Current Eye Research, Early Online, 1–8, 2014

! Informa Healthcare USA, Inc.

ISSN: 0271-3683 print / 1460-2202 online
DOI: 10.3109/02713683.2014.971929

RESEARCH REPORT

Evaluation of Meibomian Gland Dysfunction and

Local Distribution of Meibomian Gland Atrophy
by Non-contact Infrared Meibography
David Finis, Philipp Ackermann, Nadja Pischel, Claudia König, Jasmin Hayajneh,
Maria Borrelli, Stefan Schrader* and Gerd Geerling*

Department of Ophthalmology, Heinrich-Heine-University Düsseldorf, Düsseldorf, Germany

Curr Eye Res Downloaded from informahealthcare.com by University of Toronto on 03/07/15

ABSTRACT
Aims: The main purpose of this study was to investigate the relationship between Meibomian gland atrophy
(meiboscore) and Meibomian gland expressibility. In addition, the local distribution of Meibomian gland loss
was analyzed.
Methods: A retrospective analysis of 128 patients (92 women and 36 men, 57 ± 17 years) from our dry eye clinic
was performed. Infrared meibography was performed using the Keratograph 5 M (Oculus, Wetzlar, Germany)
For personal use only.

and evaluated with a scoring system introduced by Arita et al.

Results: A significant inverse correlation between Meibomian gland atrophy measured by meibography and
expressible Meibomian glands (r = 0.197, p = 0.003) as well as between meiboscore and TBUT (r = 0.1615,
p = 0.012) was found. There also was a significant correlation between the total meiboscore and the age (r = 0.33,
p50.0001). We could find a strong and highly significant correlation between the total meiboscore and the
individual meiboscore of the upper eyelid (r = 0.905, p50.0001) and the lower eyelid (r = 0.892, p50.0001). There
was no significant difference of Meibomian gland atrophy between the individual thirds of the upper eyelid,
but for the lower eyelid, we could find a higher degree of Meibomian gland atrophy in the nasal third compared
with the middle and the temporal third (Dunn’s post hoc test, p50.0001).
Conclusions: Meibomian gland atrophy seems to be not constant over the tarsal plate but the examination of the
lower tarsus might be sufficient in most of the cases. The correlation of the meiboscore with functional dry eye
parameters suggest that in patients with detectable Meibomian gland atrophy there is also an impaired
Meibomian gland function. However, meibography seems not to be sufficient as a single test for the diagnosis of
MGD. For the future larger, prospective studies are needed to confirm these results and further evaluate the
potential of meibography in the diagnosis of MGD.
Keywords: Dry eye, meibography, Meibomian gland dysfunction, MGD, tear film

INTRODUCTION Since the Tear Film and Ocular Surface Society

Obstructive Meibomian gland dysfunction (MGD) is
the most common cause of evaporative dry eye.
About two-thirds of dry eye cases are caused by
obstructive MGD. Up to 20% of the population in
Europe and 60% of the population in Asia suffer
from MGD.1–3
(TFOS) workshop on MGD in 2011 for the first time,
there is a common definition of MGD, i.e., ‘‘. . .a
chronic, diffuse abnormality of the meibomian glands,
commonly characterized by terminal duct obstruction and/
or qualitative/quantitative changes in the glandular secre-
tion. It may result in alteration of the tear film, symptoms
of eye irritation, clinically apparent inflammation, and

*These authors contributed equally to this work.

