R

T

Ca

b

c

a

KWOC

1

cicotTifehcaad

W

1d

Contact Lens & Anterior Eye 35 (2012) 148– 154

Contents lists available at SciVerse ScienceDirect

Contact Lens & Anterior Eye

j ourna l ho me p ag e: ww w.elsev ier .com/ locate /c lae

eview

he role of heat in rubbing and massage-related corneal deformation

harles W. McMonniesa,∗, Donald R. Korbb,c, Caroline A. Blackieb,c

School of Optometry and Vision Science, University of New South Wales, Kensington 2052, AustraliaKorb Associates, 400 Commonwealth Ave., Unit 2, Boston, MA 02215, United StatesTear Science, Inc., Morrisville, NC, United States

r t i c l e i n f o

eywords:arm compresses

cular massageorneal deformation

a b s t r a c t

Purpose: To examine the role of elevated corneal temperature in the development of rubbing/massage-related corneal deformation and the possibility that warm compresses in the management of meibomiangland dysfunction or chalazion could contribute to such adverse responses.Methods: With reference to reports of corneal deformation associated with meibomian gland dysfunction,chalazion, dacryocystoceles and post-trabeculectomy, the mechanisms for increased corneal temperaturedue to ocular massage, especially when combined with warm compresses are examined.Results: Several mechanisms for rubbing/massage to elevate corneal temperature have been described,apart from the application of warm compresses or other forms of heat.Conclusions: Raised corneal temperature helps to explain corneal deformation which develops in asso-ciation with rubbing or massage in conditions such as keratoconus, chalazion, post-trabeculectomy,post-laser assisted in situ keratomileusis, post-graft and dacryocystoceles. When combined with warmcompresses or other methods of heat delivery to the eye, the elevation of corneal temperature appearsto explain how meibomian gland dysfunction treatment involving warm compresses and massage could

induce rubbing-related deformation. Patients whose management involves iatrogenic ocular massageappear to require screening for risk of corneal deformation. Risk may be increased for patients with aconcurrent habit of rubbing their eyes abnormally in response to allergic itch for example. It appears to bepossible to modify ocular massage techniques to reduce the risk of corneal deformation. Careful tutoringand follow-up using corneal topography appears to be required when massage is prescribed, especially

with Britis

when used in conjunction© 2012

. Introduction

Massage of the lids with a finger or glass rod either alone, or inombination with local applications of medications, may be usedn the treatment of chronic marginal blepharitis, meibomianitis,hronic palpebral conjunctivitis, trachoma and in the managementf glaucoma [1]. Ocular massage through the lids is also a commonechnique employed to aid filtration after trabeculectomy [2,3].he treatment of meibomian gland dysfunction (MGD) frequentlynvolves massage to promote lipid secretion from obstructed glandsollowing applications of warm compresses to the lids [4–7]. How-ver, massage of the lids and globe are not without risk and chronicabits of abnormal rubbing are associated with the development oforneal deformation and keratoconus in apparently normal eyes [8]

s well as in cases of post-graft recurrence of keratoconus [9,10]. Inddition, chalazia-related rubbing can be associated with cornealeformation [11] and conservative management involving warm

∗ Corresponding author at: 77 Cliff Avenue, Northbridge, Sydney, New Southales 2000, Australia. Tel.: +61 2 9958 3046; fax: +61 2 9958 3012.

E-mail address: c.mcmonnies@unsw.edu.au (C.W. McMonnies).

367-0484/$ – see front matter © 2012 British Contact Lens Association. Published by Elsoi:10.1016/j.clae.2012.01.001

heat application.h Contact Lens Association. Published by Elsevier Ltd. All rights reserved.

compresses and massage [12] may interfere with recovery of nor-mal corneal shape. The following case report describes the onset ofcorneal distortion in a patient who was advised to use massage andwarm compresses as therapy for MGD.

A 24-year-old Caucasian male had developed severe dry eyeover the prior two years.

Subsequently, his ability to read comfortably had become veryrestricted. The remainder of the history was normal as was cornealtopography (Fig. 1). Standardized meibomian gland diagnosticexpression [13] was performed and indicated only one and twoglands yielding liquid secretion (OD and OS respectively) [14]. Dueto the normal appearance of the eyelids and the absence of obvioussigns of MGD (aside from the expression results showing definitereduced gland function), a diagnosis of non-obvious obstructivemeibomian gland dysfunction was made [15]. The treatment planincluded in-office manual therapeutic gland expression [15] andat-home warm compress therapy: 15 min applications of warmcompresses, bid, followed by self expression with finger/thumb

squeezing of the lower eyelid glands.

At the 4-week follow-up, dry eye symptoms had decreased andreading time without discomfort had significantly increased. Therefractive and corneal findings were unchanged from baseline. The

evier Ltd. All rights reserved.

C.W. McMonnies et al. / Contact Lens & Anterior Eye 35 (2012) 148– 154 149

F eep island nasally and inferiorly presentation after seven weeks of MGD therapy (below).

pobnsbtarlfpcsha

urwfiwatttat(rh

Table 1Refractive changes for the right eye marking the development and resolution ofcorneal deformation which may have been associated with warm compress andmassage therapy as well spontaneous eye rubbing in response to ocular itch.

