C L I N I C A L A N D E X P E R I M E N TA L

RESEARCH PAPER

Longitudinal changes in Langerhans cell density of the cornea and

conjunctiva in contact lens-induced dry eye

Clin Exp Optom 2017; 100: 3340 DOI:10.1111/cxo.12399

Yahya Alzahrani MScOptom Background: The aim was to determine longitudinal changes in Langerhans cell density
Luisa H Colorado OD (LCD) in the human cornea and conjunctiva during asymptomatic and symptomatic contact
Nicola Pritchard BAppSc(Optom) PhD lens wear.
Nathan Efron PhD DSc Methods: Twenty-ve participants with contact lens-induced dry eye (CLIDE) and 35 without
Institute of Health and Biomedical Innovation, and School CLIDE (NO-CLIDE), diagnosed using a range of symptom questionnaires and objective tests
of Optometry and Vision Science, Queensland University (tear lm break up, cotton thread tear test and corneal staining) were enrolled. The central
of Technology, Kelvin Grove, Queensland, Australia
cornea and nasal bulbar conjunctiva were examined using a Heidelberg laser scanning
E-mail: n.efron@qut.edu.au
confocal microscope at baseline and following one, four and 24 weeks wear of daily disposable
hydrogel contact lenses. Twenty-three non-contact lens-wearing controls were also examined.
Langerhans cells were counted manually from randomly selected images.
Results: In the cornea, mean and standard error of the mean LCD was greater after one week
of lens wear in CLIDE (55 7 cells/mm2) versus NO-CLIDE (43 4 cells/mm2) (p = 0.041) and
controls (27 4 cells/mm2) (p < 0.001). LCD was also greater in NO-CLIDE versus controls
(p = 0.010). At week 4, LCD was greater in CLIDE (41 6 cells/mm2) versus controls (27 4
cells/mm2) (p = 0.004). There were no other signicant differences between groups at weeks
four or 24. In the conjunctiva, LCD was greater after one week of lens wear in CLIDE (17 1
cells/mm2) (p = 0.003) and NO-CLIDE (17 3 cells/mm2) (p = 0.001) versus controls (7 1
Submitted: 5 October 2015 cells/mm2). There were no signicant differences between groups at weeks four or 24.
Revised: 21 December 2015 Conclusions: The initial transient increase in corneal and conjunctival LCD in CLIDE (versus
Accepted for publication: 23 December 2015 NO-CLIDE) suggests an inammatory component in the aetiology of this condition.

Key words: conjunctiva, contact lens, cornea, dry eye, inammation, Langerhans cells

Contact lens discomfort is a major unsolved
problem in ophthalmic science. Between 28
and 50 per cent of contact lens wearers expe-
rience discomfort associated with contact
lens wear, according to population-based
studies.1 It is a primary reason for dissatisfac-
tion with, and discontinuation from, lens
wear. Contact lens wearers use various terms
to describe symptoms of discomfort, but the
overwhelming majority categorise this
discomfort as dryness.1
The cause of contact lens-induced dry eye
(CLIDE) is not well understood and is likely
to be multifactorial. Factors such as eyelid
and ocular surface anatomy, blinking charac-
teristics, systemic factors, hormonal inu-
ences, lens-induced changes to the tear lm
and ocular surface, lens design, lens material,
lens surface characteristics, such as wettability
and lubricity, environmental factors, and pre-
existing ocular pathology, such as dry eye, are
all thought to have the potential to contribute
to CLIDE.2
There are two general lines of enquiry that
could shed further light on the question of
the degree to which contact lens wear and
especially CLIDE is inammatory.3 At a
clinical level, the primary sign of inamma-
tion that has been used to study the impact
of contact lens wear on the eye is limbal and
bulbar conjunctival redness.4 In fact, the rst
two clinical reports of contact lens wearing
trials on humans, conducted independently
in the late 1880s by Adolf Fick5 and August
Mller,6 used conjunctival redness as a mea-
sure of the physiological impact of lens wear.
Cytochemical analysis of ocular tissues
provides a more direct measure of the inam-
matory status of the eye during contact lens
wear. Inammatory mediators, such as
histamine, granulocyte macrophage colony-
stimulating factor, tumour necrosis factor al-
pha, interleukin-6, interleukin-8, leukotriene
B4, matrix metallopeptidase-9 and epidermal
growth factor can be assayed from the tear
lm.7,8 These mediators are altered by daily
and extended wear of rigid and soft contact
lenses and so serve as markers for contact
lens-induced ocular inammation.7,8
The laser scanning confocal microscope
(LSCM) has found considerable value in
assessing the response of the anterior ocular
structures to contact lens wear at a cellular
level.9 Various studies have used this instru-
ment to observe inammatory cells, which
are presumed to be Langerhans cells in both
the central and peripheral cornea10 as well as
the bulbar11 and palpebral12 conjunctiva.
They are about 15 m in diameter and
appear in various forms. Immature
Langerhans cells either lack dendrites or
have small dendritic processes. Mature
Langerhans cells develop long interdigitating
dendrites and can reside alone or form a
wire net-like structure, the latter appearance
being especially prevalent in the conjunc-
tiva.10 Cross-sectional studies have shown an
increase in Langerhans cell density (LCD)
in the cornea1315 and conjunctiva16 in

