Optic Disc Evaluation in Optic Neuropathies

The Optic Disc Assessment Project

Evelyn C. O’Neill, MRCOphth, MSc,1 Helen V. Danesh-Meyer, FRANZCO, MD,3 George X. Y. Kong, MBBS,1
Alex W. Hewitt, MBBS, PhD,1 Michael A. Coote, FRANZCO,1 David A. Mackey, FRANZCO, MD,1,2
Jonathan G. Crowston, FRANZCO, PhD,1 on behalf of the Optic Nerve Study Group*

Objective: Optic nerve morphology is affected by genetic and acquired disease. Glaucoma is the most
common optic neuropathy; autosomal-dominant optic atrophy (ADOA) and Leber’s hereditary optic neuropathy
(LHON) are the most prevalent hereditary optic neuropathies. These 3 entities can exhibit similar topographical
changes at the optic nerve head. Both ADOA and LHON have been reported to be misdiagnosed as glaucoma.
Our aim was to determine whether glaucoma subspecialists and neuro-ophthalmologists can distinguish these
diagnoses on optic disc assessment alone.
Design: Observational study.
Participants: Twenty-three optic nerve experts.
Methods: We randomized and masked 60 high-resolution stereoscopic optic disc photographs (15 ADOA
images, 15 LHON, 15 glaucoma, and 15 normal controls). Experts were asked to assess the discs on 12
conventional topographic features and assign a presumptive diagnosis. Intra- and interanalysis was performed
using the index of qualitative variation and absolute deviation.
Main Outcome Measures: Can glaucoma specialists and neuro-ophthalmologists distinguish among the
disease entities by optic nerve head phenotype.
Results: The correct diagnosis was identified in 85%, 75%, 27%, and 16% of the normal, glaucoma, ADOA, and
LHON disc groups, respectively. The proportion of correct diagnoses within the ADOA and LHON groups was
significantly lower than both normal and glaucomatous (P⬍0.001). Where glaucoma was chosen as the most likely
diagnosis, 61% were glaucomatous, 34% were pathologic but nonglaucomatous discs, and 5% were normal. There
was greater agreement for individual parameters assessed within the normal disc set when compared with pathologic
discs (P⬍0.05). The only parameter to have a significantly greater agreement within the glaucomatous disc set when
compared with ADOA or LHON disc sets was pallor, whereby experts agreed on is absence in the glaucomatous discs
but were not in agreement on its presence or its absence in the ADOA and LHON discs (P⬍0.01).
Conclusions: Optic neuropathies can result in similar topographic changes at the optic disc, particularly in
late-stage disease, making it difficult to differentiate ADOA and LHON from glaucoma based on disc assessment
alone. Other clinical parameters such as acuity, color vision, history of visual loss, and family history are required
to make an accurate diagnosis.
Financial Disclosure(s): The authors have no proprietary or commercial interest in any of the materials
discussed in this article. Ophthalmology 2011;118:964 –970 © 2011 by the American Academy of Ophthalmology.

*Group members listed online in Appendix 1 (available at http://aaojournal.org).

Optic neuropathies are a degenerative process of the optic
nerve and can be hereditary or acquired.1Among the inher-
ited optic neuropathies, the expression of the same genetic
defect can often lead to diverse phenotypes, and, con-
versely, different genetic defects can cause similar or even
identical phenotypes.1 Glaucoma is the most common ac-
quired optic neuropathy with autosomal-dominant optic at-
rophy (ADOA) and Leber’s hereditary optic neuropathy
(LHON), the most common hereditary optic neuropa-
thies.1,2 Optic nerve head rim loss has been reported in both
inherited LHON and ADOA and is also a feature of ac-
quired optic nerve disease.3-6
Worldwide, 4.5 million people are functionally blind
from glaucoma.7 Primary open-angle glaucoma (POAG) is
the most prevalent type of glaucoma. Evidence is accumu-
lating that glaucoma is inherited as a complex trait, in which
phenotype expression results from interactions between
genes and environmental factors.8-10 In one third of POAG
cases, intraocular pressure (IOP) remains within a statisti-
cally normal range, and patients are defined as having
normal tension glaucoma (NTG).11,12
Clinically, glaucoma is an optic neuropathy with specific
characteristic optic nerve head and visual field changes.
Morphologically, glaucomatous optic atrophy is classically
associated with neuroretinal rim loss with enlargement and
excavation of the cup area, manifesting as an increased
cup:disc ratio (CDR) and corresponding retinal nerve fiber
layer (RNFL) defects and reduced visibility of the RNFL,

964 © 2011 by the American Academy of Ophthalmology ISSN 0161-6420/11/$–see front matter
Published by Elsevier Inc. doi:10.1016/j.ophtha.2010.09.002
 O’Neill et al 䡠 Optic Disc Assessment in Optic Neuropathies: ODAP Project
optic disc hemorrhages, and angulations in the course of
existing blood vessels. The remaining neuroretinal rim ap-
pears pink.1,13
Clinically, LHON and ADOA share several characteris-
tics. Despite remarkable differences in the natural history of
these 2 diseases, their clinical end points are very similar.
