan.9 Sequential stereo photographs were
obtained during the screening examination, and
Universitat Munchen. Protected by copyright.
the results were analysed using computer-­assisted
planimetry.20 In the current study of ophthalmo-
logically normal eyes of Kumejima study partic-
ipants, the effects of the previously mentioned
factors associated with glaucoma development and/
or progression5,7–14on the neuroretinal rim area
were re-­evaluated, directing attention onto those of
which correlations with rim area have not yet been
reported. Several previous studies have indicated
that the process of glaucomatous damage might
differ between the superior and inferior halves of
the optic disc.4 21 22 Therefore, the neuroretinal rim
was divided into the anatomic superior and inferior
halves of the neuroretinal rim by a line connecting
the fovea and disc centre (fovea–disc centre axis)
and analysed separately.
Methods
Population sampling
The Kumejima study was performed according
to the tenets of the Declaration of Helsinki and
regional regulations. All participants provided
written informed consent before the examinations.
All residents aged 40 years or older in Kumejima, a
southwest island in Okinawa, Japan, were encour-
aged to participate based on the official household
registration database,9 and 4632 residents were
eligible for the study.
Examinations and diagnosis
The general screening examination consisted of
a structured interview about occupation, health
history, surgery and trauma histories and smoking
habits, and body weight, height and systemic
blood pressure measurements. The ophthalmic
examinations included measurements of the best-­
corrected visual acuity (BCVA), refraction using
an auto-­ refractometer (ARK-730, Topcon), IOP
using a Goldmann applanation tonometer, CCT
1
 Clinical science
using a specular microscope (SP-2000, Topcon), anterior
chamber depth (ACD) and AL using the IOLMaster (Carl Zeiss
Meditec, Dublin, CA), and slit-­lamp examination, gonioscopy,
ophthalmoscopy, fundus photography and VF testing.9 A pair
of sequential stereoscopic ONH photographs at a parallax of
about 8 degrees (30 degree angle of view) and non-­stereoscopic
fundus photographs (45 degree angle of view) were obtained
using a digital non-­ mydriatic fundus camera (TRC-­ NW7,
Topcon, Tokyo, Japan) in both eyes of the subjects. The periph-
eral ACD was scored according to the van Herick method and
the gonioscopic findings according to Shaffer’s grading system.
The VF was examined using frequency-­ doubling technology
(FDT) perimetry with the C-20–1 screening programme (Carl
Zeiss Meditec).
Participants were referred for a definitive examination if they
were suspected of having ocular abnormalities including glau-
coma after meeting one or more of the following criteria during
the screening examination: corrected BCVA≤20/30, IOP ≥
19 mm Hg, vertical cup/disc (v-­C/D) ratio ≥ 0.6, superior (11–1
o’clock hours) or inferior (5–7 o’clock hours) rim width/disc
diameter ratio ≤0.2, bilateral asymmetry of the v-­C/D≥0.2, a
nerve fibre layer defect or disc haemorrhage, abnormal findings
on slit-­lamp examination or fundus photographs, angle width of
van Herick grade ≤2 and at least one abnormal test point in the
FDT C-20–1 test results. The definitive examination included
detailed slit-­lamp, gonioscopy, and fundus examinations and VF
testing with the Humphrey Field Analyzer Central 24–2 Swedish
interactive threshold algorithm standard programme (Carl Zeiss
Meditec).