2016 Optometry Australia Clinical and Experimental

Clinical andOptometry 100.1
Experimental January 2017
Optometry 2016
1
33
Langerhans cells
cells in
in contact
contactlens
lenswear
wearAlzahrani,
Alzahrani,Colorado,
Colorado,Pritchard
response to contact lens wear; however, longi-
tudinal studies of LCD in response to contact
lens wear have not been undertaken.
This study sought to determine:
1. longitudinal changes in LCD during
asymptomatic contact lens wear and
2. whether there is an increased LCD associ-
ated with CLIDE.
A higher LCD may suggest in vivo cytological
evidence of an inammatory basis for CLIDE.
METHODS
Ametropic participants who were willing to
wear contact lenses for six months were
enrolled and stratied into those with CLIDE
and those without CLIDE (NO-CLIDE) (see
below). A control group of non-contact lens
wearers was also enrolled. LCD in the cornea
and bulbar conjunctiva were monitored
using a LSCM in all three groups at baseline
(when contact lenses were dispensed) and
after one, four and 24 weeks.
Sample size calculation (G*Power 3.1)17 to
detect a difference of 10 cells/mm2 at a signif-
icance level of p 0.05, assuming a measure-
ment standard deviation of 8 cells/mm2,
revealed that a minimum of 12 participants
per group were required to be able to reject,
with 80 per cent power, the null hypothesis
that the population group means are equal.
Allowing for participant drop-out over the
24 weeks of the study, the minimum number
of participants recruited per contact lens
subgroup was set at 25.
Ethics approval for this study was obtained
from the Queensland University of Technol-
ogy Human Research Ethics Committee (ap-
proval number 1300000117). The research
followed the tenets of the Declaration of Hel-
sinki and informed consent was obtained
from the subjects after explanation of the na-
ture and possible consequences of the study.
Participants
Potential participants were recruited from
the staff and student population of the
Queensland University of Technology,
Brisbane, Australia. Inclusion criteria were
as follows: good general health (self-re-
ported), no contact lens wear for the previous
six months, age ranging from 18 to 50 years
and willingness to be tted with, and if
suitable to wear, daily disposable lenses on a
regular basis for 24 weeks.
Clinical and
and Experimental
ExperimentalOptometry
Optometry2016
100.1 January 2017
2
34
Pritchard and
and Efron
Efron
The following exclusion criteria were
applied: recent ocular inammation, history
of ocular trauma or surgery, currently using
ocular medication with the exception of
non-preservative articial tear supplements,
systemic disease that may affect the cornea
or conjunctiva, hypertension, diabetes, dry
eye, pregnancy or breastfeeding.
Additional exclusion criteria for the con-
tact lens wearing group were as follows: astig-
matism of more than 1.50 D, myopia more
than -7.00 D or hyperopia more than +2.00 D.
Participants entered into the study who
were interested in wearing contact lenses
were tted with daily disposable lenses and
given instructions in their use. After one week
of lens wear, participants were stratied into
CLIDE and NO-CLIDE. Recruitment contin-
ued until there were at least 25 participants
in each contact lens subgroup. Twenty-three
non-contact lens wearing control participants
were also recruited.
At each visit, participants were examined
using a variety of dry eye examination tech-
niques. All examinations were performed in
the morning between 7:00 hours to 12:00
hours to avoid any potential confounding
inuence of diurnal variation.
Dry eye assessment
DRY EYE QUESTIONNAIRE-5 (DEQ-5)
This is a self-administrated questionnaire
consisting of ve questions about frequency
and intensity of eye discomfort and eye
dryness and frequency of watery eyes.18
Scores for each question are assigned, rang-
ing from Never (score of 0) to Constantly
(4) or Very intense (5). A total score of more
than six (out of 22) indicates dry eye.
CONTACT LENS DRY EYE QUESTIONNAIRE-
8 (CLDEQ-8)
The Contact Lens Dry Eye Questionnaire-8
was developed specically to assess dry eye
symptoms associated with contact lens wear.19
This questionnaire consists of eight questions
which are then analysed to reveal a score out
of 37, where higher scores indicate greater
discomfort. In order to stratify our cohort
into CLIDE and NO-CLIDE, we adopted a
threshold score of more than 17 as indicating
CLIDE (and up to 17 indicating NO-CLIDE),
based upon the validation study of Chalmers
and colleagues,19 who demonstrated that a
mean score of 17.4 or less was associated with
an overall opinion rating of fair or better.
NON-INVASIVE TEAR BREAK-UP TIME
(NITBUT) TEST
This is a non-invasive measure of the capacity
of the tear lm to remain intact on the sur-
face of the cornea.20 The right and left eyes
were assessed and the average of the three
readings per eye was taken as the mean value.
The participant was positioned behind a
keratometer and asked to blink once and
then refrain from blinking. The time elapsed
until the rst sign of any distortion of the im-
age of the keratometer mires (NITBUT) was
recorded using an electronic timing device.
A reading of less than nine seconds was con-
sidered as indicating dry eye.21
PHENOL RED THREAD TEST (PRTT)
This is a test of the volume of tears in the
lower cul de sac, based on the Hamano and
colleagues22 cotton test. One end of a yellow
cotton thread impregnated with phenol red
(Pcot-test, Tianjin Jingming New Technology
Development Ltd, Tianjin Hi-Tech Industrial
Park, China) is draped over the temporal as-
pect of the lower eyelid and into the lower
conjunctival sac for 20 seconds. The greater
the length of the thread that becomes wet
with tears, indicated by that portion of the
thread turning red, the greater is the tear vol-
ume. A wet length of 10 mm or less was con-
sidered to indicate dry eye.
CORNEAL FLUORESCEIN STAINING (CFS)
A drop of saline was instilled on a uorescein-
impregnated strip, which was then touched
gently on the lower bulbar conjunctiva. The
cobalt blue light on the slit-lamp bio-
microscope and yellow observation lter were
used to evaluate the extent of corneal stain-
ing with reference to the validated Efron
grading scale.23 The extent of staining was
graded from 0 (normal) to 4 (severe stain-
ing). A grade of staining of two or more was
considered as being abnormal.23
Diagnosis of dry dye
The following tests were conducted as part of
the baseline examination protocol on all
those recruited for potential participation in
the study: DEQ-5, NITBUT, PRTT and CFS.
In accordance with the recommendation of
various authors24,25 that a combined battery
of subjective and objective tests should be
used to diagnose dry eye, we adopted the
criteria that participants who passed DEQ-5
and at least one of the objective dry eye tests
(NITBUT, PRTT or CFS) were considered el-
igible for inclusion in the study.
2016 Optometry Australia