Clinically, in both there is optic atrophy and pallor with
severe loss of central vision and relative sparing of the
peripheral field. This is due to the selective and early
involvement of the papillomacular bundle. In both LHON
and ADOA, as with other nonglaucomatous optic neuropa-
thies, the severity of optic disc pallor generally depends on
the extent of loss of the papillomacular bundle. Nonglauco-
matous rim loss and optic nerve head excavation has fre-
quently been reported as a sign of chronicity in both dis-
eases.4,14,15 In some instances, this has been reported in up
to 89% of ADOA subjects.3
Glaucoma, LHON, and ADOA all have multiple over-
lapping clinical features. Specifically, in late disease, these
3 clinical entities all exhibit similar topographic changes at
the optic nerve head. Previously, cases in which both LHON
and ADOA have been misdiagnosed as NTG on disc ap-
pearance alone have been documented3,15,16 and LHON has
been reported to develop in a family with a previous diag-
nosis of NTG.17 Few studies, however, specifically deter-
mined whether disease entities can be distinguished on the
basis of optic nerve appearance alone.
The purpose of this study was to highlight the similarities
in anatomic changes and the difficulties in accurately dis-
tinguishing between these forms of neuropathy on ophthal-
moscopy alone. We specifically asked whether it is possible
for glaucoma specialists and neuro-ophthalmologists to dis-
tinguish among the disease entities by optic nerve head
phenotype.
Materials and Methods
Participants
Thirty clinicians from the Americas, Europe, and Australasia with
expertise in optic disc examination (glaucoma specialists and
neuro-ophthalmologists) were invited to participate in this evalu-
ation. Twenty-three participants completed the disc assessment.
Before commencing the assessment, all participants were given
instructions on receiving their access codes to the restricted site.
They were given no details of medical history, other examination
findings, or test results. There was no time restriction enforced for
image review. Examiners were not told of the percentage of
individuals with each diagnosis.
Image Collection and Selection
Simultaneous stereoscopic optic disc photographs were taken (Ni-
dek fundus camera 3-Dx/F; Nidek, Gamagori, Japan) and then
digitalized at a high resolution (2102⫻1435 pixels; 2900 pixels per
inch; 16-bit color) using a Nikon CoolScan IV ED slide scanner
(Nikon Corp, Tokyo, Japan). All images were collected with full
informed consent and taken under pharmacologic mydriasis. Insti-
tutional review board approval was obtained for this project and
the research was conducted in accordance with the Declaration of
Helsinki.
The image set of 60 color stereoscopic optic disc images
included 15 normal subjects, 15 glaucoma subjects, 15 ADOA
subjects, and 15 LHON subjects selected in a random, masked
fashion from spreadsheet lists from 4 image banks of normal
controls, clinically proven glaucoma, and clinically and genetically
proven ADOA and LHON (Fig 1; available online at http://
aaojournal.org). Selection was from numerical lists with no prior
viewing of the images.
Our glaucoma disc images were from a list of patients with
functional evidence of glaucoma (i.e., visual field loss consistent
with glaucoma) as per the paper by Foster et al on glaucoma
definition18 and a Glaucoma Inheritance Study Tasmania (GIST)
severity score of ⬎0.70. As previously described, the GIST sever-
ity score assigns a value to the findings of optic disc assessment,
visual field deficit and elevated IOP whereby scores of 0.7, 0.8,
0.9, and 1.0 confer disease of mild, moderate, severe, and very
severe phenotypes, respectively.19 Given the de-emphasis of IOP
as a diagnostic criterion,18 GIST severity scores were also calcu-
lated with the IOP points removed. All glaucoma cases included in
this study were of moderate, severe, or very severe phenotypes.
Our ADOA and LHON images were from a list of subjects with
clinically diagnosed and genetically confirmed disease and with
stable visual acuities of ⱕ6/24. The average time from diagnosis to
image acquisition among the ADOA group was 30.5 years (range,
10 – 69). The visual acuities of the ADOA patients at time of image
acquisition ranged from 6/24 to count fingers. The average time
from image acquisition after acute vision loss among the LHON
group was 16.8 years (range, 6 months to 54 years). All ADOA
and LHON patients included had stable chronic disease with visual
acuities of ⱕ6/24. Each ADOA patient harbored a pathogenic
OPA1 mutation as described previously.20 All identifying infor-
mation was removed from each stereoscopic image before grading.
Exclusion within any of the groups was based on poor image
quality alone.
