Content uploaded by Miki Uchino
Author content
All content in this area was uploaded by Miki Uchino on Jan 18, 2017
Content may be subject to copyright.
ORIGINAL ARTICLE
Baseline profiles of ocular surface and tear dynamics after allogeneic
hematopoietic stem cell transplantation in patients with or without chronic
GVHD-related dry eye
Y Wang
1,2
, Y Ogawa
1
, M Dogru
3,4
, Y Tatematsu
1
, M Uchino
1
, M Kamoi
1
, N Okada
1
, S Okamoto
5
and K Tsubota
1
1
Department of Ophthalmology, Keio University School of Medicine, Tokyo, Japan;
2
Department of Ophthalmology, The Eye and
ENT Hospital of Fudan University School of Medicine, Shanghai, China;
3
Department of Ophthalmology, Tokyo Dental College,
Chiba, Japan;
4
J&J Ocular Surface and Visual Optics Department, Keio University School of Medicine, Tokyo, Japan and
5
Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan
We evaluated ocular surface alterations in allogeneic
hematopoietic stem cell transplantation (HSCT) recipi-
ents with or without chronic GVHD-related dry eye in
a prospective study. Fifty eyes of 25 post-HSCT patients
and 28 eyes of 14 age-matched healthy controls were
included. Meibomian gland (MG) obstruction, tear
evaporation rate, corneal sensitivity (CS), Schirmer test-
I, tear break-up time (BUT) and ocular surface vital
staining were examined. Conjunctival impression and
brush cytology specimens were collected to evaluate the
goblet cell density (GCD) and the inflammatory cell
numbers. Obvious MG obstruction, decreased CS and
enhanced tear evaporation rate were found in post-HSCT
patients compared with normal controls. In addition,
decreased conjunctival GCD, increased conjunctival
squamous metaplasia and inflammatory cells were noted
in cGVHD-related dry eyes compared with normal
controls and post-HSCT without dry eye subjects.
Furthermore, the conjunctival inflammatory cells were
significantly higher in severe dry eyes compared with mild
dry eyes (P¼0.03). We found comprehensive ocular
surface alteration in post-HSCT patients, regardless of
whether they had cGVHD-related dry eye or not. The
results suggest that the extent of inflammatory process
seems to have a pivotal role in the outcome of the
cGVHD-related dry eye.
Bone Marrow Transplantation (2010) 45, 1077–1083;
doi:10.1038/bmt.2009.312; published online 9 November 2009
Keywords: allogeneic hematopoietic stem cell trans-
plantation (HSCT); dry eye; impression cytology; brush
cytology; meibomian gland; tear evaporation
Introduction
Chronic GVHD is a major complication of allogeneic
hematopoietic stem cell transplantation (HSCT).
1
Ocular
surface is one of the target tissues of cGVHD. About 50%
of patients develop dry eye or experience a worsening of the
pre-existing dry eye after HSCT.
2
Dry eye is a distinctive
sign and symptom for the diagnosis of cGVHD.
1
However,
the pathogenesis of dry eye associated with cGVHD is still
unclear, and effective treatments have not yet been
established.
3
Pathogenic studies of dry eye associated with
cGVHD depend on the lacrimal gland and conjunctival
biopsy.
4–6
It is impossible to follow the alterations of the
ocular surface pathologic process after HSCT by repeated
biopsy. On the other hand, impression cytology and brush
cytology are widely used methods to evaluate the ocular
surface pathologic changes.
7
They are noninvasive, repea-
table, and useful in following the changes in the ocular
surface.
8,9
However, there are few reports on impression
cytology changes and brush cytology characteristics in
patients with cGVHD-related dry eye.
10
On the other hand,
the conditioning regimen including total body irradiation
and high incidence of meibomian gland dysfunction
(MGD) in post-GVHD patients contributes to the ocular
surface and tear function changes.
However, there is no report comparing the tear functions
and ocular surface alterations between post-HSCT patients
with or without dry eye. In a previous study,
2
we noticed
there were two types of dry eye after HSCT. One had
severe ocular surface and tear function damage with
decreased reflex tearing that occurred soon after the onset
of dry eye, whereas the other was mild with normal
reflex tearing. There are no data comparing the ocular
Received 29 June 2009; revised 11 August 2009; accepted 12 August
2009; published online 9 November 2009
Correspondence: Dr Y Ogawa, Department of Ophthalmology, Keio
University, School of Medicine, Shinanomachi 35, Shinjuku-ku, Tokyo
160-8582, Japan.