One week after the baseline examination,
CLDEQ-8 was used to assess those in the
contact lens-wearing group. Those failing
the CLDEQ-8 test (score greater than 17)
were assigned to the CLIDE subgroup and
the remainder formed the NO-CLIDE
subgroup. The subjective (CLDEQ-8) and
objective (NITBUT, PRTT or CFS) dry eye
tests were repeated four and 24 weeks from
the baseline examination.
Laser scanning confocal
microscopy
Images of Langerhans cells were captured
using a Heidelberg LSCM (HRT3) in
combination with a Rostock Corneal Module
(Heidelberg Engineering GmbH, Heidel-
berg, Germany). A new disposable Perspex
applanating cap (Tomocap) was used for each
participant. Before tting the Tomocap to the
Rostock Corneal Module, it was lled with a
GenTeal Gel (Novartis Pharmaceuticals
Australia Pty. Limited, Sydney, NSW,
Australia). The gel facilitates an optical
coupling of the Rostock Corneal Module
objective lens with the back surface of the
Tomocap. An anaesthetic drop (0.4%
oxybuprocaine hydrochloride; Chauvin Phar-
maceuticals Ltd., London, UK) was instilled
prior to examination.
Participants presented to the laboratory
wearing their contact lenses and the follow-
ing procedure was undertaken within ve mi-
nutes of lens removal on one randomly
selected eye. The cornea was scanned rst.
With the head securely positioned in the chin
and brow rest, the face of the Tomocap was
brought into gentle contact with the central
region of the cornea. When capturing the
images, the participant was advised to xate
on a target located directly in front of the
contralateral eye. Accurate positioning of
the Tomocap on the cornea was facilitated
by a side-mounted charge-coupled device
(CCD) camera that transfers a magnied
and real-time image onto a computer display
screen (Figure 1). The Tomocap was moved
slightly in a vertical and a horizontal direc-
tion, while the LSCM was focused at the level
of the sub-basal nerve plexus, approximately
60 m from the epithelial surface, which is
the known location of Langerhans cells in
the cornea.
After scanning the central cornea, the par-
ticipant was instructed to sit back for a brief
time before scanning the conjunctiva. In all
participants, the face of the Tomocap was
brought into gentle contact with the nasal
2016 Optometry Australia
Langerhans
Langerhanscells
cellsin
incontact
contact lens
lens wear Alzahrani, Colorado,Pritchard
Alzahrani, Colorado, Pritchardand
Figure 1. View captured from side-
mounted charge-coupled device camera of
Heidelberg laser scanning confocal micro-
scope, showing the plano face of the
Perspex Tomocap (A) gently contacting
the cornea (B)
conjunctiva, about two to four millimetres
from the limbus. Again, accurate positioning
was facilitated by the side-mounted CCD cam-
era (Figure 2). The participant was advised to
xate on a target located in the horizontal
plane but approximately 45 degrees to the
left or right of the midline of the contralateral
eye, depending on the eye being scanned.
The Tomocap was moved slightly in a vertical
and a horizontal direction, while the LSCM
was focused at various depths between 20 to
150 m from the epithelial surface, which is
the known location of Langerhans cells in
the conjunctiva.
Image analysis
Approximately 100 digital images were cap-
tured during each examination. Pilot studies
revealed that the analysis of ve and six im-
ages from the cornea and conjunctiva, re-
spectively, provided repeatability for
determining LCD with a standard deviation
of 8 cells/mm2 and an intra-class correla-
tion coefcient of 0.95. Thus, ve and six ran-
domly selected, high-quality images that were
overlapping less than 20 per cent were
analysed to determine corneal and conjuncti-
val LCDs, respectively. The number of
Langerhans cells was counted using the in-
built general-utility manual counting tool of
the Heidelberg instrument and data were
expressed as LCD, which is the mean of the
number of cells per square millimetre from
the images. The L-method was used to select
which cells were included in the count,
Clinical and Experimental
andEfron
Efron
whereby those cells partially cut off at the su-
perior and right hand borders of the eld
were included in the count. The operator un-
dertaking image selection and analysis was
masked with respect to the assigned group
of participants.
Contact lenses
All participants in the contact lens groups
were tted with Biomedics 1 day Extra daily
disposable soft contact lenses (CooperVision,
Pleasanton, California, USA). A hydrogel lens
was used in this experiment (the choice of
brand was arbitrary) as this lens category will
impart a greater physiological stress, at least
in terms of hyopxia, than a silicone hydrogel
lens. The Biomedics 1 day Extra lenses are
made from the hydrogel material oculclon
D. These lenses had the following parame-
ters and characteristics: water content 55 per
cent, diameter 14.2 mm, base curve 8.6 or
8.8 mm, centre thickness (at -3.00 D)
0.07 mm, power range -10.00 to +6.00 D, oxy-
gen permeability (Dk) 19 10-11 cm2 mlO2/
s ml mmHg, oxygen transmissibility (Dk/t; at
-3.00D) 27 10-9 cm mlO2/s ml mmHg and
light blue handling tint. Lens power was
determined by subjective refraction and any
necessary vertex correction was applied.
Participants were trained in the use of daily
disposable contact lenses and were provided
with a leaet and video recording that
explained contact lens insertion, removal
and care.
Figure 2. View captured from side-
mounted charge-coupled device camera of
Heidelberg laser scanning confocal micro-
scope, showing the plano face of the
Perspex Tomocap (A) gently contacting
the bulbar conjunctiva (B)
Clinical andOptometry 100.1
Experimental January 2017
Optometry 2016
3
35
Langerhans cells
cells in
in contact
contactlens
lenswear
wearAlzahrani,
Alzahrani,Colorado,
Colorado,Pritchard
Pritchard and
and Efron
Efron