Received 9 August 2013; revised 29 August 2014; accepted 28 September 2014; published online 20 October 2014
Correspondence: Dr David Finis, Department of Ophthalmology, Heinrich-Heine-University Düsseldorf, Mooren Str. 5, 40225 Düsseldorf,
Germany. E-mail: David.Finis@med.uni-duesseldorf.de

1
 2 D. Finis et al.
ocular surface disease’’.3,4 A lot of different tests have
been proposed for the diagnosis of MGD; however, no
clear diagnostic criteria have been defined yet. The
most common diagnostic tools are the assessment of
lid margin changes and diagnostic expression of the
Meibomian glands. If the diagnostic expression is
performed with the fingers, less than 10 Meibomian
glands of the lower eyelid yielding secretion indicate
the diagnosis of obstructive MGD.5 As the force
applied to the eye lid by this method is very variable,
Korb et al.5,6 developed a hand-held device, with
which a defined pressure can be applied to the eyelid.
For this standardized expression, 4 expressible
glands of the lower eyelid suggest a high likelihood
of obstructive MGD.
The obstruction of Meibomian glands is mainly
Curr Eye Res Downloaded from informahealthcare.com by University of Toronto on 03/07/15
caused by an increased keratinization of the epider-
mal epithelium in the excretory ducts.7,8 In addition,
a change in lipid composition of Meibomian gland
secretion causes increased viscosity and thus further
reduced discharge of Meibum.9–11 This can be
exacerbated by an increased number of commensal
bacteria on the lid margin which influence the lipid
composition by production of lipases and ester-
ases.7,12 As a consequence, the tear film lipid layer
becomes thinner and destabilizes, resulting in an
For personal use only.
evaporative dry eye. Second, the ongoing production
of secretions causes an elevated intraglandular pres-
sure resulting in atrophy of the secretory acini.2,8
Atrophy of Meibomian glands can be visualized
using meibography. Initially, a light source applied to
either the palpebral or the conjunctival side of the eye
lid was used to for Meibomian gland visualization by
trans illumination. Recently, infrared cameras were
introduced to detect a Meibomian gland loss non-
invasively.5,13,14 Several scoring systems have been
developed to quantify Meibomian gland loss.5 Some
of them are based on the examination of the lower
eyelid only whereas some scoring systems evaluate
both the upper and the lower eyelid. However, up to
now, there are no data concerning the local distribu-
tion of Meibomian gland loss. Therefore, it is not
certain, whether it is sufficient to examine just the
lower eyelid or if it is necessary to examine the upper
tarsus, as well. In addition, it is still unclear what
diagnostic information a loss of Meibomian glands
provides and what impact a loss of part of the
Meibomian glands has on the function of the remain-
ing glands. So far it has been shown that meibography
scores as an expression of loss of functional
Meibomian glands are correlated with age and inverse
correlated with tear break-up-time (BUT)15 but there
are no data on the correlation of meibography and
diagnostic Meibomian gland expression as an import-
ant test for obstructive MGD.5
Therefore, the main purpose of this study was to
investigate the relationship between Meibomian
gland atrophy measured by infrared meibography
and expressible Meibomian glands as a test of
meibomian gland function. In addition, an analysis
of the local distribution of Meibomian gland loss was
performed to determine, whether Meibomian gland
atrophy follows a certain pattern and if the examin-
ation of only the lower eyelid is sufficient for the
diagnosis of MGD.
MATERIALS AND METHODS
A retrospective analysis of data from 128 patients
(252 eyes, age 57 ± 17 years, 92 women and 36 men)
from the dry eye clinic of the Department of
Ophthalmology of the University of Duesseldorf
was performed between April 2012 and March 2013.
All examinations were part of the routine diagnostic
and were analyzed anonymously. This study followed
the tenets of the declaration of Helsinki.
Inclusion criteria were as follows: (a) at least
18 years of age and (b) an evaluable meibography of
the upper and lower eyelid.
Exclusion criteria were as follows: (a) Ocular
surgery or trauma within three months before exam-
ination, (b) any eyelid abnormalities, and (c) systemic
diseases resulting in dry eye.
Before clinical examination, patients were
required to complete a symptom questionnaire
[Ocular-surface-disease-index (OSDI)]. This question-
naire gives a range of 0 (no symptoms) till 100 (severe
symptoms).
Meibography was performed using the
Keratograph 5 M Õ (Oculus, Wetzlar, Germany). The
evaluation was based on the criteria proposed by Arita
et al.15 First, the Meibomian glands of the upper eyelid
were visualized and graded. No atrophy corresponds
to 0 point, less than one-third to one point, more than
one-third to two points, and more than two-thirds to
three points. After that the glands of the lower eyelid
were visualized and graded in the same way. The
values for the upper (MS UEL) and lower eyelid (MS
LEL) are then added to the total meiboscore (MS total)
with a range from 0 to 6 (Figure 1).
In addition, a separate analysis of the nasal third of
the lower eyelid (MS LEL nasal), the middle third of
the lower eyelid (MS LEL middle), and the temporal
third of the lower eyelid (MS LEL temporal) was
performed. No atrophy in an individual in nasal third
corresponded to 0 point, a partial degree of atrophy to
one point, and total atrophy to two points. The same
analysis was performed for the nasal third of the upper
eyelid (MS UEL nasal), the middle third of the upper
eyelid (MS UEL middle), and the temporal third of the
upper eyelid (MS UEL temporal) (Figure 2). To get the
most objective results, all meibography images were
analyzed by two different investigators.
The number of expressible Meibomian glands was
quantified using the Meibomian-gland-evaluatorÕ
Current Eye Research
 Non-contact Meibography in the Diagnosis of MGD 3
Curr Eye Res Downloaded from informahealthcare.com by University of Toronto on 03/07/15
FIGURE 1 Grading of meibography according to Arita et al. (A: lower eyelid; B: upper eyelid): 1, no loss of Meibomian glands (0
point); 2, less than one-third loss of Meibomian glands (one point); 3, less than two-thirds loss of Meibomian glands (two points); 4,
more than two-thirds loss of Meibomian glands (three points).
For personal use only.
FIGURE 2 Grading of meibography of the individual thirds of
the eyelids.
(TearScience Inc., Morrisville, NC). With this hand-
held instrument, a defined pressure was applied to
the lateral, the middle, and the medial third of the
lower eyelid and the number of expressible
glands was counted. This method was first described
by Korb et al.6
! 2014 Informa Healthcare USA, Inc.
Non-invasive tear film break up time (TBUT) was
automatically detected with the Keratograph 5 MÕ
(Oculus, Wetzlar, Germany). For further analysis, the
time of the first tear film break-up was used.
Lipid layer thickness was measured using the
LipiViewÕ interferometer (TearScience Inc.,
Morrisville, NC). This method has been described in
detail by Blackie et al.16 In brief, the patient is asked
to look into a camera that records a 20-s video of the
interference patterns of the tear film. These patterns
are then analyzed and a value in interferometric color
units (ICUs) is generated, wherein one ICU approxi-
mately reflects 1 nm of lipid layer thickness.
Ocular surface staining was measured 2 min after
instillation of a single drop of fluorescein (Fluoreszein
SE Thilo, Alcon Pharma GmbH, Freiburg im Breisgau,
Germany). The staining was evaluated using a slit
lamp with a cobalt blue exciter filter and then graded
according to the Oxford grading system. The cornea
and two conjunctival zones (nasal and temporal)
were graded, with a score of 0–5 per zone according
to a visual analogue scale, resulting in a total range of
0–15.5,17
Lid margin parallel conjunctival folds (LIPCOFs)
were graded as described by Hoeh et al.18 No
permanent existing lid margin parallel conjunctival
fold was graded as 0; a small fold, lower than a
normal tear meniscus was graded as 1; a significant
fold up to the height of the normal tear meniscus was
graded as 2; a large fold, multiple wrinkled, exceeding
 4 D. Finis et al.