E-mail: yoko@sc.itc.keio.ac.jp or Dr M Dogru, J&J Ocular Surface
and Visual Optics Department, Keio University School of Medicine,
Shinanomachi 35, Shinjuku-ku, Tokyo 160-8582, Japan.
E-mail: muratodooru@yahoo.com
Bone Marrow Transplantation (2010) 45, 1077–1083
&2010 Macmillan Publishers Limited All rights reserved 0268-3369/10 $32.00
www.nature.com/bmt
surface and tear function differences between these two
types of dry eye.
Patients and methods
Patients
Fifty eyes of 25 patients who underwent HSCT were
enrolled at the dry eye clinic at Keio University from
January 2006 to December 2006. Included were 20 eyes of
10 patients (5 males and 5 females; range, 30–66; median,
50 years) with cGVHD-related severe dry eye, 20 eyes of 10
patients (6 males and 4 females; range, 37–62; median, 51
years) with cGVHD-related mild dry eye and 10 eyes of 5
patients (3 males and 2 females; range, 39–50; median, 45
years) without dry eye. All the patients had no previous
conjunctival or corneal disease or infections or other ocular
disease at clinical examination. Twenty-eight eyes of 14
healthy subjects (10 males and 4 females; range, 20–70;
median, 39 years) were also recruited as normal controls.
The control subjects did not have any history of ocular or
systemic disease or a history of topical eye drops or contact
lens use that would alter the ocular surface as well.
According to the global diagnostic criteria of dry eye,
and the severity grading of the Dry Eye Workshop
Report 2007,
11,12
we diagnosed the patients as having dry
eye when patients had any sign of tear film instability (tear
break-up time (BUT) p5 s, Schirmer test p5 mm), any
abnormality of the ocular surface (Rose Bengal score X3,
Fluorescein score X1) and/or symptoms of ocular irrita-
tion. Severe dry eye was defined as previously described
2,13
In brief, patients were diagnosed as having severe dry eye if
the Schirmer test with nasal stimulation (reflex tearing) was
p10 mm, and the FS and RB scores were X3 points and/or
grade 3 and 4 according to the DEWS report 2007. The
study was carried out in accordance with the principles
of the Declaration of Helsinki. Informed consents and
ethics board reviews for the examination procedure were
obtained.
Clinical examinations
The ocular surface was examined by the double vital
staining method. Two microliters of a preservative-free
combination of 1% Rose Bengal and 1% fluorescein was
instilled in the conjunctival sac by a micropipette.
14
The
staining of Rose Bengal was scored for the temporal and
nasal conjunctiva and the cornea, on a scale of 0–3 points.
Fluorescein staining score also ranged between 0 and 9
points, but only for the cornea.
15
The BUT value was
measured three times at the time of double staining, and the
mean value was used for calculation. Schirmer 1 test was
performed with standardized strips of filter paper (Alcon
Inc., Fort Worth, TX, USA). To evaluate the obstruction
of the MG orifice, digital pressure was applied on the
tarsus. The expression of meibomian secretion (meibum)
was scored as follows:
16
grade 0, clear meibum is easily
expressed; grade 1, cloudy meibum is expressed with mild
pressure; grade 2, cloudy meibum is expressed with more
than moderate pressure; and grade 3, meibum cannot be
expressed even with hard pressure.
Tear evaporimetry
The tear evaporation was measured with the evaporimeter
(KAO Corporation, Tokyo, Japan).
17
Briefly, the eyecup of
the evaporimeter tightly covered the subject’s eye, and then
the device measured the tear evaporation rate in both eyes-
closed and eyes-open conditions. In this way, we can
eliminate the evaporation from the eyelid. The computer
system calculated the difference between these two condi-
tions and gave the tear evaporation rate. The unit of tear
evaporation rate is 10
7
g/cm
2
s.
Corneal sensitivity
Measurement of corneal sensitivity (CS) was performed
using a Cochet-Bonnet aesthesiometer. The measurements
were begun with the nylon filament fully extended. The tip
of the nylon filament was applied perpendicularly to the
surface of the cornea making certain not to touch the
eyelashes and was pushed until the fiber’s first visible
bending. The length of the fiber was gradually decreased
until a blink reflex was observed. The length was recorded
in units of millimeters. Measurements were taken from the
central cornea and the mean of the measurements was
recorded as the CS reading of that eye.