Statistical approach
Normality of data relating to demographic
and clinical characteristics was examined
using the ShapiroWilk test and the most
appropriate test was applied for each analy-
sis. To compare the demographic and clin-
ical characteristics between group pairs,
parametric data were analysed using an
independent sample t-test and non-para-
metric data were analysed using an 2 test
and MannWhitney U-test.
A linear mixed model was used to examine
changes over time in LCD and whether the
changes were different among the three test
groups (CLIDE, NO-CLIDE and controls).
Although our data set displayed relatively
high kurtosis, a linear mixed model was
deemed most suitable for our longitudinal
data set, bearing in mind that these models
are robust to normality violations.26 A re-
peated-measures multivariate analysis of vari-
ance (using Pillais trace and Wilks lambda)
was performed for the groups and include
the factors, age and sex. Separate analyses
were undertaken for data relating to the cor-
nea and conjuctiva.
Tests of association between LCD and level
of severity as indicated by any of the dry eye
assessment techniques (DEQ-5, CLDEQ-8,
NITBUT, PRTT and CFS) at the one week
(when there was a maximum inammatory
response) were conducted using Spearmans
rank correlation test.
Data were analysed using IBM SPSS Statis-
tics for Windows, Version 21.0 (IBM Corpora-
tion, Armonk, New York, USA). A p-value of
less than 0.05 was considered to be statistically Figure 3. Flow diagram of study. The control group was recruited separately from the con-
signicant.
tact lens-induced dry eye (CLIDE) and NO-CLIDE groups.

RESULTS
A ow diagram showing the number of
participants recruited and enrolled into the
study, discontinuing from the study (and the
reasons for this) and nally examined, is
presented in Figure 3. After screening 110
potential study participants, 92 were enrolled
into the study. Sixty participants were tted
with contact lenses and 23 served as controls.
When signs and symptoms of CLIDE were
assessed among contact lens wearers at one
week, the group assignment comprised 25
CLIDE and 35 NO-CLIDE. Sixteen partici-
pants (eight CLIDE, six NO-CLIDE and two
controls) dropped out of the study over the
24-week study period, leaving 17 CLIDE, 29
NO-CLIDE and 21 controls who completed
the study.
The mean and standard deviation (SD) lens
powers tted to participants were: R -1.75
1.99 D, L -1.64 1.80 D for CLIDE and R
-1.70 1.95 D, L -1.63 1.79 D for NO-CLIDE.
Demographic and clinical data of the three
study groups at baseline and week 24 are
shown in Table 1, together with paired
differences among groups. The groups were
age-matched but were not effectively gender
balanced, with a signicantly higher propor-
tion of males among controls versus CLIDE
and NO-CLIDE. The signicantly higher
mean CLDEQ-8 scores for CLIDE versus
NO-CLIDE at baseline constituted the basis
for group assignment.
Changes in corneal LCD
Both immature and mature Langerhans cell
phenotypes were observed at the level of the
basal epithelium and sub-basal nerve plexus
of the cornea (Figure 4). There was no
change in corneal LCD over time in the
control group (p = 0.272) throughout the 24-
week observation period.
Figure 5 shows the mean and standard
error of the mean (SEM) of corneal LCD at
baseline, one, four and 24 weeks for each of
the three groups. At baseline, there were no
differences in corneal LCD overall or
between any paired combinations of CLIDE
(24 3 cells/mm2), NO-CLIDE (24 3 cells/
mm2) and controls (26 4 cells/mm2). At
week 1, LCD was greater in CLIDE (55 7
cells/mm2) versus NO-CLIDE (43 4 cells/
mm2) (p = 0.041) and controls (27 4 cells/
mm2) (p < 0.001). LCD was also greater in
NO-CLIDE (p = 0.041) versus controls
(p = 0.01). At week 4, LCD was greater in

Clinical and
and Experimental
ExperimentalOptometry
Optometry2016
100.1 January 2017 2016 Optometry Australia
4
36
 Parameter Baseline 24 weeks p-value

CLIDE NO-CLIDE Control CLIDE NO-CLIDE Control AvB AvC BvC DvE DvF EvF AvD BvE CvF
(A) (B) (C) (D) (E) (F)

Male/female 7/18 10/25 19/4 15/12 8/21 17/4 0.96* <0.001* <0.001* 0.98* 0.001* <0.001* 0.63* 0.61* 0.88*

2016 Optometry Australia

(n)
Age (years) 28.5 6.4 32.1 9.8 30.0 8.0 28.1 5.9 31.2 9.3 30.3 8.0 0.10 0.54 0.34
DCLWT (h) 8.3 3.9 9.8 3.9 0.38
DEQ-5 (0-22) 4.5 1.8 3.8 2.1 1.9 2.1 2.2 2.3 0.18+ 0.002+ 0.01+ 0.64#
CLDEQ-8 21.4 4.2 11.6 3.6 19.6 9.0 10.6 6.2 <0.001 <0.001 0.12 0.43
(037)
NITBUT 13.4 5.7 11.9 5.6 12.7 6.1 9.7 5.9 11.0 4.7 10.0 4.3 0.31+ 0.67+ 0.58+ 0.12+ 0.87+ 0.52+ 0.24# 0.52# 0.10#
(seconds)
CFS (04) 0.4 0.5 0.3 0.5 0.5 0.6 1.0 0.8 0.7 0.6 0.3 0.4 0.51+ 0.73+ 0.30+ 0.17+ 0.01+ 0.13+ 0.006# 0.008# 0.58#