Optic Disc Evaluation in Optic Neuropathies
System
An online, Internet-based, restricted-access system using Cold-
fusion (Adobe Systems, San Jose, CA) and a database Structured
Query Language technology was developed to deliver the disc
images and disc evaluation outcomes to the participants. The
previously acquired images were stored as low-compression (high-
quality) JPEG images of equal size (1200⫻970 pixels) and mag-
nification. The images were shown to participants via web browser
at recommended screen resolution of 1280⫻960 pixels at 32-bit
color (Fig 2; available online at http://aaojournal.org). The images
were then viewed on screen using a handheld Screen-Vu Stereo
Viewer (Eyesupply USA, Inc., Tampa, FL) or with adjustable
3-dimensional prism glasses (Berezin, Inc., Mission Viejo, CA).
For each disc image (and masked to diagnosis), participants
were asked to assess the following optic disc characteristics: disc
size, disc shape, disc tilt, degree of parapapillary atrophy, vertical
CDR, cup shape, cup depth, RNFL loss, the presence or absence of
hemorrhages, neuroretinal rim pallor, and a gray crescent, and to
give a presumptive diagnosis. All variables were rated according to
the categorical rating scale detailed in Table 1 (available online at
http://aaojournal.org). Participants were then asked to state the
“most likely” diagnosis. All answers were recorded verbatim im-
mediately after completion of each disc image.
Statistical Analysis
An intra- and interanalysis of the 4 clinical subgroups (normal,
glaucomatous, ADOA, and LHON) was performed. To analyze the
differential spread of responses for individual discs, the index of
965
 Ophthalmology Volume 118, Number 5, May 2011

qualitative variation was used to measure statistical dispersion for

nominal variables (diagnosis, disc shape, tilt, cup shape, RNFL
loss, hemorrhages, pallor, and gray crescent) and the absolute
deviation from the mode was used for ordinal variables (disc size,
degree of parapapillary atrophy, vertical CDR, cup depth). In brief,
index of qualitative variation is a numerical value ranging between
0 and ⫹1 that indicates the amount of dispersion of values in a
nominal categorical variable.21 Absolute deviation is the average
of the deviation of all values away from the mode, thereby giving
a numerical indicator of spread for ordinal variables. Larger values
of index of qualitative variation or absolute deviation indicate
greater spread among expert observers in their assessment of the
studied disc characteristic. Significant dispersions or deviations
were expressed as a P-value. All statistical analysis was performed
with SPSS operating software (version 15.00; SPSS Inc., Chicago,
IL).

Results
The correctly assigned diagnosis on disc appearance alone was
identified in 85%, 75%, 27%, and 16% of the normal, glaucoma,
ADOA, and LHON disc subgroups, respectively (Fig 3A). There
was no difference between the 2 expert groups in the proportions
of correctly assigned diagnoses (Fig 3B). Neuro-ophthalmologists
assigned the correct diagnosis in 83%, 74%, 27%, and 16% of the
normal, glaucoma, ADOA, and LHON disc subgroups, respec-
tively, and glaucoma specialists in 87%, 76%, 26%, and 15% of
the normal, glaucoma, ADOA, and LHON disc subgroups, respec-
tively. The proportion of correct diagnoses within the ADOA and
LHON groups was significantly lower than both the normal
and glaucoma groups (P⬍0.001; Fig 3A). However, the ADOA and
LHON discs were correctly identified as abnormal in 84% and
95% of cases, respectively.
Within the ADOA and LHON subgroups, where experts com-
mitted to an alternate diagnosis, the diagnoses chosen in the
ADOA group were glaucoma, normal, and LHON at 48%, 33%,
and 19%, respectively. Of the misclassified LHON optic discs, the
proportion of alternate diagnoses nominated were ADOA, glau-
coma, and normal at 52%, 39%, and 9%, respectively (Fig 3C).
Thus, glaucoma accounted for a substantial proportion of alternate
diagnosis in both of the inherited optic neuropathy groups. There
was no difference between the 2 expert groups (P⬎0.05).