the height of a normal tear meniscus was graded as TABLE 1 Characteristics of the study collective.
3 and a large fold which extends over the inner lid
25% 75% p-value
margin was graded as 4. Percentile Median Percentile KS
Statistical analysis was performed using Microsoft
Excel 2010 and GraphPad Prism version 6 (GraphPad Age 47 58 70 0.0033
Inc., San Diego, CA). For normally distributed values LLT (nm) 47 64 89 50.0001
Tear meniscus 0.2 0.3 0.4 50.0001
and for non-normally distributed values, the linear height (mm)
Pearson’s correlation coefficient and the linear correl- TBUT (s) 3 6 13.5 50.0001
ation coefficient by Spearman were calculated, Schirmer 5 10 17.5 50.0001
respectively. A p-value of 50.05 was considered LIPCOF 2 3 3 50.0001
statistically significant. The testing for normal distri- Staining 0 3 4.5 50.0001
Expr. glands 2 4 7 50.0001
bution was performed by using the Kolmogorov– OSDI 32.1 45.8 61.4 0.035
Smirnov test. A p-value40.05 was considered as not MS UEL temporal 1 1 1 50.0001
statistically significantly different to the normal MS UEL middle 1 1 1 50.0001
distribution. For analysis of the local distribution of MS UEL nasal 1 1 1 50.0001
the meiboscore and comparison between individual MS UEL 1 1 2 50.0001
MS LEL temporal 0 1 1 50.0001
Curr Eye Res Downloaded from informahealthcare.com by University of Toronto on 03/07/15