PRTT 18.0 7.6 19.9 8.0 22.7 8.4 9.4 4.9 16.8 8.4 18.8 7.7 0.35 0.36 0.17 0.003 <0.001 0.38 0.001 0.12 0.10
(mm/20s)
DCLWT: daily contact lens wear time, DEQ-5: dry eye questionaire-5, CLDEQ-8: contact lens dry eye questionaire-8, NITBUT: non-invasive tear break-up time, CFS: corneal uorescein staining,
PRTT: phenol red thread test, CLIDE: contact lens-induced dry eye, NO-CLIDE: no contact lens-induced dry eye.
Results are expressed as mean SD, or counts for categorical variables.
Signicant differences (p < 0.05) are shown in bold.
* test.
Langerhans

Independent sample t-test.
+
MannWhitney U-test.
Langerhanscells

CLDEQ-8 baseline data is that collected after 1 week of contact lens wear.
cellsin

Table 1. Demographic and clinical data at baseline and 24 week visits of three study groups and pairwise differences
incontact
contact lens
lens wear Alzahrani,

epithelial surface.

Clinical and Experimental

(28 4 cells/mm2).
Alzahrani, Colorado,

Clinical andOptometry
standard error of the mean.

Experimental
Colorado,Pritchard

100.1
Pritchardand

Optometry
andEfron

pants over 24 weeks. Error bars indicate

univariate, within-group analyses of corneal

(up to 30 years and more than 30 years),
NO-CLIDE (29 3 cells/mm2) and controls
combinations of CLIDE (35 8 cells/mm2),
CLIDE (41 6 cells/mm2) versus controls
the cornea of contact lens-induced dry eye
cate immature Langerhans cells with short

5
between time and sex or between the time
LCD showed no signicant interaction
cells with long dendrites. Thin arrows indi-
scopic image of Langerhans cells in cornea.

2016
When participants were grouped by age
24, there were no differences in corneal
between CLIDE versus NO-CLIDE. At week
(27 4 cells/mm2) (p = 0.004); however,
(CLIDE), NO-CLIDE and control partici-
tured at a depth of 63 m from the corneal

CLIDE (35 4 cells/mm2) versus controls or

there were no differences between NO-
Thick arrows indicate mature Langerhans

LCD overall or between any paired

Figure 5. Langerhans cell density (LCD) in
or absent dendrites. This image was cap-
Figure 4. Laser scanning confocal micro-