18,19
Conjunctival impression cytology
The impression cytology samples were obtained under
topical anesthesia with 0.4% oxybuprocaine. A piece of
cellulose acetate filter paper (Millipore HAWP 304, Bed-
ford, MA, USA) was put on the temporal bulbar
conjunctiva and gently pressed by forceps for several
seconds. The specimens were fixed with 10% formalin
neutral buffer solution and stained with Periodic Acid-
Schiff (PAS). Five nonoverlapping areas of 400 magni-
fication were randomly selected and photographed. The
goblet cell density (GCD) was reported as cells per square
millimeter. The conjunctival epithelial squamous metapla-
sia was evaluated according to Nelson’s grading scheme.
20
Conjunctival brush cytology
The brush cytology samples were collected after adminis-
tration of topical anesthesia with 0.4% oxybuprocaine. The
central upper palpebral conjunctiva was gently brushed
seven times with a disposable dental brush 1.5 mm in
diameter (Dentalpro, Jacks. Co., Osaka, Japan). After
sampling, the brush was immediately put in 1 ml of Hank’s
solution and shaken several times to detach the cells
from the brush. The suspended cells were centrifuged
with cytocentrifuge at 700 r.p.m. for 10 min to make the
monolayer cell smears. The slides were stained by diff-quick
staining. We counted up to 500 cells including inflamma-
tory cells and epithelial cells in nonoverlapping fields under
microscopic observation (magnification, 400). The in-
flammation was reported as the number of inflammatory
cells in the total number of 500 brush cells.
10
Statistical analysis
The data were analyzed by Instat (GraphPad Software, San
Diego, CA, USA). Mann–Whitney U-test was used to com-
pare the onset duration of dry eye. Kruskal–Wallis H-test
Baseline profiles of ocular surface after HSCT
Y Wang et al
1078
Bone Marrow Transplantation
was used for the comparisons of clinical examination
parameters, tear evaporation rates, GCD, conjunctival
squamous metaplasia, and inflammatory cell amount. The
probability level of 5% was chosen as the statistical
significance.
Results
Demographic characteristics
Patients’ demographic characteristics were summarized in
Table 1. The onset of dry eye in cGVHD-related severe and
mild dry eye was 6.8±2.5 and 13.2±9.1 months, respec-
tively, after HSCT. The onset of dry eye in the severe dry
eye group was significantly earlier than the onset in the mild
dry eye group (P¼0.02). Nine out of 10 severe dry eye
patients had systemic cGVHD, but only 3 in 10 mild dry
eye patients had systemic cGVHD.
Clinical examination parameters
The baseline scores of CS, ocular surface vital staining and
tear function were summarized in Table 2. Obviously
decreased CS was found in post-HSCT patients either with
or without dry eye, but statistically significant decrease was
found only in the severe dry eye group. Although the mean
CS in the severe dry eye group was considerably lower than
those with mild dry eye and post-HSCT without the dry eye
groups, there was no statistically significant difference
among the three groups. Obvious MG orifice obstruction
(grade 41) was noted in 40 of 50 eyes of the post-HSCT
patients as shown in Table 3. MG orifice obstruction
degree in post-HSCT patients was statistically higher
than normal controls, but there was no significant
difference between the three post-HSCT groups. The tear
evaporation rate in normal control, post-HSCT without
dry eye, mild dry eye, and severe dry eye group was
2.2±1.53 10
7
g/cm
2
s, 4.42±2.13 10
7
,3.6
±1.66 10
7
,
Table 1 Demographic characteristics
Case no Age (years) Gender Diagnosis Systemic cGVHD Dry eye Onset (month from
HSCT to dry eye)
Month since HSCT
1 52 M MDS Lung, skin, mouth Severe 7 29
2 63 M MDS Liver, skin, mouth Severe 6 65
3 64 M MM Skin, mouth Severe 7 30
4 35 F CML Mouth Severe 7 61
5 30 F AML Mouth Severe 7 93
6 50 M AML Mouth, skin, liver Severe 3 69
7 36 M CML Mouth Severe 7 88
8 66 F MM Mouth, skin Severe 11 58
9 57 F MDS Mouth, skin Severe 10 34
10 34 F ALL (–) Severe 3 144
11 49 M ALL Lung, liver, skin Mild 9 24
12 54 M ALL Mouth, skin, intestinal Mild 12 66
13 59 F ALL () Mild 11 96
14 56 M MDS Liver Mild 36 156
15 62 F ALL () Mild 2.5 19
16 37 M MDS () Mild 5 28
17 45 M MDS () Mild 12 12
18 61 F ALL () Mild 16 36
19 51 M AML () Mild 15 36
20 58 F NHL () Mild 13 30
21 45 M AA ()()()60
22 44 F AML ()()() 120
23 39 F CML ()()() 144
24 50 M AML ()()()30
25 48 M AML ()()()3
Abbreviations: AA ¼aplastic anemia; cGVHD ¼chronic GVHD; F ¼female; HSCT ¼hematopoietic stem cell transplantation; M ¼male; MDS ¼
myelodysplastic syndrome; MM ¼multiple myeloma; NHL ¼non-Hodgkin lymphoma.