thirds of the eyelids, a Kruskal–Wallis test with MS LEL middle 0 1 1 50.0001

Dunn’s post-hoc analysis was used. A p-value of MS LEL nasal 1 1 2 50.0001
50.05 was considered statistically significant. MS LEL 1 1 2 50.0001
MS total 2 2 4 50.0001

LLT, lipid layer thickness; TBUT, Tear break-up-time; LIPCOF,

RESULTS lid margin parallel conjunctival folds; OSDI, Ocular surface
disease index; p-value KS, p-value Kolmogorov–Smirnov test.
The mean and median values of the investigated dry
eye parameters of our study collective (128 patients,
252 eyes, 92 women and 36 men) are shown in Table 1.
For personal use only.

All parameters were not normally distributed. All

meibography images were analyzed by two investi-
gators with a high concordance (r = 0.91). The median
value of the meiboscore was 2. About 56.7% of the
patients had a meiboscore  2 and 43.3% of the
patients had a meiboscore42. About 7.5% showed
absolutely no atrophy with a meiboscore of 0 and
7.1% showed a severe atrophy of the Meibomian
glands with a meiboscore 6. The relative frequency
of all meiboscore values is displayed in Figure 3.
The distribution of Meibomian gland atrophy in
our study collective was not constant over the tarsal
plate (Kruskal–Wallis test p50.0001). There was no
significant difference between the individual thirds of
the upper eyelid, but for the lower eyelid, we could
find a higher degree of Meibomian gland atrophy in
FIGURE 3 Distribution of meiboscore in the study collective.
the nasal third compared with the middle and the
temporal third (Dunn’s post hoc test, p50.0001).
There was a strong and highly significant correl-
ation between the total meiboscore and the individual p = 0.003) (Figure 4) as well as between meiboscore
meiboscore of the upper eyelid (r = 0.905, p50.0001) and TBUT (r = 0.1615, p = 0.012) (Figure 5). There
and the lower eyelid (r = 0.892, p50.0001). There were was also a significant correlation between the total
also highly significant correlations between the total meiboscore and the age (r = 0.33, p50.0001) (Figure 6).
meiboscore and the meiboscore of the individual All these parameters also correlated significantly with
thirds of the upper and the lower eyelid but with a the individual meiboscores of the upper and the lower
lower correlation coefficient of about 0.7 (for exact eyelid. For most of the individual thirds of the upper
values, see Table 2). The meiboscores of the right and and the lower eyelid, these correlations were found as
left eye also correlated significantly with a coefficient well. If only a third of one eyelid is analyzed, the
of 0.842 (p50.0001). middle third of the lower eyelid seems to be the best
We found a significant inverse correlation between choice, as the correlation coefficients for this third
Meibomian gland atrophy measured by meibography were most often higher compared with the other
and expressible Meibomian glands (r = 0.197, thirds (Table 3).
Current Eye Research
 Curr Eye Res Downloaded from informahealthcare.com by University of Toronto on 03/07/15
For personal use only.