37
January 2017
Efron
Langerhans cells
cells in
in contact
contactlens
lenswear
wearAlzahrani,
Alzahrani,Colorado,
Colorado,Pritchard
and age group. There were signicant multi-
variate effects with respect to LCD for time
(p < 0.001) and an interaction between
group and time (p < 0.001); however, there
were no signicant multivariate effects for
the study group (CLIDE/NO-CLIDE), sex,
age group, interaction between sex and time,
or interaction between age group and time.
There were no signicant correlations
between corneal LCD and level of severity as
indicated by any of the dry eye assessment
techniques (DEQ-5, CLDEQ-8, NITBUT,
PRTT and CFS) at one week.
Changes in conjunctival LCD
Both immature and mature Langerhans cell
phenotypes were observed in the stroma of
the bulbar conjunctiva (Figure 6). There was
no change in conjunctival LCD over time in
the control group (p = 0.944) throughout
the 24-week observation period.
Figure 7 shows the mean and standard
error of conjunctival LCDs at baseline, one,
four and 24 weeks for each of the three
groups. At baseline, there were no differ-
ences in conjunctival LCD overall or between
any paired combinations of CLIDE (7 1
cells/mm2), NO-CLIDE (8 2 cells/mm2)
and controls (7 1 cells/mm2). At week 1,
conjunctival LCD was greater in CLIDE (17
1 cells/mm2) (p = 0.003) and NO-CLIDE
(17 3 cells/mm2) (p = 0.001) versus controls
(7 1 cells/mm2). At week 4, there were no
differences in conjunctival LCD overall or
Figure 6. Laser scanning confocal micro-
scopic image of Langerhans cells in con-
junctiva. Arrows indicate the typically
mature Langerhans cells with long den-
drites found in the conjunctiva. This image
was captured at a depth of 20 m from the
conjunctival epithelial surface.
Clinical and
and Experimental
ExperimentalOptometry
Optometry2016
100.1 January 2017
6
38
Pritchard and
and Efron
Efron
Figure 7. Langerhans cell density (LCD) in
the conjunctiva of contact lens-induced dry
eye (CLIDE), NO-CLIDE and control par-
ticipants over 24 weeks. Error bars indi-
cate standard error of the mean.
between any paired combinations of CLIDE
(14 3 cells/mm2), NO-CLIDE (12 2 cells/
mm2) and controls (7 1 cells/mm2). At week
24, there were no differences in conjunctival
LCD overall or between any paired
combinations of CLIDE (10 3 cells/mm2),
NO-CLIDE (10 2 cells/mm2) and controls
(8 2 cells/mm2).
When participants were grouped by age
(up to 30 years and over 30 years), univariate,
within-group analyses of conjunctival LCD
showed no signicant interaction between time
and sex or between the time and age group.
There were no signicant correlations be-
tween conjunctival LCD and level of severity
as indicated by any of the dry eye assessment
techniques (DEQ-5, CLDEQ-8, NITBUT,
PRTT and CFS) at one week.
DISCUSSION
Of the 60 participants who were entered into
the contact lens arm of this study all of whom
were pre-screened to ensure that they were
free of dry eye symptoms at baseline 25 (42
per cent) developed CLIDE after one week.
This observation is broadly consistent with
previous studies, which report a prevalence
of CLIDE of between 28 and 50 per cent.1
Langerhans cells are antigen-presenting
cells. They are observed in the skin, mucous
membranes including the ocular surface,
lymph nodes and spleen. Although knowl-
edge of the precise role of Langerhans cells
is evolving,27 they are generally believed to
be essential in the processing of antigens
presented through the epithelial surface
and perform as a histocompatibility antigen
to stimulate T and B cells. These cells in turn
recruit further inammatory cells, such as
neutrophils, granulocytes and macrophages
from the surrounding tissues and the blood
vascular system, resulting in an inammatory
response with typical symptoms.27 Therefore,
activation of Langerhans cells is an early and
sensitive measure of an impending inamma-
tory process.
Langerhans cells are a subset of a broad
category of cells known as dendritic cells, so
all dendritic cells observed on the ocular
surface with the LSCM may not necessarily
be Langerhans cells. Evidence in the litera-
ture suggests that there is a direct correlation
between LSCM and immunohistochemical
observations of dendritic cells in the
cornea.28 The expression of Langerhans
cell-specic surface markers by dendritic cells
in the corneal and limbal epithelium has also
been reported.29,30 In view of this strong pre-
sumption that dendritic cells observed in the
cornea using LSCM are Langerhans cells, we
have adopted this term in the present study.
Cross-sectional studies have found
Langerhans cells to be up-regulated in the
cornea during contact lens wear.13,15 Zhivov
and colleagues15 reported that Langerhans
cells could be observed in the cornea of 59
per cent of contact lens wearers and 33 per
cent on non-lens wearing controls. In those
with Langerhans cells, LCD of the central
cornea was found to be 78 cells/mm2 in lens
wearers versus 34 cells/mm2 in non-lens wear-
ing control subjects. The generally higher
values of LCD reported by Zhivov and
colleagues15 compared with our study could
be attributed to the inclusion of all eyes in
our analysis, including those with zero LCD.
Sindt and colleagues13 reported the mean
LCD to be signicantly higher in 53 contact
lens wearers (64 cells/mm2) compared with
10 non-lens wearing controls (29 cells/
mm2). These reported values of LCD are
generally commensurate with those reported
in our study.
Efron, Al-Dossari and Pritchard11 have also
observed Langerhans cells in the bulbar con-
junctiva but were unable to differentiate LCD
between those who had been wearing contact
lenses for 10 years versus non-lens wearers.16
This observation is consistent with the
present study, which noted an increase in
conjunctival LCD after one week of lens wear
but no difference after one month of wear.
The observation in this study of an ini-
tial up-regulation in corneal LCD in
CLIDE versus controls at one and four
weeks is consistent with previous reports
of higher levels of corneal LCD in dry eye
patients. Machetta and colleagues31
2016 Optometry Australia

