Rethinking Dry Eye Disease: A Perspectiveon Clinical Implications
ANTHONY J. BRON, FRCOPHTH, F.MEDSCI,1 ALAN TOMLINSON, PHD, DSC, FCOPTOM,2
GARY N. FOULKS, MD, FACS,3 JAY S. PEPOSE, MD, PHD,4 CHRISTOPHE BAUDOUIN, MD, PHD,5
GERD GEERLING, MD, FEBO,6 KELLY K. NICHOLS, OD, MPH, PHD,7 AND MICHAEL A. LEMP, MD8
ABSTRACT Publication of the DEWS report in 2007 estab-lished the state of the science of dry eye disease (DED). Sincethat time, new evidence suggests that a rethinking oftraditional concepts of dry eye disease is in order. Specif-ically, new evidence on the epidemiology of the disease, aswell as strategies for diagnosis, have changed the under-standing of DED, which is a heterogeneous disease associ-ated with considerable variability in presentation. Theseadvances, along with implications for clinical care, aresummarized herein. The most widely used signs of DED arepoorly correlated with each other and with symptoms. Whilesymptoms are thought to be characteristic of DED, recentstudies have shown that less than 60% of subjects withother objective evidence of DED are symptomatic. Thus theuse of symptoms alone in diagnosis will likely result inmissing a significant percentage of DED patients, particu-larly with early/mild disease. This could have considerableimpact in patients undergoing cataract or refractive surgeryas patients with DED have less than optimal visual results.The most widely used objective signs for diagnosing DED allshow greater variability between eyes and in the same eyeover time compared with normal subjects. This variability isthought to be a manifestation of tear film instability whichresults in rapid breakup of the tearfilm between blinks andis an identifier of patients with DED. This feature emphasizesthe bilateral nature of the disease in most subjects notsuffering from unilateral lid or other unilateral destabilizing
Accepted for publication February 2014.
Authors’ academic affiliations: 1Professor emeritus - University of Oxford,Nuffield Laboratory of Ophthalmology, Nuffield Dept of Clinical Neurosci-ences, UK, 2Professor of Vision Sciences, Glasgow Caledonian University,Scotland, 3Emeritus Professor of Ophthalmology, University of Louisville;Editor-in-Chief, The Ocular Surface, USA, 4Professor of Clinical Ophthal-mology and Visual Sciences, Washington University School of Medicine,Director, Pepose Vision Institute, St. Louis, Missouri, USA, 5Quinze-VingtsNational Ophthalmology Hospital, and Vision Institute, University Paris 6,Paris, France, 6Professor and Chair, Department of Ophthalmology, Hein-rich-Heine-University Moorenstr. 5 40225 Düsseldorf, Germany, 7FERVProfessor (Foundation for Education and Research in Vision), The OcularSurface Institute, University of Houston, College of Optometry, Houston,Texas, USA and 8Clinical Professor of Ophthalmology, Georgetown Univer-sity, Washington DC and George Washington University, Washington DC,USA.
THE OCULAR SURFACE / APRIL 2014, VOL
surface disorders. Instability of the composition of the tearsalso occurs in dry eye disease and shows the same variancebetween eyes. Finally, elevated tear osmolarity has beenreported to be a global marker (present in both subtypesof the disease- aqueous-deficient dry eye and evaporativedry eye). Clinically, osmolarity has been shown to be thebest single metric for diagnosis of DED and is directlyrelated to increasing severity of disease. Clinical examina-tion and other assessments differentiate which subtype ofdisease is present. With effective treatment, the tear os-molarity returns to normal, and its variability between eyesand with time disappears. Other promising markers includeobjective measures of visual deficits, proinflammatorymolecular markers and other molecular markers, specific toeach disease subtype, and panels of tear proteins. As yet,however, no single protein or panel of markers has beenshown to discriminate between the major forms of DED.With the advent of new tests and technology, improvedendpoints for clinical trials may be established, which inturn may allow new therapeutic agents to emerge in theforeseeable future. Accurate recognition of disease is nowpossible and successful management of DED appears to bewithin our grasp, for a majority of our patients.
This article was developed from a roundtable meeting held on December 1-2,2012, andorganizedbyMediTechMedia.Themeetingwas supported byanun-restricted educational grant from TearLab, Inc, who had no input into themeeting or content of this article. Support for this publication was providedby TearLab, Inc. Complete disclosure of authors’ sources of support and rela-tionships with commercial companies are provided at the end of the text.).
Corresponding Author: Anthony J. Bron, FRCOphth, F.MedSci, ProfessorEmeritus - University of Oxford, Nuffield Laboratory of Ophthalmology,Nuffield Dept of Clinical Neurosciences, UK. E-mail address: [email protected]
1. The Blink Cycle and Formation of the Tear FilmLipid Layer
2. Refreshment of the Lipid Film
3. The Blink Rate
4. Optical Performance of the Tear Film
D. Tear Osmolarity
1. The Thickness and Quality of Tear Film LipidLayer
2. Palpebral Aperture Width and GlobeProminence
3. Tear Film StabilitydTear Film Breakup
4. Effect of Ambient Environment
E. The Influence of Tear Film Stability
1. Tear Breakup
2. Blink Rate
III. Classification, Epidemiology, and Natural History ofDED
A. Epidemiology and Predisposing Factors for DED
1. Dispositional and External EnvironmentalFactors
2. Systemic Disease
B. Natural History
C. Relationship of Natural History to Epidemiologyand to the Relevance of Diagnostic Tests andTreatments across the Spectrum of Disease
IV. Inflammation in the Pathogenesis of DED and Path-ways to the Vicious Cycle
A. Inflammation in DED
B. Inflammatory Events Involved in the Initiation ofDED
1. T Cells and NK Cells as Drivers of Inflammation
2. Deregulation of Regulatory T cells (Tregs) andIL-17eProducing T cells
3. APCs and Toll-Like Receptors (TLRs)
B. Markers of Inflammation in DED
1. Matrix Metalloproteinase-9 (MMP-9)
2. HLA-DR
3. Cytokines and Chemokines
D. Anti-inflammatory Therapies for DED
1. Cyclosporine A
2. Corticosteroids and Doxycycline
3. Resolution of Inflammation with Resolvins
4. Novel Agents
V. Assessment Issues: Symptoms and Signs
A. Challenges Associated with Assessment in DED
B. The Role of Tear Film Instability
VI. New Applications of Point-of-Care Tests
A. Contact Lens Wear
B. Refractive Surgery
C. Cataract Surgery
D. Glaucoma Therapy
VII. New Tests on the Horizon
A. Interferometry
B. Assessment of Tear Film
C. Measuring Visual Acuity and Contrast Sensitivity
D. Meniscus Measurement
E. Quantitative Proteomics
F. Evaluating the Meibomian Gland
G. Patient-Reported Outcomes (PROs)
RETHINKING DRY EYE DISEASE / Bron, et al
S2 THE OCULAR SURFACE / APRIL 2014, VO
I. INTRODUCTIONA. Overview
I n 2007, the report of the International Dry EyeWorkshop (DEWS) served as a comprehensive re-view of dry eye disease (DED), its pathogenesis, nat-
ural history, and methods used to diagnose the condition.1-6
Although this report represented the state of the art in2007, a number of important research and clinical develop-ments have transpired since then. Accordingly, there is aneed to reassess some aspects of the DEWS conclusions andrecommendations in light of recent advances in diagnosis.
B. Objectives of This PublicationThis report is not intended to be as comprehensive as the
DEWS report but rather, aims to provide an update on dryeye for the practicing clinician. At the time of theDEWS report,many advances were imminent, and a further aim of this paperis to demonstrate their importance. One concept retained fromthe DEWS report and emphasized here is that of the lacrimalfunctional unit (LFU) maintains ocular surface homeostatisby regulating tear flow, thus conserving the tear film andcorneal transparency. The LFU consists of the cornea, conjunc-tiva, lacrimal and meibomian glands and the lacrimal drainagesystem, connected reflexly by a neural network. Its failure torespond adequately to dessicating stress is a key initator ofdry eye.7
One of the major challenges in the dry eye field is in theproper assessment of DED, a multifactorial condition. Animportant objective is to review recent advances in diagnosisand in grading severity and to consider their implications forpatient selection criteria for clinical trials. Articles on inflam-mation in DED have continued to extend our knowledge ofthis important subject. In addition, since publication of theDEWS Report in 2007, over 140 articles on tear osmolarityhave appeared in the peer-reviewed literature, making it the
Figure 1. Expression of meibum from meibomian glands. Photocourtesy of Jay Pepose.
RETHINKING DRY EYE DISEASE / Bron, et al
most researched area of the newer approaches to DED. Thesestudies have not only elucidated the role of tear osmolarity butare relevant to other aspects of the disease. These studiescomprise a significant portion of this report. Lastly, an impor-tant goal of this publication is to challenge some commonlyheld misconceptions about DED, including that: (1) DED issynonymous with the aqueous-deficient subtype (ADDE);(2) DED is a symptom-driven disease; (3) there are no signif-icant visual sequelae with DED; (4) variability in objectivesigns and symptoms imposes problems for clinical practiceand research; and (5) DED is simply a part of the burden ofold age. These misconceptions hamper clinical managementof DED and need to be re-evaluated.
The panelists who served as authors of this perspectivewere selected for their familiarity with the previous DEWSReport and their published clinical and research activitiesrelated to DED. Although specific authors were assignedto lead each major section of this report, all authors partic-ipated in a roundtable meeting prior to compilation of thereport (with the exceptions of Professors Baudouin whosubmitted extensive written material regarding inflamma-tion and Professor Nichols who provided critical commen-tary and review of the manuscript). All authors areconsultants to or on the Academic Research Board of Tear-Lab, Inc. Each author reviewed the entire report and rec-ommended additions or changes to the content andlanguage. Each author confirmed their agreement andassent to the final document. Disclosures of potential con-flict of interest are provided in the acknowledgementappendix.
II. TEAR PHYSIOLOGY AND PATHOPHYSIOLOGYA. The Ocular Surface and the Sources of Tears
Whether the eyes are open or closed, the ocular surfaceis continuously bathed by the tears, which are secreted bythe lacrimal and meibomian glands, with additional contri-butions from the conjunctiva.8 The ocular surface consists ofthe epithelium and subjacent tissues of the cornea andlimbus and the bulbar and tarsal conjunctiva, extending tothe mucocutaneous junctions of the lid margins.
The lacrimal glands are exocrine glands whose main,palpebral, and accessory parts secrete an aqueous fluidinto the upper and outer conjunctival fornices.8 The lacrimalsecretion is responsible for the bulk of the tear volume andflow8,9 with a smaller portion resulting from conjunctivalsecretion.10 Tear fluid is distributed to and mixed with thepreocular film during the blink and is then lost by outflowfrom the tear menisci via the nasolacrimal system. Wateris further lost by evaporation from the whole of the exposedpreocular tear film.8,11
The meibomian glands, embedded in the tarsal plates,are tubuloacinar, holocrine glands, whose acini dischargetheir entire contents into their ducts in the process of secre-tion.12,13 They lie in parallel rows, with their duct orificesopening onto the lid margins, anterior to the mucocuta-neous junction. The meibomian secretion (meibum) isdelivered into a shallow, lid margin pool, or reservoir,
THE OCULAR SURFACE / APRIL 2014, VOL
from which it can be spread onto the anterior surface ofthe preocular tear film with each blink to form its lipid layer,which performs a key role in retarding water evaporationfrom the exposed ocular surface.12,13
The meibomian glands secrete a lipid mixture with amelting range between 19.5 and 32.9�C, which is liquid atbody temperature and on the lid margins.12 Normally, clearoil (meibum) can be expressed from the meibomian orificesby pressing on the glands through the lids (Figure 1).Expressibility is greatest nasally and least temporally.14 De-livery of oil to the lid margin reservoir occurs in partthrough a steady secretory process and in part through therelease of small aliquots of meibum with each blink. Thelid reservoir contains at least 30 times the amount of lipidpresent on the surface of the tear film (approximately 300mg vs 10 mg, respectively).15,16 The excretory process formeibomian oil is not fully known, but meibomian lipidmay migrate from the reservoirs over the lid margin skinand lashes, further serving to prevent tear film contamina-tion by sebaceous skin lipids (sebum), which can otherwisedestabilize the tear film.17 This has implications for meibo-mian gland dysfunction (MGD), when reduced oil deliveryto the lid margin or changes in chemical composition canbreak down this barrier.