Time of Measurement Refraction result Best correctedvisual acuity

Baseline + 0.50 −0.25 × 180 20/15After seven weeks of warm compress + 0.50 −1.00 × 45 20/25

ig. 1. Corneal map showing right eye changes from baseline (top) to the definite st

atient was advised to continue the home program and return inne month. He called three weeks later and reported a constantlur OD. The OD corneal map showed a definite steep islandasally and inferiorly (Fig. 1). The OS findings had not changedignificantly. The possibility of early keratoconus was consideredased on corneal topography, distorted keratometric reading,he development of a 1.00 diopter oblique refractive cylinder,

reduction of best corrected acuity of two lines, and distortedetinoscopic and ophthalmoscopic reflexes. There were no slitamp signs of keratoconus. However, a habit of knuckle rubbingor extended periods of time had developed in response to itchrovocation which became apparent following episodes of warmompresses. Interestingly, experimental eye rubbing has beenhown to induce ocular itch [16], and a similar mechanism mayave contributed to his itch symptoms following warm compresspplications.

The patient was instructed to cease eye rubbing/massage and tose gentle cold compresses and/or ice to control acute episodes ofubbing provocation. Patanol was also prescribed bid; the patientas advised to limit gland expression to squeezing glands betweennger and thumb following gently applied warm compresses. Heas followed at two, four and eight weeks with corneal findings and

cuity found to have progressively returned to baseline. The refrac-ive findings OD for these visits are shown in Table 1. It is possiblehat the corneal changes and vision loss observed were, in part dueo the combination of massage associated with warm compressesnd itch-related eye rubbing. That adverse changes were limited

o the right eye might have been associated with his right handbeing dominant and stronger) delivering more rubbing/massageelated corneal trauma [17]. In hindsight, instructions shouldave been provided to avoid rubbing, especially following warm

therapy and itch-related rubbing8 weeks after cessation of rubbing + 0.25 −0.50 × 15 20/15

compress application, if as discussed below, raised corneal tem-perature increases the risk of corneal deformation.

Ordinarily, with patients carefully monitored to ensure compli-ance with instructions, post-trabeculectomy ocular massage is safeand effective [2], however, non-compliance with massage instruc-tions may be associated with adverse responses and numerouscomplications have been reported including suture rupture, blebrupture, hyphema, iris incarceration, endophthalmitis secondary tobleb rupture, late wound dehiscence in a corneal graft, subretinalhemorrhage, choroidal hemorrhage, wound dehiscence, hypotony,and shallowing of the anterior chamber [3]. Excessive massageforce can lead to corneal or scleral ectasia [3] and a case of post-trabeculectomy massage which resulted in keratectasia, has beenreported [18]. Another case report described corneal ectasia associ-ated with massage of dacryocystoceles [19]. However, as describedabove, apart from patients who exceed the prescribed frequency

and/or duration of massage, a patient who complies with mas-sage instructions may be at greater risk if, as in the case describedabove, they have or develop a habit of rubbing their eyes in additionto the prescribed regimen. Corneal deformations associated with

1 ens &

mdcbcov

2i

itdthvcsttssutmtjcvtcatrc1otCcbfmtauawtotabtcMpdelpw

50 C.W. McMonnies et al. / Contact L

assage/rubbing, which have been recorded in associated withifferent clinical circumstances, raise questions regarding whichommon mechanisms may be involved. Corneal deformation haseen linked to elevated corneal temperature [8] and warrants aloser examination, especially the possibility that the applicationf warm compresses could contribute to corneal temperature ele-ation and increase the risk of deformation.

. How massage elevates corneal temperature andncreases the risk of deformation

Corneal temperature may increase with lid closure, rapid blink-ng, increased tearing, ambient temperature, aqueous humouremperature and with blood supply [20]. The apparent partialependence of corneal temperature on blood flow is indicated byhe finding that, compared to the central cornea, temperature isigher in the extreme peripheral cornea adjacent to the limbalasculature [21]. Consequently, indirect radiometric methods oforneal temperature measurement give higher readings than mea-urements recorded at the centre, because of the contribution fromears and the peripheral cornea, which is about 0.5 ◦C warmer thanhe central cornea [20]. Additional dependence on blood flow isuggested by the finding that temperature increases with eye clo-ure and closer proximity of the cornea to the vasculature of thepper lid palpebral conjunctiva as well as reduced evaporation ofhe tear layer [22]. When eyes are passively open and during a nor-

al blink, the upper lid appears to have contact with the corneahat is limited to the lid wiper [23]. In particular, the tarsal con-unctiva may not contact the cornea [24]. However, during warmompress application, especially if combined with massage, theascular tarsal conjunctiva is compressed into unnatural close con-act with the cornea and associated increased heat transfer to theornea may occur. Corneal epithelial temperature of 33.7 ◦C [20]nd blood temperature of 38 ◦C [25] is consistent with transfer fromhe palpebral conjunctiva to the pre-corneal tear layer and ante-ior cornea, especially when the conjunctiva is pressed against theornea. Under open eye conditions, evaporation carries off about3% of the heat lost from the surface of the eye [26]. About 46%f the heat is carried away by convection (heat transfer directlyo the surrounding air), and 41% is dissipated by radiation [26].onsequently, apart from heat transfer from the lids to the cornea,losed eye conditions during warm compress application coulde expected to raise corneal temperature by inhibiting heat lossrom the anterior surface. Under normal conditions, corneal stro-