Of the images where glaucoma was selected as the most likely
diagnosis, 61% did actually have confirmed glaucoma. Of the
remaining, 19% were ADOA discs, 15% were LHON discs, and
5% were normal discs (Fig 4A). There was no difference between
glaucoma experts and neuro-ophthalmologists. Among the glau-
coma specialists, where glaucoma was chosen as the most likely
diagnosis, 56% were glaucoma discs and 44% were nonglauco-
matous. Among the neuro-ophthalmologists, 65% were glaucoma
discs and 35% were nonglaucomatous. Thus, glaucoma experts
tended to classify discs as glaucomatous among nonglaucomatous
discs more often than neuro-ophthalmologists, but this difference
was not significant (P ⫽ 0.91; Fig 4B). Figure 3. A, Proportion of correct and incorrect diagnoses in each sub-
Assessment of individual disc parameters revealed higher lev- group of disease for all consultants. (Correct in 85% of normal discs, 75%
els of agreement among the expert group within the normal disc set of glaucomatous discs, 27% of ADOA discs and 16% of LHON discs.) B,
when compared with pathologic discs (glaucoma, ADOA, and Proportion of correct and incorrect diagnoses in each subgroup of disease
LHON). This reached statistical significance for disc size for neuro-ophthalmologists and glaucoma specialists. There was no differ-
(P⬍0.01), CDR (P⬍0.001), cup shape (P⬍0.001), cup depth ence between the 2 groups of specialists. C, Proportion of alternate
(P⬍0.05), pallor (P⬍0.001), the presence or absence of a gray diagnoses within the ADOA and LHON subgroups. Alternate diagnoses
crescent (P⬍0.001), peripapillary atrophy (P⬍0.01), and RNFL chosen for the ADOA discs: 48% glaucoma, 33% normal, and 19%
(P⬍0.001; Fig 5). Thus, these parameters were more easily LHON. Alternate diagnoses chosen in the LHON discs: 52% ADOA,
assessed and with higher levels of agreement within the normal 39% glaucoma, and 9% normal. ADOA ⫽ autosomal-dominant optic
disc set when compared with all pathologic discs irrespective of atrophy; LHON ⫽ Leber’s hereditary optic neuropathy; Neuro ⫽ neuro-
disease entity. There was no difference between the 2 expert ophthalmology.

966
 O’Neill et al 䡠 Optic Disc Assessment in Optic Neuropathies: ODAP Project
Figure 4. Proportion of underlying disease entity where glaucoma was the
chosen diagnosis. A, All experts: 61% were glaucoma discs, 19% were
ADOA discs, 15% were LHON discs, and 5% were normal discs. B,
Comparing neuro-ophthalmology specialists with glaucoma specialists.
Among the neuro-ophthalmologists, 65% were glaucoma discs, 17% were
ADOA discs, 13% were LHON discs, and 5% were normal discs. Among
the glaucoma specialists, 56% were glaucoma discs, 21% were ADOA
discs, 18% were LHON discs, and 5% were normal discs. There was no
difference between the 2 specialty groups. ADOA ⫽ autosomal-dominant
optic atrophy; LHON ⫽ Leber’s hereditary optic neuropathy.
groups in levels of agreement in the assessment of any of the
disc parameters.
Interestingly, analysis of the pathologic discs showed there was
a significantly greater proportion of discs deemed to have shallow
cups among the ADOA and LHON discs sets (34% and 49%,
respectively) compared with the glaucomatous discs (9%; P⬍0.01)
and a significantly greater proportion of discs classified as having
deep cups among the glaucoma discs (49%) compared with either
ADOA or LHON (16% and 11%, respectively; P⬍0.01).
It also showed that assessment of the color of the rim as the
only disc parameter to have a significantly greater level of agree-
ment within the glaucomatous discs when compared with the
ADOA and LHON discs (P⬍0.01). There was no significant
difference between the pathologic groups for levels of agreement
Figure 5. Levels of agreement assessing the normal discs and the patho-
logic discs (⫾ standard error of the mean [SEM]). There were higher levels
of agreement for all parameters assessed within the normal disc set when
compared with the pathologic discs. *P⬍0.05. Av Dev ⫽ average devia-
tion; CDR ⫽ cup:disc ratio; IQV ⫽ index of qualitative variation; NFL ⫽
nerve fiber layer; PPA ⫽ peripapillary atrophy.
for any of the other parameters assessed (Fig 6). Furthermore,
there was no significant difference between glaucoma specialists
and neuro-ophthalmologistss in their ability to diagnose normal
(P ⫽ 0.42), glaucoma (P ⫽ 0.75), ADOA (P ⫽ 0.81), and LHON
(P ⫽ 0.24).
Discussion
Optic neuropathies of varying etiologies have multiple over-
lapping clinical characteristics. Glaucoma, LHON, and
ADOA can exhibit similar morphologic changes at the optic
nerve head and instances whereby both LHON and ADOA
have been misclassified as glaucoma on disc appearance
have been reported.3,15-17 Recently, it has been postulated
that abnormalities in mitochondrial function as demon-
Figure 6. Level of agreement assessing glaucomatous, ADOA, and LHON
discs (⫾ standard error of the mean [SEM]). Pallor is the only parameter
where there was a statistically significant higher level of agreement among
the glaucoma discs compared with either LHON or ADOA. *P⬍0.05. Av
Dev ⫽ average deviation; ADOA ⫽ autosomal-dominant optic atrophy;
CDR ⫽ cup:disc ratio; IQV ⫽ index of qualitative variation; LHON ⫽
Leber’s hereditary optic neuropathy; NFL ⫽ nerve fiber layer; PPA ⫽
peripapillary atrophy.