The conjunctival epithelial cells make a small contribu-tion to the aqueous part of tears, and the conjunctival gobletcells add a gel mucin component (MUC5AC), which lubri-cates the area between the lids and globe, particularly in thelid-wiper region of the lid margins, where they impinge onthe ocular surface.10,18
B. The Tear Film and Tear Film StructureThe preocular tear film covers both the exposed conjunc-
tiva and the cornea.19 Film thickness over the conjunctiva isdifficult to measure because of conjunctival rugosity andconjunctival folds, but the precorneal tear film, measuredinterferometrically, is approximately 3 mm.20 The precornealtear film has 3 structural components: a superficial lipid layer,an aqueous layer, and a deep mucin layer, with the latter
related directly to the epithelial surface. Holly suggested thattear mucin (referring to goblet cell, gel mucin) is separatedinto 2 phases: a deep phase of high viscosity containingmost of the macromolecules and associated with the epithe-lium, and a superficial phase containing mucin in dilute solu-tion, associated with the lipid layer.21 Recently it has beensuggested that there is no distinct segregation of gel mucinwithin the aqueous subphase.22,23
1. The Tear Film Lipid LayerThe tear film lipid layer arises from the shallow reservoir
of lipid at the lid margins and is spread onto the tear filmwith each blink. It has a mean thickness of 42 nm (15-157nm)24 and plays a key role in retarding tear evaporationfrom the exposed ocular surface.25 Its deepest aspect iscomposed of polar lipids and some long-chain fatty acids,a few molecules thick, which interface with the aqueous sub-phase of the tear film. This represents about 5% to 15% ofthe total lipid composition.26 Some proteins and glycopro-teins, such as lipocalins, lysozyme, and mucin, are thoughtto be intercalated with the lipid layer and enhance its stabil-ity.27-30
2. The Aqueous LayerThe aqueous layer forms the bulk of the tear film and
contains salts and a large range of proteins derived fromthe lacrimal gland and conjunctiva. Proteins include growthfactors such as epidermal growth factor and hepatocytegrowth factor, important to the maintenance of the epithe-lium.31-33 There are also defense proteins, such as lysozyme,lactoferrin, surfactant protein-D, and trefoil peptide, con-cerned with innate immunity, and immunoglobulin (Ig) A,of lacrimal plasma cell origin, involved in adaptive immu-nity.34-36 Levels of these proteins may be decreased inDED, making the eye more vulnerable to infection. Addi-tional proteins, such as albumin, normally present at lowconcentration are derived from the plasma and may beincreased in DED as a result of inflammation caused byan increase in both vascular capillary permeability andthat of the surface layer of the conjunctival epithelium.36
3. The Mucin LayerThe goblet cells of the conjunctiva secrete a gel mucin
(MUC5AC) into the tears, which interacts with the glycoca-lyx and performs a lubricating function between the lids andglobe.23 This mucin layer may maintain wettability of theocular surface where the glycocalyx is defective.21 It alsotraps shed epithelial cells, inflammatory cells, and debristo form a mucous thread within the lower conjunctival sac.37
4. The Tear MenisciTear menisci are strips of tear fluid lying at the junction
of the lid margins and the globe formed by surface tensionforces as the lid margins separate from one another in theupstroke of the blink.17 A negative hydrostatic pressurewithin the menisci, responsible for their concave externalsurfaces, draws water from the tear film as it forms in the
S4 THE OCULAR SURFACE / APRIL 2014, VO
upstroke of the blink, leading to segregation of the meniscifrom the tear film.38,39 This negative pressure within themenisci opposes the continuous flow of aqueous tears intothe puncta, so that drainage is limited to roughly the first2 s of the blink interval.38-40 The volume of the menisci isdirectly related to the total volume of the tear fluid41 andindirectly to the lacrimal secretory rate.9 For this reason,the height and radius of curvature of the tear menisci arereduced in aqueous-deficient DED and their measurement,in the lower meniscus, is used in DED diagnosis(Figure 2).42-46
C. Blinking and Tear Film DynamicsSpontaneous blinking consists of brief bilateral, synchro-
nous opening and closing of the palpebral aperture by thedescent and ascent of the upper lid.47-49 It refreshes thetear film in response to the environment and to changesin mental state and eye movements. Its key role is to preventdesiccation and it also contributes to the mechanism of teardrainage, meibomian oil delivery and, importantly, toremoval of debris from the ocular surface.
1. The Blink Cycle and Formation of the Tear Film LipidLayerBlinking plays a key role in tear dynamics by spreading,
mixing, and distributing the tears and by refreshing the tearfilm. The blink cycle consists of the blink itself, and an inter-blink period, or blink interval, during which evaporativewater loss occurs.50 The tear film lipid layer is formed inthe upstroke of each blink, when lipid from the lowermeibomian reservoir spreads upward over the aqueous sub-phase of the preocular tear film.19 It has been suggested thatthinning of the lipid layer superiorly at this time creates alocal rise in surface tension, which is the driving force forlipid layer spreading.19 It is thought that this is achievedinitially by an interaction between polar meibomian lipidsand the aqueous phase of the tear film, reinforced by thepresence of intercalated proteins, such as lipocalin.30,51 Dur-ing spreading, the polar lipid layer is considered to act as acarrier for the spreading of the nonpolar lipid fraction.25
In the normal eye, spreading of the tear film lipid layercan be observed clinically by interference video microscopy,where it is seen as an upwardly moving front of horizontallydisposed colored fringes (Figure 3). The lipid film spreadsrapidly at first (about 10 mm/s),52 although it lags markedlybehind the upper lid, and then slows to stabilize asymptoti-cally after 1 s or more, long after completion of theblink.52-54 Spread time of the lipid layer may be reduced inboth aqueous and lipid tear deficiency. In the presence of ameibomian lipid deficiency, these fringes assume a verticalpattern, which may be diagnostically useful.53 Brown andDervichian (1969) considered that surface tension forcesgenerated by the spreading lipid layer dragged aqueous up-ward in a secondary manner, causing thickening of theaqueous subphase inferiorly.55 This has been questioned byKing-Smith et al, who observed that about 1 second afterthe blink, the central tear film thins.19
Figure 3. Spreading of the lipid layer viewed by interferometry.Representative, sequential, interference images of the spreadingtear-film lipid layer, captured every 0.05 s after a blink, by video inter-ferometry. Spreading increases from left to right and from abovedownward. From Yokoi N, Yamada H, Mizukusa Y, et al. Rheology of tearfilm lipid layer spread in normal and aqueous tear-deficient dry eyes.Invest Ophthalmol Vis Sci. 2008;49(12):5319-5324.
Figure 2. Meniscus volume imaged using a video meniscometer in a:(A) normal eye, (B) dry eye, and (C) dry eye with punctal plugs. Pleasenote the different widths. From Yokoi N, Komuro A. Non-invasivemethods of assessing the tear film. Exp Eye Res. 2004;78(3):399-407.
RETHINKING DRY EYE DISEASE / Bron, et al
2. Refreshment of the Lipid FilmIn the down phase of the blink, the lipid film is com-
pressed completely and then restored in the up phase.
THE OCULAR SURFACE / APRIL 2014, VOL
When it is viewed by interference microscopy, the patternsof colored fringes can show a strong resemblance to oneanother from blink to blink, but nonetheless show a stepwisedegradation over a series of blinks, implying a gradual lossof structural integrity. The pattern eventually changesabruptly, suggesting that the lipid layer has been re-formed by mixing with meibum from the lid marginreservoirs12
3. The Blink RateWide variations in reported blink rate have been seen in
normal adults, probably reflecting individual variation andthe influence of environmental and experimental conditions.Blink rate is strongly influenced by mental state, attention,activity, exposure of the ocular surface, and by environ-mental conditions. Reported blink rates vary from lessthan 1 per min to as high as 176 per min.56 In controlledenvironmental conditions (eg, at about 22�C with humidityof 40.0%), blink rate in normal adults ranges between 15 and20 per min.50,57,58 Tsubota reported a blink rate of 14.3 permin in adults with healthy eyes, representing a blink cycle of4.2 s (ie, a blink interval of 4.0 s and a blink time of 0.2 s).58
Blink rate is increased reflexively in response to ocularsurface irritation or exposure to high winds and ocular sur-face cooling. It also increases in DED, where it is thought tobe compensatory. However, Himebaugh et al noted that inpatients with DED, despite the increase in blink rate,when tear film breakup occurred within the blink interval,the rate of evolution of the breakup and its final area wasgreater than in participants with healthy eyes.59 Blink ratefalls during a number of common tasks requiring visual con-centration, and the increased evaporative loss may act as atrigger for DED.50 Because of the importance of blink rate
Figure 4. Higher-order wavefront aberrations after blinking in normal(C) and dry (B) eyes. From Montes-Mico R, Alio JL, Charman WN. Dy-namic changes in the tear film in dry eyes. Invest Ophthalmol Vis Sci.2005;46(5):1615-1619.
RETHINKING DRY EYE DISEASE / Bron, et al
to tear osmolarity, tear film stability, and vision, it will bediscussed in detail in a later section.
4. Optical Performance of the Tear FilmThe normal optical system is not perfect and shows
higher-order aberrations (HOAs) because of the propertiesof the lens, cornea, and tear film. These HOAs have beenrecorded by wavefront aberrometry to demonstrate dynamicchanges in optical properties of the tear film. Wavefront ab-erration data have also been derived directly from the sur-face of the tear film by converting videokeratoscopic datainto wavefront aberration coefficients.60,61 Wavefront aber-ration data can be processed further to gain some insightinto the quality of retinal imaging,60,62-64 but this has beenachieved more directly using a narrow-beam, double-passsystem.64
Wavefront aberrometry studies have shown that, inhealthy eyes, optical quality of the tear film improvessteadily during the blink interval. In one study, during blinksuppression, HOAs reached a minimum at 6.1 � 0.5 s aftera blink and rose steadily thereafter. The same pattern wasobserved in a group of patients with DED but with an earlierminimum at 2.9 � 0.4 s.61 These values were significantlydifferent between the groups after 4 s. In both groups, totalaberration minima were highly correlated with tear breakupvalues (Figure 4), suggesting that the inflection point is asso-ciated closely with the breakdown of the tear film.61 In onestudy in which modulation transfer function (MTF) wascalculated, the inflection point correlated well with breakuptime, from which it was inferred that the MTF minimum oc-curs immediately preceding or at the time of tear filmbreakup.62
D. Tear OsmolarityIn healthy participants tested in open-eye laboratory
conditions tear osmolarity was maintained within narrowlimits. Tomlinson reported a value of 302 � 9.7 mOsm/Lbased on data from several studies65 and importantly,between-eye differences were also limited (normal value,6.9 � 5.9 mOsm/L).66 The narrow range of values betweenparticipants reflects the influence of homeostatic mecha-nisms, and blink interval (ie, blink rate), as the chief modi-fier of evaporation, likely determines the bilateral set pointof tear osmolarity.67 Mathematical modeling suggests thatthere is a small osmolar differential between the tear filmand menisci, with tear film osmolarity higher than that ofthe menisci in the steady state.11 This may relate to the ratioof the tear film thickness to its surface area compared withthat of the menisci, and to tear mixing and flow in themenisci in the early phase of the interblink interval.68
Modeling considerations suggest that in DED, this differen-tial is amplified; a tear sample taken from the meniscus mayunderestimate that of the tears over the surface of the eye.11
Once breakup has occurred, this differential could be ampli-fied by a wave of hyperosmolarity spreading over the surfacefrom the point of breakup, which may not be fully reflectedby meniscus measurements. Evaporation during the blink
S6 THE OCULAR SURFACE / APRIL 2014, VO
interval causes a measurable thinning of the tear film, anda consequent rise in tear film osmolarity is predicted.69
With each blink, the tear film mixes partially with tarsal-fornical fluid of lower osmolarity, so the osmolarity of thepreocular film is assumed to be lower just after the blinkthan just before.69 However, tear osmolarity in the menisciis remarkably stable in healthy eyes.66
1. The Thickness and Quality of Tear Film Lipid LayerThe rate of water loss from the eye is hypothesized to be
strongly influenced by the normal tear film lipid layer,thereby reducing evaporative loss from the eye. Expressionof meibum in normal eyes leads to thickening of the tearfilm lipid layer70,71 and to a reduction of evaporation inboth healthy individuals and patients with DED.72,73
When the quality or integrity of the tear film lipid layer isdeficient, as judged by interferometry, evaporative lossmay be increased.74 A similar increase might be anticipatedwith the slow spreading and vertical streaking of the lipidlayer occurring in MGD.52,75
2. Palpebral Aperture Width and Globe ProminenceTsubota and Nakamori examined the effect of gaze po-
sition on evaporation rate (at 40% humidity and a blinkrate of 30 per min) and showed that evaporative loss is3.4 and 2.5 times greater, respectively, when looking upand straight ahead than when looking down, not only pereye but also per unit area of the exposed ocular surface.57
3. Tear Film StabilitydTear Film BreakupBecause of its importance to image formation, stability
of the precorneal tear film has attracted considerable atten-tion. Clinically, stability is assessed by the tear film breakuptest (TBUT), which can be conducted either after fluoresceininstillation (the fluorescein breakup test [FBUT]) or nonin-vasively (the noninvasive breakup test [NIBUT]).53,67,76-79
The test assesses the time required for a random break toappear in the tear film after a spontaneous blink within
the blink interval during blink suppression (ie, a deliberatelyextended blink interval).