The details of the disc, fundus, VF and glaucoma diagnosis
have been reported.9 The glaucoma diagnosis was based on
the clinical records obtained during all examinations and VFs,
and the International Society of Geographic and Epidemiologic
Ophthalmology Criteria.23
Planimetry on stereoscopic fundus photographs
The details of the current planimetric method were reported
previously.20 An experienced ophthalmologist (TT) re-­ eval-
uated all stereo photographs. While stereoscopically viewing
the optic disc, the clinical disc contour was determined by a
series of seven points with spline interpolation, and the cup
contour, defined as the point of change in the slope from the
cup wall to the neural rim, was determined as a closed curve
by an unlimited number of points placed on the computer
monitor using a computer mouse. The fovea also was deter-
mined. The disc contour was drawn automatically as an ellipse
fitted to the seven points with the least mean square method,
and the disc centre was calculated automatically as the centre
of gravity of the disc area. The β-PPA area was characterised by
visible sclera and large choroidal vessels owing to the absence
of retinal pigment epithelium and also determined as a closed
curve by an unlimited number of points placed on the outer
boundary of the β-zone and that of the peripapillary scleral
ring on a computer monitor. After correcting for magnifica-
tion by corneal curvature, AL and refractive error according to
the formula provided by the manufacturer,20 the planimetric
parameters, disc area, disc ovality (long axis of the ellipse fitted
to the disc contour/short one), disc torsion (deviation of the
long axis of the fitted ellipse from the reference line vertical to
the fovea–disc centre axis), the superior and inferior halves rim
areas divided by the fovea–disc centre axis, β-PPA area (mm2)
and its angular extent6 and the fovea–disc centre distance were
calculated automatically.
2 Iwase A, et al. Br J Ophthalmol 2019;0:1–5. doi:10.1136/bjophthalmol-2019-314696
Br J Ophthalmol: first published as 10.1136/bjophthalmol-2019-314696 on 15 November 2019. Downloaded from http://bjo.bmj.com/ on November 25, 2019 at Medizinische Lesehalle
Data analysis
The variables used for adjustment in the multiple regression
analysis for the superior or inferior half rim area were factors
previously reported as being related to glaucoma prevalence or
progression5 7–14 and those reported to be related to rim area in
population-­based studies,15–19 that is, height, BMI, age, gender,
mean OPP (2/3×mean brachial artery blood pressure—IOP),
CCT, IOP, AL, disc area, β-PPA area, disc ovality, disc torsion, the
fovea–disc centre distance which was reported to be correlated
with the parapapillary β/γ zone area24 and use of systemic antihy-
pertensive drugs, which reportedly affected rim area in subjects
without glaucoma25 and glaucoma progression.11 A familial
history of POAG was not adopted, since none of the normal
subjects involved in the current analysis did not report a familial
history of POAG, and mean (brachial artery) blood pressure
(mBP) instead of mean OPP was used in the current analysis,
since Pearson’s correlation coefficient between mean OPP and
mBP in the current subjects was 0.957 (p<0.001) and inclusion
of both IOP and mean OPP results in double counting of IOP in
a same equation. The data analyses were performed using SPSS
software (V.21.0J for Windows, SPSS Japan Inc., Tokyo, Japan);
p<0.05 was considered significant.
Universitat Munchen. Protected by copyright.
Results
In the Kumejima study, 3762 (participation rate, 81.2%) of
the 4632 eligible residents aged 40 years or older underwent
a screening examination. The 3762 participants were younger
than the 870 non-­participants (59.1±14.9 vs 61.8±14.0 years,
respectively; p<0.001, unpaired t-­test), and there were more
women than men among the participants (female/male ratio,
1929/1833 vs 315/555, respectively; p<0.001, χ2 test).18
Eyes excluded from the current analyses were eyes where
acceptable stereo fundus photographs could not be obtained
for various reasons (376 right and 421 left eyes), pseudopakic
or aphakic eyes (453 right and 443 left eyes), those with optic
disc diseases or anomalies or retinal or brain diseases (184 right
and 181 left eyes), those with spherical equivalent refraction <
–8.0 or >+5.0 diopters (14 right and 11 left eyes), those with
primary angle closure, glaucoma or glaucoma suspects (279 right
and 272 left eyes), non-­glaucomatous fellow eyes of subjects
with glaucoma or glaucoma suspect (90 right and 118 left eyes)
and eyes of which foveal location could not be determined with
confidence (six right and eight left eyes).
When both eyes of a subject met the inclusion criteria, one eye
was randomly chosen. As a result, 2474 eyes of 2474 subjects
(one randomly chosen eye of 2194 subjects and one eye of 280
subjects) were included for analyses (table 1).