Table 2 The scores of tear functions, corneal sensitivity and vital stainings
Tear evaporation rate
(10
7
g/cm
2
s)
CS (mm) Schirmer test (mm) BUT (s) FS (points) RB (points)
Normal controls 2.2±1.53 60 16.35±11.82 8.92±3.17 0.54±0.66 0.13±0.34
Post-HSCT without dry eye 4.42±2.13 57.5±4.63 14.7±10.34 10 0.3±0.95 0.3±0.95
cGVHD-related mild dry eye 3.6±1.66 57.25±4.16 13.06±11.05 4.85±2.18
a,b
2.4±1.93
a,b
1.8±1.85
a
cGVHD-related severe dry eye 5.98±3.61
a
54.98±7.75
a
2.45±2.28
a,b,c
2.68±1.4
a,b
5.6±2.56
a,b,c
5.55±2.06
a,b,c
Abbreviations: BUT ¼tear break-up time; cGVHD ¼chronic GVHD; FS ¼fluorescein score; HSCT ¼hematopoietic stem cell transplantation; RB ¼Rose
Bengal score.
a
Po0.05, compared with normal controls, Kruskal–Wallis test.
b
Po0.05, compared with post-HSCT without dry eye patients, Kruskal–Wallis test.
c
Po0.05, compared with cGVHD-related mild dry eye patients, Kruskal–Wallis test.
Baseline profiles of ocular surface after HSCT
Y Wang et al
1079
Bone Marrow Transplantation
and 5.98±3.61 10
7
g/cm
2
s, respectively. Although the
mean tear evaporation rate in mild dry eye and post-HSCT
without dry eye patients was higher than in normal
controls, statistically increased tear evaporation was found
only in cGVHD-related severe dry eye patients (Po0.001).
Conjunctival impression cytology
Conjunctival specimens from normal controls and post-
HSCT without dry eye subjects showed plenty of goblet
cells and mucin pick up (Figure 1). The goblet cell
densities in these two groups were 1313.13±733.82 and
1030±433.14 cells/mm
2
. The mean GCD in the cGVHD-
related mild and severe dry eye groups was 706.49±583.52
and 396.36±381.00 cells/mm
2
. Both were obviously lower
than the former two groups without dry eye (Table 4).
Moreover, significant conjunctival epithelial squamous
metaplasia was noted in severe dry eye patients. The mean
grades of squamous metaplasia in normal control, post-
HSCT without dry eye, and mild dry eye groups were
0.70±0.46, 0.71±0.52, and 0.72±0.56, respectively. There
was no statistical difference among the three groups.
However, the average grade of squamous metaplasia in
severe dry eye subjects was 1.61±0.72, which was
significantly higher than that in the other three groups
(Table 4). Except decreased GCD, the PAS staining also
showed inflammation in some impression cytology speci-
mens from the cGVHD-related severe dry eye and mild dry
eye patients (Figures 2 and 3). In addition, the PAS staining
also indicated the intense inflammatory cell infiltration that
frequently appeared with the abnormal mucin conglom-
eration.
