CLINICAL INVESTIGATION

Association of contact lens-related allergic conjunctivitiswith changes in the morphology of meibomian glands

Reiko Arita Kouzo Itoh Shuji Maeda

Koshi Maeda Atsuo Tomidokoro

Shiro Amano

Received: 24 March 2011 / Accepted: 5 October 2011 / Published online: 23 November 2011

Japanese Ophthalmological Society 2011

Abstract

Purpose To observe morphological changes in the mei-

bomian glands of patients with contact lens-related allergic

conjunctivitis (CLAC) and to assess the relations between

the morphological changes and eyelid and tear film

parameters.

Methods We observed subjects in four groups: 64 eyes of

64 contact lens (CL) wearers with CLAC, 77 eyes of 77 CL

wearers without CLAC, 55 eyes of 55 patients with

perennial allergic conjunctivitis (perennial AC), and 47

eyes of 47 healthy volunteers. The following tests were

performed: slit-lamp examination, measurement of tear

film breakup time, grading of morphological changes in

meibomian glands (meiboscore) as assessed by noncontact

meibography, grading of meibomian gland distortion in

meibography, tear production as assessed by Schirmers

I test, and grading of meibum expression.

Results The mean score for meibomian gland distortion

was significantly higher in the CL wearers with CLAC than

in the CL wearers without CLAC (p \ 0.0001); it was alsosignificantly higher in the non-CL wearers with perennial

AC than in the non-CL wearers without perennial AC

(p \ 0.0001). There was no significant difference between

the mean scores for meibomian gland distortion of the non-

CL wearers with perennial AC and the CL wearers with

CLAC (p = 0.27). The score for meibomian gland distor-

tion was significantly positively correlated with the mei-

bum score in the CL wearers with CLAC and with the

meiboscore in the CL wearers without CLAC.

Conclusion CLAC is associated with an increase in

meibomian gland distortion. Allergic reaction, rather than

CL wear, appears to be responsible for the increase in

meibomian gland distortion in patients with CLAC.

Keywords Contact lens Allergic conjunctivitis Meibomian glands Meibography

Introduction

Contact lens-related allergic conjunctivitis (CLAC) is a

common cause of ocular discomfort in contact lens wearers,

which sometimes causes contact lens (CL) intolerance.

CLAC inflammation induces various reactions and symp-

toms, such as papillary formation, conjunctival hyperemia,

mucus discharge, and itchy sensations. Severe CLAC

involves the formation of giant papillary conjunctivitis

(GPC), which is reported to be associated with decreased

meibomian gland function [1]. Mathers and Billborough [1]

report that CL wearers with GPC have significantly more

meibomian gland dropout and greater meibum viscosity than

those without GPC. Moreover, allergic conjunctivitis (AC)

without CL is also reported to be associated with tear film

instability and dry eye [2, 3]. Toda et al. [2] report that AC

decreases goblet cell density, which leads to tear film insta-

bility. Suzuki et al. [3] report that seasonal AC is associated

with advanced tear instability and thickening of the tear film

lipid layer. Because the conjunctival tissue lies adjacent to

R. Arita (&) K. ItohItoh Clinic, 626-11 Minaminakano, Minuma-ku, Saitama,

Saitama 337-0042, Japan

e-mail: ritoh@za2.so-net.ne.jp

R. Arita A. Tomidokoro S. AmanoDepartment of Ophthalmology, University of Tokyo School

of Medicine, Tokyo, Japan

S. Maeda K. MaedaMaeda Ophthalmic Clinic, 3-30 Nakamachi,

Aizuwakamatsu, Fukushima, Japan

123

Jpn J Ophthalmol (2012) 56:1419

DOI 10.1007/s10384-011-0103-6

the meibomian glands, CLAC may also affect the meibo-

mian glands [4, 5], resulting in tear film instability. The

purpose of the present study was to observe the morpho-

logical changes in the meibomian glands of patients with

CLAC using noncontact meibography [6], and to assess the

relations between the morphological changes and eyelid and

tear film parameters.

Subjects and methods

Subjects

The subjects comprised four groups: 64 eyes of 64 CL

wearers with CLAC (32 men and 32 women; mean age

31.1 9.4 years), 77 eyes of 77 CL wearers without

CLAC (31 men and 46 women; 32.5 8.1 years old), 55

eyes of 55 patients with perennial AC (29 men and 26

women; 32.3 15.0 years), and 47 eyes of 47 healthy

volunteers (22 men and 25 women; 32.8 9.5 years).