In healthy individuals, values reported for the TBUT arewell beyond the normal blink interval recorded in standardconditions of temperature and humidity. At first glance, thiswould suggest that the tear film is completely stable in suchindividuals during the blink interval and that its role inretarding evaporation is maintained throughout the blinkcycle.67,80-82 However, tear breakup does occur in healthyindividuals, although to a lesser extent than in patientswith DED.67,81 The relationship between the blink intervaland the breakup time can be captured in an individual bymeasuring the ocular protection index (OPI), which is thebreakup time divided by the blink interval.83 OPI of < 1 in-dicates that breakup is occurring within the blink interval.83
With a value of � 1, the breakup time recorded with blinksuppression exceeds the blink interval, and the eye is pro-tected from desiccation throughout the blink cycle.83 Inearly DED, OPI is initially > 1 and nears 1 as tear osmolar-ity increases, either by raised evaporation rate (in evapora-tive dry eye disease [EDE]) or by evaporation from areduced preocular tear film volume (in ADDE). Later, asthe disease progresses and OPI falls below 1, hyperosmolar-ity is amplified locally by the increase in evaporation at thesite of tear breakup, heightened by the defective lipid layer,and expanding centrifugally as the breakup enlarges. Thismechanism will likely raise osmolarity within the exposedepithelial cells locally, subjacent to the breakup, with the ef-fect being greater the earlier the onset of breakup. Modelingconsiderations suggest that a wave of hyperosmolarity,whose peak is at the starting point of the breakup, spreadsoutward as the breakup expands, reaching a level that de-pends on the duration of breakup and ambient conditions.84
In the regions surrounding the breakup, osmolarity is alsoincreased but to a more modest level over the same timeperiod.84 Local tear instability, giving rise to local drying,can be an independent starting point for tear hyperosmolar-ity and DED. Tear hyperosmolarity can be initiated by a lossof ocular surface wettability, leading to tear breakup andan OPI of < 1. The resulting dry eye has been referred toas an “extrinsic” form of EDE,1 but a better term is ocularsurfaceerelated EDE.
4. Effect of Ambient EnvironmentCertain environmental conditions increase evaporative
loss and are risk factors for DED. Evaporation is increasedin conditions of low humidity and increased airflow overthe surface of the eye.50,85,86 Such conditions may be com-bined and may also occur in natural, outdoor conditions.The effect of the environment on evaporation is the basisfor providing goggles or water-conserving spectacles forthe prevention or treatment of dry eye states.
E. The Influence of Tear Film StabilityNormally, tear dynamics are tightly regulated to main-
tain a smooth and optically effective precorneal tear filmwhile performing a range of tasks,87 but in DED, these
THE OCULAR SURFACE / APRIL 2014, VOL
homeostatic mechanisms break down, giving rise to tearfilm instability and visual disturbances.1 Clinically, patientswith DED experience blurred and fluctuating vision,decreased contrast sensitivity, and increased forward scatterand glare. These, together with symptoms of discomfortduring visual tasks, impair vision-related daily activitiesand quality of life.88
1. Tear BreakupThe importance of the tear film for optical imaging is
well known.87 Many approaches have been adopted to studyits stability and influence on visual function. Particular in-terest has been directed to the influence of DED on thesefunctions. Tear film breakup within the blink interval is acause of visual degradation, and its character and timecourse have been studied in detail in contact lenswearers.89,90 The effect of precorneal tear film breakup onvision may be due to variations in film thickness andrupture of the tear film and, in DED, partly due to uncov-ered surface irregularities and light-scattering opacities inthe surface epithelium. Visual acuity is a standard measureof visual function but does not provide a full account of vi-sual performance. However, a functional measure of visualacuity has been developed,91-93 and broader measures of vi-sual function have been used, such as contrast sensi-tivity,87,94-97 glare disability,95 and scatter index98dall ofwhich have been shown to be negatively impacted inDED.99-101
2. Blink RateDesiccating stresses influence tear film thickness and sta-
bility and vary from moment to moment. Eye movementsmay be constrained while performing selected visual tasksor in special environments in ways that influence the evap-oration rate and tear stability. Factors that modify the blinkrate, and hence the blink interval during which evaporationoccurs, likely influence visual function, particularly when theblink rate falls. A fall in blink rate has been documentedduring everyday visual tasks, such as working at a videodisplay terminal, reading in downgaze, operating monitor-based and handheld video games, and performingsurgery.102-104 In these situations, both gaze position anddifficulty of the visual task are important determinants ofblink rate.
Increased blink intervals have implications for vision,evaporation, and ocular surface desiccation. However, theeffects of downgaze on tear stability and visual functionare difficult to compute. For a given palpebral aperturesize, an increase in evaporative loss is expected on the basisof an increased blink interval. However, a reduction inpalpebral aperture size reduces evaporation at a given blinkrate.105-107In a study by Shaw et al, significant regions oftopographic disturbance were observed in young healthyparticipants in downgaze; these were greater for larger an-gles than for small (40� vs 20�) and were greater duringtasks that required deliberate eye movements comparedwith tasks without eye movements.106 These factors have
been attributed to the effects of lid margin pressure, partic-ularly of the upper lid, on corneal shape.Summary: Tear Physiology and Pathophysiology
d The lacrimal functional unit consists of the cornea, con-junctiva, lacrimal glands, meibomian glands, lids, andlacrimal drainage system and their reflex connections.
d Secreted tears are distributed to and mixed with thepreocular tear film and menisci with each blink andthen lost by evaporation and drainage from themenisci through the nasolacrimal passage.
d The lipid, aqueous, and mucin components of the tearfilm interact to maintain stability of the tear film,which maintains the integrity of the ocular surfaceepithelium.
d Secretion from the main and accessary lacrimal glandsaccounts for most of the tear volume, with the remainderderiving from active secretion by the conjunctiva.
d A negative hydrostatic pressure within the meniscidrives tears from the tear film into the menisci inthe first 100 ms after a blink.
d Increased evaporation and/or reduced production oftears leads to increased osmolarity and loss of the tears’homeostatic balance.
d The volume of the tear menisci is directly related to thetotal volume of tears. The tear osmolarity in themenisci, while not identical to that in the preoculartear film, is representative of it.
d Desiccating environmental stress from wind, lowambient humidity, and decreased blinking exacerbatestear evaporation. Looking upward or straight ahead isassociated with 2.5 to 3.4 times greater evaporative tearloss than a downward gaze.
d The tear film becomes stable again after the first sec-ond postblink, and stability increases for the following5 s, remaining stable until the next blink in healthy in-dividuals. In many patients with DED, tear filmbreakup begins much earlier postblink, usually within2 to 3 s or sooner. In healthy individuals, the tear filmusually remains stable over the entire interblink inter-val. In patients with DED, as the tear film begins tobreak up, there is a spike in tear osmolarity leadingto further instability and drying. This instability inthe tear film and the composition of the tear is a hall-mark of DED.
III. CLASSIFICATION, EPIDEMIOLOGY, AND NATURALHISTORY OF DED
A. Epidemiology and Predisposing Factors for DEDDED affects 5% to 34% of all people globally, and preva-
lence increases with age.5,108-116 In a retrospective analysisof 224 patients with DED, 159 (71%) were found to have 1of the 3 main subtypes: EDE (35%), ADDE (10%), and mixedMGD/ADDE (25%).108 The remaining 65 patients (29%)were not found to have clear evidence of ADDE or EDEand were classified as “dry eye: non-ADDE and non-EDE.”108
Regardless of DED subtype, downstream pathologicalevents contributing to ocular surface damage are similar.
S8 THE OCULAR SURFACE / APRIL 2014, VO
DED is an ocular surface disorder with a complicated andmultifactorial etiology.1 Numerous factors are hypothesizedto predispose one to DED, including contact lens wear,autoimmune disease, meibomian gland dysfunction, femalegender, low humidity, and over-the-counter medicationuse.114,117-126 This section will review predisposing factorsfor DED, as well as DED epidemiology and natural history.
1. Dispositional and External Environmental FactorsFemale gender is one of the most common predisposing
factors for DED.113,115,116,126,127 One population-based studyfound that the prevalence of DED was higher in women vsmen (17.0% vs 11.1%; P < .001).128 Among US veterans,women had a significantly higher risk of DED vs men(22% vs 12%; relative risk ¼ 2.40).127 Specifically, DED isassociated with older women, as DED incidence increaseswith age.126,129 In both women and men aged > 80 years,DED prevalence was 19.0%.128 The age differential inwomen may be related to the increased use of hormonereplacement therapy with increasing age and the significantrisk of developing DED with hormone replacement therapyuse.126,129 Decreased androgen production with age is likelyan important contributing factor to age-related DED.5,130,131
Lending support to this theory is the observation of anincreased incidence of DED among women with prematureovarian failure.132
Demographic and clinical factors (eg, age, gender, lowandrogen levels) are strong predisposing factors, but per-sonal and external environmental factors also contributeto DED development and progression. Visual activities,including computer use, are associated with abnormal teardynamics and DED symptoms.103,133,134 This may be theresult of a decreased rate of blinking and increased tearevaporation associated with use of visual displays.103 Ocularaperture was also increased during computer use comparedwith reading a book. This could contribute to increased tearevaporation caused by instability of the tear film across thelarger ocular surface, which is caused by thinning of mucinand lipid layers.103
Low environmental humidity was found to contribute tosymptoms of DED, both in patients and in healthy con-trols.135 Patients with DED had greater tear evaporation ratesthan healthy controls at both 5% and 40% relative humiditybut not at 70%, at which the tear evaporation rate was zeroin both groups.135 The most extreme example of low environ-mental humidity experienced bymost individuals is that of anaircraft cabin during a long flight. A minimum relativehumidity of 20% for aircraft cabins is recommended.136 How-ever, one study found that short-haul flights had minimumrelative humidity levels of 16.3% and that this decreased to6.7% for long-haul flights.136 The authors predicted that thelow relative humidity in an aircraft cabin could be a sourceof irritation and discomfort for passengers.136 High roomtemperature and air velocity as well as indoor air pollution(eg, the presence of lipophilic volatile organic compounds)may also contribute to DED symptoms by causing alterationsin the precorneal tear film.137
Contact lens wearers also experience ocular discomfort,most frequently described as dryness, as well as decreasedcorneal sensitivity.122,138-140 Their condition is often referredto as contact lenserelated dry eye.117 Discomfort associatedwith contact lens wear accounts for up to 50% of all discon-tinuations.141-144 One study found that 43% of soft contactlens wearers experienced DED symptoms, with 28% havingmoderate to severe symptoms.122 However, only about 15%of nonecontact lens wearers experienced symptoms.122
Contact lens discomfort is the subject of a recent TearFilm and Ocular Surface society sponsored monograph.145
Dryness and discomfort associated with contact lenswear can be successfully managed with many therapies,including hypo-osmotic saline drops, multipurpose disin-fecting solution, ocular lubricants, hydroxypropyl celluloseophthalmic inserts, and refitting a patient with siliconehydrogel soft contact lenses.146-151 The relationship be-tween contact lens use and increased tear osmolarity is stillunclear, as some studies have reported increased tear os-molarity while the lens is being worn and immediatelyfollowing removal of the lens138,152-154; other studieshave not found a strong association.139,155-159 Decreasedcorneal sensitivity among contact lens wearers was alsoassociated with tear hyperosmolarity.138 Hypo-osmotic sa-line drops reduced tear osmolarity associated with contactlens dry eye and may contribute to improved ocular com-fort with use.150 A reduction in lipid layer thickness and/orrapid prelens tear film thinning have also been suggestedas contributing factors to contact lens dry eye.117 It isimportant to note that contact lens discomfort may berelated to the lens and lens care system, but it also mightbe an indication of underlying ocular surface disease;thus, the discomfort may or may not be eliminatedfollowing lens removal.