Brush cytology
There was no inflammatory cell in the brush cytology
specimens from normal controls. In contrast, a different
extent of inflammatory cell infiltration was found in the
specimens from post-HSCT patients (Figure 4). The mean
number of inflammatory cells in 500 brush cells in post-
HSCT without dry eye, mild dry eye, and severe dry eye
specimens were 5.44±6.04 cells, 14.64±9.75 cells, and
22.64±11.69 cells, respectively. The mean inflammatory
cell numbers in both cGVHD-related mild and severe dry
eye specimens were significantly higher than in normal
controls and post-HSCT without the dry eye group
(Po0.001). Moreover, the inflammatory cell number in
the severe dry eye group was statistically higher than in the
mild dry eye group (P¼0.03).
Discussion
In this study, we evaluated the detailed baseline profiles of
ocular surface and tear function alterations in post-HSCT
patients with or without dry eye disease. We found
obviously decreased CS in post-HSCT subjects either with
or without dry eye disease. Although the reduction of CS in
severe dry eye patients seemed to be more prominent, there
were no statistical differences compared with post-HSCT
without dry eye and mild dry eye patients. A reduction of
CS has been reported in dry eye patients.
18,19
We also noted
decreased CS in cGVHD-related dry eye patients in our
previous study.
3
Considering the conditioning regimens
before HSCT, such as total body irradiation, which
includes orbital irradiation, we thought decreased CS in
cGVHD-related dry eye patients may not be because of the
dry eye pathologic process. Therefore, we recruited post-
HSCT without dry eye subjects in this study. According to
the present results, decreased CS was obvious even in post-
HSCT without dry eye patients. Our study suggested that
the conditioning regimens before HSCT may be more
responsible for the decreased CS in cGVHD-related dry eye
disease.
Moreover, increased MG obstruction grade was found in
post-HSCT both with and without dry eye patients.
Consistent with this, an increased tendency in the tear
evaporation rate was noted in post-HSCT patients.
However, the statistical increase was found only in the
severe dry eye patients. MGs produce lipid material that
spread and cover the ocular surface during the blink to
keep the tear film stable and to reduce the tear evaporation.
The dysfunction of the MG can induce evaporative
dry eye.
21
On the other hand, decreased tear production
Table 3 Comparison of orifice obstruction grade of meibomian gland
Orifice obstruction Normal controls Post-HSCT without dry eye cGVHD-related mild dry eye cGVHD-related severe dry eye
Grade 0 26 (92.86%) 2 (20%) 0 2 (10%)
Grade 1 2 (7.14%) 2 (20%) 4 (20%) 0
Grade 2 0 2 (20%) 7 (35%) 5 (25%)
Grade 3 0 4 (40%) 9 (45%) 13 (65%)
Abbreviations: cGVHD ¼chronic GVHD; HSCT ¼hematopoietic stem cell transplantation.
Figure 1 Representative conjunctival impression cytology specimens
from a 50-year-old male, post-HSCT without dry eye subject. Note plenty
goblet cells (black arrow) and mucin pick up (yellow arrows). Periodic acid-
Schiff (PAS) staining, magnification, 400.
Baseline profiles of ocular surface after HSCT
Y Wang et al
1080
Bone Marrow Transplantation
can also cause an increase in the tear evaporation rate.
22
Taken together, these findings suggest that increased MGD
in post-HSCT patients cause enhanced tear evaporation in
both groups of patients with or without dry eye, and
enhanced tear evaporation acting together with the
decreased tear production, induced the enhancement of
ocular surface changes and tear function changes in
cGVHD-related severe dry eye patients.
In our previous study,
2
we noticed the fact that there
were two types of dry eye after HSCT. One had severe
ocular surface and tear function alterations with decreased
reflex tearing, which occurred soon after the onset of dry
eye. Another was mild with normal reflex tearing. In this
study, we performed a further comparison about the
difference between these two types of dry eye. We found
that the onset of cGVHD-related severe dry eye was
obviously earlier than that of mild dry eye. In our patients,
severe dry eye occurred around 6.8±2.5 months after
HSCT, but the onset of mild dry eye was around 13.2±9.1
months after HSCT. There was one patient in whom the
mild dry eye occurred 3 years after the HSCT. Moreover,
most severe dry eye patients had systemic cGVHD, whereas
only a few patients in the mild dry group had systemic
cGVHD. Those findings indicated the different pathologic
processes in cGVHD-related severe and mild dry eye
disease.