967
 Ophthalmology Volume 118, Number 5, May 2011
strated in both LHON and ADOA may also contribute to the
underlying pathogenic mechanism in POAG; thus, it could
potentially account for the similarities in end-stage morpho-
logic features at the optic nerve head.22,23
The paradigm of mitochondrial optic neuropathies is
LHON, where a primary role for mitochondrial dysfunction
is demonstrated through both maternal inheritance and spe-
cific mutations in mitochondrial DNA (mtDNA).22 Three
primary LHON mtDNA point mutations—at loci positions
11778/ND4, 3460/ND1, and 14484/ND6 —are pathogenic
in ⬎90% of cases,22,24-26 with the 11778 mutation the most
prevalent.26 The prevalence and incidence of visual loss
from LHON worldwide remains unknown, but among indi-
viduals in the northeast of England, there is a prevalence of
a primary LHON-associated mtDNA mutation of 11.82 per
100 000 individuals and a minimal prevalence of visual loss
of 3.22 per 100 000 individuals.27
Considered the most common form of inherited (non-
glaucomatous) optic neuropathy,28 ADOA has a prevalence
of 1:12 000 to 1:50 000.29 It has been linked primarily with
2 different loci, the large majority to chromosome 3q and
the nuclear gene OPA1.22,30,31 This gene encodes a dy-
namin-related protein localized to mitochondria.30,32 This
protein is closely related to a family of proteins involved in
mitochondrial network organization and it is thought that
OPA1 expression may play a major role in the maintaining
the structural integrity of mitochondrial cristae.22
These findings suggest that ADOA and LHON are linked
by mitochondrial pathogenetic mechanisms.22,24,25 Interest-
ingly, polymorphisms in the OPA1 gene have been found in
NTG patients, raising the possibility that NTG is also a
heritable neuropathy demonstrating similar mitochondrial
dysfunction, but altered phenotype to ADOA, with differing
clinical presentations and functional visual loss.33 No asso-
ciation between OPA1 polymorphisms and high IOP POAG
has been found.34,35 However, it has been hypothesized that
abnormalities in mitochondrial function as found in LHON
and ADOA may be part of the underlying pathogenic mech-
anism in both POAG and NTG.22,23,36
In our study, experts easily identified the normal and
glaucomatous discs within the disc set. Within the ADOA
and LHON subgroups, they identified the discs as abnormal;
however, in the majority of cases they had failed to assign
the correct diagnosis on disc appearance alone, with glau-
coma chosen as the presumed diagnosis in the majority
(48% of presumed alternate diagnoses in the ADOA sub-
group and 39% in LHON subgroup). The expert clinicians
showed high level of agreement on all disc parameters
assessed for normal discs but poor interobserver agreement
for all parameters assessed in the pathologic discs. Pallor
was the only physical parameter to show a significant dif-
ference in levels of agreement between the glaucoma,
ADOA, and LHON groups with higher levels of agreement
on the presence or absence and distribution of pallor in the
glaucomatous subgroup (P⬍0.01). Interestingly, where ex-
perts agreed on the presence or absence of pallor, they
tended to choose the correct diagnosis. In the majority of
cases, experts agreed that pallor was absent in the glauco-
matous discs and diagnosed the disc correctly, despite dis-
agreeing on multiple other parameters. However, within the
968
ADOA and LHON groups, on a disc-by-disc basis there was
large variability in assessment of pallor and whether it was
present or absent. When the assessors agreed on the absence
of pallor in the glaucoma discs, they tended to make an
accurate diagnosis; when they disagreed on the presence or
absence of pallor among the ADOA and LHON discs, they
tended to misclassify the disc. Overall glaucoma, ADOA,
and LHON discs had similar levels of disagreement for all
other parameters assessed.
In both inherited LHON and ADOA, optic nerve head
rim loss has been reported.3-6 Fournier et al3 assessed optic
nerve head photos of patients with a clinical diagnosis of
ADOA. The primary aim was to identify features differen-
tiating disc excavation in this disorder from glaucomatous
optic nerve cupping. They found 89% of their ADOA sub-
jects showed significant neuroretinal rim thinning (CDR ⬎
0.5). They also demonstrated that ADOA patients have
significant large discs. They believe that the combination of
these 2 factors—rim thinning and disc size— could poten-
tially lead the examiner to the erroneous diagnosis of NTG.
However, all of the ADOA patients had pallor of the neu-
roretinal rim, which does not typically occur in glaucoma.