al temperature appears to be maintained by heat transfer fromhe tear film and the limbal vasculature, but primarily from thequeous humour [20]. For a sample of 40 normal subjects, andsing a direct thermo-electrical method of recording, the temper-ture of the central corneal epithelium was found to be 33.7 ◦Chen the mean core body temperature was 36.7 ◦C and ambient

emperature was 23.1 ◦C [20]. However, at ambient temperaturesf 32–34 ◦C anterior corneal temperature was found to increaseo 36.5 ◦C [27]. Assuming that the aqueous humour temperaturet the corneal endothelial surface is only slightly less than coreody temperature under normal levels of anterior uveal hyperemia,hese findings [20,27] suggest a gradient of temperature with theornea being warmer by about 3 ◦C at the posterior surface. Foreibomian gland dysfunction therapy, these mechanisms for tem-

erature increases would be in addition to any heat transferredirectly from warm compresses through the lids to the cornea. For

xample, when 45 ◦C warm compresses were applied gently to theids, without massage, and replaced every two minutes, the tem-erature of the palpebral surface of the lower lid was raised to 40 ◦Cithin 6–8 min [28].

Anterior Eye 35 (2012) 148– 154

3. Temporary vision changes associated with massage

Warm (45 ◦C) compresses replaced every 2 min for 30 min, andapplied with the least pressure consistent with uniform contactto the closed eyelids (specifically without massage) were foundto induce transient visual degradation but without significantchanges in topography [29]. Gentle experimental rubbing in ahorizontal direction, for 1 min using the distal end of an indexfinger, and without warm compress application, resulted in thedevelopment of 0.50–0.75 D of astigmatism [30]. The changes dis-sipated quickly [30] and may have been due to transient regionalchanges in epithelial thickness [31]. Corneal temperature mayneed to be raised significantly for significant corneal distortion tooccur.

4. Other mechanisms for elevating corneal temperature

Corneal temperature may be raised by the effects of mas-sage by other mechanisms. Marked inflammatory palpebralconjunctival responses to experimental rubbing (without warmcompresses) were found to be sustained over several hours inrats and rabbits [32,33]. In addition to warm compress heattransfer to the cornea, palpebral conjunctival inflammation asso-ciated with massage may give rise to a correspondingly greatertransfer of heat to the cornea during and subsequent to themassage episode. Anterior uveal hyperemia may also occur inresponse to lid massage and heat application with increasedaqueous flow and temperature [34], and corresponding increasesin posterior corneal temperature. There is also the possibilitythat the viscoelastic properties of the cornea could give riseto another source of massage-related heat generation. The vis-cous or damping properties of viscoelastic materials give rise toenergy loss and its dissipation as heat [35]. This means that aviscoelastic material loses energy when a load is applied andthen removed (cyclic loading) [36]. Accordingly, hysteresis isobserved in the stress–strain curve of viscoelastic materials likethe cornea, with the area between the loops which is createdby the difference between the loading and the (delayed) unload-ing responses, being an indication of the energy lost, and theheat generated within the cornea during loading/unloading cycles[36]. Massaging the lids involves a cyclic pattern of unloadingand increased loading of the cornea. The area of applicationof massaging force is unloaded when the cornea is indented,with curvature transfer and elevated intraocular pressure (IOP)resulting in increased loading and compensatory bulging inareas adjacent to the indentation [8]. Heat may be generatedwithin the cornea by these unloading/increased loading cycles.The rate of energy dissipation is greater at higher cycle fre-quencies [37], so that vigorous massage of the eye might beassociated with higher stromal temperature due to higher lev-els of energy dissipation and associated higher levels of heatgeneration.

5. Possible mechanisms for massage and elevated cornealtemperature to induce deformation

Biomechanical properties of collagen structures dependstrongly on temperature. For example, the creep/elongation rate ofhuman digital tendon is greater with increasing temperature [38].Bending of the cornea is facilitated by a reduction in shear modulus

[39]. Resistance to corneal bending may be reduced mechanicallywhen massaging agitation (cyclic unloading and loading) inducesa thixotropic reduction in viscosity of the proteoglycan gel groundsubstance [8]. In normal corneas, 20 s of ‘normal’ self-inflicted

ens &

rsAtsadocamaeculaCcpatr

6t

m(teebmseaatccacgpcfilmmlatiq(ttttaote

C.W. McMonnies et al. / Contact L

ubbing, which was intended to mimic ‘itchy eye’ rubbing, washown to reduce corneal hysteresis and resistance factor [40].

second 20 s episode of rubbing, 2 min later, further reducedhose biomechanical corneal properties [40]. However, any or allources of increased corneal temperature discussed above maylso reduce ground substance viscosity in addition to a reductionue to the mechanical thixotropic process. This combinationf mechanical and temperature-related ground substance vis-osity reduction may help to facilitate corneal bending throughn associated greater reduction in the ground substance shearodulus and glue function [8]. Failure to recover normal shape

fter corneal bending could be associated with the development ofctasia [8]. In addition, after massaging ceases, recovery of normalorneal temperature may be delayed whilst conjunctival and/orveal hyperemia persists. Blood flow is a major determinant of

id cooling after application of warm compresses, [28] but thevascular cornea may, accordingly, retain heat for longer periods.onsequently, warm compress-related mechanisms for elevatingorneal temperature might combine with massage-related tem-erature increases to create a sustained period of increased risk fordverse consequences. Any additional rubbing that occurs duringhis period (in response to itch [8] for example), may increase theisk of corneal deformation or keratectasia development.