The reproducibility of the disc and rim measurements using
the current planimetric method was reported.20 To estimate
the reproducibility of the β-PPA, disc torsion, disc ovality and
disc–fovea distance measurements, the current investigator (TT)
conducted a preliminary study in a separate group of 49 eyes
of 49 normal subjects of which 51% had β-PPA to evaluate the
intersession intraclass correlation coefficients. For the β-PPA
area, disc torsion, disc ovality and disc–fovea distance measure-
ments, it was 0.904 (95% CI 0.815 to 0.980), 0.896 (0.810
to 0.944), 0.776 (0.665 to 861) and 0.954 (0.927 to 0.983),
respectively.
The averages±SD of the demographic and planimetric data
of the current ophthalmologically normal Japanese subject eyes
were listed in table 1. Pearson’s correlation coefficients>0.60
were seen between the AL and spherical equivalent refraction
(r=0.601), and between the β-PPA area and its angular extent
 Clinical science

Br J Ophthalmol: first published as 10.1136/bjophthalmol-2019-314696 on 15 November 2019. Downloaded from http://bjo.bmj.com/ on November 25, 2019 at Medizinische Lesehalle
Table 1  Demographic data from eligible 2474 subject eyes Table 2  Factors significantly correlating with superior and/or
Women/men 1232/1242 inferior half rim area
Age (years) 57.3 (12.1) Superior half rim area Inferior half rim area
Height (cm) 156.1 (9.1) (mm2) (mm2)
Body mass index 25.1 (3.6) Coefficient P value Coefficient P value
Mean blood pressure (mm Hg) 99.4 (15.3) Age (years) −0.0016 <0.001 −0.0015 <0.001
Use of systemic antihypertensives 694/2474 Mean blood pressure 0.0005 0.020 0.0002 0.234
Intraocular pressure (mm Hg) 14.8 (3.0) (mm Hg)
Spherical equivalent error (diopters) 0.09 (1.69) Gender (male=0, 0.0088 0.012 0.0019 0.585
Axial length (mm) 23.4 (0.9) female=1)
Central corneal thickness (µm) 515 (33) Axial length (mm) −0.0071 <0.001 −0.0048 0.004
Disc area (mm )2
2.53 (0.50) CCT (mm) 0.189 0.008 0.0931 0.184
Rim area (mm2) 1.67 (0.30) IOP (mm Hg) −0.0026 0.004 −0.0025 0.006
 Superior-­half rim area* (mm ) 2
0.82 (0.15) Disc area (mm2) 0.181 <0.001 0.224 <0.001
 Inferior-­half rim area* (mm2) 0.84 (0.16) Disc–fovea distance (mm) 0.0191 0.013 0.0182 0.016
β-PPA area (mm ) 2
0.45 (0.66) Use of systemic 0.0018 0.562 0.0009 0.766
antihypertensives
 Superior-h­ alf β-­PPA area* (mm2) 0.19 (0.33)
Height (cm) −0.0004 0.319 −0.0002 0.566
 Inferior-­half β-­PPA area* (mm2) 0.26 (0.38)
Body mass index 0.0001 0.855 0.0009 0.233
β-­PPA angular extent (degrees) 95 (86)
Disc ovality −0.0195 0.647 0.0356 0.395
Disc ovality 1.11 (0.06)
Disc torsion (degrees) 0.00002 0.819 −0.00004 0.581
Disc–fovea distance (mm) 4.7 (0.3)
Superior-h­ alf β-PPA area 0.0116 0.166  
Disc torsion (degrees)† –17.6 (35.0)
(mm2)

Universitat Munchen. Protected by copyright.

The data are expressed as the mean (SD). The values of systemic parameters
and ocular parameters were obtained from 2474 eyes from 2474 subjects (one
randomly chosen eye of 2194, 166 right eyes of 166 subjects, and 114 left eyes of
114 subjects). Disc ovality = long axis of the ellipse fitted to the disc contour/the
short axis.
*Divided by a fovea–disc center axis as a reference line.
†A positive value indicates inferotemporal torsion and a negative values indicates
superotemporal torsion.
β-PPA, β-peripapillary atrophy.
(r=0.821). Thus, spherical equivalent refraction and β-PPA’s
angular extent were excluded from explanatory variables.