To understand the potential correlation between tear os-molarity and contact lens dry eye, a study was performed toinvestigate differences in tear osmolarity and dry eye symp-toms between wearers of soft and rigid gas-permeable con-tact lenses. Patients with 2 different types of soft contactlenses and 2 different types of rigid contact lenses had signif-icantly higher tear osmolarity after a period of contact lenswear than before wearing the contact lenses. Although thedifferences in tear osmolarity were significant before and af-ter contact lens use, there were no significant differences intear osmolarity associated with use of a particular type ofcontact lens (soft vs rigid).152
2. Systemic DiseaseMany systemic diseases, particularly those of an autoim-
mune or immune-driven nature, are significantly associatedwith development of DED. Sjögren syndrome, an autoim-mune disease that affects the lacrimal and salivary glands,is strongly associated with development of DED.124 In areport from a tertiary referral center practice, it was esti-mated that patients with Sjögren syndrome account forapproximately 11% of all DED.124 Similar to the increasedincidence of DED in women with premature ovarian failure,
THE OCULAR SURFACE / APRIL 2014, VOL
DED symptoms in patients with Sjögren syndrome havebeen associated with decreased estrogens.160
Similar to other populations of patients with DED, asso-ciations between DED clinical signs and symptoms may notbe obvious in patients with Sjögren syndrome.118 A study ofDED in patients with Sjögren syndrome found no signifi-cant correlation between Ocular Surface Disease Index(OSDI) and Schirmer test results.118 However, there was amoderate association between tear osmolarity and a lowSchirmer score, and a negative correlation between tear os-molarity and the OSDI in this patient population.118 A com-parison of tear film osmolarity between patients withSjögren syndrome, patients with rheumatoid arthritis(RA), and healthy controls found that osmolarity was signif-icantly higher in patients with Sjögren syndrome but not pa-tients with RA compared with healthy controls. (SchargusM, et al. IOVS 2010;51(5):ARVO E-abstract 6250) Addition-ally, patients with Sjögren syndrome, as predicted, hadsignificantly higher dry eye severity scores (derived from aseries of tests, including Schirmer tests, the OSDI, fluores-cein, and Lissamine green staining) compared with patientswith RA and healthy controls. (Schargus M, et al. IOVS2010;51(5):ARVO E-abstract 6250).
Patients with Graves disease, an autoimmune syndromecharacterized by a hyperactive thyroid, have symptoms ofocular discomfort similar to those in patients withDED.119,120 In approximately half of patients with Gravesdisease, ocular discomfort was associated with tear hyperos-molarity.119 Increased palpebral fissure width in patientswith Graves disease was found to be a significant predictorof tear hyperosmolarity.119 Tear breakup time was alsofound to be significantly shorter in patients with Graves dis-ease.120 Although there was a significantly higher tear osmo-larity in patients with Graves disease compared with healthycontrols, there was no statistically significant correlation be-tween the mean palpebral fissure width and increased tearosmolarity.120 Additionally, there was no significant correla-tion between tear osmolarity and degree of proptosis.120
Therefore, the increased degree of proptosis and increasedinterpalpebral fissure width may contribute to Graves disea-seeassociated DED through evaporative tear loss, but thereis no significant association between the degree of thesemeasures and the degree of tear osmolarity as measuredbilaterally by the freezing-point depression technique.120
DED is the most common complication of chronic graft-versus-host disease (cGVHD) resulting from hematopoieticstem cell transplant, with up to 62.5% of patients developingocular pathology.123 The rate of tear turnover in patientswith cGVHD-associated DED was similar to that in patientswith Sjögren syndrome and significantly lower than in pa-tients with MGD.123 This may reflect the greater severityof DED in patients with Sjögren syndrome and cGVHDrather than intrinsic, disease-specific differences in tearturnover rate.123 However, tear evaporation and meibomiangland dropout rates in patients with cGVHD with DEDwere similar to those in patients with MGD and significantlyhigher than in patients with Sjögren syndrome.123
Figure 5. Comparison of the contributing factors to tear film hyperosmolarity in ADDE and EDE. From Bron AJ, Yokoi N, Gafney E, Tiffany JM.Predicted phenotypes of dry eye: Proposed consequences of its natural history. Ocul Surf. 2009;7(2):78-92.
RETHINKING DRY EYE DISEASE / Bron, et al
Additionally, approximately 70% of the patients withcGVHD or MGD had an unstable lipid layer comparedwith approximately 50% of the patients with Sjögren syn-drome.123 Tear osmolarity and tear volume on the ocularsurface were not found to be different between any of thegroups tested.123 However, a separate study found a signifi-cant correlation between tear film osmolarity and GVHDscore. (Schargus M, et al. IOVS 2011;52(5): ARVO E-ab-stract 3793).
These results suggest that in patients with cGVHD-associated DED, all aspects of tear physiology are affected,unlike in MGD or Sjögren syndrome, where individual as-pects of tear physiology take prominence.123
Pseudoexfoliation syndrome is an ocular disorder char-acterized by deposits of pseudoexfoliative material at the pu-pillary border and in a characteristic target-like pattern onthe anterior lens capsule. This material is also found inthe lacrimal glands and conjunctiva.161 The pathology ofthis disease leads to development of glaucoma in approxi-mately 50% of patients.161 Compared with healthy individ-uals, patients with pseudoexfoliation syndrome haveincreased tear osmolarity.161 This raises the possibility thatpseudoexfoliation syndrome is a risk factor for DED.161
B. Natural HistoryTear film hyperosmolarity is considered to be a central
factor in development of ocular surface damage inDED.1,130 Tear film hyperosmolarity results from differentcontributing factors, depending on the subtype of DED(Figure 5).1,130 In ADDE, a reduction in lacrimal secretiondominates the clinical picture, leading to tear hyperosmolar-ity. Different authors have reported either an increased162-164
or a decreased evaporation rate.85,165,166 Decreased lacrimal
S10 THE OCULAR SURFACE / APRIL 2014, V
flow can come about in several different ways.1,130 Theseinclude Sjögren syndrome, non-Sjögren dry eye, lacrimalgland obstruction, reflex hyposecretion, and use of systemicdrugs, such as antihistamines, anticholinergics, and some an-tidepressants.5,130 Most patients with DED do not have pureADDE because by itself, it accounts for only about 10% ofDED cases.108
In EDE, tear evaporation is increased compared to thatin healthy individuals when measured under standard con-ditions of humidity, airflow, and temperature.1,130,167,168
Evaporation may be increased by intrinsic or extrinsic fac-tors.1,130 Intrinsic, lid-related factors may include insuffi-cient meibomian oil delivery or quality, as in MGD, lowblink rates or blink speed, a widened palpebral aperture,or disorders of lid congruity.1,130 Extrinsic factors, affectingocular surface wettability and tear stability, include contactlens wear, topical use of anesthetics and preservatives,allergic conjunctivitis,1 and vitamin A deficiency.169
Although a number of factors can contribute to the develop-ment of EDE, MGD is likely to be the most commoncause.167
Mixed, or hybrid, types of DED also occur (Figure 5).108
These involve a combination of both ADDE and EDE.108 Inboth Sjögren syndrome and cGVHD, DED results from acombination of inflammatory lacrimal gland damage andMGD.1,123,124 Additonally, in cGVHD, there may belacrimal gland duct obstruction.123 It has also been sug-gested that damage to sensory corneal nerve terminals inadvanced MGD-related EDE, by removing the sensory driveto a compensatory lacrimal secretion, could lead to a func-tional lacrimal deficiency. This would give rise to an organicEDE combined with a functional ADDE.130 Similarly, it hasbeen proposed that in advanced aqueous deficiency, the
barrier properties of the tear film lipid layer may be greatlyimpaired by defective spreading, resulting in a functionalEDE accompanying an organic ADDE.
Tear hyperosmolarity resulting from decreased lacrimalflow or increased evaporation in combination with accom-panying tear film instability contributes to ocular surfacedamage through a cascade of downstream events. Hyperos-molarity stimulates signal transduction cascades in conjunc-tival epithelial cells, resulting in the production ofproinflammatory cytokines, cell death via apoptosis, andloss of goblet cells.170-174 Signaling cascades, including theJNK and ERK, MAPK, and NF-kB pathways, are activatedby hyperosmolar conditions and result in the productionof proinflammatory cytokines in tears.170-173 Apoptotic celldeath of corneal epithelial cells resulting from tear hyperos-molarity is mediated by these signaling pathways.174 Tearhyperosmolarity also contributes to a loss of goblet cells,resulting in decreased mucin production and subsequentinstability of the tear film.175
Inflammation caused by tear film hyperosmolarity mayfurther contribute to DED pathology. It has been proposedthat proinflammatory mediators and other sources of desic-cating corneal stress stimulate compensatory reflex lacrimaltear secretion via the corneal trigeminal nerves.130 Withmore advanced disease, it is conceivable that excessive reflexcompensation could then result in neurogenic inflammationwithin the lacrimal gland, leading to a loss of secretoryresponsiveness176 and, additionally, a subsequent loss ofcorneal sensitivity could further decrease lacrimal flow andthereby accentuate disease severity.130,176 Thus, the naturalhistory of DED appears to represent a self-perpetuatingcycle in which lacrimal insufficiency contributes to ocularsurface inflammation and vice versa, with the net resultbeing ocular surface damage.1,130
C. Relationship of Natural History to Epidemiologyand to the Relevance of Diagnostic Tests andTreatments across the Spectrum of DiseaseDue to the differing etiologies of the DED subtypes,
various diagnostic tests may be more or less effective at diag-nosing disease in ADDE or EDE subtypes. Additionally, treat-ments for DEDmay be effective to varying degrees dependingon the subtype of disease. A comparison of Schirmer tests inpatients with ADDE vs patients with EDE revealed that withor without the use of anesthesia, ADDE patients had signifi-cantly more positive test results and EDE patients had signif-icantly fewer positive results than healthy controls, suggestingthat Schirmer tests may be of greater diagnostic value forADDE patients vs EDE patients.177 This may be explainedby the fact that decreased tear production is not the primaryissue in EDE as it is in ADDE and, because of compensatorymechanisms, there may be increased tear production in pa-tients with EDE.1,130,177,178
Although Schirmer values were found to be significantlydifferent between patients with ADDE and those with EDE(including MGD), other parameters of ocular surface diseasewere not found to be different.179 In a study of 40 patients with
THE OCULAR SURFACE / APRIL 2014, VOL.
DED (25MGD, 15 ADDE), TBUTwas not found to be signif-icantly different between the 2 groups.179 Self-reported ocularsymptoms and lid-margin abnormalities were also notdifferent between MGD and ADDE patients.179 However,there were significant differences noted in diagnostic param-eters relating to meibomian gland function between ADDEand MGD patients.179 Meibomian gland changes and mei-bum scores (volume and quality ofmeibum) were both signif-icantly better for ADDE patients compared with MGDpatients.179 Measures of punctate epithelial keratopathywere significantly higher for ADDE patients vs MGD pa-tients.179 Although it had been proposed that the combinationof lid-margin abnormality, ocular symptoms, and morpho-logical changes ofmeibomian glands could be used for the dif-ferential diagnosis of ADDE vs MGD, these tests proved to beonly 68% sensitive and 80% specific for differentiating the twoDED subtypes.179,180
It has been reported that Sjögren syndrome DED can bedifferentiated from non-Sjögren syndrome ADDE by use ofrose bengal conjunctival staining in addition to symptoms ofdryness of the mouth, as patients with Sjögren syndrometend to have increased temporal staining compared with pa-tients without Sjögren syndrome.181 This may be represen-tative of the increased severity of DED in patients withSjögren syndrome compared with other patients withADDE rather than being a marker for Sjögren syndrome.181
The differences in the value of various diagnostic pa-rameters for diagnosing different subtypes of DED as wellas the tendency of these parameters to shift with timenecessitate a reliable diagnostic measure that is sensitivefor the detection of all subtypes of DED, remains rela-tively consistent over time, and changes with responseto treatment. Numerous factors, intrinsic and external,contribute to the development of DED,117,119,125,126,182,183
and only a diagnostic measure that addresses all of theseparameters may be of clinical utility.Summary: DED Definition, Classification, and NaturalHistory
d DED is a condition affecting 20% or more of the pop-ulation in North America, Europe, and Asia. Mostcases fall into the mild to moderate category andshow a female predominance. Sex differences inDED prevalence decrease with advancing years.
d The 2 major subtypes are ADDE resulting from adecrease in lacrimal gland secretion and EDE, in whichthere is excessive evaporative water loss. Current evi-dence suggests that the most common form of EDEis MGD, and this is the most common form of DED,affecting a large percentage of the aging population.
d Numerous factors, intrinsic and external, contribute tothe development of DED through disruption inlacrimal production and/or the stability and barrierfunction of the tear film lipid layer.
d Current evidence suggests that EDE is more commonthan ADDE, and MGD is the most common cause ofEDE, which creates nomenclature and diagnosticchallenges.
d Regardless of the etiology or subtype of DED, tear filmhyperosmolarity induces ocular surface damage andinflammation, which represents a final commonpathway of DED pathogenesis.
d Because of its role in DED pathogenesis, osmolarityserves as a global marker for disease and is bettersuited to capture the totality of disease than testsmore specific to one subtype.