For further comparison, we performed conjunctival
impression cytology to evaluate the alterations and
differences in GCD and squamous metaplasia in these
two types of dry eye disease. GCD and squamous
metaplasia are two parameters that were widely used to
evaluate the ocular surface epithelial condition in dry eye
and other ocular surface disease.
18,23,24
In addition, goblet
cell content has been reported to be a sensitive indicator of
primary ocular surface disease.
25,26
However, the report
concerning the conjunctival impression cytology character-
istics in cGVHD-related dry eye disease was still rare.
8
In
this study, we found that both cGVHD-related mild and
severe dry eye specimens showed significantly decreased
GCD compared with normal controls and post-HSCT
without dry eye specimens. Moreover, the mean GCD in
severe dry eye patients was only about half of the density in
mild dry eye patients with decrease in goblet cell numbers
along with increased squamous metaplasia and keratiniza-
tion of the ocular surface.
27
In cGVHD-related mild dry
eye, although the GCD decreased, there was no obvious
squamous metaplasia. However, high grades of squamous
metaplasia with a further decrease in goblet cell numbers
were found in severe dry eye patients. On the basis of these
findings, we confirmed GCD to be a sensitive indicator for
evaluating the extent of cGVHD-related dry eye disease.
Except for decreased GCD and squamous metaplasia, some
impression cytology specimens from cGVHD-related dry
eye patients showed inflammatory cell infiltration in the
conjunctival epithelium. These intense inflammatory areas
often appeared in the area with clustered abnormal mucin.
It indicates that the inflammation process involves the
pathologic changes of cGVHD-related dry eye, which may
influence the secretion and physiological characteristics of
the ocular surface mucin.
For revealing the inflammation status in cGVHD-related
dry eye disease and comparing the inflammation extent
between cGVHD-related severe and mild dry eye disease,
we collected the conjunctival brush cytology specimens and
calculated the amount of inflammatory cells. We found
considerably increased inflammatory cell numbers in both
cGVHD-related severe dry eye and mild dry eye patients
compared with normal controls and post-HSCT without
dry eye subjects. Moreover, the number of inflammatory
cells in severe dry eye specimens was significantly higher
than in mild dry eye specimens. Recently, increased
evidence suggests that dry eye is an inflammation-related
disease.
28
Our previous study also found many inflamma-
tory markers expressed in biopsy samples of the conjuncti-
va and lacrimal gland from cGVHD-related dry eye
patients.
4–6
The present findings confirm that inflam-
mation is involved in the pathogenesis of cGVHD-related
dry eye. Moreover, our results suggest that the extent of
Figure 2 Representative conjunctival impression cytology specimens
from a 37-year-old male, cGVHD-related mild dry eye patient. Note
decreased goblet cell number (black arrow), and the conjunctival
epithelium has no obvious keratinization. Periodic acid-Schiff (PAS)
staining, magnification, 400.
Table 4 Comparison of conjunctival GCD and epithelium squamous metaplasia
Normal controls Post-HSCT without dry eye cGVHD-related mild dry eye cGVHD-related severe dry eye
GCD (cells/mm
2
) 1313.13±733.82 1030±433.14 706.49±583.52
a
396.36±381.00
a,b
Squamous metaplasia (Nelson’s) 0.70±0.46 0.71±0.52 0.72±0.56 1.61±0.72
a,b,c
Abbreviations: cGVHD ¼chronic GVHD; GCD ¼goblet cell density; HSCT ¼hematopoietic stem cell transplantation.
a
Po0.05, compared with normal controls, Kruskal–Wallis test.
b
Po0.05, compared with post-HCT without dry eye patients, Kruskal–Wallis test.
c
Po0.05, compared with cGVHD-related mild dry eye patients, Kruskal–Wallis test.
Baseline profiles of ocular surface after HSCT
Y Wang et al
1081
Bone Marrow Transplantation
inflammation may be responsible for the different extent of
pathologic damage in cGVHD-related mild and severe dry
eye disease.
As the time we collected the brush cytology and
impression cytology samples was relatively far from the
onset of the dry eye, it is hard to distinguish whether the
inflammation is the consequence or a cause of cGVHD-
related dry eye. However, these two techniques are
relatively noninvasive and repeatable examinations. They
are very suitable to monitor the dynamic changes of the
ocular surface epithelium and inflammation after HSCT.