In normal Japanese eyes, older age, longer AL and higher
IOP significantly negatively affected the superior (p<0.001,
p<0.001, p=0.004) and inferior halves rim areas (p<0.001,
p=0.004, p=0.006), while greater disc area and disc–fovea
distance significantly positively affected the superior (p<0.001,
p=0.013) and inferior halves rim areas (p<0.001, p=0.016).
On the other hand, higher mBP (p=0.020) and thicker CCT
(p=0.008) significantly positively and male gender (p=0.012)
negatively affected only the superior half rim area (table 2).
Discussion
The current mean disc, rim and β-PPA area in the 2474 ophthal-
mologically normal Japanese subjects averaged 2.53, 1.67 and
0.45 mm2, which were within the range of those photograph-
ically determined in previous population-­based studies.15–17 19
The current study showed that the rim area in otherwise normal
eyes was affected by many of the reported factors associated with
glaucoma development and/or progression, and that impacts of
some of them were different between the superior and inferior
half rims.
Age and IOP are the most evident risk factors for glaucoma.7
Population-­based studies using laser scanning tomography15 or
optical coherence tomography (OCT)16 found that older subjects
with normal eyes had a smaller rim area, and the IOP correlated
negatively with the rim area in normal eyes. The current study
photographically confirmed those results. A prospective longi-
tudinal study has shown that lower OPP is a risk factor for the
progression of glaucoma.11 The current study first showed that a
Iwase A, et al. Br J Ophthalmol 2019;0:1–5. doi:10.1136/bjophthalmol-2019-314696
Inferior-­half β-PPA area   −0.0145 0.056
(mm2)
CCT, central corneal thickness; IOP, intraocular pressure; β-PPA, β-peripapillary
atrophy.
smaller superior half rim area, but not the inferior half rim area,
was associated significantly with lower mBP which should have a
very high correlation with OPP (r=0.957 in the current subjects)
in normal subjects. Previous studies have shown that the lower
hemifield is more likely to be damaged in patients with POAG
and diabetes mellitus21 and in those with normal IOP and find-
ings indicative of ischaemic changes in brain magnetic resonance
image.26 Further, it was recently reported in POAG that the
tissue blood velocity in the superior sector of the ONH showed
significant correlation with the change rate of VF damage in the
corresponding VF, while that in the inferior sector of the ONH
did not.27 It is known that non-­arteritic ischaemic optic neurop-
athy is more likely to be associated with inferior VF damage,28 29
which coincided with the circle of Haller and Zinn morpholog-
ically provided an altitudinal blood supply to the retrolaminar
optic nerve.30 Taken together, the current result may suggest that
superior-­half rim is relatively more perfusion-­dependent, and
that normal eyes with a smaller superior rim area are relatively
more vulnerable to perfusion-­related insults. The current study
also first found that men had a significantly smaller superior rim
area, which agreed with the results of population-­based studies
that male gender is a related factor for POAG.9 12 A thinner
central cornea is also an important risk factor for glaucoma.10
The current study is the first to report that a thinner CCT signifi-
cantly correlated with a smaller superior half rim area in other-
wise normal eyes. Although the exact mechanisms for a thinner
CCT as a risk factor for glaucoma development10 are unknown,
a significant association of a thinner central cornea, male gender
and low OPP with a smaller superior half rim area after adjust-
ment for other confounding factors in normal eyes may suggest
that the superior half rim is more likely to be vulnerable to glau-
comatous insults related to these non-­IOP factors than the infe-
rior half rim.