!
TABLE 2 Correlations between the total meiboscore and partial meiboscores of the eyelids.

MS total versus MS total versus MS total versus MS total versus MS total versus MS total versus MS total versus MS total versus
Spearman r MS UEL temporal MS UEL middle MS UEL nasal MS UEL MS LEL temporal MS LEL middle MS LEL nasal MS LEL

r 0.6897 0.6395 0.5995 0.905 0.7073 0.7148 0.686 0.8922

95% confidence 0.6163–0.7513 0.5573–0.7093 0.5109–0.6755 0.8789–0.9257 0.6369–0.7661 0.6460–0.7722 0.6116–0.7484 0.8628–0.9156
interval
p (two-tailed) 50.0001 50.0001 50.0001 50.0001 50.0001 50.0001 50.0001 50.0001

2014 Informa Healthcare USA, Inc.

MS total, meiboscore according to Arita et al.; MS UEL, meiboscore of the upper eyelid; MS LEL, meiboscore of the lower eyelid; nasal, middle, and temporal refer to the individual third
of the eyelids.

Arita et al.) and age.

to Arita et al.) and tear film break-up-time.
to Arita et al.) and expressible Meibomian glands.

of the international workshop on MGD (report of

 4 expressible glands according to the suggestions
In Table 4, the sensitivity and specificity values for

For this analysis, a significant MGD was defined as

different cut-off values of the meiboscore are listed.
FIGURE 6 Correlation between meiboscore (according to
Non-contact Meibography in the Diagnosis of MGD 5

FIGURE 5 Inverse correlation between meiboscore (according

FIGURE 4 Inverse correlation between meiboscore (according
 6 D. Finis et al.

MS total, meiboscore according to Arita et al.; MS UEL, meiboscore of the upper eyelid; MS LEL, meiboscore of the lower eyelid; nasal, middle, and temporal refer to the individual third
0.2122–0.4392
TABLE 4 Sensitivity and specificity for different cut-off values

0.06431

0.03271
MS total
of the meiboscore to detect MGD Meiboscore as defined by

0.3222 to

0.2851 to
Arita et al., MGD defined as  4 expressible Meibomian glands.

0.3304

50.0001
0.1966

0.0029
0.1615

0.0117
Sensitivity Specificity
0.1866–0.4183 Meiboscore  1 0.963 0.108
Meiboscore  2 0.836 0.247
Meiboscore  3 0.493 0.645

0.04709

0.001405
MS LEL

0.3077 to

0.2545 to
Meiboscore  4 0.284 0.796
50.0001
0.1806

0.0066
0.1287
Meiboscore  5 0.164 0.903
0.307

0.046
0.2595–0.4803
MS LEL nasal

the diagnosis subcommittee)5 based on the data of

0.1203
0.09880

0.01044
0.1713 to

0.2471 to
Korb et al.6
0.03691
0.3752

50.0001

0.5835

0.0633
Curr Eye Res Downloaded from informahealthcare.com by University of Toronto on 03/07/15

DISCUSSION
MS LEL middle

0.2328–0.4577

In this retrospective analysis, Meibomian gland loss

0.02598
50.0001
0.1331
0.3845 to

0.2804 to

was graded as described by Arita et al.15 Recently, a

0.3503

50.0001
0.2633

0.1558

0.0157

few studies were published that used automated,

software-based grading of meibography.19–21 There
TABLE 3 Correlations between age, TBUT (Tear break-up-time), expressible glands and the different meiboscores.

might be advantages of automated analysis of

Meibomian gland loss, as such a method might be
MS LEL temporal

0.1302–0.3703

more objective. However, one potential problem of the

0.05845

0.07420

automated evaluation might be the difficulty to

0.3191 to

0.1862 to
0.05695
For personal use only.