Perennial AC was diagnosed both in the non-CL wearers

and the CL wearers based on complaints of an itchy sen-

sation and observation of papillary formation in the con-

junctiva. Exclusion criteria for the AC and control groups

included blepharitis, obvious eyelid or ocular surface dis-

orders, CL wear, continuous eye drop use, history of eye

surgery, and either systemic or ocular diseases that could

interfere with either tear film production or function. Data

used in this study were obtained from the right eye of each

subject. When the right eye (but not the left eye) was

excluded from the study, data from the left eye were used.

Written informed consent was obtained from all subjects

before examination. When subjects were younger than

20 years of age, written informed consent was obtained

from their parents. This study was approved by the insti-

tutional review board of Itoh Clinic and adhered to the

tenets of the Declaration of Helsinki.

Examinations

After CL removal, the following tests were performed

sequentially: a slit-lamp examination of the cornea and

conjunctiva both before and after fluorescein staining,

measurement of tear film breakup time (BUT), meibogra-

phy, tear production by Schirmers I test, and grading of

meibum expression. Superficial punctate keratopathy in the

cornea was scored from 0 to 3. Tear film BUT was mea-

sured three times consecutively after the instillation of

fluorescein, and the median value was adopted. Using the

noncontact meibography system, the upper and lower

eyelids were everted and the meibomian glands were

observed. Partial or complete loss of the meibomian glands

was scored using the following grades (meiboscore) for

each eyelid, as previously described [6]: grade 0: no loss of

meibomian glands, grade 1: area of loss \1/3 of the totalarea of meibomian glands, grade 2: area of loss between

1/3 and 2/3 of the total area, grade 3: area of loss [2/3 ofthe total area. Meiboscores for the upper and lower eyelids

were summed to obtain a score between 0 and 6 for each

eye. The presence of distortion was determined when dis-

tortion of [45 in at least one meibomian gland in theupper or lower eyelid was confirmed by meibography.

Meibomian gland (MG) distortion score was graded for

MG distortion in the upper eyelid from 0 to 2; grade 0: no

distortion of the meibomian glands; grade 1: 14 meibo-

mian glands with distortion; grade 2: more than five MGs

with distortion (Fig. 1). Tear film production was evaluated

by Schirmers test without applying topical anesthetic.

Digital pressure was applied to the upper tarsus, and the

degree of ease with which meibomian secretion (meibum)

was expressed was evaluated semiquantitatively as follows

[7]: grade 0, clear meibum that was expressed easily; grade

1, cloudy meibum that was expressed with mild pressure;

grade 2, cloudy meibum that was expressed with more than

moderate pressure; and grade 3, meibum that was not

expressed, even with hard pressure. The unpaired t test and

MannWhitney U test were used to compare variables

between the two groups. Spearman correlation analysis was

used to examine the correlation between the MG distortion

score and the other parameters in subjects wearing contact

lens both with and without allergic conjunctivitis. Kruskal

Wallis test or analysis of variance (ANOVA) was used

to compare variables among three or more groups. The

TukeyKramer test was used as a post hoc test after

ANOVA. The MannWhitney U test with Bonferronis

correction for multiple comparisons was used as a post hoc

test after the KruskalWallis test. A probability level of

\0.05 was considered statistically significant.

Statement of ethics

We certify that all applicable institutional and govern-

mental regulations concerning the ethical use of human

volunteers were followed during this research.

Results

Mean age (p = 0.72) and sex ratio (p = 0.50) did not

differ significantly among the four groups. In the CLAC

group, 16 patients used rigid gas-permeable lenses and 48

patients used hydrogel lenses (conventional hydrogel len-

ses, n = 8; disposable hydrogel lenses, n = 40). In the

CL wearers without CLAC, 28 patients used rigid

Allergy and meibomian glands 15

123

gas-permeable lenses and 49 used hydrogel lenses (conven-

tional hydrogel lenses, n = 11; disposable hydrogel lenses,

n = 38). The ratio of rigid gas-permeable lens wearers to

hydrogel lens wearers did not differ significantly between CL

wearers with and those without CLAC (p = 0.22). The

average duration of CL wear among CL wearers with CLAC

was 11.2 8.2 years, and 13.1 7.3 years in those without

CLAC (p = 0.16).