3
 Clinical science
Because of the close correlation between the refraction and AL
and relative independence of the AL from the optical properties
of the cornea and lens, we adopted the AL as an explanatory vari-
able in the current analysis. Previous reports on the effect of AL or
myopic refraction on the rim area in normal eyes yielded conflicting
results; some reported that a larger rim area was associated with
myopic refraction1 or longer AL,17 while another reported a nega-
tive correlation between the AL and rim area.15 The AL has a
strong effect on the magnification of the fundus images, and the
measurements of the fundus structures are relatively smaller in eyes
with a longer AL, if the magnification correction of the fundus
photograph is incomplete. The current study where the correction
of the magnification considered the refraction, corneal curvature
and AL20 confirmed that a longer AL was associated with a smaller
rim area in the superior and inferior half discs in normal eyes,
which agreed with the results of population-­based studies that
reported the relative vulnerability of myopic eyes to glaucoma.8 9 In
contrast to AL elongation, a greater disc–fovea distance positively
affected the superior and inferior half rim areas. Positive correla-
tion between disc–fovea distance and rim area seems paradoxical,
since AL which showed negative correlation with rim area report-
edly showing positive correlation with disc–fovea distance.24 In the
current subjects, it was also confirmed that disc–fovea distance and
AL had positive correlation (r=0.111, p<0.001). This paradoxical
correlation may be partly explained by assuming that more elon-
gated AL of an eye ball makes the angle between disc and fovea in a
cross-­section steeper and consequently making apparent length of
disc–fovea distance estimated on a film of a fundus camera shorter.
But, this speculation is not compatible with positive correlation
between AL and apparent disc–fovea distance found in the current
subjects. Several unknown factors may be involved in the rela-
tionship between disc–fovea distance and AL or rim area, and it
seems difficult to further elucidate this relationship only using
planimetric data of fundus photos. Methods allowing more direct
measurement of disc–fovea distance and neural components of the
ONH would be needed to further elucidate this apparently para-
doxical relationship between disc–fovea distance and rim and/or
AL. Practical implication of the current result of significant posi-
tive correlation between disc–fovea distance rim would be that the
disc–fovea distance is advised to corrected for interindividual or
intergroup comparison of the neuroretinal rim area.
A limitation of the Kumejima study was the substantial differ-
ence in the female-­to-­male ratio between the participants and non-­
participants (1.05 vs 0.57).9 The mean age of the current subjects
(56.8 years) was younger than that of the Kumejima study partic-
ipants (61.8 years),9 since they were selected because of the avail-
ability of good-­quality disc stereo photographs. These biases might
be at least partly managed by including gender and age as covari-
ates in the multivariate analysis. The currently used β-PPA area was
determined photographically and thus probably included γ-PPA
without Bruch’s membrane.24 β-PPA area should have been better
evaluated using OCT, which was unfortunately unavailable, when
the Kumejima Study was conducted. In routine clinical practice,
however, the correlation between known glaucoma risk factors and
photographically determined β-PPA size would be still of practical
and clinical use for physicians, since it seems rather exceptional to
adopt OCT-­based β-PPA in routine clinical practices.
In summary, we evaluated the effects of reported factors asso-
ciated with glaucoma development and/or progression on the
neuroretinal rim area in ophthalmologically normal eyes in a
population-­ based setting. A smaller rim area correlated signifi-
cantly with older age, smaller disc area, higher IOP and longer AL,
and a smaller superior half rim area with lower mBP, thinner CCT
and male gender after adjustment for other confounding factors,
4 Iwase A, et al. Br J Ophthalmol 2019;0:1–5. doi:10.1136/bjophthalmol-2019-314696
Br J Ophthalmol: first published as 10.1136/bjophthalmol-2019-314696 on 15 November 2019. Downloaded from http://bjo.bmj.com/ on November 25, 2019 at Medizinische Lesehalle
and the effects of CCT, mBP and gender were those first identi-
fied in normal eyes and different between the superior and infe-
rior rims. Further, a greater disc–fovea distance was first found to
correlate with greater superior and inferior halves rim area. These
results may be useful both in screening subjects being at higher risk
for developing POAG in future from general population and in
studying pathogenesis of onset of POAG.
Contributors  AI: conceptualisation, data curation, formal analysis, investigation,
methodology, writing—original draft, writing—review and editing. SS:
conceptualisation, data curation, funding acquisition, investigation, methodology,
project administration, resources. KT: data analysis. TT: data analysis. MA:
conceptualisation, data curation, formal analysis, funding acquisition, investigation,
methodology, project administration, resources, writing—original draft, writing—
review and editing.