IV. INFLAMMATION IN THE PATHOGENESIS OF DEDAND PATHWAYS TO THE VICIOUS CYCLE
A. Inflammation in DEDDED is a chronic condition with a multifactorial etiol-
ogy.1 In addition to the contributing factors of aqueoustear deficiency and excessive evaporation, inflammation isa principal factor contributing to disease pathology andassociated tissue damage.1 Since the DEWS report in 2007,much has been learned about the role of the inflammatoryprocess and its contribution to DED pathogenesis.6 This sec-tion will detail the major inflammatory processes involved inDED, hallmarks of DED-induced ocular inflammation, andanti-inflammatory therapies, with a focus on recent under-standing since the DEWS report.
B. Inflammatory Events Involved in the Initiation ofDED
1. T Cells and NK Cells as Drivers of InflammationTear hyperosmolarity is a major contributor to symp-
toms of irritation in DED, acting both directly and by initi-ating inflammatory events in the surface epithelial cells.6
This inflammatory process is amplified through the actionsof activated T cells recruited to the conjunctiva.184-186 Sternet al found that most inflammatory infiltrates in conjunc-tival biopsies from patients with DED with and withoutSjögren syndrome were CD4 positive (CD4þ) T cells.185
Many of the cells in conjunctival tissue from patients withDED expressed human leukocyte antigen DR (HLA-DR),a marker of immune activation and antigen presentation.185
In addition to lymphocytes, the epithelial cells of the con-junctiva also expressed HLA-DR.185,187 It was suggestedthat CD4þ T cells contribute to ocular inflammationthrough production of proinflammatory cytokines,185 andresearch using mouse models of DED have demonstratedthat interferon (IFN)-g secretion by infiltrating CD4þ Tcells in the conjunctiva is a major source of ocularinflammation.184,188,189
Desiccating stress, as used in experimental DED,resulted in a decrease in CD8þ T cells and an increase inCD4þ T cells, with associated loss of goblet cells.184,188,189
IFN-g was detected in the goblet celledense areas of theconjunctiva after induction of experimental DED in micebut not in control mice.184,188 Further supporting the roleof IFN-g in DED, there was no loss of goblet cells in theconjunctiva of IFN-g knockout mice after induction ofexperimental DED, and treatment of healthy mice with sub-conjunctival IFN-g resulted in goblet-cell destruction.184
Migration of CD4þ cells into goblet-cell zones resulted in
S12 THE OCULAR SURFACE / APRIL 2014, V
increased concentrations of IFN-g in tears,188 and theseIFN-geproducing CD4þ inflammatory infiltrates wereassociated with increased apoptosis in the corneal epithe-lium.188,190 In the draining lymph nodes of mice withDED, T cells expressed more markers of activation (CD69and CD154) and shifted to more of an IFN-geproducing(Th1) subtype with an associated decrease in interleukin(IL)-4eproducing (Th2) subtypes.186 These IFN-geproducing T cells were CD4þ and highly proliferative.186
Although T cells are thought to be a primary driver ofinflammation in DED,184,185 animal models have shownthat early in the disease process, there is an infiltration ofNK cells into the conjunctiva.191 This is also associatedwith an increase in IFN-gesecreting NK cells in the drain-ing lymph nodes.191 Depletion of NK cells or IFN-g inmouse models inhibits maturation of lymph node antigen-presenting cells (APCs) and amelioration of disease.191
These results suggest a critical role for IFN-g productionby T and NK cells in DED-related pathology.
2. Deregulation of Regulatory T cells (Tregs) and IL-17eProducing T cellsIn addition to increased proinflammatory activities of
conjunctival T and NK cells,184-186,191 immunosuppressiveTregs were found to have decreased functionality inDED.192 There were no differences found in the frequenciesof CD4þ, CD25þ, Foxp3þ Tregs as a proportion of CD4þcells or total lymph node cells, but Tregs from mice withDED were significantly impaired in their abilities to suppressthe proliferation of primed and naive T cells.192 Conversely,increased frequencies of IL-17eproducing T cells were foundin DED mice vs healthy controls.192,193 Proliferating, primedT cells isolated from DED mice were mostly IL-17eproducing CD4þ T cells.192 Desiccating stress inducedsignificantly higher transcript levels of IL-17 and associatedcytokines and receptors, such as IL-23, IL-23R, IL-6, IL-22,and transforming growth factor (TGF)-b.194 Lymph nodesisolated from DED mice have significantly (4- to 5-fold)increased frequencies of IL-17eproducing CD4þ T cells vshealthy mice.192 This is of significance in relation to tissuedamage, as the ocular surface is more vulnerable to IL-17ein-duced inflammatory damage because of the constitutiveexpression of IL-17 receptors by the corneal and conjunctivalepithelium.192 Additionally, increased IL-17 mRNA is pro-duced in the conjunctiva of DED mice vs healthy controls.192
AntieIL-17 therapy in DED mice results in recovery ofTregs’ suppressive capacities and reduced frequencies ofIL-17eproducing CD4þ cells.
3. APCs and Toll-Like Receptors (TLRs)Although T cells are the primary drivers of inflamma-
tion associated with DED, they require activation by profes-sional APCs, such as dendritic cells (DCs).195 In a mousemodel of DED, depletion of APCs from the conjunctiva pre-vented CD4þ T-cellemediated inflammation.195 Anotherroute to T-cell activation involves the migration of APCsto draining lymph nodes.192 An association was found
between increased percentages of DCs in the draining cervi-cal lymph nodes and acute cytokine production in responseto DED.195 These DCs were found to be mostly activated asdetermined by an increased expression of major histocom-patibility complex (MHC) II and other surface markers ofactivation,195 and DC accumulation in the conjunctivalepithelium preceded CD4þ T-cell arrival and activation(CD69 expression).195 Depletion of DCs via clodronatetreatment resulted in decreased CD4þ T-cell infiltration ofthe conjunctiva as well as an absence of ocular surface tissuedamage.195 CD4þ T cells infiltrating the conjunctiva need toencounter APCs to become completely activated andgenerate pathogenic proinflammatory responses.195 Thesedata suggest a process by which desiccating stress activatesconjunctival epithelial DCs, prompting them to activate Tcells and manifesting in tissue-damaging inflammatory re-sponses.188,192-196
DCs and many other cell types express TLRs, a family ofhighly conserved pattern-recognition receptors that recog-nize a variety of pathogen-associated microbial motifs.197
The result of TLR expression and stimulation is an intracel-lular signaling cascade that results in downstream cytokineproduction.197 In a mouse model of Sjögren syndrome,increased expression of TLR-4 and TLR-5 was found inthe cornea and lacrimal glands. (Christopherson PL, et al.IOVS 2005;46 (5): ARVO E-abstract 4462) It is unknownwhether the increased TLR expression associated withocular inflammation is a cause or effect.197 Also, the polarityof TLR expression remains to be determined, and this couldcontribute to ocular inflammation, particularly if TLRexpression is localized mainly to the epithelial surface, facil-itating the interaction of TLRs with commensalmicroflora.197
B. Markers of Inflammation in DED1. Matrix Metalloproteinase-9 (MMP-9)
Expression of MMP-9, a protein with important roles inwound healing and inflammation, was found to be signifi-cantly elevated in tears from patients with DED.198 MMP-9 expression is increased in response to desiccating stresson the corneal epithelium, and levels of MMP-9 increasewith increasing severity of disease.189,198 Additionally, inone study, significantly higher levels of MMP9, IL6, IL1b,TNFa, and TGFB1 mRNA were found in conjunctivalepithelial cells isolated from patients with DED vs healthycontrols.198 However, levels of another metalloproteinase,MMP3, were the same in patients with DED and healthycontrols.198 MMP-9 levels in tears from patients withDED were correlated with clinical parameters, includingsymptom scores and decreased low-contrast visual acuity,and negatively correlated with tear film breakup time.198
MMP-9 expression is elevated under conditionsmimicking DED.199 Using an in vitro model of DED (exper-imentally induced dry eye in vitro, EDEV), MMP-9 expres-sion, along with TNFa, MUC4, and DEFB2, was found to beelevated under conditions of low humidity and high temper-ature.199 After 24 hours of EDEV induction, MMP-9
THE OCULAR SURFACE / APRIL 2014, VOL.
expression was observed in nearly all layers of the epithe-lium.199 MMP-9 expression was reduced by treatmentwith tear substitutes, including those containing dexameth-asone, sodium hyaluronate, carboxymethylcellulose sodium(CMC), and TS-polysaccharide, and this was associatedwith a restoration of corneal epithelial morphology.199
MMP-9 expression has been proposed as a potential toolfor DED diagnosis, but it is unlikely to be specific to DEDand may be more representative of nonspecific ocular tissuedamage or remodeling.198,200 An association exists betweenincreased MMP-9 expression and DED as mice that areknock-out models for MMP-9 expression are more resistantto experimentally induced DED.201 Although increasedMMP-9 expression was found in vitro to be associatedwith increased hyperosmolarity, elevated MMP-9 expressionin tears was also found in patients with acanthamoeba kera-titis, herpetic keratitis, ocular rosacea, and keratoco-nus.170,200,202 Human corneal epithelial cells cultured inmedium containing sodium chloride for 24 hours expressedincreasing amounts of MMP-9 in a sodium chloride concen-trationedependent fashion, implicating tear hyperosmolar-ity as a cause of MMP-9 expression in DED.170 MMP-9elevation has also been detected in response to cornealwounding.203
These data suggest that MMP-9 expression in tears ofpatients with DED may be more representative of tissueremodeling in the corneal epithelium in response to damage,rather than a specific diagnostic marker for DED.203 Addi-tionally, since there is a concomitant reduction in diseaseseverity in mice lacking MMP-9, targeting MMP-9 expres-sion on the ocular surface may lessen the severity ofDED.170,201
2. HLA-DRExpression of the class II MHC protein HLA-DR is also
increased on the ocular surface in response to inflammationand tear hyperosmolarity.187,204-207 HLA-DR plays a criticalrole in initiation of immune responses, including those inthe ocular epithelium, via antigen presentation to T cells.204
In studies of conjunctival epithelial cells from patients withDED, there was no significant correlation between clinicalseverity and HLA-DR expression, which could be attributedto the relatively small numbers of patients with severe DEDor Sjögren syndrome.187,204,206,207 Although activated DCsupregulate their expression of HLA-DR, impression samplesfrom patients with DED revealed that most HLA-DRþ cellswere not DCs or other professional APCs and instead dis-played an epithelial phenotype.204 These results suggestthat, in addition to infiltration of DCs into the conjunctivalepithelium,195 antigen presentation to CD4þ T cells mayoccur via epithelial cells in the conjunctiva in response todesiccating stress.