reported a higher LCD in the central cor-
nea of patients with Sjgren syndrome
dry eye (SSDE) (79 cells/mm2) compared
to those without dry eye (22 cells/mm2)
(p = 0.0031). In a similar study, Lin and
colleagues32 reported a higher LCD in
the central cornea of patients with both
SSDE (128 cells/mm2) and non-SSDE (90
cells/mm2) compared to those without
dry eye (35 cells/mm2). We failed to nd
evidence of an up-regulation of conjuncti-
val LCD in CLIDE (versus NO-CLIDE) at
any stage in this longitudinal study,
suggesting that the cornea is a more
sensitive barometer of ocular inamma-
tion associated with CLIDE.
The concept of adaptation has long been
discussed in the contact lens eld. It is an ill-
dened term and refers to various phenom-
ena, such as physical, visual and psychological
processes. Short-term adaptation describes
the subjective perception of lenses becoming
more comfortable in the brief period of
seconds or minutes following the initial
instantaneous discomfort experienced
during insertion.32 Long-term adaptation
describes the subjective perception of lenses
becoming more comfortable in the days,
weeks or months following the commence-
ment of lens wear.33 Adaptation is more evi-
dent in respect of the wearing of rigid
contact lenses, which have a greater physical
impact on the anterior ocular structures than
soft lenses.34
The basis for adaptation may be, in part,
psychological (that is, getting used to the
persistent discomfort).35 The physiological
basis of adaptation is poorly understood,35
although recent evidence from studies of
up-regulation of tear components and cor-
neal inammatory cells during contact lens
wear are providing possible answers.
Markoulli and colleagues36 noticed a signi-
cant increase in matrix metalloproteinase-9
in the tear lm on awakening in people wear-
ing extended wear contact lenses. They sug-
gested that return of this inammatory
mediator to baseline by one month indicates
an adaptive process.
Using LSCM, Zhivov and colleagues15
observed that central corneal LCD values of
144, 122 and 15 cells/mm2 in people who
had been wearing contact lenses for less than
ve years, ve to 10 years and more than
10 years, respectively. They suggested that this
indicates long-term adaptation to contact
lens wear. We observed an initial increase in
corneal and conjunctival LCD in CLIDE
and NO-CLIDE after one week of lens wear,
2016 Optometry Australia
Langerhans
Langerhanscells
cellsin
incontact
contact lens
lens wear Alzahrani, Colorado,Pritchard
followed by an apparent progressive reduc-
tion in LCD over the next 23 weeks. This nd-
ing provides further physiological evidence of
an adaptation to contact lens wear. This
hypothesis must be tempered by the fact that,
in our study, ve participants discontinued
due to contact lens discomfort from the
CLIDE group versus only one from the NO-
CLIDE group; had there been no discontinu-
ations, higher LCD values may have been
recorded in the CLIDE group at week 24.
Our failure to demonstrate signicant
correlations between corneal and conjuncti-
val LCD and any of the ve tests for dry eye
(DEQ-5, CLDEQ-8, NITBUT, PRTT and
CFS) precludes us from developing predictive
models for LCD. This lack of correlation
perhaps also supports the notion that the
sensation of discomfort, which is subjectively
reported as dryness, might not actually
represent a true aqueous deciency but
rather may arise from complex neural coding
of other adverse sensory inputs.
The strengths of this study are as follows.
1. Our cohort was carefully characterised in
respect of those with and without CLIDE,
as well as a contemporaneous control
group of non-lens wearers.
2. Experimental bias was minimised by
masking the operator assessing LCD from
the clinical status of the participants.
3. All participants wore the same lens type
for the same length of time, thus exclud-
ing these factors as possible confounding
variables.
4. The experiment was sufciently powered
to detect differences in LCD between the
assigned groups.
This study has a limitation related to the
gender imbalance of CLIDE and NO-CLIDE
versus the control group, which may have
confounded comparisons between those
groups, if signicant gender differences exist
in the parameters assessed in this work. An-
other study limitation is that the non-lens
wearing control group was recruited sepa-
rately, rather than having been randomly
assigned from a single combined recruitment
cohort of potential study participants.
The limitation of LSCM for unambiguously
identifying the observed bright features in
LSCM images as Langerhans cells is another
weakness, notwithstanding the histochemical
studies discussed above which strongly sug-
gest that these dendritic cells are Langerhans
cells.2729 Immature forms of Langerhans
Clinical and Experimental
Alzahrani, Colorado, Pritchardand
andEfron
Efron
cells, which have short or absent dendrites,
could be misidentied for other inamma-
tory cell types known to reside in ocular
surface tissues, such a monocytes and
polymorphs. Even if these bright features,
which we have identied as Langerhans cells,
are some other cell type involved in the
immunologic cascade of events, our conclu-
sions are essentially unaltered because our
observations support the overriding principle
of inammatory cell up-regulation in CLIDE.
Misidentication of tissue features as
Langerhans cells would constitute a random
error that would manifest without bias across
all groups.
Although the use of only one lens type lim-
ited data variability, this approach precluded
examination of the impact of different
surface characteristics (for example, lubricity
and wettability) that may have impacted com-
fort and thus LCD. The study was limited to
six months, so the impact of longer periods
of lens wear is unknown.
In conclusion, CLIDE is associated with an
initial up-regulation of Langerhans cells in
the cornea, which suggests an inammatory
basis for this condition during the initial
phases of contact lens wear. The apparent re-
duction in LCD following this initial increase
suggests a form of physiological adaptation to
lens wear. Assessment of LCD using laser
scanning confocal microscopy offers new in-
sights into the pathophysiology of the ocular
tissues in response to contact lens wear. As
well, this approach introduces the potential
to non-invasively and reiteratively monitor
the inammatory status of the cornea and
conjunctiva in vivo in response to various lens
materials and designs and different modali-
ties of contact lens wear. In this way, clinicians