Funding  Grant-­in-­Aid for Scientific Research by the Ministry of Health, Labour
and Welfare of Japan (H18-­Sensory-­General-001) and a Grant-­in-­Aid for Scientific
Research (C) 17591845 from the Ministry of Education, Culture, Sports, Science and
Technology, Japan, and funds from the Japan National Society for the Prevention of
Blindness, Tokyo.
Competing interests  None declared.
Patient consent for publication  Not required.
Ethics approval  The ethics board of the regional council approved the study
protocol.
Provenance and peer review  Not commissioned; externally peer reviewed.
Universitat Munchen. Protected by copyright.
Data availability statement  The data are from the Kumejima Study and are
owned by the Japanese Glaucoma Society (JGS). Data access is strictly restricted
to researchers who are members of the JGS and are accepted by the society. Data
is restricted to JGS members due to a contract agreed between Kumejima Town
and the JGS, following the municipal law of Kumejima Town for protecting private
information. To manage the epidemiological data which the JGS gathered, the JGS
has the Epidemiology Study Group Data Center (JGS Data Analysis Group, Tajimi
Branch),
ORCID iDs
Aiko Iwase http://​orcid.​org/​0000-​0001-​5950-​4260
Makoto Araie http://​orcid.​org/​0000-​0001-​9025-​3973
References
1 Jonas JB, Nguyen XN, Gusek GC, et al. Parapapillary chorioretinal atrophy in
normal and glaucoma eyes. I. Morphometric data. Invest Ophthalmol Vis Sci
1989;30:908–18.
2 Jonas JB, Gründler AE. Correlation between mean visual field loss and
morphometric optic disk variables in the open-­angle glaucomas. Am J Ophthalmol
1997;124:488–97.
3 Emdadi A, Kono Y, Sample PA, et al. Parapapillary atrophy in patients with focal visual
field loss. Am J Ophthalmol 1999;128:595–600.
4 Kawano J, Tomidokoro A, Mayama C, et al. Correlation between hemifield visual field
damage and corresponding parapapillary atrophy in normal-­tension glaucoma. Am J
Ophthalmol 2006;142:40–5.
5 Teng CC, De Moraes CGV, Prata TS, et al. β-zone parapapillary atrophy and the
velocity of glaucoma progression. Ophthalmology 2010;117:909–15.
6 Sugiyama K, Tomita G, Kitazawa Y, et al. The associations of optic disc hemorrhage
with retinal nerve fiber layer defect and peripapillary atrophy in normal-­tension
glaucoma. Ophthalmology 1997;104:1926–33.
7 Ernest PJ, Schouten JS, Beckers HJ, et al. An evidence-­based review of prognostic
factors for glaucomatous visual field progression. Ophthalmology 2013;120:512–9.
8 Marcus MW, de Vries MM, Junoy Montolio FG, et al. Myopia as a risk factor for
open-­angle glaucoma: a systematic review and meta-­analysis. Ophthalmology
2011;118:1989–94.
9 Yamamoto S, Sawaguchi S, Iwase A, et al. Primary open-­angle glaucoma in a
population associated with high prevalence of primary angle-­closure glaucoma.
Ophthalmology 2014;121:1558–65.
10 Gordon MO, Torri V, Miglior S, et al. Validated prediction model for the development of
primary open-­angle glaucoma in individuals with ocular hypertension. Ophthalmology
2007;114:10–19.
11 De Moraes CG, Liebmann JM, Greenfield DS, et al. Risk factors for visual field
progression in the low-­pressure glaucoma treatment study. Am J Ophthalmol
2012;154:702–11.
12 Rudnicka AR, Mt-­Isa S, Owen CG, et al. Variations in primary open-­angle glaucoma
prevalence by age, gender, and race: a Bayesian meta-­analysis. Invest Ophthalmol Vis
Sci 2006;47:4254–61.

13 Park H-­YL, Lee K, Park CK. Optic disc torsion direction predicts the location
of glaucomatous damage in normal-­tension glaucoma patients with myopia.
Ophthalmology 2012;119:1844–51.