3. Cytokines and ChemokinesExpression of proinflammatory cytokines and chemo-
kines is a hallmark of ocular inflammation in DED.208-210
Recently, multiplex assays have facilitated the detection of
cytokines and chemokines in tears from patients with DED,and there have been attempts to correlate their expressionwith clinical measures.208-210 One study of 30 healthyvolunteers found a significant positive correlation betweenproinflammatory cytokine levels (IL-1a, -1b, -6, and -8and TNF-a)/MMPs (MMPs 1, 2, 7, 9, and 10) and increasedtear osmolarity.210 However, no association was found be-tween any cytokine or MMP and TBUT or OSDI.210 A sepa-rate investigation comparing TNF-a and IL-6 in tears frompatients with Sjögren syndrome and non-SjögrenerelatedDED found that whereas TNF-a was elevated comparedwith controls, there was no significant difference in produc-tion between Sjögren- and non-Sjögrenerelated DED.211
Levels of IL-6 were significantly elevated in the tears of pa-tients with Sjögren-related DED compared with non-Sjög-ren subtypes and correlated with TBUT, goblet celldensity, tear clearance, and Schirmer test and corneal fluo-rescein staining scores.211
Tears collected from 23 EDE patients with mild to mod-erate disease were found to have elevated levels of IL-8, IL-1Ra, fractalkine/CX3CL1, IP10, vascular endothelial growthfactor, and epidermal growth factor compared with healthycontrols.208 These cytokines and chemokines were found tocorrelate with pain (using a symptom of discomfort ques-tionnaire) and clinical measures of disease, including tearstability, tear production, and ocular surface integrity.208 Acomparison of cytokine production in the tears of patientswith DED revealed differences between the various subtypes(ADDE, EDE, mixed).212 Using antibody microarrays,Boehm et al determined that patients with ADDE and mixedDED subtypes had higher levels of the proinflammatory cy-tokines TNF-a, IL-6, and IL-1b than those with EDE.212
These relative differences could be attributed to between-group differences in disease severity, and it is possible thatonly qualitative differences will be discriminatory. A studyof the cytokine/chemokine profile in the tears of 133 pa-tients with DED found significantly increased levels of cyto-kines and chemokines, including IL-1b, IL-6, monocytechemotactic protein-1, IL-16, IL-33, TGF-a, granulocytecolony-stimulating factor, CX3CL1, and CXCL5 comparedwith healthy controls.213 A significant decrease in IL-4, IL-12p40, IL-17A, and IFN-g was also seen in patients withDED compared with healthy controls.213
The diversity in cytokine profiles detected in samplesfrom patients with DED does not facilitate a simple correla-tion between any 1 cytokine or group of cytokines and thepresence or severity of disease.208-210,213 The cytokineresponse may vary as a result of autoimmune diseasecontributing to ocular pathology (as in the case of Sjögrensyndrome) or between ADDE and EDE.211,212 The methodsof tear collection and measurement of the cytokine responsecould also impact the results. Some studies have employedLuminex-based analysis of tear samples and others haveused antibody microarrays or mRNA analysis by real-timepolymerase chain reaction.198,209,210 Tears have beencollected using Schirmer strips or capillary tubes, and thisvariability in tear collection methodology could potentially
S14 THE OCULAR SURFACE / APRIL 2014, V
yield different cytokine responses resulting from responsesinduced via the collection method itself.198,208-210
D. Anti-inflammatory Therapies for DED1. Cyclosporine A
Because of the inflammation-driven nature of DED pa-thology,1 anti-inflammatory therapies have been exploredfor the treatment of DED. Cyclosporine A 0.05% (CsA;Restasis�, Allergan, Irvine, CA) was approved for the topicaltreatment of DED by the US Food and Drug Administration(FDA) in 2002.214,215 Two randomized clinical trials estab-lished the safety and efficacy of CsA for the treatment ofmoderate to severe DED.216,217 Expression of HLA-DRwas significantly reduced after 3 or 6 months of topicalcyclosporine 0.05% or 6 months of topical cyclosporine0.1%.205 In response to the clinical observation that somepatients do not adequately respond to twice-daily CsA treat-ment, a recent trial focused on the utility of using high-frequency (3-4 times daily) CsA for treatment of chronicDED.218 After a 2-month course of high-frequency topicalCsA, 68.2% of patients who were previously unresponsiveto less aggressive CsA therapy, including patients withocular GVHD and Sjögren syndrome, exhibited improve-ments in disease symptoms.218 No additional adverse eventswere recorded beyond those seen with reduced-frequencydosing.218 This study demonstrated that patients with se-vere, chronic DED with multiple etiologies who are unre-sponsive to initial CsA treatment could benefit fromincreased frequency of topical application.218
2. Corticosteroids and DoxycyclineCorticosteroids and doxycycline have also been used to
reduce DED-associated ocular inflammation and alleviatesymptoms. Topical corticosteroid therapy has been utilizedfor the successful treatment of Sjögren syndromeerelatedDED.219,220 Daily treatment with methylprednisoloneimproved both subjective and objective DED factors andwas safe for long-term treatment.219 However, a previousstudy suggested that long-term treatment with methylpred-nisolone was associated with an increase in the developmentof cataracts.220 In vitro and in vivo studies with doxycyclinehave demonstrated that topical treatment is effective atreducing MMP-9 and inflammatory cytokine expres-sion.170,221 This reduction in MMP-9 expression was attrib-uted to prevention of its induction in response tohyperosmolarity.170 Doxycycline may exert its anti-inflammatory effects via scavenging of reactive oxygen spe-cies and inhibition of phospholipase A2.222
3. Resolution of Inflammation with ResolvinsResolvins are molecules derived from omega-3 fatty
acids that have a potent ability to mitigate inflammatory re-sponses.223,224 Studies of resolvins in mouse models of DEDhave revealed that they can decrease inflammation andimprove tear production.223,224 Treatment of DED micewith topical resolvin E1 restored epithelial cell density andresulted in an increased tear production of 60% compared
with vehicle-treated control mice.224 There were significantreductions in the numbers of infiltrating CD11bþ APCsand CD4þ T cells after resolvin E1 treatment.224 Addition-ally, treatment with either resolvin E1 or D1 resulted indecreased goblet-cell secretion, which is associated withdamaging inflammation in DED.223 These results suggestthat resolvins may be effective at improving disease-relatedinflammation and increasing tear production but may bemore effective for ADDE than EDE.
Because of the anti-inflammatory and disease-ameliorating properties observed in models of DED, oraltreatment with resolvin precursors (omega-3 and omega-6fatty acids) has been explored.225,226 In a randomized con-trol trial, patients receiving thrice-daily oral omega-3 andomega-6 fatty acids displayed a significant reduction inHLA-DR expression on conjunctival epithelial cells.225 Sinceexpression of HLA-DR is associated with ocular surfaceinflammation and tear hyperosmolarity, oral treatmentwith omega-3 and omega-6 fatty acids may facilitate resolu-tion of some inflammatory DED symptoms.225,226 A recentstudy found that omega-3 and omega-6 fatty acid supple-mentation significantly reduced HLA DR in patients withDED compared with placebo.225 Consistent with these re-sults, dietary supplementation with omega-3 fatty acids for90 days resulted in amelioration of symptoms in 70% of pa-tients.227 There was no associated change in meibum lipidcomposition, but Schirmer testing and fluorophotometrysuggested increased tear production with treatment.227 Sincelack of tear production is a primary issue of ADDE, andmeibum lipids were unaffected by oral omega-3 fatty acidtreatment, this treatment would be expected to benefitADDE patients more than EDE patients.1,227
4. Novel AgentsSeveral preclinical studies have highlighted the potential
utility of novel agents for DED. Animal studies have demon-strated the efficacy of topical trehalose, a naturally occurringnonreducing disaccharide of glucose, for treatment ofDED.228 DED mice treated with topical trehalose dropsfor 3 weeks showed reduction in inflammatory responsesas measured by IL-1b, IL-2, IL-6, and IL-17 mRNA produc-tion in the conjunctiva.228 Additionally, conjunctival expres-sion of MMP-9 and TNF-a was reduced with trehalosetreatment.228
In a mouse model of Sjögren syndrome, topical treat-ment with an anti-CD4 monoclonal antibody was effectiveat reducing infiltration of mononuclear cells into thelacrimal glands.229 Reduction in autoimmune pathology inthe lacrimal glands was associated with an inhibition ofCD4 T-cell activation rather than removal of T cells.229
However, this was not associated with increased tear volumein anti-CD4etreated mice.229 These data suggest the poten-tial utility of trehalose drops and anti-CD4 monoclonal an-tibodies for the treatment of DED.Summary: Inflammation in Pathogenesis of DED
d The pathophysiology of DED is a complex processmediated by both primary and secondary
THE OCULAR SURFACE / APRIL 2014, VOL.
inflammation of the lacrimal tissue and the ocular sur-face, the severity of which may depend on the precip-itating etiology.
d Inflammation leads to a vicious cycle that reinforces it-self with each repetition, leading to an increasingseverity of clinical expression of disease.
d Increased osmolarity of the tear film stimulates inflam-mation of the ocular surface through alteration ofepithelial immune receptors and APCs, with subse-quent recruitment of immunocompetent lymphocytes.
d Systemic immune disorders, such as Sjögren syn-drome, stimulate potent inflammatory cells and cyto-kines that provoke more extensive inflammation thatresults in greater clinical expression of DED.
d Inflammatory mediators that can contribute to the tis-sue damage, including MMPs, are produced and maybe measured as a marker of inflammation. Althoughthere is expression of many proinflammatory proteinsin the tears, there is no singular chemical entity orcombination of them that has been shown to indicatedisease severity over the entire range of disease. HLA-DR expression in conjunctival epithelial cells has beenshown to correlate with increased tear osmolarity inpatients with systemic disease.
d Anti-inflammatory therapies (corticosteroids, cyclo-sporine, resolvins) can modulate the degree of inflam-mation in DED and ameliorate the damaging effects ofthe inflammation on the ocular surface.
V. ASSESSMENT ISSUES: SYMPTOMS AND SIGNSThere are many tests available for the assessment of
DED. Some of the most commonly used tests include themeasurement of tear osmolarity, Schirmer testing, meibo-mian grading, TBUT, corneal and conjunctival staining,and self-report measures of symptoms, including OSDI,the Standard Patient Evaluation of Eye Dryness (SPEED),the Dry Eye Questionnaire (DEQ), and the Contact LensDry Eye Questionnaire (CLDEQ). However, many of theavailable tests are poorly reproducible and less reliable inthe diagnosis of mild to moderate disease. This sectionwill evaluate the literature on assessing signs and symptomsin DED.
Assessments serve several important purposes, one ofwhich is to allow for accurate diagnosis, which requiresidentification of appropriate referent values. A variety ofdifferent diagnostic thresholds have been proposed for tearosmolarity, with studies suggesting a range from 305 to316 mOsm/L as a diagnostic threshold.65,66,230,231 The differ-ence in diagnostic threshold values observed in these studiesmay reflect use of populations with differing severities. Ahigher number (316 mOsm/L) was suggested as a referentbased on a meta-analysis of studies, most of which includedpatients with more severe disease.65 In a more recent studywith a larger number of patients with mild disease, 305mOsm/L was suggested.230 In a study more representativeof the entire spectrum of disease, 308 mOsm/L was sug-gested as the most sensitive cutoff value.66
Figure 6. Changes in commonly used tests for DED over time following treatment, which began at visit 3. Mean osmolarity values (C) and highervalue of tear osmolarity (B). From Sullivan BD, Crews LA, Sonmez B, et al. Clinical utility of objective tests for dry eye disease: Variability over time andimplications for clinical trials and disease management. Cornea 2012;31(9):1000-1008.
RETHINKING DRY EYE DISEASE / Bron, et al
Use of clinical tests is an important strategy for guidingdisease management.4,6,232 With regard to monitoring effectsof treatment, it is important that signs be able to predictchanges in symptoms. In one study, a variety of tests,including bilateral tear osmolarity, Schirmer test, TBUT,staining, meibomian grading, and self-report (measured us-ing the OSDI) were measured over 3 months; patients thenreceived topical cyclosporine and were evaluated for 3 moremonths.233 Of all tests performed, tear osmolarity was consid-erably less variable vs corneal and conjunctival staining andmeibomian grading over a 3-month period. Following treat-ment, there was a significant decrease in osmolarity and a sig-nificant decrease in inter-eye variability (indicatingimprovement in tear film stability), which was followed by atrend toward improved symptoms. In contrast, other signsmeasured did not demonstrate a change following treatment(Figure 6). Similar results have been shown after treatmentwith hyaluronic acid and punctal plugs.234,235
The availability of a validated diagnostic test wouldgreatly benefit the clinical trial process, both from a stand-point of determining appropriate patients for inclusionand from the perspective of being able to detect responseto treatment.233 Many existing tests suffer from a numberof biases, including spectrum bias, with many tests perform-ing better in patients with moderate to severe disease, whichcould in part be the result of floor and ceiling effects of thetests. Treatment effects may be demonstrated by tests spe-cific to samples of patients with subtypes of DED, forexample tear evaporation measurements may be appropriateto evaluate treatment of EDE with emulsion eye drops.236 In
S16 THE OCULAR SURFACE / APRIL 2014, V
contrast, osmolarity appears to be the best marker across alllevels of disease severity as well as in different subtypes ofDED.237
A. Challenges Associated with Assessment in DEDAlthough there are many tests available, results of these
tests often conflict with one another. Modest and inconsis-tent correlations are often observed. In an analysis of datafrom 2 separate studies, seven commonly used signs andsymptoms were evaluated (tear osmolarity, TBUT, Schirmertest, corneal and conjunctival staining, meibomian grading,and OSDI) using independent components analysis.238 Allcorrelations were modest, with none over r2 ¼ 0.17. In addi-tion, symptoms alone are insufficient to assess DED. Arecent review found that correlations of DED symptomsand objective tests ranged from r2 ¼ 0.07 to 0.84, with themajority of correlations falling within the range of 0.2 to0.3.239 A recently presented poster at the 2013 AnnualMeeting of the Association for Research in Vision andOphthalmology reported on an analysis of a small numberof patients (n ¼ 11) that found that conjunctival staining,tear turnover rate, and tear volume all correlated signifi-cantly with symptoms, although no information on diseaseseverity was available. (Saigal S, et al. IOVS. 2013;54:ARVO E-abstract 4361).