Moreover, the impression cytology and brush cytology
samples can also be used to perform immunohistochemical
staining, enzyme-linked immunosorbent assay, flow cyto-
metry, and mRNA expression analysis.
9,10,29,30
Therefore,
they are also useful to monitor the pathologic progress in
cGVHD-related dry eye disease and helpful for investigat-
ing the etiology of cGVHD-related dry eye disease.
In this study, we used the tear evaporimetry, MG
expression examination, impression cytology, and brush
cytology to give a comprehensive evaluation of the changes
of ocular surface and tear functions in patients with
cGVHD-related mild and severe dry eye disease, and
compared the results with healthy controls and post-HSCT
patients without dry eye disease. According to the findings
in this study, we speculated that the extent of the
inflammatory process seems to have a pivotal role in the
outcome of cGVHD-related dry eye disease with changes in
tear evaporation, CS and GCD acting as determinants of
the differences of the ocular surface healthy status.
In conclusion, our present data provide the baseline data
of each type of dry eye disease associated with cGVHD
using these methods. These data are also useful for future
therapeutic evaluation.
Conflict of interest
The authors declare no conflict of interest.
Acknowledgements
This work was presented in part at the 60th Japan Congress of
Clinical Ophthalmology, 5–8 October 2006, Kyoto, Japan and
ARVO 2007, Fort Lauderdale, USA. This work was supported
by a grant from the Japanese Ministry of Education, Science,
Sports and Culture #20592058.
References
1 Filipovich AH, Weisdorf D, Pavletic S, Socie G, Wingard JR,
Lee SJ et al. National Institutes of Health consensus
development project on criteria for clinical trials in chronic
Figure 3 Representative conjunctival impression cytology specimens from a 52-year-old male cGVHD-related severe dry eye patient. (a) Note obvious
squamous meterplasia and decreased GCD. Black arrows indicate the small goblet cells with decreased cytoplasmic mucin. (b) Note inflammatory
infiltration (black arrows) in the conjunctival epithelium. Periodic acid-Schiff (PAS) staining, magnification, 400.
Figure 4 Representative brush cytology (BC) specimens. (a) Specimen from a 37-year-old male, cGVHD-related mild dry eye patient. Note inflammatory
cells (black arrow). (b) Specimen from a 66-year-old female cGVHD-related severe dry eye patient. Note obviously inflammatory cells (black arrows) and
keratinized conjunctival epithelial cells (yellow arrow). Diff-quick staining, magnification, 400.
Baseline profiles of ocular surface after HSCT
Y Wang et al
1082
Bone Marrow Transplantation
graft-versus-host disease: I. Diagnosis and staging working
group report. Biol Blood Marrow Transplant 2005; 11:
945–956.
2 Ogawa Y, Okamoto S, Wakui M, Watanabe R, Yamada M,
Yoshino M et al. Dry eye after hematopoietic stem cell
transplantation. Br J Ophthalmol 1999; 83: 1125–1130.
3 Ogawa Y, Okamoto S, Mori T, Yamada M, Mashima Y,
Watanabe R et al. Autologous serum eye drops for the
treatment of severe dry eye in patients with chronic
graft-versus-host disease. Bone Marrow Transplant 2003; 31:
579–583.
4 Ogawa Y, Yamazaki K, Kuwana M, Mashima Y,
Nakamura Y, Ishida S et al. A significant role of stromal
fibroblasts in rapidly progressive dry eye in patients with
chronic GVHD. Invest Ophthalmol Vis Sci 2001; 42: 111–119.
5 Ogawa Y, Kuwana M, Yamazaki K, Mashima Y, Yamada M,
Mori T et al. Periductal area as the primary site for T-cell
activation in lacrimal gland chronic graft-versus-host disease.
Invest Ophthalmol Vis Sci 2003; 44: 1888–1896.
6 Rojas B, Cuhna R, Zafirakis P, Ramirez JM, Lizan-garciia M,
Zhao T et al. Cell populations and adhesion molecules
expression in conjunctiva before and after bone marrow
transplantation. Exp Eye Res 2005; 81: 313–325.
7 Singh R, Joseph A, Umapathy T, Tint NL, Dua HS.
Impression cytology of the ocular surface. Br J Ophthalmol
2005; 89: 1655–1659.
8 Dogru M, Okada N, Asano-Kato N, Tanaka M, Igarashi A,
Takano Y et al. Atopic ocular surface disease: implications
on tear function and ocular surface mucins. Cornea 2005; 24:
S18–S23.