14 Lee JE, Sung KR, Lee JY, et al. Implications of optic disc tilt in the progression of
primary open-­angle glaucoma. Invest Ophthalmol Vis Sci 2015;56:6925–31.
15 Abe H, Shirakashi M, Tsutsumi T, et al. Laser scanning tomography of optic discs
of the normal Japanese population in a population-­based setting. Ophthalmology
2009;116:223–30.
16 Cheung CY, Chen D, Wong TY, et al. Determinants of quantitative optic nerve
measurements using spectral domain optical coherence tomography in a
population-­based sample of non-­glaucomatous subjects. Invest Ophthalmol Vis Sci
2011;52:9629–35.
17 Wu R-­Y, Wong T-Y­ , Zheng Y-­F, et al. Influence of refractive error on optic disc
topographic parameters: the Singapore Malay eye study. Am J Ophthalmol
2011;152:81–6.
18 Bourne RRA, Foster PJ, Bunce C, et al. The morphology of the optic nerve head in the
Singaporean Chinese population (the Tanjong Pagar study): part 1--optic nerve head
morphology. Br J Ophthalmol 2008;92:303–9.
19 Zheng Y, Cheung CYL, Wong TY, et al. Influence of height, weight, and body mass
index on optic disc parameters. Invest Ophthalmol Vis Sci 2010;51:2998–3002.
20 Saito H, Tsutsumi T, Iwase A, et al. Correlation of disc morphology quantified on
stereophotographs to results by Heidelberg retina tomograph II, GDx variable corneal
compensation, and visual field tests. Ophthalmology 2010;117:282–9.
21 Zeiter JH, Shin DH, Baek NH. Visual field defects in diabetic patients with primary
open-­angle glaucoma. Am J Ophthalmol 1991;111:581–4.
Iwase A, et al. Br J Ophthalmol 2019;0:1–5. doi:10.1136/bjophthalmol-2019-314696
Clinical science
Br J Ophthalmol: first published as 10.1136/bjophthalmol-2019-314696 on 15 November 2019. Downloaded from http://bjo.bmj.com/ on November 25, 2019 at Medizinische Lesehalle
22 Stirling RJ, Pawson P, Brimlow GM, et al. Patients with ocular hypertension have
abnormal point scotopic thresholds in the superior hemifield. Invest Ophthalmol Vis
Sci 1996;37:1608–17.
23 Foster PJ, Buhrmann R, Quigley HA. The definition and classification of glaucoma in
prevalence surveys. Br J Ophthalmol 2002;86:238–42.
24 Jonas RA, Wang YX, Yang H, et al. Optic Disc-­Fovea distance, axial length and
parapapillary zones. The Beijing eye study 2011. PLoS One 2015;10:e0138701.
25 Topouzis F, Coleman AL, Harris A, et al. Association of blood pressure status with
the optic disk structure in non-­glaucoma subjects: the Thessaloniki eye study. Am J
Ophthalmol 2006;142:60–7.
26 Suzuki J, Tomidokoro A, Araie M, et al. Visual field damage in normal-­tension
glaucoma patients with or without ischemic changes in cerebral magnetic resonance
imaging. Jpn J Ophthalmol 2004;48:340–4.
27 Kiyota N, Shiga Y, Yasuda M, et al. Sectoral differences in the association of optic
nerve head blood flow and glaucomatous visual field defect severity and progression.
Invest Ophthalmol Vis Sci 2019;60:2650–8.
28 Hayreh SS, Zimmerman B. Visual field abnormalities in nonarteritic anterior
ischemic optic neuropathy: their pattern and prevalence at initial examination. Arch
Ophthalmol 2005;123:1554–62.
29 Kerr NM, Chew SSSL, Danesh-­Meyer HV. Non-­arteritic anterior ischaemic optic
neuropathy: a review and update. J Clin Neurosci 2009;16:994–1000.
30 Olver JM, Spalton DJ, McCartney AC. Microvascular study of the retrolaminar optic
nerve in man: the possible significance in anterior ischaemic optic neuropathy. Eye
1990;4:7–24.
Universitat Munchen. Protected by copyright.
5