0.2542

50.0001
0.1921

0.3807

display the entire surface of the tarsus in the same

0.004

focus. As the analysis of the individual thirds of the

upper and the lower eyelid in this study showed that
Meibomian gland atrophy is not constant over the
0.2189–0.4457

tarsal plate, the evaluation of whole tarsus seems to be

0.07779

0.07609
MS UEL

critical. Therefore, in this study, meibography pictures

0.3353 to

0.3255 to

were rated manually by two investigators with a high

0.3371

50.0001
0.2102

0.0015
0.2041

0.0014

concordance, confirming the reproducibility and the

reliability of this procedure.
In this study, collectively there was a significant
0.1337–0.3725
MS UEL nasal

higher atrophy in the nasal compared with the middle

0.05706

0.01039

and the temporal third of the lower eyelid.

0.3167 to

0.2461 to
50.0001
0.1902

0.0042

0.0633
0.1198

Interestingly, these data seem to be in contrast to the

0.257

data by Korb et al., who described, in an analysis of

the diagnostic expression of the Meibomian glands,
that there were a higher number of expressible glands
MS UEL middle

0.1706–0.4046

in the nasal third compared with the other thirds of

0.002119
0.1333

the lower eyelid.6 However, in the present study, there

0.3842 to

0.2538 to
50.0001
0.2632

50.0001

0.0472

was no significant correlation (r = 0.04, p = 0.584,

0.292

0.128

Table 3) between the meiboscore of the nasal third of

the lower eyelid and the number of expressible
glands, but a highly significant correlation for the
0.2010–0.4306
MS UEL temp

middle third (r = 0.26, p50.0001). This difference

0.05610

0.007695

might be caused by the fact that it can sometimes be

0.3158 to

0.2486 to
0.3205

50.0001
0.1893

0.0044
0.1225

0.0576

easier to display the middle third of the lower eyelid

at a high level of sharpness compared with the nasal
and the temporal third. On one hand, the higher grade
of atrophy found in the nasal third may be caused by
of the eyelids.
TBUT versus
Expr. glands

the more difficult imaging in that part of the eyelid.