Comparison of the meibomian gland changes and tear

film-related factors among the four groups

Table 1 shows the results of the examinations in the four

groups. There were no significant differences in the mei-

bum scores (p = 0.25, KruskalWallis test), meiboscores

(p = 0.051, KruskalWallis test), or Schirmers values (p =

0.16, ANOVA) among the four groups. The BUT

(p = 0.0002, ANOVA) and superficial punctate keratopathy

scores (p \ 0.0001, KruskalWallis test) differed signifi-cantly among the four groups. The BUT was significantly

shorter in the CL wearers without CLAC than in the normal

volunteers (p = 0.0002, TukeyKramer test). The superficial

punctate keratopathy scores were significantly lower in the

normal volunteers than in the patients with perennial AC

(p = 0.0076) and in the CL wearers without CLAC

(p \ 0.0001).In all four groups, all MG distortions were observed in

the upper eyelid (Fig. 2). Table 2 shows a comparison of

the MG distortion scores. The mean MG distortion score

was significantly higher in the CL wearers with CLAC than

in the CL wearers without CLAC (p \ 0.0001). The meanMG distortion scores were significantly higher in the non-

CL wearers with perennial AC than in the non-CL wearers

without perennial AC (p \ 0.0001). There was no signifi-cant difference between the mean MG distortion scores of

the non-CL wearers with perennial AC and the CL wearers

with CLAC (p = 0.27).

In the CL wearers with CLAC, there was no significant

difference between the mean MG distortion scores of the

16 rigid gas-permeable lens wearers (0.31 0.70) and the

48 hydrogel lens wearers (0.63 0.84) (p = 0.15); nor

was there any significant difference between the mean MG

distortion scores of the 28 rigid gas-permeable lens wearers

(0.036 0.19) without CLAC and the 49 hydrogel lens

wearers (0.12 0.39) without CLAC (p = 0.29).

Correlations between the meibomian gland distortion

score and other factors in CL wearers

Table 3 shows the results of a Spearman correlation anal-

ysis between the MG distortion scores and the other

parameters in contact lens wearers with or without allergic

conjunctivitis. The MG distortion score showed signifi-

cantly positive correlations with meibum score in CL

wearers with CLAC and with meiboscore in CL wearers

without CLAC.

grade 0 grade 1

grade 2

Fig. 1 Representative cases ofeach grade of meibomian gland

distortion, scored using the

following grades: grade 0, nodistortion; grade 1, 14distortions;grade 2, more thanfive distortions

16 R. Arita et al.

123

Discussion

Although AC is reported to be associated with tear film

instability and dry eye [2, 3], morphological changes of the

MG in patients with AC have not been investigated. In the

present study, we examined MG morphology using non-

contact meibography in patients with CLAC and perennial

AC, and assessed the relations between observed mor-

phological changes and eyelid and tear film parameters.

Table 1 Mean (standard deviation) values of parameters in each group

Parameters CL wear ? CL wear - p value

AC? (n = 64) AC- (n = 77) AC? (n = 55) AC- (n = 47)

BUT (s) 4.3 2.1 4.1 2.1 5.2 2.4 5.3 2.1 0.0002

Schirmers value (mm) 15.0 8.9 21.3 10.5 17.5 10.2 15.9 10.1 0.17

SPK score 0.56 0.53 0.52 0.53 0.38 0.51 0.10 0.31 \0.0001Meibum score 0.34 0.54 0.30 0.51 0.57 0.63 0.21 0.46 0.25

Meiboscore 1.8 1.5 1.5 0.9 1.3 1.5 0.7 1.1 0.051

The p values refer to the superficial punctate keratopathy (SPK) scores, meibum scores, meiboscores (KruskalWallis test), BUTs, andSchirmers value (ANOVA) among the four groups.

CL contact lens, AC allergic conjunctivitis, BUT tear film breakup time, SPK superficial punctate keratopathy

Fig. 2ah Representativemeibography results. a Animage with meibography and

b a color photograph of a CLwearer with contact lens-related

allergic conjunctivitis (CLAC).