who have access to this technology may be
assisted in clinical decision making, as well
as the contact lens industry in developing
contact lenses with material characteristics
that are less physiologically challenging to
the eye and afford greater levels of comfort.
ACKNOWLEDGEMENTS
Yahya Alzahrani was supported by a Saudi
Arabian Government Scholarship. The con-
tact lenses used in this study were kindly
supplied by CooperVision Australia.
REFRENCES
1. Dumbleton K, Caffery B, Dogru M et al. The TFOS
International Workshop on Contact Lens Discom-
fort: report of the subcommittee on epidemiology.
Invest Ophthalmol Vis Sci 2013; 54: TFOS20TFOS36.
Clinical andOptometry 100.1
Experimental January 2017
Optometry 2016
7
39
Langerhans cells
cells in
in contact
contactlens
lenswear
wearAlzahrani,
Alzahrani,Colorado,
Colorado,Pritchard
Pritchard and
2. Nichols JJ, Willcox MD, Bron AJ et al. The TFOS In-
ternational Workshop on Contact Lens Discomfort:
executive summary. Invest Ophthalmol Vis Sci 2013;
54: TFOS7TFOS13.
3. Efron N. Is contact lens wear inammatory? Br J
Ophthalmol 2012; 96: 14471448.
4. Papas E. On the relationship between soft contact
lens oxygen transmissibility and induced limbal
hyperaemia. Exp Eye Res 1998; 67: 125131.
5. Efron N, Pearson RM. Centenary celebration of
Ficks Eine Contactbrille. Arch Ophthalmol 1988;
106: 13701377.
6. Pearson RM, Efron N. Hundredth anniversary of Au-
gust Mullers inaugural dissertation on contact
lenses. Surv Ophthalmol 1989; 34: 133141.
7. Gonzalez-Perez J, Villa-Collar C, Sobrino Moreiras T
et al. Tear lm inammatory mediators during con-
tinuous wear of contact lenses and corneal refractive
therapy. Br J Ophthalmol 2012; 96: 10921098.
8. Kallinikos P, Morgan P, Efron N. Assessment of
stromal keratocytes and tear lm inammatory
mediators during extended wear of contact lenses.
Cornea 2006; 25: 110.
9. Efron N. Contact lens-induced changes in the ante-
rior eye as observed in vivo with the confocal
microscope. Prog Retin Eye Res 2007; 26: 398436.
10. Zhivov A, Stave J, Vollmar B, Guthoff R. In vivo con-
focal microscopic evaluation of Langerhans cell
density and distribution in the normal human cor-
neal epithelium. Graefes Arch Clin Exp Ophthalmol
2005; 243: 10561061.
11. Efron N, Al-Dossari M, Pritchard N. In vivo confocal
microscopy of the bulbar conjunctiva. Clin Experiment
Ophthalmol 2009; 37: 335344.
12. Efron N, Al-Dossari M, Pritchard N. In vivo confocal
microscopy of the palpebral conjunctiva and tarsal
plate. Optom Vis Sci 2009; 86: E1303E 1308.
13. Sindt CW, Grout TK, Critser DB, Kern JR, Meadows
DL. Dendritic immune cell densities in the central
cornea associated with soft contact lens types and
lens care solution types: a pilot study. Clin Ophthalmol
2012; 6: 511519.
Clinical and
and Experimental
ExperimentalOptometry
Optometry2016
100.1 January 2017
8
40
and Efron
Efron
14. Su PY, Hu FR, Chen YM, Han JH, Chen WL.
Dendritiform cells found in central cornea by in-
vivo confocal microscopy in a patient with mixed
bacterial keratitis. Ocul Immunol Inamm 2006; 14:
241244.
15. Zhivov A, Stave J, Vollmar B, Guthoff R. In vivo con-
focal microscopic evaluation of Langerhans cell
density and distribution in the corneal epithelium
of healthy volunteers and contact lens wearers. Cor-
nea 2007; 26: 4754.
16. Efron N, Al-Dossari M, Pritchard N. Confocal mi-
croscopy of the bulbar conjunctiva in contact lens
wear. Cornea 2010; 29: 4352.
17. Faul F, Erdfelder E, Lang A-G, Buchner A. G*Power
3: A exible statistical power analysis program for
the social, behavioral, and biomedical sciences.
Behav Res Methods 2007; 39: 175191.
18. Chalmers RL, Begley CG, Caffery B. Validation of
the 5-Item Dry Eye Questionnaire (DEQ-5): Discrim-
ination across self-assessed severity and aqueous tear
decient dry eye diagnoses. Cont Lens Anterior Eye
2010; 33: 5560.
19. Chalmers RL, Begley CG, Moody K, Hickson-Curran
SB. Contact Lens Dry Eye Questionnaire-8 (CLDEQ-
8) and opinion of contact lens performance. Optom
Vis Sci 2012; 89: 14351442.
20. Mengher LS, Bron AJ, Tonge SR, Gilbert DJ. A non-in-
vasive instrument for clinical assessment of the pre-
corneal tear lm stability. Curr Eye Res 1985; 4: 17.
21. Morris S. How to grade dry eyes. Rev Optom 2006;
143: 68.
22. Hamano H, Hori M, Hamano T et al. A new method
for measuring tears. CLAO J 1983; 9: 281289.
23. Efron N. Grading scales for contact lens complica-
tions. Ophthalmic Physiol Opt 1998; 18: 182186.
24. Nichols JJ, Sinnott LT. Tear lm, contact lens, and
patient-related factors associated with contact lens-
related dry eye. Invest Ophthalmol Vis Sci 2006; 47:
13191328.
25. Pult H, Murphy PJ, Purslow C. A novel method to
predict the dry eye symptoms in new contact lens
wearers. Optom Vis Sci 2009; 86: E1042E1050.
26. Fitzmaurice GN, Laird N, Ware J. Applied
Longitudinal Analysis. Hoboken, NJ: John Wiley
& Sons, 2012.
27. Romani N, Brunner PM, Stingl G. Changing views of
the role of Langerhans cells. J Invest Dermatol 2012;
132: 872881.
28. Mayer WJ, Mackert MJ, Kranebitter N et al. Distribu-
tion of antigen presenting cells in the human
cornea: correlation of in vivo confocal microscopy
and immunohistochemistry in different pathologic
entities. Curr Eye Res 2012; 37: 10121018.
29. Chen W, Hara K, Tian Q, Zhao K, Yoshitomi T. Exis-
tence of small slow-cycling Langerhans cells in the
limbal basal epithelium that express ABCG2. Exp
Eye Res 2007; 84: 626634.
30. Mayer WJ, Irschick UM, Moser P et al. Characteri-
zation of antigen-presenting cells in fresh and
cultured human corneas using novel dendritic
cell markers. Invest Ophthalmol Vis Sci 2007; 48:
44594467.
31. Machetta F, Fea AM, Actis AG, de Sanctis U,
Daimasso P, Grignolo FM. In vivo confocal micro-
scopic evaluation of corneal langerhans cells in dry
eye patients. Open Ophthalmol J 2014; 8: 5159.
32. Lin H, Li W, Dong N et al. Changes in corneal
epithelial layer inammatory cells in aqueous tear-
decient dry eye. Invest Ophthalmol Vis Sci
2010;51:122128.
33. Dumbleton KA, Woods CA, Jones LW, Fonn D.
Comfort and adaptation to silicone hydrogel
lenses for daily wear. Eye Contact Lens 2008; 34:
215223.
34. Jones-Jordan LA, Walline JJ, Mutti DO et al. Gas
permeable and soft contact lens wear in children.
Optom Vis Sci 2010; 87: 414420.
35. McMonnies CW. Psychological and other mecha-
nisms for end-of-day soft lens symptoms. Optom Vis
Sci 2013; 90: e175e181.
36. Markoulli M, Papas E, Cole N, Holden BA. Effect of
contact lens wear on the diurnal prole of matrix
metalloproteinase-9 in tears. Optom Vis Sci 2013; 90:
419429.
2016 Optometry Australia