They concluded pallor of the remaining neuroretinal rim to
be the main differentiating feature of optic nerve head
morphology between ADOA and glaucoma patients. Al-
though the assessors were not masked to diagnosis, these
results are in agreement with ours.3
Similarly, Votruba et al15 studied the optic nerve mor-
phology, in optic disc photographs of genetically confirmed
ADOA, and compared the features with known/previously
documented morphology of NTG discs. On examination, all
of their ADOA discs exhibited significant pallor, with 52%
having temporal disc pallor and 48% global disc pallor. A
large proportion of the ADOA discs also exhibited disc
excavation with 48% having a CDR of ⬎0.5 and 79% of the
discs demonstrating shallow excavation or saucerization of
the disc, particularly the temporal sector. They concluded
that the most significant findings in ADOA subjects were
pallor of the neuroretinal rim and shallow excavation or
shelving of the disc, which is not seen in NTG subjects.
Furthermore, using the Heidelberg Retinal Tomograph to
assess the optic nerve heads of LHON, NTG, and normal
subjects, Mashima et al4 found that the Heidelberg Retinal
Tomograph classification program correctly classified 87%
of the NTG discs as glaucomatous, but misclassified 73% of
the LHON discs as glaucomatous, compared with only 8%
misclassification of normal discs. They found the main
clinical difference was that LHON patients demonstrated
global optical nerve head pallor, whereas the NTG patients
had loss of the neuroretinal rim without pallor.4
Each of these studies highlighted the similarities of optic
disc appearance in for both LHON and ADOA and glau-
coma and the potential for misclassification without careful
clinical examination and additional clinical history and in-
vestigations (such as visual acuity, color vision, history of
visual loss, and family history).
Similarly, in our cohort, although experts could easily
identify the ADOA and LHON discs as abnormal, they had
difficulty assigning the correct diagnosis on disc assessment
alone. The theory of posterior bowing of the lamina cribrosa
 O’Neill et al 䡠 Optic Disc Assessment in Optic Neuropathies: ODAP Project
in glaucoma thought secondary to increased IOP is well
known.37 Interestingly, in keeping with this, our study
found a significantly greater proportion of discs were clas-
sified as deep among the glaucoma discs compared with
either ADOA or LHON, and similarly a significantly larger
proportion were classified as having shallow cups among
the ADOA and LHON discs compared with the glaucoma
discs; however, this did not affect or aid in classification of
the discs within the ADOA or LHON groups.
When glaucoma was chosen as the most likely diagnosis,
only 61% of these discs were indeed from glaucoma pa-
tients; of the remaining 39%, 19% were ADOA discs and
15% were LHON discs. Our findings further support the
body of evidence suggesting that these disease entities have
similar appearance at the optic nerve head.
Recently, Kim et al38 described the ocular coherence
tomography (OCT) RNFL profiles in ADOA eyes and its
possible role in differentiating these discs from glaucoma-
tous ones (particularly NTG subjects). They found that the
OCT could distinguish the 2 clinical entities by the pattern
of RNFL damage. The temporal measurements were outside
the 95% normal limit in 90% of ADOA eyes assessed and
a high degree of symmetry existed between both right and
left eyes. This is in keeping with histopathologic results,
which suggest that selective loss of the retinal ganglion
cells, in particular from the macular area, is the primary
abnormality in ADOA,39 whereas in glaucoma this is spared
until the end stage of disease. Given that ADOA and LHON
can often mimic glaucoma in disc appearance and excava-
tion of the optic nerve head, the use of OCT may enable
clinicians more accurately distinguish the 2 entities in dif-
ficult cases and examination of color photographs is clearly
not sufficient (even among expert observers) to distinguish
these entities.
This study has several limitations. First, participants
were only shown disc images from 1 eye, not both. Both
ADOA and LHON occur bilaterally and presentation of
both optic nerve images may have aided in diagnosis. Sec-
ond, 2 different stereo-viewers were used by our experts, a
handheld Screen-Vu Stereo Viewer (Eyesupply USA, Inc.)
and adjustable 3-dimensional prism glasses (Berezin, Inc.),
which may have been a potential source of bias. Finally,
experts were asked to make a presumptive diagnosis on disc
morphology alone; no other clinical information was sup-
plied. This does not reflect true clinical practice because
both presentation and clinical examination (both visual acu-
ity and pattern of field loss) is very different in these 3
entities.
In conclusion, optic neuropathies of varied etiologies can
result in similar topographic changes at the optic nerve
head. Rim loss is a common feature of both hereditary optic
neuropathies and glaucoma. We found, among an expert
group, that normal and glaucomatous discs were easily
identified, and that ADOA and LHON discs were correctly
identified as abnormal. However, there was significant dif-
ficulty in assigning the correct diagnosis on disc appearance
alone among the hereditary optic neuropathies. Our findings
highlight that it is difficult to distinguish among these en-
tities, particularly in the later stages of the disease, using
morphologic features at the optic nerve head alone. We
also show that optic discs in LHON and ADOA can have
glaucoma-like morphologic change; therefore, clinicians
must consider other clinical parameters such as acuity, color
vision, history of visual loss, and family history in making
accurate diagnoses. Newer imaging modalities, such as
OCT, may also aid in distinguishing these entities. It re-
mains to be determined whether a mitochondrial dysfunc-
tion similar to that of ADOA and LHON could occur in
glaucoma.