Meibomian gland distortion is

observed. Meibomian gland

distortion score: 2. Papilla

formation and edematous

conjunctiva are observed. c Ameibography image and d acolor photograph of a CL

wearer without CLAC. A

shortening of the meibomian

glands is observed without any

distortion. Neither papilla

formation nor edematous

conjunctiva are observed. e Ameibography image and f acolor photograph of a non-CL

wearer with perennial allergic

conjunctivitis (perennial AC).

Meibomian gland distortion is

observed. Meibomian gland

distortion score: 2. Papilla

formation is also observed. g Ameibography image and h acolor photograph of a normal

volunteer. No meibomian gland

distortion is observed. Neither

papilla formation nor edematous

conjunctiva are observed

Table 2 Mean (standard deviation) meibomian gland distortionscore in each group

AC? AC- p value

CL wear - 0.75 0.89 0.11 0.31 \0.0001CL wear ? 0.55 0.82 0.091 0.33 \0.0001p value 0.27 0.61

CL contact lens, AC allergic conjunctivitis

Allergy and meibomian glands 17

123

In the present study, MG distortion was observed sig-

nificantly more frequently in patients with CLAC than in

CL wearers without CLAC. Moreover, MG distortion

was more frequent in patients with perennial AC than in

normal volunteers.

There was no significant difference in the frequency of

MG distortion between patients with perennial AC and the

CLAC group or between CL wearers without CLAC and

the normal volunteers. These findings suggest that AC, but

not CL wear, is associated with MG distortion. Although

the exact mechanism underlying the association between

AC and MG distortion is unclear, inflammatory changes in

the conjunctival tissue might induce pressure on the MG in

the tarsus [4, 5], resulting in tear film instability, and

leading to cornification and obstruction. Histologic exam-

ination of the distorted MG in patients with AC may be

necessary to elucidate the relevant mechanism. Because

patients with AC frequently rub their eyes, the rubbing

could induce MG distortion. Furthermore, because distor-

tion was confined to the upper eyelid and was not present in

the lower eyelid, the rubbing quite possibly led to the

distortion of the MG ducts of the upper eyelid [8].

There was a marginal difference in meiboscores among

the four groups (p = 0.051). The meiboscores of the CL

wearers both with and without CLAC tended to be higher

than those of the non-CL wearers. These findings are in

agreement with the findings of our previous study: that CL

wear is associated with a decrease in the number of MGs

[9]. On the other hand, as shown in the present study, CL

wear is not associated with MG distortion. Therefore, CL

wear is likely to decrease the number of meibomian glands,

but it is not likely to induce meibomian gland distortion.

The meibum score was significantly higher in the CL

wearers with CLAC than in the CL wearers without CLAC.

Moreover, in our recent study [8], the meibum score was

significantly higher in perennial AC patients with MG

distortion than in perennial AC patients without MG dis-

tortion. According to a previous report, distortion occurs in

the first stage of morphological changes of the MG [10].

Taken together, these findings indicate that MG distortion

might decrease meibum expression during blinking.

The abovementioned distortion was related to the mei-

bum score, but not to the meiboscore. It is likely that the

meiboscore reflects not the changes in shape but the lost

area of the MG [6].

Although the BUT and superficial punctate keratopathy

scores differed significantly among the four groups in the

present study, these differences existed between the normal

volunteer group and the other groups. Because CL wear

causes tear film instability [11], it is reasonable that BUT

was shorter in the CL wearers without CLAC than in the

normal volunteers. The average BUT in the control group

was short (5.3 s). People who typically have a shorter

BUT, such as workers who frequently use a video display

terminal, might have formed part of the control group [12].

It is also reasonable that the superficial punctate keratop-

athy score was lower in the normal volunteers than in the

CL wearers without CLAC and the perennial AC group,

because both CL wear and the presence of AC could induce

superficial punctate keratopathy.

In the present study, most of the MG distortions were

observed in the upper eyelids. This finding is consistent

with the fact that allergic inflammation such as GPC is

observed in the upper tarsal conjunctiva more often than in

other areas. Additionally, MG are much longer in the upper

eyelids than in the lower eyelids, which may explain the

greater incidence of MG distortion in the upper eyelids.