Age versus

versus

On the other hand, a possible explanation might be

that the glands with the highest workload are the first
glands to become atrophic.
Current Eye Research
 Non-contact Meibography in the Diagnosis of MGD 7
Although Meibomian gland lost was not constant
over the tarsal surface in this study, there was a strong
correlation between meiboscores of the upper and the
lower eyelid and the total meiboscore, indicating that
the examination of the lower eyelid might be suffi-
cient for screening purposes in the clinical routine. For
the meiboscore of the middle third of the lower eyelid,
we could still find a strong connection with the total
meiboscore as well as with functional parameters like
TBUT or expressible glands. However, the correlation
coefficient was lower (r = 0.71) compared with the
complete lower eyelid (r = 0.89), indicating the neces-
sity to examine the complete lower eyelid.
Nonetheless, in unclear cases of MGD, the complete
examination of both the upper and the lower eyelid
should always be performed. Although there was a
Curr Eye Res Downloaded from informahealthcare.com by University of Toronto on 03/07/15
strong correlation between left and right eyes in this
study and a significant asymmetry between the eyes is
uncommon for MGD,3 we suggest the examination of
both eyes, to rule out other differential diagnoses.
The inverse correlation between Meibomian gland
loss and expressible Meibomian glands as well as
TBUT as functional parameters is of particular
importance. In the present study, the number of
expressible Meibomian glands was detected with the
method described by Korb et al.6 They used a
For personal use only.
handheld instrument to apply a certain pressure on
the eyelid. The main advantage of this instrument is
that the high variability of diagnostic expression
between different investigators caused by variance
of the applied force to the eyelid is minimized.5,6
Obstructive MGD is the main reason for an evapora-
tive dry eye but clinical signs like obvious blepharitis
are often absent. Therefore, the diagnostic expression
can be considered to be the most important functional
test in the diagnosis of obstructive MGD.5 The
negative correlation between expressible Meibomian
glands and Meibomian gland atrophy found by
meibography strongly indicates that meibography is
a useful test in the diagnosis of MGD and underlines
that Meibomian gland loss is associated with an
impaired Meibomian gland function.
The inverse correlation between meibography and
TBUT as another functional parameter was already
described by Arita et al. and Ban et al.15,19 Our data
show similar results using non-invasive and auto-
mated TBUT measurements. However, non-invasive
TBUT usually shows higher values than fluorescein-
TBUT because of the destabilizing effect of fluorescein
applied on the tear film. While it has also been
suggested that non-invasive and fluorescein-TBUT
might measure different phenomena,5 our data are in
accordance with the results of Arita and Ban et al.
However, our data also shows that for TBUT values
510 s, there is no correlation between TBUT and
meiboscore (Figure 6). This indicates that the detec-
tion of Meibomian gland loss by meibography is not
a suitable single test to detect MGD. As shown in
! 2014 Informa Healthcare USA, Inc.
Table 4, there is no cut-off value with a satisfactory
sensitivity and specificity for the detection of MGD as
defined by  4 expressible Meibomian glands. For
comparison, determination of tear-osmolarity with a
cut-off value of 316 mOsm/L was found to yield a
sensitivity of 59%, a specificity of 94%, and a overall
predictive accuracy of 89% for the diagnosis of dry
eye syndrome.5 Also other tests like, for example,
thermography might be better single test for the
diagnosis of dry eye.22,23 In our study, collectively
33.3% (5 of 15) of the eyes without Meibomian gland
atrophy (meiboscore 0) showed less than four express-
ible glands, which underlines that MGD can occur
without atrophy of the Meibomian glands, and
that this group of patients would be overlooked, if
only meibography would be used for the diagnosis
of MGD.
One major problem of all studies quantifying
Meibomian gland drop out is the possible confound-
ing of age, since the correlation of Meibomian gland
loss and aging in healthy individuals has been well
established.15,19 In our study of dry eye patients, we
also found a correlation between Meibomian gland
loss and aging. Therefore, a confounding of age
cannot be excluded in this study.
In summary, our data show that Meibomian gland
atrophy is not constantly distributed over the tarsal
plate but the examination of the lower tarsus seems to
be sufficient for the evaluation of Meibomian gland
atrophy. The correlation of the meiboscore with
expressible Meibomian glands and TBUT suggest
that in patients with detectable Meibomian gland