References
1. O’Neill EC, Mackey DA, Connell PP, et al. The optic nerve
head in hereditary optic neuropathies. Nat Rev Neurol 2009;
5:277– 87.
2. Newman NJ, Biousse V. Hereditary optic neuropathies. Eye
(Lond) 2004;18:1144 – 60.
3. Fournier AV, Damji KF, Epstein DL, Pollock SC. Disc exca-
vation in dominant optic atrophy: differentiation from normal
tension glaucoma. Ophthalmology 2001;108:1595– 602.
4. Mashima Y, Kimura I, Yamamoto Y, et al. Optic disc exca-
vation in the atrophic stage of Leber’s hereditary optic
neuropathy: comparison with normal tension glaucoma.
Graefes Arch Clin Exp Ophthalmol 2003;241:75– 80.
5. Trobe JD, Glaser JS, Cassady J, et al. Nonglaucomatous
excavation of the optic disc. Arch Ophthalmol 1980;98:
1046 –50.
6. Trobe JD, Glaser JS, Cassady JC. Optic atrophy: differential
diagnosis by fundus observation alone. Arch Ophthalmol
1980;98:1040 –5.
7. Resnikoff S, Pascolini D, Etya’ale D, et al. Global data on
visual impairment in the year 2002. Bull World Health Organ
2004;82:844 –51.
8. Allingham RR, Liu Y, Rhee DJ. The genetics of primary
open-angle glaucoma: a review. Exp Eye Res 2009;88:
837– 44.
9. Wiggs JL. Genes associated with human glaucoma. Ophthal-
mol Clin North Am 2005;18:335– 43, v.
10. Wiggs JL. Genetic etiologies of glaucoma. Arch Ophthalmol
2007;125:30 –7.
11. Klein BE, Klein R, Sponsel WE, et al. Prevalence of glau-
coma. The Beaver Dam Eye Study. Ophthalmology 1992;99:
1499 –504.
12. Sommer A, Tielsch JM, Katz J, et al. Relationship between
intraocular pressure and primary open angle glaucoma among
white and black Americans. The Baltimore Eye Survey. Arch
Ophthalmol 1991;109:1090 –5.
13. Jonas JB, Budde WM, Panda-Jonas S. Ophthalmoscopic
evaluation of the optic nerve head. Surv Ophthalmol 1999;
43:293–320.
14. Ortiz RG, Newman NJ, Manoukian SV, et al. Optic disk
cupping and electrocardiographic abnormalities in an Ameri-
can pedigree with Leber’s hereditary optic neuropathy. Am J
Ophthalmol 1992;113:561– 6.
15. Votruba M, Thiselton D, Bhattacharya SS. Optic disc mor-
phology of patients with OPA1 autosomal dominant optic
atrophy. Br J Ophthalmol 2003;87:48 –53.
16. Lauer SA, Ackerman J, Sunness J, et al. Leber’s optic atrophy
with myopia masquerading as glaucoma: case report. Ann
Ophthalmol 1985;17:146 – 8.
17. Weiner NC, Newman NJ, Lessell S, et al. Atypical Leber’s
hereditary optic neuropathy with molecular confirmation.
Arch Neurology 1993;50:470 –3.
969
 Ophthalmology Volume 118, Number 5, May 2011
18. Foster PJ, Buhrmann R, Quigley HA, Johnson GJ. The defi-
nition and classification of glaucoma in prevalence surveys.
Br J Ophthalmol 2002;86:238 – 42.
19. Coote MA, McCartney PJ, Wilkinson RM, Mackey DA. The
‘GIST’ score: ranking glaucoma for genetic studies. Glau-
coma Inheritance Study of Tasmania. Ophthalmic Genet 1996;
17:199 –208.
20. Cohn AC, Toomes C, Potter C, et al. Autosomal dominant
optic atrophy: penetrance and expressivity in patients with
OPA1 mutations. Am J Ophthalmol 2007;143:656 – 62.
21. Wilcox AR. Indices of Qualitative Variation. 1967:12. Report
no. ORNL-TM—1919. Available at: http://www.ornl.gov/
info/reports/1967/3445605133753.pdf. Accessed August
31, 2010.
22. Carelli V, Ross-Cisneros FN, Sadun AA. Mitochondrial dys-
function as a cause of optic neuropathies. Prog Retin Eye Res
2004;23:53– 89.