. Additional factors associated with raised cornealemperature and increased susceptibility to distortion

A change in corneal temperature of 1.0 ◦C will alter theetabolic rate by approximately 10% [20]. For example, a windy

16 kph/10 mph) −15 ◦C environment can reduce tear tempera-ures to such low levels (about 5 ◦C), that metabolism in thepithelial cells will cease [26]. The effect of temperature onnzyme activity is complex and varies with the thermal sta-ility of the enzyme and substrate [41]. As temperature rises,olecular motion and hence collisions between enzymes and

ubstrates speed up [42]. Accordingly, raised temperature gen-rally increases enzyme activity, and the elevation in rate islmost linear up to 40 ◦C [41]. Denaturing of enzymes startst temperatures above 40 ◦C and the activity rate commenceso slow accordingly [41]. Abnormal enzyme function in sus-eptible corneas and any associated loss of, or alteration toollagen and/or ground substance [43–45], might be acceler-ted during a period of raised temperature. For example, bovineorneal collagen denatures at approximately 39 ◦C [46]. Colla-en denatures into gelatin with the triple helix structure andhysico-chemical properties of collagen disappearing [47]. Suchhanges may increase susceptibility to deformation. However, thending that Cathepsin G is capable of degrading denatured col-

agen (gelatin) [48], raises the possibility that corneal collagenight be degraded by either or both denaturation and enzy-atic processes at higher than homeostatic temperatures. The

evel of cathepsin G is increased in keratoconus corneas [44]nd may contribute to heightened gelatin digesting activities. Theemperature at which collagen denatures is reduced if collagens exposed to sinusoidally fluctuating strain at a constant fre-uency of 1 Hz [47]. As described above, massaging mechanicscycles of loading and unloading) may induce this type of fluc-uating strain. The possibility of corneal shape changes appearso be greater when rubbing/massage-related mechanical cornealrauma occurs under conditions of increased corneal tempera-ure. As suggested by the case described above, warm compress

pplication, repeated over several weeks in the managementf MGD, may have contributed to significant episodic eleva-ion of stromal corneal temperature which, when combined withye rubbing, may have facilitated the development of corneal

Anterior Eye 35 (2012) 148– 154 151

deformation. Warm compress and massage therapy for the conser-vative management of chalazia [12] may, for example, carry similarrisks [11].

7. Other applications for therapeutic massage

Despite the popularity of prescribed post-trabeculectomy ocu-lar massage, there is no consensus amongst physicians regardingtechnique [3]. The two most commonly used techniques are toapply pressure through the lower lid for varying lengths of timewith the patient looking up, or to apply force through the upperlid to the posterior edge of the scleral flap [3]. For example, 10 smassage episodes, four times a day, may be recommended, tomaintain improved aqueous drainage for a duration of about fourmonths [2]. However, that post-graft recurrence of keratoconus canbe associated with histories of abnormal rubbing [9,10], suggeststhat rubbing trauma can induce ectasia in normal (donor) corneas.The alternative explanation, that the donor cornea carries a highsusceptibility to the development of keratoconus, which is onlymanifest post-graft, may be appropriate in some cases [49] butcould be inappropriate for all rubbing-related cases of post-graftectasia [9,10]. Again, it is possible that a grafted cornea, depend-ing on the state of corneal wound healing, is more susceptibleto this complication, even when the donor cornea carries no riskfor keratoconus. Post-laser assisted in situ keratomileusis (LASIK)ectasia can occur in the absence of apparent risk factors [50] andmight be explained in part by abnormal rubbing [8,50]. Late onsetidiopathic diffuse lamellar keratitis [51,52] after LASIK, may alsoinvolve wounding by rubbing trauma. Measurements of the cohe-sive tensile strength of human LASIK corneal wounds indicated aweak central and paracentral stromal scar that averages only 2.4%of the normal corneal strength [53]. These laboratory findings cor-relate well with a clinical series, which indicated that LASIK flapscould be lifted without complications during re-treatments, up to8.4 years after initial surgery [53]. A region of reduced interlamel-lar cohesive strength at the ablation interface may increase thedegree of rubbing-related shear strain in the post-LASIK cornea.Any foreign matter or tissue debris trapped at the interface mayincrease the risk of an inflammatory reaction to rubbing-relatedshear strain trauma. Again, depending on the level of wound heal-ing activity, massage trauma which is not significant for the normalpre-LASIK cornea may be traumatic for the post-LASIK cornea [8].LASIK patients who follow their surgeon’s instructions to avoid eyerubbing may relapse to normal, or abnormal rubbing, years later.Such events could help explain late onset post-LASIK ectasia or dif-fuse lamellar keratitis. Similarly, iatrogenic massage could have acontributory role in these complications, perhaps especially if heatapplication is involved.