Moreover, many patients with DED report no symptomswhen presenting to clinics for eye care. In one study, only57% of patients with DED reported symptoms consistentwith DED.238 Thus, objective measures retain significantimportance in managing DED.
Figure 7. Variability of commonly used tests for DED across the spectrum of disease. Severity is calculated as a composite index based on an in-dependent component analysis. Sullivan BD, Whitmer D, Nichols KK, et al. An objective approach to dry eye disease severity. Invest Ophthalmol Vis Sci2010;51(12):6125e30.
RETHINKING DRY EYE DISEASE / Bron, et al
Test results are highly variable, in part because of mea-surement error and in part because of the disease itself(Figure 7).237,240 Sensitivity and specificity of single testsare often low, although combinations of tests may perform
THE OCULAR SURFACE / APRIL 2014, VOL.
better.241 However, it has been noted that as a single test, os-molarity is the best predictor of DED.
DED is a multifactorial disease for which no standardizedapproach to assessing disease severity exists, although a
number of strategies have been examined and advocated. Onestrategy is to average a number of tests, with all tests beingweighted equally. This approach was used by Sullivan et al,who used independent components analysis (ICA) to createa more objective composite index of disease severity. ICA co-efficients were used to determine the composite; these deter-minations resulted in an equal weighting, indicating that eachof the measurements were uncorrelated and thus indepen-dent. The results of this analysis showed that osmolaritywas most closely correlated with disease severity and alsoshowed the lowest variability. In contrast to the ICA method,using combinations of clinical indices resulted in 63% of pa-tients being poorly classified.237 Moreover, the ICA methodremoves the criticism that one test may be unduly influencingthe composite. However, one criticism of this approach is thatit assumes that all signs contribute equally to disease severity.Alternatively, a weighted approach using a linear or geometricprogression can be used, whereby higher weights are given tomeasures that may be better indicators of severity, includingTBUT, MGD grading, and corneal staining. No clinical evi-dence exists, however, to suggest the best strategy for weight-ing the respective tests.
In a recent paper, authors reported on a study in whichthey compared an independent component analysis (ICA)approach to dry eye patients with that of latent class analysis(LCA).242 This latter test is a theoretical mathematicalmodel to discern characteristics of an ideal test and thentest characteristics of other tests and how closely theyapproximate this ideal. They reported that the ICAapproach demonstrated the superiority of tear osmolarity,meibomian gland sign and symptoms, while the LCA resultsfavored the Schirmer test and breakup time. The LCAapproach depends heavily on assumptions concerning clin-ical variables, including the assumption that tests are nor-mally distributed (which is not true for most of theobjective tests for DED), and that patients could be readilysegmented into healthy and diseased states (which is alsonot true for dry eye, where a continuum of severity is farmore appropriate). LCA also required square-root transfor-mation of Schirmer’s and TBUT prior to processing. Thisheavy reliance on clinically spurious assumptions and datamanipulation prior to analysis may well have led to theirconclusions that Schirmer’s & TBUT are superior diagnos-tics (the only two variables to be transformed in their study).The Schirmer test is likely best reserved for identifying pa-tients with deficiencies in aqueous tear production, whichhas been shown to be a small subset of the overall dry eyepopulation. This is reinforced by clinical experience withthe limitations of the Schirmer test. Instead, methods thathighlight meibomian gland dysfunction such as tear osmo-larity and meibomian gland orifice obstruction, and byextension, composite indices based on independent compo-nents analysis, likely better emphasize the underlying etiol-ogy of dry eye disease.
In another recent article, Amparo et al. reported thatcorneal staining was a superior efficacy endpoint in clinicaltrials for dry eye therapeutics and that tear osmolarity was
S18 THE OCULAR SURFACE / APRIL 2014, V
inferior.243 This highly biased study demonstrated funda-mental statistical design flaws resulting in selection bias,e.g. where subjects were qualified only on the basis of highOSDI and corneal staining scores while excluding inputfrom tear osmolarity. Accordingly, results reported byAmparo et al. were dominated by statistical artifacts suchas regression to the mean, and changes observed in patientswere not prospectively controlled and not attributable totreatment effect. A subsequent, unbiased analysis of adifferent longitudinal dry eye study comparing tear osmolar-ity and corneal staining revealed quite the opposite - that thechange in symptoms and change in tear osmolarity weremore synchronized than the changes in staining. A rebuttalto the Amparo paper is contained in an editorial in the jour-nal that summarizes these results.244
A larger review of current thinking on the subject ofclinical trial design including appropriate designs for trialsinvolving DED, is available,245
B. The Role of Tear Film InstabilityThe term tear film instability, generally relates to the
integrity of the tear film. As described in the DEWS report,tear film instability, epitomized by early tear film break up,together with tear hyperosmolarity, represent the two coreunderlying mechanisms of DED.1 Also, osmolar homeosta-sis of the tears becomes progressively unstable withadvancing DED, showing increasing temporal variation astear osmolarity rises. This is depicted in Figure 8. However,tear osmolarity measured from the lower meniscus may notreflect osmolarity on the corneal surface over the interblinkinterval; the latter is predicted to be higher, especially in pa-tients with DED.11 Transient increases in tear osmolaritymay be observed under conditions of tear instability. Theamount of increase in osmolarity and its potential effecton epithelium and nerves has been examined experimen-tally.246 Results showed that tear instability was associatedwith ratings of discomfort and symptoms, including percep-tions of burning and stinging. Similarly, exposure to hyper-osmolar solution (approximately 800-900 mOsm/L) alsogenerated perceptions of discomfort. In vitro analysisshowed that a 600-mOsm/L solution resulted in the activa-tion of mitogen-activated protein kinase. Tear film insta-bility may also be associated with visual aberrations.247-249
One manifestation of tear film instability is inter-eyevariability in tear osmolarity. Absolute difference in inter-eye tear osmolarity has been shown to increase with severityand can be a useful strategy to differentiate healthy individ-uals from patients with DED (Figure 9).66 Variability in os-molarity was strongly related to the higher osmolarity valueof the two eyes when bilateral measurements were taken,which is the recommendation for clinical practice. Usingthe higher osmolarity value of bilateral measurements alsoincreases the likelihood of correct diagnosis. (Eldridge DC,et al. IOVS 2012;53(5): ARVO E-abstract 3379).Summary: Assessment Issues: Symptoms and Signs
d DED is best viewed as a disease of the lacrimal func-tional unit, causing a loss of tear homeostasis. Its
Figure 8. The variability of tear osmolarity in successive measurements, rises with increasing osmolarity. Serial tear osmolarity measurements wereperformed on each eye; four separated by 15min followed by four separated by 1min, at each of three visits. Comparison of each of the four graphsillustrates the increasing fluctuation of tear osmolarity as osmolarity increases. Normal (non-dry eye) subjects are demonstrated in the upper left paneland the dry eye patients of increasing severity are demonstrated in the successive panels. Adapted from Keech A, Senchyna M, Jones L. Current EyeResearch. 2013;38:428-436.
RETHINKING DRY EYE DISEASE / Bron, et al
hallmarks are tear instability, including early tear filmbreakup, and tear hyperosmolarity, with increasingfluctuations with advancing disease.
d There is poor correlation between signs and symp-toms. Signs reflect either only one subtype of DEDand/or provide information that is independent ofother signs but reflect a specific aspect of the diseaseappearing at a different stage of development.
d Tear osmolarity is a globalmarker of DED reflecting dis-ease severity across the entire spectrum of the disease as
Figure 9. Absolute difference in inter-eye osmolarity increases withDED disease severity. From Lemp MA, Bron AJ, Baudouin C, et al. Tearosmolarity in the diagnosis and management of dry eye disease. Am JOphthalmol. 2011;151(5):792-798.e1.
THE OCULAR SURFACE / APRIL 2014, VOL.
well as different disease subtypes. Tear osmolarity hasbeen shown to decrease with effective treatment withsymptoms tending to follow afterwards.
d Symptoms alone are not sufficient to diagnose diseaseas many patients with clear objective evidence of dis-ease are asymptomatic.
VI. NEW APPLICATIONS OF POINT-OF-CARE TESTSThe routine clinical evaluation of patients with dry eye
disease includes testing for tear stability and ocular surfacestaining by the use of instilled fluorescein, measurement ofchanges in tear volume with time using Schirmer strips,and evaluation of meibomian gland status by transillumina-tion of the eyelid but has not included those tests that pre-viously required laboratory evaluation, such as measurementof tear osmolarity or presence of inflammatory mediators.The advent of point-of-care testing for osmolarity and mea-surement of at least one protein provides a new opportunityto diagnose and monitor DED in a number of clinical situ-ations. This section will review the literature on DED insome of these contexts and identify a possible role forpoint-of-care testing.
A. Contact Lens WearContact lens wear may contribute to a large proportion
of EDE cases, as it has been shown to promote evapora-tion.18,165 In a recent study, contact lens wearers had blink-ing parameters and tear film stability evaluated whileperforming tasks that required varying degrees of visualconcentration (eg, listening to music or playing a video
game) with or without wearing contact lenses.104 When per-forming the tasks that required a higher degree of visualconcentration, the interblink interval increased, as did thenumber of incomplete blinks. Interestingly, the authorsconcluded that incomplete blinks may play a larger rolethan overall blink frequency. Contact lens wear was associ-ated with tear film instability, increased blink rate, andocular irritation when performing the visual task. In a studyexamining factors that may predispose patients to contactlenserelated DED, it was shown that a number of factorswere predictive, including female sex, lenses with high watercontent, rapid prelens tear film thinning time, and tearosmolarity.117
B. Refractive SurgeryRefractive surgery may contribute to DED, as it has
been associated with decreases in corneal sensation andtear film stability.250,251 In a study of patients who under-went LASIK, participants were evaluated for DED 2 weeksbefore surgery, then 1 week, 3 months, and 9 months aftersurgery. At 9 months after surgery, patients were dividedinto 2 groups: chronic dry eye group (CDEG) and nonedry eye group (NDEG).252 Most diagnostic markers forDED were elevated following surgery but returned to pre-surgical levels at either the 3- or 9-month visit. Preopera-tive Schirmer test (without anesthesia) was determined tobe a predictor of tear breakup time at 9 months after sur-gery, suggesting that preoperative tear volume may influ-ence recovery following LASIK. Similarly, presurgicaltear osmolarity has been shown to be predictive ofrefractive outcomes. (Eldridge DC, et al. IOVS 2012;53(5) E-abstract 1286).
Patients with elevated osmolarity prior to surgery expe-rienced worse functional visual outcomes after refractivesurgery. Although the hyperosmolar patients had poorer un-corrected vision outcomes, these same “at-risk” patientscould not be identified preoperatively using vital dye stain-ing alone. Patients treated preoperatively with preservedartificial tears containing hyaluronic acid had a faster recov-ery of normal tear osmolarity than patients not receivingpretreatment. In another study of patients undergoing LA-SIK or LASEK, mean tear osmolarity was elevated at 12months following surgery; however, mean Schirmer testswere not significantly different.253 Additionally, a recent re-view showed that DED is one of the most common compli-cations following refractive surgery.254
There are a number of important clinical implications ofthese findings. First, refractive surgery has been associatedwith alterations in corneal sensitivity and changes in tearphysiology. Presurgical variables (eg, Schirmer, osmolarity)have been shown to be predictive of visual outcomes and pa-tient satisfaction. Importantly, from a clinical standpoint,these data indicate that identifying patients with DED priorto surgery then treating them appropriately postsurgicallycan improve visual outcomes. Thus, point-of-care testingfor osmolarity may be of value in improving clinicaloutcomes.
S20 THE OCULAR SURFACE / APRIL 2014, V
C. Cataract SurgeryBoth DED and cataract formation are common among
elderly people. One study of elderly patients undergoingphacoemulsification assigned patients to receive either tearlubricant (carboxymethylcellulose sodium [CMC]), saline,or no additional drops outside the standard of care for phaco-emulsification.255 Results showed that phacoemulsificationhad an adverse effect on all measured parameters of cornealsensitivity and tear physiology immediately following sur-gery. Tear physiology improved after 1 month, but cornealsensitivity did not recover to presurgical levels until 3 monthsafter treatment. Saline and CMC were not found to be effec-tive at improving measured parameters.255 In a retrospectivestudy, dissatisfaction after multifocal intraocular lens implan-tation was examined. Amajority of the complaints concernedunwanted visual symptoms, including blurred vision andphotic phenomena. Dry eye disease accounted for 15% of pa-tients reporting blurred vision.256
Based on the contribution of DED to blurred vision anddissatisfaction with outcomes from multifocal intraocularlens (IOL) implantation, one study evaluated the efficacyof cyclosporine therapy in this context. Patients undergoingbilateral phacoemulsification with IOL implantationreceived artificial tears in 1 eye and cyclosporine 0.05% inthe other. Drops were given from 1 month before surgeryuntil 2 months after the second surgery.257 Treatmentwith cyclosporine resulted in improvements in both cor-rected and uncorrected distance visual acuity as well ascorneal staining, as measured 2 months postoperatively.Cyclosporine treatment also improved TBUT, contrastsensitivity, and conjunctival staining.