23. Kong GY, Van Bergen NJ, Trounce IA, Crowston JG. Mito-
chondrial dysfunction and glaucoma. J Glaucoma 2009;18:
93–100.
24. Carelli V, La Morgia C, Iommarini L, et al. Mitochondrial
optic neuropathies: how two genomes may kill the same cell
type? Biosci Rep 2007;27:173– 84.
25. Carelli V, Ross-Cisneros FN, Sadun AA. Optic nerve degen-
eration and mitochondrial dysfunction: genetic and acquired
optic neuropathies. Neurochem Int 2002;40:573– 84.
26. Mackey DA, Oostra RJ, Rosenberg T, et al. Primary patho-
genic mtDNA mutations in multigeneration pedigrees with
Leber hereditary optic neuropathy. Am J Hum Genet 1996;
59:481–5.
27. Man PY, Griffiths PG, Brown DT, et al. The epidemiology of
Leber hereditary optic neuropathy in the North East of Eng-
land. Am J Hum Genet 2003;72:333–9.
28. Votruba M, Moore AT, Bhattacharya SS. Clinical features,
molecular genetics, and pathophysiology of dominant optic
atrophy. J Med Genet 1998;35:793– 800.
29. Kjer B, Eiberg H, Kjer P, Rosenberg T. Dominant optic
atrophy mapped to chromosome 3q region. II. Clinical and
Footnotes and Financial Disclosures
Originally received: April 12, 2010.
Final revision: September 8, 2010.
Accepted: September 8, 2010.
Available online: December 3, 2010. Manuscript no. 2010-538.
1
Centre for Eye Research Australia, Royal Victorian Eye and Ear Hospital,
University of Melbourne, Melbourne, Australia.
2 yahoo.com.
The Lions Eye Institute at the University of Western Australia.
3
Department of Ophthalmology, University of Auckland, Auckland, New
Zealand.
Supported by the Cranbourne Trust and Helen McPherson Trust Grant; CERA
receives Operational Infrastructure Support from the Victorian Government.
970
epidemiological aspects. Acta Ophthalmol Scand 1996;
74:3–7.
30. Alexander C, Votruba M, Pesch UE, et al. OPA1, encoding a
dynamin-related GTPase, is mutated in autosomal dominant
optic atrophy linked to chromosome 3q28. Nat Genet 2000;
26:211–5.
31. Eiberg H, Kjer B, Kjer P, Rosenberg T. Dominant optic
atrophy (OPA1) mapped to chromosome 3q region. I. Linkage
analysis. Hum Mol Genet 1994;3:977– 80.
32. Delettre C, Lenaers G, Griffoin JM, et al. Nuclear gene
OPA1, encoding a mitochondrial dynamin-related protein,
is mutated in dominant optic atrophy. Nat Genet 2000;26:
207–10.
33. Aung T, Ocaka L, Ebenezer ND, et al. A major marker for
normal tension glaucoma: association with polymorphisms in
the OPA1 gene. Hum Genet 2002;110:52– 6.
34. Aung T, Ocaka L, Ebenezer ND, et al. Investigating the
association between OPA1 polymorphisms and glaucoma:
comparison between normal tension and high tension
primary open angle glaucoma. Hum Genet 2002;110:
513– 4.
35. Liu Y, Schmidt S, Qin X, et al. No association between OPA1
polymorphisms and primary open-angle glaucoma in three
different populations. Mol Vis [serial online] 2007;13:2137–
41. Available at: http://www.molvis.org/molvis/v13/a242/.
Accessed August 31, 2010.
36. Abu-Amero KK, Bosley TM. Mitochondrial abnormalities in
patients with LHON-like optic neuropathies. Invest Ophthal-
mol Vis Sci 2006;47:4211–20.
37. Hayreh SS, Pe’er J, Zimmerman MB. Morphologic changes in
chronic high-pressure experimental glaucoma in rhesus mon-
keys. J Glaucoma 1999;8:56 –71.
38. Kim TW, Hwang JM. Stratus OCT in dominant optic atrophy:
features differentiating it from glaucoma. J Glaucoma 2007;
16:655– 8.
39. Johnston PB, Gaster RN, Smith VC, Tripathi RC. A clinico-
pathologic study of autosomal dominant optic atrophy. Am J
Ophthalmol 1979;88:868 –75.
Financial Disclosure(s):
The authors have no proprietary or commercial interest in any of the
materials discussed in this article.
Correspondence:
Dr. Evelyn C. O’Neill, Centre for Eye Research Australia, 32 Gisborne
Street, East Melbourne, VIC 3002, Australia. E-mail: evelynoneill@
*A complete listing of The Optic Nerve Study Group membership is
available at http://aaojournal.org.