8. Massage-related IOP elevations and their possiblecontribution to corneal deformation

Intraocular pressure is raised by any contact that applanates orindents the ocular surface and displaces intraocular fluid [54]. Astudy of the application of experimental digital forces to humaneyes indicated that, for an open eye with normal IOP of 15 mmHg,‘light’ and ‘firm’ digital forces increased IOP by approximately 100%and 300% respectively [55]. For these measurements, the distal endof an index finger was used to apply the ‘light’ and ‘firm’ forces tothe temporal sclera, through the adnexal skin [55]. It is possible thatthe distending forces due to elevated IOP loading may contribute to

the development of corneal deformation, especially in susceptiblethinner and/or less rigid corneas. As described above, the risk maybe greater when corneal temperature is raised above homeostaticlevels.

1 ens &

9

pcTcogeowmtbioaIfelosmht[ttpOsocbelIb

1d

aosttctticwAqRm

1

i

52 C.W. McMonnies et al. / Contact L

. Factors which may increase risk of corneal deformation

Massaging a cornea that is already compromised may com-ound the initial trauma especially if the cornea does not get thehance to recover in the periods between massage episodes [32].his hypothesis is supported by the findings that corneal biome-hanical properties are progressively altered by repeated episodesf rubbing [40]. Ten-year follow-up data for a cohort of post-cornealraft patients indicated that complications included rejectionpisodes, suture-related problems, glaucoma, infection, recurrencef original disease and cataract surgery [56]. Trauma associatedith any rubbing/massage episode, iatrogenic or spontaneous,ay contribute to the development of these complications. Some-

imes, rather than finger massage, electric massage devices maye employed for post-trabeculectomy patients [3]. Their use may

nvolve a controlled number of applications and a moderate levelf massaging force [3]. However, sometimes a device that supplies

continuous level of vibratory mechanical force is used to reduceOP [57]. It may be possible for the user to increase the level oforce provided through the lids by tilting the device toward theyes [57]. Consequently, the level of force delivered through theids may depend on the prominence of the globe relative to therbit, as well as the preference or tolerance of the user to the sen-ation experienced. These factors suggest that such massage forcesight be much higher or lower for some individuals. After 10 min of

eat application with warm moist air in the treatment of MGD, lidemperature was raised by 1 ◦C and corneal temperature by 1.5 ◦C58]. An infrared warming device was found to raise lid and cornealemperature to 39.7 ◦C after the recommended 5 min of applica-ion [59]. Several web sites advertise electric massagers which areromoted for ocular use, and which are available to the public.ne device provides heating combined with the vibratory mas-

age and set times of treatment duration up to 15 min. The levelf heating and mechanical trauma with such devices may be suffi-ient to induce corneal changes. Other web-sites sell wheat-filledags which can be heated and, in addition to delivering heat to theyes, involve a sustained period of ocular compression through theids and associated corneal applanation/indentation and elevatedOP. People using any of these types of devices might not have theenefit of professional advice about risks associated with them.

0. Screening for higher susceptibility to cornealeformation

Assessment of risk of adverse responses to prescribed heatingnd/or massage may indicate the need for modification of the rec-mmended techniques to reduce those risks. Family history andcreening by anterior and posterior corneal topography, similar tohat employed for prospective LASIK patients may be appropriateo detect those with significant risk for the development of kerato-onus or other form of corneal deformation [60]. Reduced cornealhickness may be associated with increased keratectasia risk. A his-ory of chronic habits of abnormal rubbing in response to allergictch or other provocation, suggests the possibility that associatedorneal trauma could be compounded with any trauma associatedith prescribed massage, especially if heat application is involved.s discussed above, risks of adverse responses may be higher subse-uent to corneal surgery, depending on the level of wound healing.isk of massage-related complications may be greater in warm cli-ates and associated elevated anterior corneal temperature [61].

1. Reducing risk of corneal deformation

Prescribed massage for post-trabeculectomy patients shouldnvolve a standardized tutorial in ocular massage technique [2,3].