9 Miyoshi T, Fukagawa K, Shimmura S, Fujishima H, Takano
Y, Takamura E et al. Interleukin-8 concentrations in
conjunctival epithelium brush cytology samples correlate with
neutrophil, eosinophil infiltration, and corneal damage.
Cornea 2001; 20: 743–747.
10 Aronni S, Cortes M, Sacchetti M, Lambiase A, Micera A,
Sgrulletta R et al. Upregulation of ICAM-1 expression in the
conjunctiva of patients with chronic graft-versus-host disease.
Eur J Ophthalmol 2006; 16: 17–23.
11 Lemp MA. Report of the national eye institute/industry
workshop on clinical trials in dry eyes. CLAO J 1995; 21:
221–232.
12 Report of the International Dry Eye Work Shop. Ocul Surf
2007; 5: 75–92.
13 Tsubota K. The importance of the Schirmer test with nasal
stimulation test. Am J Ophthalmol 1991; 111: 106–108.
14 Toda I, Tsubota K. Practical double vital staining for ocular
surface evaluation. Cornea 1993; 12: 366–368.
15 Dogru M, Asano-Kato N, Tanaka M, Igarashi A, Shimmura
S, Shimazaki J et al. Ocular surface and MUC5AC alterations
in atopic patients with corneal shield ulcers. Curr Eye Res
2005; 30: 897–908.
16 Shimazaki J, Goto E, Ono M, Shimmura S, Tsubota K.
Meibomian gland dysfunction in patients with Sjogren
syndrome. Ophthalmology 1998; 105: 1485–1488.
17 Goto E, Endo K, Suzuki A, Fujikura Y, Matsumoto Y,
Tsubota K. Tear evaporation dynamics in normal subjects and
subjects with obstructive meibomian gland dysfunction. Invest
Ophthalmol Vis Sci 2003; 44: 533–539.
18 Xu KP, Yagi Y, Tsubota K. Decrease in corneal sensitivity and
change in tear function in dry eye. Cornea 1996; 15: 235–239.
19 Bourcier T, Acosta MC, Borderie V, Borra
´s F, Gallar J, Bury
Tet al. Decreased corneal sensitivity in patients with dry eye.
Invest Ophthalmol Vis Sci 2005; 46: 2341–2345.
20 Nelson JD. Impression cytology. Cornea 1988; 7: 71–81.
21 McCulley JP, Shine WE. The lipid layer of tears: dependent on
meibomian gland function. Exp Eye Res 2004; 78: 361–365.
22 Mathers W. Evaporation from the ocular surface. Exp Eye Res
2004; 78: 389–394.
23 Tseng SC. Staging of conjunctival squamous metaplasia by
impression cytology. Ophthalmology 1985; 92: 728–733.
24 Aragona P, Ferreri G, Micali A, Puzzolo D. Morphological
changes of the conjunctival epithelium in contact lens wearers
evaluated by impression cytology. Eye 1998; 12: 461–466.
25 Tseng SC, Hirst LW, Maumenee AE, Kenyon KR, Sun TT,
Green WR. Possible mechanisms for the loss of goblet cells in
mucin-deficient disorders. Ophthalmology 1984; 91: 545–552.
26 Kinoshita S, Kiorpes TC, Friend J, Thoft RA. Goblet cell
density in ocular surface disease. A better indicator than tear
mucin. Arch Ophthalmol 1983; 101: 1284–1287.
27 Gipson IK, Hori Y, Argueso P. Character of ocular surface
mucins and their alteration in dry eye disease. Ocul Surf 2004;
2: 131–147.
28 Stern ME, Pflugfelder SC. Inflammation in dry eye. Ocul Surf
2004; 2: 124–130.
29 Tsubota K, Fujihara T, Saito K, Takeuchi T. Conjunctival
epithelium expression of HLA-DR in dry eye patients.
Ophthalmologica 1999; 213: 16–19.
30 Rolando M, Barabino S, Mingari C, Moretti S, Giuffrida S,
Calabria G. Distribution of conjunctival HLA-DR expression
and the pathogenesis of damage in early dry eyes. Cornea 2005;
24: 951–954.
Baseline profiles of ocular surface after HSCT
Y Wang et al
1083
Bone Marrow Transplantation