There are two limitations of our study. One is that we

did not observe the morphology of meibomian glands

before either CL wear or AC. We did not examine whether

those who had originally had duct distortions tended to

suffer from AC. The other is that we did not examine the

correlation between the MG duct distortions and either the

quality or the quantity of meibum, as we graded the mei-

bum score only in the center part of the upper eyelid, and

graded the meibomian gland distortion score in the whole

Table 3 Correlations of meibomian gland distortion score with other parameters in contact lens wearers with or without allergic conjunctivitis

Parameters CL? AC? group CL? AC- group

R value p value R value p value

BUT -0.16 0.19 0.080 0.49

Schirmers -0.16 0.21 -0.20 0.78

SPK score 0.028 0.82 0.083 0.47

Meibum score 0.40 0.0017 0.14 0.24

Meiboscore 0.030 0.81 0.27 0.019

CL wear length -0.0080 0.53 -0.021 0.85

R value represents correlation coefficient obtained using Spearman correlation analysis

BUT tear film breakup time, SPK superficial punctate keratopathy

18 R. Arita et al.

123

upper eyelid. Therefore, we did not confirm whether a

meibomian gland with duct distortion had any

dysfunctions.

In a future study, we plan to observe whether the MG

distortion decreases with the treatment of allergic

conjunctivitis.

In conclusion, increased distortion and therefore allergic

reaction, and not CL, appears to be responsible for the

increase in MG distortion in patients with CLAC and with

perennial allergic conjunctivitis.

References

1. Mathers WD, Billborough M. Meibomian gland function and

giant papillary conjunctivitis. Am J Ophthalmol. 1992;114:

18892.

2. Toda I, Shimazaki J, Tsubota K. Dry eye with only decreased tear

break-up time is sometimes associated allergic conjunctivitis.

Ophthalmology. 1995;102:3029.

3. Suzuki S, Goto E, Dogru M, Asano-Kato N, Matsumoto Y, Hara

Y, et al. Tear film lipid layer alteration in allergic conjunctivitis.

Cornea. 2006;25:27780.

4. Knop E, Knop N. Meibomian glands: part IV. Functional inter-

actions in the pathogenesis of meibomian gland dysfunction

(MGD). Ophthalmologe. 2009;106:9807 (in German).

5. Knop E, Knop N, Millar T, Obata H, Sullivan DA. The interna-

tional workshop on meibomian gland dysfunction: report of the

subcommittee for anatomy, physiology and pathophysiology of the

meibomian gland. Invest Ophthalmol Vis Sci. 2011;52:193878.

6. Arita R, Itoh K, Inoue K, Amano S. Noncontact infrared meib-

ography to document age-related changes of the meibomian

glands in a normal population. Ophthalmology. 2008;115:9115.

7. Shimazaki J, Goto E, Ono M, Shimmura S, Tsubota K. Meibo-

mian gland dysfunction in patients with Sjogren syndrome.

Ophthalmology. 1998;105:14858.

8. Arita R, Itoh K, Maeda S, Maeda K, Furuta A, Tomidokoro A,

et al. Meibomian gland duct distortion in patients with perennial

allergic conjunctivitis. Cornea. 2010;29:85860.

9. Arita R, Itoh K, Inoue K, Kuchiba A, Yamaguchi T, Amano S.

Contact lens wear is associated with decrease of meibomian

glands. Ophthalmology. 2009;116:37984.

10. Mathers WD, Shield WJ, Sachdev MS, Petroll WM, Jester JV.

Meibomian gland dysfunction in chronic blepharitis. Cornea.

1991;10:27785.

11. Nichols JJ, Sinnott LT. Tear film, contact lens, and patient-related

factors associated with contact lens-related dry eye. Invest Oph-

thalmol Vis Sci. 2006;47:131928.

12. Tsubota K, Nakamori K. Dry eyes and video display terminals.

N Eng J Med. 1993;328:584.

Allergy and meibomian glands 19

123

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

c.10384_2011_Article_103.pdfAssociation of contact lens-related allergic conjunctivitis with changes in the morphology of meibomian glandsAbstractPurposeMethodsResultsConclusionIntroductionSubjects and methodsSubjectsExaminationsStatement of ethicsResultsComparison of the meibomian gland changes and tear film-related factors among the four groupsCorrelations between the meibomian gland distortion score and other factors in CL wearersDiscussionReferences