Anterior Eye 35 (2012) 148– 154

Post-trabeculectomy patients may be able to reduce risk by direct-ing gaze so that the cornea is not directly affected by digital ormassage trauma [2,3]. Patients using warm compresses and mas-sage for MGD or chalazion management could reduce the risk ofcorneal trauma by using similar techniques. For example, massag-ing the upper or lower lids with the eyes in appropriate infraductionand supraduction should allow the meibomian glands to receiveencouragement to express lipid, without exposing the cornea tomechanical forces directly. Using a mirror with head tilted up forthe upper lid and down for the lower lid gives better control foravoiding forces being directed to the cornea. The force used can bebeneficially applied along the MGs in the direction of the orifices(as in a blink-induced milking fashion [4]). Alternatively, groups ofglands can be squeezed between index finger and thumb to helpinduce flow from the orifices [62]. Asymptomatic non-contact lenswearing subjects over 40 years of age were found to have a mean of5.5 meibomian glands yielding liquid secretion per lower lid [62].However, patients with symptoms of evaporative dry eye havebeen found to have a significantly lower number of functioningglands [13]. Treatment which increases the number of functioningglands by only 1 could have the potential to improve lipid secre-tion from a lower lid by more than 20%. Although in-office glandexpression can be painful [62], clearing just one or two glands couldmake a significant improvement in tear stability and symptomrelief. Delivery of oil to the lid margin depends on the muscu-lar action of the lids during blinking [4,63,64]. Deliberate blinkspromote secretion from unobstructed glands [4,63,65]. Depend-ing on pre-existing blink efficiency, there may be great potentialfor improvement in frequency and completeness of blinking toincrease and maintain lipid secretion from all functioning glandsincluding those whose function has been restored. Patient educa-tion and guidance on appropriate blink efficiency exercises whichimprove blink completeness may greatly enhance the benefits ofimproved meibomian gland function [63,66]. Failure to considerblink efficiency may significantly undermine efforts to improveMG function. When scleral or lid massage is prescribed for post-trabeculectomy patients, careful monitoring of compliance and thepossibility of adverse responses is an essential component of sub-sequent management [2,3]. The same caveat is appropriate in thetreatment of MGD with corneal topography having an importantrole in detecting adverse changes.

12. Conclusions

Increased corneal susceptibility to adverse shape changes fromprescribed massage may be associated with elevated cornealtemperature. Utilising heat and therapeutic occupation facilitatesdeformation of tissue more effectively resulting in tissue elongationand deformation [67]. Several mechanisms for massage-relatedcorneal temperature elevation combined with warm compressesor other methods of heat application, explain how meibomiangland dysfunction treatment could facilitate the development ofrubbing-related corneal deformation. Apparently for the first time,such a case is described in this review. Raised corneal temper-ature may be significant for explaining corneal deformation inassociation with rubbing/massage in conditions such as chalazion,post-trabeculectomy, post-laser assisted in situ keratomileusisand dacryocystoceles. The risk may also be higher for corneaswith concurrent post-surgical wound healing subsequent to graft,refractive surgery or trabeculectomy. These associations are consis-tent with the alteration to corneal biomechanical properties found

following experimental rubbing. Elevated temperatures may besustained in the avascular cornea with associated risk of increasedmetabolic rate and enzyme activity, and possible loss, or otheradverse alteration to collagen or ground substance. Consequently,

ens &

tarwwe

F

p

t

Bcfer

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

C.W. McMonnies et al. / Contact L

he risk may be higher when spontaneous rubbing occurs inddition to prescribed massage. Consideration of risk factors forubbing/massage-related adverse changes appears to be indicatedhen prescribing therapeutic massage. Safer massage techniques,hich reduce corneal exposure to mechanical trauma can be

mployed.

inancial interests

There has been no funding or other form of financial supportrovided in relation to the preparation of this report.

McMonnies CW has no financial interests to declare in relationo this report.

Korb DR is an inventor or co-inventor of Soothe XP and Systanealance, products used for the treatment of dry eye. He is also aofounder and has a financial interest in TearScience Inc, a companyocused on the treatment of evaporative dry eye. Blackie CA is anmployee of Tear Science Inc. None of these interests have a directelation to the contents of this report.

eferences

[1] Daily RK, Daily L. Physical therapy. In: Dunlap EA, editor. Gordon’s medicalmanagement of ocular disease. New York: Medical Book Department, Harper& Row Publishers; 1976. p. 6–7.

[2] McIlraith I, Buys Y, Campbell RJ, Trope GE. Ocular massage for intraocular pres-sure control after Ahmed valve insertion. Can J Ophthalmol 2008;43:48–52.

[3] Gouws P, Buys YM, Rachmiel R, Trope GE, Fresco BB. Finger massage versus anovel massage device after trabeculotomy. Can J Ophthalmol 2008;43:222–4.

[4] Linton RG, Curnow DH, Riley WJ. The meibobian glands. Br J Ophthalmol1961;45:718–23.

[5] Driver P, Lemp M. Meibomian gland dysfunction. Surv Ophthalmol1995–1996;40:343–67.

[6] Korb DR, Henriquez AS. Meibomian gland dysfunction and contact lens intol-erance. J Am Optom Assoc 1980;51:243–51.

[7] Romero JM, Biser SA, Perry HD, Levinson DH, Doshi SJ, Terraciano A, et al.Treating Meibomian gland dysfunction. Eye Cont Lens 2004;30:14–9.

[8] McMonnies CW. Mechanisms of rubbing-related corneal trauma in kerato-conus. Cornea 2009;28:607–15.

[9] Koenig SB. Bilateral recurrent self-induced keratoconus. Eye Cont Lens2008;34:343–4.

10] Yeniad B, Alparslan N, Akarcay K. Eye rubbing as an apparent cause of recurrentkeratonus. Cornea 2009;28:477–9.

11] Cosar CB, Rapuano CJ, Cohen EJ, Laibson PR. Chalazion as a cause of decreasedvision after LASIK. Cornea 2001;20(8):890–2.