atrophy there is also an impaired Meibomian gland
function. However, due to its limited sensitivity and
specificity, meibography alone seems not to be suffi-
cient as a single test, but has to be interpreted in the
context of other clinical parameters such as express-
ible glands and TBUT for the diagnosis of MGD. In
the future, larger, prospective studies are needed to
further evaluate the potential of meibography in the
diagnosis of MGD.
DECLARATION OF INTEREST
The Keratograph 5 M used in this study was provided
by the company Oculus. David Finis has received
travel reimbursement from the company Oculus.
Otherwise there are no conflicts of interest related to
the present work for any of the authors.
REFERENCES
1. Schaumberg DA, Nichols JJ, Papas EB, Tong L, Uchino M,
Nichols KK. The international workshop on meibomian
gland dysfunction: report of the subcommittee on the
 8 D. Finis et al.
epidemiology of, and associated risk factors for, MGD.
Invest Ophthalmol Vis Sci 2011;52:1994–2005.
2. Finis D, Schrader S, Geerling G. Meibomian gland dys-
function. Klin Monbl Augenheilkd 2012;229:506–513.
3. Nichols KK, Foulks GN, Bron AJ, Glasgow BJ, Dogru M,
Tsubota K, et al. The international workshop on meibo-
mian gland dysfunction: executive summary. Invest
Ophthalmol Vis Sci 2011;52:1922–1929.
4. Geerling G, Tauber J, Baudouin C, Goto E, Matsumoto Y,
O’Brien T, et al. The international workshop on meibo-
mian gland dysfunction: report of the subcommittee
on management and treatment of meibomian gland
dysfunction. Invest Ophthalmol Vis Sci 2011;52:2050–2064.
5. Tomlinson A, Bron AJ, Korb DR, Amano S, Paugh JR,
Pearce EI, et al. The international workshop on meibomian
gland dysfunction: report of the diagnosis subcommittee.
Invest Ophthalmol Vis Sci 2011;52:2006–2049.
6. Korb DR, Blackie CA. Meibomian gland diagnostic
expressibility: correlation with dry eye symptoms and
gland location. Cornea 2008;27:1142–1147.
Curr Eye Res Downloaded from informahealthcare.com by University of Toronto on 03/07/15
7. Knop E, Knop N. Meibomian glands: Part IV.
Functional interactions in the pathogenesis of meibomian
gland dysfunction (MGD). Ophthalmologe 2009;106:
980–987.
8. Knop E, Knop N, Millar T, Obata H, Sullivan DA. The
international workshop on meibomian gland dysfunction:
report of the subcommittee on anatomy, physiology,
and pathophysiology of the meibomian gland. Invest
Ophthalmol Vis Sci 2011;52:1938–1978.
9. Auw-Haedrich C, Reinhard T. Chronic blepharitis.
Pathogenesis, clinical features, and therapy.
For personal use only.
Ophthalmologe 2007;104:817–826; quiz 827–828.
10. McCulley JP, Shine WE. Meibomian secretions in chronic
blepharitis. Adv Exp Med Biol 1998;438:319–326.
11. Green-Church KB, Butovich I, Willcox M, Borchman D,
Paulsen F, Barabino S, et al. The international workshop on
meibomian gland dysfunction: report of the subcommit-
tee on tear film lipids and lipid-protein interactions in
health and disease. Invest Ophthalmol Vis Sci 2011;52:
1979–1993.
12. O’Brien TP. The role of bacteria in blepharitis. Ocul Surf
2009;7:S21–S22.
13. Hwang HS, Park CW, Joo C-K. Novel noncontact
meibography with anterior segment optical coherence
tomography: Hosik meibography. Cornea 2013;32:40–43.
14. Arita R, Itoh K, Maeda S, Maeda K, Amano S. A newly
developed noninvasive and mobile pen-shaped meibogra-
phy system. Cornea 2013;32:242–247.
15. Arita R, Itoh K, Inoue K, Amano S. Noncontact infrared
meibography to document age-related changes of the
meibomian glands in a normal population.
Ophthalmology 2008;115:911–915.
16. Blackie CA, Solomon JD, Scaffidi RC, Greiner JV,
Lemp MA, Korb DR. The relationship between dry eye
symptoms and lipid layer thickness. Cornea 2009;28:
789–794.
17. The definition and classification of dry eye disease: report
of the Definition and Classification Subcommittee of the
International Dry Eye WorkShop (2007). Ocul Surf 2007;5:
75–92..
18. Brewitt H, Zierhut M. Trockenes auge: anatomie, physio-
logie, pathophysiologie, diagnostik, therapie. 1. Aufl.
Kaden, R;2001:81–85.
19. Ban Y, Shimazaki-Den S, Tsubota K, Shimazaki J.
Morphological evaluation of meibomian glands using
noncontact infrared meibography. Ocul Surf 2013;11:47–53.
20. Koh YW, Celik T, Lee HK, Petznick A, Tong L. Detection of
meibomian glands and classification of meibography
images. J Biomed Opt 2012;17:086008.
21. Pult H, Riede-Pult B. Comparison of subjective grading
and objective assessment in meibography. Cont Lens
Anterior Eye 2013;36:22–27.
22. Gonnermann J, Maier A-KB, Klein JP, Bertelmann E,
Pleyer U, Klamann MKJ. Evaluation of ocular surface
temperature in patients with pterygium. Curr Eye Res
2014;39:359–364.
23. Klamann MKJ, Maier A-KB, Gonnermann J, Klein JP,
Pleyer U. Measurement of dynamic ocular surface tem-
perature in healthy subjects using a new thermography
device. Curr Eye Res 2012;37:678–683.
Current Eye Research