12] Wilson LA. External diseases of the eye. New York: Harper & Row Publishers;1979. p. 45.

13] Korb DR, Blackie CA. Meibomian gland diagnostic expressibility: correlationwith dry eye symptoms and gland location. Cornea 2008;27:1142–7.

14] Blackie CA, Korb DR, Knop E, Bedi R, Knop N, Holland EJ. Non-obvious obstruc-tive meibomian gland dysfunction (NOMGD). Cornea 2010;29(12):1333–45.

15] Korb DR, Blackie CA. Meibomian gland therapeutic expression: quantifyingthe applied pressure and the limitation of resulting pain. Eye Cont Lens2011;37:298–301.

16] Raizman MB, Rothman JS, Maroun F, Rand WM. Effect of eye rubbing on signsand symptoms of allergic conjunctivitis in cat-sensitive individuals. Ophthal-mology 2000;107:2158–61.

17] McMonnies CW, Boneham GC. Keratoconus allergy, itch, eye-rubbing and handdominance. Clin Exp Ophthalmol 2003;86:376–84.

18] Baldassare RD, Brunette I, Desjardins DC, Amyot M. Corneal ectasia secondaryto excessive ocular massage following trabeculectomy with 5-fluorourcil. CanJ Ophthalmol 1996;31:252–4.

19] Lucarelli MJ, DeBry P. Corneal ectasia associated with massage of dacryocysto-celes. Cornea 2002;21:419–20.

20] Horven I. Corneal temperature in normal subjects and arterial occlusive disease.Acta Ophthalmol (Copenh) 1975;53:863–74.

21] Lychovskii LM. The corneal temperature as an index of intensity of the circula-tion in the anterior segment of the eye. Byull Eksp Biol Med 1967;63:16–8.

22] Hill RM. Curiosities of the contact lens. Chicago: Professional Press; 1981. p.66–8.

23] Korb DR, Greiner JV, Herman JP, Hebert E, Finnemore V, Exford J, et al. Lid-wiper epitheliopathy and dry eye symptoms in contact lens wearers. CLAOJ2002;28:211–6.

24] Kessing SV. A new division of the conjunctiva on the basis of X-ray examination.Acta Ophthalmol 1967;45:680–3.

25] Adler FH. Physiology of the eye. St. Louis: The C.V. Mosby Company; 1959. p.137–53.

26] Fatt I. Physiology of the eye. Boston: Butterworths; 1978. p. 210.

[

Anterior Eye 35 (2012) 148– 154 153

27] Morgan PB, Tullo AB, Efron N. Infrared thermography of the tear film in dry eye.Eye 1995;9:615–8.

28] Blackie CA, Solomon JD, Greiner JV, Holmes M, Korb DR. Inner eyelid surfacetemperature as a function of warm compress methodology. Optom Vis Sci2008;85:675–83.

29] Solomon JD, Case CL, Greiner JV, Blackie CA, Herman JP, Korb DR. Warm com-press induced visual degradation and Fischer-Schweitzer polygonal reflex.Optom Vis Sci 2007;84:580–7.

30] Mansour AM, Haddad RS. Corneal topography after ocular rubbing. Cornea2002;21:756–8.

31] McMonnies CW, Alharbi A, Boneham GC. Epithithelial responses to rubbing-related mechanical forces. Cornea 2010;29:1223–31.

32] Greiner JV, Leahy CD, Welter DA, Hearn SL, Weidman TA, Korb DR. Histopathol-ogy of the ocular surface after eye rubbing. Cornea 1997;16:327–32.

33] Greiner JV, Peace DG, Baird RS, Allansmith MR. Effects of eye rubbing on the con-junctiva as a model of ocular inflammation. Am J Ophthalmol 1985;100:45–50.

34] Mapstone R. Determinants of corneal temperature. Br J Ophthalmol1968;52:729–41.

35] Palladino JL, Davis RB. Biomechanics. In: Enderle JD, Blanchard SM, Bronzino JD,editors. Introduction to biomedical engineering. New York: Elsevier AcademicPress; 2005. p. 159.

36] Knudson D. Fundamentals of biomechanics. New York: Kluwer Aca-demic/Plenum Publishers; 2003. p. 69–78.

37] Bonifasi-Lista C, Lake SP, Small MS, Weiss JA. Viscoelastic properties of thehuman medial collateral ligament under longitudinal, transverse and shearloading. J Orthop Res 2005;23:67–76.

38] Greene PR, McMahon TA. Scleral creep vs. temperature and pressure in vitro.Exp Eye Res 1979;29:527–37.

39] Dupps WJ. Biomechanical modeling of corneal ectasia. J Refract Surg2005;21:186–90.

40] Liu W-C, Lee S-M, Graham A, Lin MC. Effects of eye-rubbing and breath holding-holding on corneal biomechanical properties and intraocular pressure. Cornea2011, doi:10.1097/ICO.obo13e3182032b21.

41] Peterson ME, Daniel RM, Danson MJ, Eisenthal R. The dependence of enzymeactivity on temperature: determination and validation of parameters. BiochemJ 2007;402:331–7.

42] Langridge J. Thermal responses of mutant enzymes and temperature limits togrowth. Mol Gen Genet 1968;103:116–26.

43] Wilson SE, He Y-G, Qian JW, Li Q, McDowall AW, Vital M, et al. Epithelial injuryinduces keratocyte apoptosis: hypothesized role for the interleukin-1 systemin the modulation of corneal tissue organization and wound healing. Exp EyeRes 1996;62:325–37.

44] Zhou L, Sawaguchi S, Twining SS, Sugar J, Feder RS, Yue BYJT. Expression ofdegradative enzymes and protease inhibitors in corneas with keratoconus.Invest Ophthalmol Vis Sci 1998;39:1117–24.

45] Kenney MC, Brown DJ. The cascade hypothesis of keratoconus. Cont Lens AntEye 2003;26:139–46.

46] Lewis M, Dubin M, Arandahl V. Physical properties of bovine corneal collagen.Exp Eye Res 1967;6:57–69.

47] Li Y, Li Y, Zongliang D, Li G. Comparison of dynamic denaturation temperature ofcollagen with its static denaturation temperature and the configuration charac-teristics in collagen denaturation processes. Thermochim Acta 2008;469:71–6.

48] Capodici C, Berg RA. Cathepsin G degrades denatured collagen. Inflammation1989;13:137–45.

49] Unal M, Yucel I, Akar Y, Akkoyunlu G, Gultekin I. Recurrence of keratoconus intwo corneal grafts after penetrating keratoplasty. Cornea 2007;26:362–4.

50] Binder P. Analysis of ectasia after laser in situ keratomileusis: risk factors. JCataract Refract Surg 2007;33:1530–8.

51] Jin GJC, Lyle WA, Merkley KH. Late-onset idiopathic diffuse lamellar keratitisafter laser in situ keratomileusis. J Cataract Refract Surg 2005;31:435–7.

52] Wheeldon CE, Hadden OB, Niederer RL, McGhee CNJ. Presumed late diffuselamellar keratitis progressing to interface fluid syndrome. J Cataract RefractSurg 2008;34:322–6.

53] Schmack I, Dawson DG, McCarey BE, Waring GO, Grossniklaus HE, EdelhauserHF. Cohesive tensile strength of human LASIK wounds with histologic, ultra-structural and clinical correlations. J Refract Surg 2005;21:433–45.

54] Gloster J. The displacement of intra-ocular fluid during tonometry. Int Ophthal-mol Clin 1965;5:952–7.

55] McMonnies CW, Boneham GC. Corneal curvature stability with increasedintraocular pressure. Eye Cont Lens 2007;33:124–9.

56] Claesson M, Armitage W. Ten-year follow-up of graft survival and visual out-come after penetrating keratoplasty in Sweden. Cornea 2009;28:1124–9.

57] Lam AK, Chen D. Effect of ocular massage on intraocular pressure and cornealbiomechanics. Eye 2007;21:1245–6.

58] Matsumoto Y, Dogru M, Goto E, Ishida R, Kojima T, Onguchi T, et al. Efficacyof a new warm moist air device on tear functions of patients with simpleMeibomian gland dysfunction. Cornea 2006;25:644–50.

59] Goto E, Monden Y, Takano Y, Mori A, Shimmura S, Shimazaki J, et al. Treatmentof non-inflamed obstructive meibomian gland dysfunction by an infrared warmcompression device. Br J Ophthalmol 2002;86:1403–7.

60] Li X, Yang H, Rabinowitz YS. Keratoconus: classification scheme

based on videokeratography and clinical signs. J Cataract Refract Surg2009;35:1597–603.

61] Kessel L, Johnson L, Arvidsson H, Larsen M. The relationship between bodyand ambient temperature and corneal temperature. Invest Ophthalmol Vis Sci2010;51:6593–7.

1 ens &

[

[

[

[

[

54 C.W. McMonnies et al. / Contact L

62] Korb DR, Blackie CA. The number of Meibomian glands yielding liquid secretion(MGYLS) as a function of age and type of contact lens wear. Invest OphthalmolVis Sci 2010;51:D876. E-Abstract 6248.

63] Korb DR, Baron DF, Herman JP, Finnemore VM, Exford JM, Londono Her-mosa J, et al. Tear film lipid layer thickness as a function of blinking. Cornea1994;13:354–9.

64] Tiffany JM, Chew CKS, Bron AJ. Delivery of meibomian gland oils using theClinical Meibomianomometer. Exp Eye Res 1994;59:S111.

[

Anterior Eye 35 (2012) 148– 154

65] Bron AJ, Tiffany JM, Gouveia SM, Yokoi N, Voon LW. Functional aspects of thetear film lipid layer. Exp Eye Res 2004;78:347–60.

66] McMonnies CW. Incomplete blinking: exposure keratopathy, lid wiper epithe-

liopathy, dry eye, refractive surgery, and dry contact lenses. Cont Lens Ant Eye2007;30:37–51.

67] Bracciano AG. Physical agent modalities: theory and application for theoccupational therapist. 2nd ed. New Jersey: Slack Incorporated; 2008.p. 105.