This emerging body of literature suggests that DED maydevelop or be exacerbated by cataract surgery. Treatmentwith artificial tears may not be sufficient to improve out-comes. However, treatment with cyclosporine, both beforeand after phacoemulsification and IOL implantation, cansignificantly improve outcomes.
D. Glaucoma TherapyOcular surface disease, including DED, is common
among patients with glaucoma, with recent studies suggest-ing that 42% of glaucoma patients also have an ocular sur-face disease, many of which are severe.258,259 Treatments forglaucoma have been linked to increases in ocular surfacediseases,260 with changes seen following chronic treatmentwith topical glaucoma medications, hypothesized to occurbecause of the preservatives in the medications (benzalko-nium chloride). However, the effect of the antiglaucomamolecule itself cannot always be ruled out.259 Topical glau-coma therapy has been shown to induce a variety of inflam-matory changes and to increase tear osmolarity.261,262
Glaucoma filtering blebs have also been associated withindices of DED, including alterations in Schirmer scores,staining, and TBUT.263
Summary: New Applications of Existing Point-of-Care Testsd Cataract and refractive surgery, contact lens wear, andglaucoma treatment may exacerbate ocular surface
problems but can be identified preoperatively throughpoint-of-care testing.
d The newly available point-of-care test for measuringelevated osmolarity, which is a global marker forDED, has been reported to be the best single DEDmetric.
d Preoperative identification of patients with compro-mised tear function or subclinical ocular surfaceinflammation allows preoperative management of thetear film and ocular surface to improve postoperativeintegrity of the ocular surface with improved visualoutcome.
d DED may develop or be exacerbated by cataract sur-gery. Treatment with cyclosporine, both before and af-ter phacoemulsification and IOL implantation, cansignificantly improve outcomes.
d Many topical glaucoma therapies, some containingpreservatives, may exacerbate underlying.
DED by altering tear secretion or damaging the ocularsurface.
VII. NEW TESTS ON THE HORIZONOlder/established tests can be invasive, and many are
useful only in diagnosing severe DED. A recent analysis ofthe individual diagnostic components attempted to improvethe methodology of determining DED severity.237 Tear filmosmolarity was found to be the best marker of diseaseseverity, while other tests including Schirmer, TBUT,corneal staining, meibomian dysfunction assessment, andconjunctival staining were informative in the more severeforms of disease.237 Recent emphasis has been on the devel-opment of noninvasive assessments of DED that span therange of severity, allowing for early detection. Newermethods are frequently examined in conjunction with estab-lished methods to determine the level of correlation withthese tests and with symptoms.
A. InterferometryThis approach has been used to grade the severity of the
ADDE subtype ofDED.264,265Oldermethods of interferometrywere not done at point of care, but newer methods can be usedin the office. Interferometry measures tear-film and lipid-layerthickness noninvasively based on wavelength-dependentfringes.266 Interferometric tear film thickness measurementrevealed impaired precorneal tear-film formation in eyeswith ADDE. Interferometry may be helpful for evaluatingaqueous tear deficiency in conjunction with other DED exam-inations. Lipiview (TearScience,Morrisville, North Carolina) isa noninvasive ocular surface interferometer that captures liveimages of tear film, allowing quantitative analysis of> 1 billiondata points of an interferometric image [http://www.tearscience.com/physician/in-officeprocedure/lipid-science/].Lipiview can be used in conjunction with a meibomian glandevaluator and thermal pulsation system (LipiFlow; Tear-Science) for evaluating and treating EDE. Although thetechnology is promising, limited research has been publishedto date.
THE OCULAR SURFACE / APRIL 2014, VOL.
B. Assessment of Tear FilmThe corrected lipid layer stabilization time is a novel,
objective clinical parameter.267 One study examined the cor-relation of this parameter with commonly performed clin-ical tests. There was significant correlation with TBUT(Spearman r ¼ �0.485; P < .01), Schirmer test withoutanesthesia (r ¼ �0.431; P < .01), OSDI (r ¼ 0.498; P <.01), and lissamine green staining score (r ¼ 0.379; P <.05). However, there was no correlation with osmolarity.
Dynamic recording of double-pass retinal images64 is anevaluation that may allow for early detection of DED. Thetear film is the first medium that modifies the optical pathof light, and tear film changes in between blinks affect thequality of the retinal image. This noninvasive test gives theability to more directly measure changes in the retinal pointspread function over time and assess the impact of DED onboth wavefront aberrations and scatter. Dynamic recordingof double-pass retinal images is a sensitive method that maybe useful in detecting mild symptoms of DED and/or togauge its severity.
C. Measuring Visual Acuity and Contrast SensitivityThe dry ocular surface/aberrant tear film in patients with
DED may affect vision; however, visual acuity is not easilyquantifiable in DED patients, as patients report decreasedvisual acuity and complain of blurred vision but appearnormal on conventional visual acuity tests.268 Many effortshave been made to clinically measure visual function inDED patients and were recently reviewed by Ridder.99 Anoptimal test of visual function in the context of DED wouldbe one that can assess visual performance at multiple timesfollowing a blink, as most patients complain of only transi-tory changes in vision, and increased blinking and extendedviewing time can improve reading of a Snellen or ETDRStest chart.99 Similarly, contrast sensitivity tests, particularlystatic charts, may not detect differences between DED pa-tients and healthy individuals, and contrast sensitivitytesting has led to conflicting results in DED patients, whichmay be the result of temporal fluctuations in visual perfor-mance related to tear break-up time. Ridder et al used amethod whereby computer-generated sine-wave gratingswere presented at a set time after the blink, which decreasedvariability and showed that DED patients exhibit decreasedcontrast sensitivity when the tear film breaks up.99
Functional visual acuity (FVA) measures changes incontinuous visual acuity over time during a period of sus-tained eye opening.91 DED with punctate epithelial keratop-athy showed significant deterioration of visual function andoptical quality vs DED without SPK and healthy eyes by vi-sual maintenance ratio (VMR), the ratio between FVA andbaseline visual acuity (P < .05).92 DED without SPK showedminor visual deterioration compared with healthy eyes byVMR (P < .05).92 Recently, FVA measurements have beenperformed under natural blinking conditions.269 Addition-ally, the interblink interval visual acuity decay test is analternative to FVA that measures change in vision overtime between blinks. DED patients maintained best
corrected visual acuity for a shorter duration vs healthyindividuals.100
D. Meniscus MeasurementOptical coherence tomography (OCT) can be used to
noninvasively evaluate tear meniscus based on the relation-ship between tear volume and tear meniscus curva-ture.41,270-272 In one study, the upper tear meniscus wassmaller in patients with ADDE vs healthy controls (measuredvia real-time OCT), with the lower tear meniscus radiusand height demonstrating potential in diagnosis of aqueoustear deficiency.45 Lower tear meniscus measurement withfrequency domain OCT also correlated well with DED symp-toms and Schirmer score273 via noninvasive technique. OCTmeniscus area was positively correlated with Schirmer scoreand negatively correlated with symptoms (both P < .01).273
E. Quantitative ProteomicsDetection of tear-specific biomarkers of inflammation
(eg, MMP-9, IL-6, IL-1B)200,274 can be used for diagnosis,treatment decisions, and monitoring of DED. Isobaric tagfor relative and absolute quantitation (iTRAQ) technologyis one of the newest tools for quantitative mass spectrom-etry275 that can detect tear proteins. Tear samples foriTRAQ were collected using Schirmer strips275 which, alongwith symptom scores, allowed for categorization into non-DED, mild DED, moderate to severe DED and mixedDED subgroups. iTRAQ analysis revealed differences inprotein ratios between healthy participants and patientswith DED. Furthermore, unique proteins were associatedwith each subgroup of DED, and a greater number of pro-teins were downregulated in moderate to severe DED vsmild DED. Further evaluation in larger studies, in conjunc-tion with a full battery of diagnostic tests, is needed toconfirm and further elucidate the value of iTRAQ in tearproteomics.
InflammaDry is a new rapid immunoassay by RapidPathogen Screening, Inc (Sarasota, Florida) that measuresMMP-9 levels via an in-office test.276 This test can providepoint-of-care information on elevated MMP-9, helping toidentify patients with “inflammatory dry eye” and deter-mining which patients may respond to anti-inflammatorytherapy.276 In a test of 143 patients and 63 healthy controls,InflammaDry showed a sensitivity and specificity of 85%and 94%, respectively, and positive and negative predictivevalues of 73% and 97%, respectively. However, in this study,to be classified as having DED, patients had to meet all ofthe following criteria: OSDI increased to 13 or more,Schirmer test < than 10 mm in 5 min, reduced TBUT (<10 s), and presence of staining, suggesting that these wererelatively severe cases of DED.277
Electrophoresis of tear proteins278 can be used forroutine analysis of high-risk groups (eg, contact lenswearers, computer users), which could help with early detec-tion of DED and monitoring of therapy. The Sebia auto-mated system (Hyrys-Hydrasys, Sebia, France) hasimproved resolution and sensitivity over SDS-PAGE (Life
S22 THE OCULAR SURFACE / APRIL 2014, V
Technologies, Grand Island, New York), allowing the quan-tification of many proteins in a single analysis using 5 mL oftears collected via noninvasive procedure.278
A test measuring lactoferrin in human tears at point ofcare is now approved.279 Lactoferrin is thought to have arole in the regulation of ocular surface inflammation.280
One study demonstrated that measuring lactoferrin, inconjunction with the Schirmer test, resulted in satisfactorytest sensitivity and false-positive test rates.281
F. Evaluating the Meibomian GlandWhen evaluating meibomian glands, several factors
should be considered, including meibomian gland dropout,meibomian gland expressibility, and morphologic lidchanges.232 According to a recent review, a number of stra-tegies are available to perform meibography, includingtransilluminating meibography, noncontact meibography,infrared photography, and confocal microscopy.282
Although the transillumination method may be the simplest,it requires technical experience to perform and is associatedwith patient discomfort. By contrast, noncontact methodsare now available and offer a more patient-friendly strategyfor meibomian gland imaging.282,283 Although measurementvia meibography is now clinically available, it remains un-clear whether regeneration of atrophied or dropped-outglands is possible. As a diagnostic test, however, meibogra-phy is an objective measurement. An alternative approachis the use of a MeibopenTM for imaging of glandular struc-ture. Development of automated meibography gradingschemes will further improve the utility of this procedure.284
In vivo laser scanning confocal microscopy is a noninva-sive potential alternative to impression or brush cytologythat is useful as a supplementary diagnostic tool.285,286 Anew generation confocal microscope (Heidelberg Retina To-mography/Rostock Cornea Module [Heidelberg Engineer-ing, Heidelberg, Germany]) has been used in a number ofstudies. Kojima et al examined the mean individual epithe-lial cell area and nucleocytoplasmic ratios and determinedthat these were worse in patients with DED (Sjögren syn-drome) vs controls (P < .0001).285 No significant differenceswere found when the two examination techniques (impres-sion cytology and confocal microscopy) were compared; infact, the two techniques were correlated (P < .0001).285
Wakamatsu examined Sjögren syndrome DED and non-Sjögren syndrome DED286 and found that tear quantity, sta-bility, and vital staining scores were significantly worse inpatients with Sjögren syndrome DED or non-Sjögren syn-drome DED vs controls (P < .001 and P < .05, respectively).Additionally, the density of conjunctival and corneal inflam-matory infiltrates was significantly higher in eyes of patientswith Sjögren syndrome DED or non-Sjögren syndromeDED vs control eyes (P < .001). Conjunctival inflammatorycell densities negatively correlated with tear stability andquantity and positively correlated with vital staining scores.Conjunctival epithelial cell densities were significantly lowerin patients with Sjögren syndrome DED and non-Sjögrensyndrome DED vs controls (P < .05). The density of
epithelial cysts was significantly higher in patients with Sjög-ren syndrome vs controls (P < .001).286 Furthermore,in vivo laser scanning confocal microscopy revealedmorphological and inflammatory changes in meibomianglands and showed patterns of abnormalities not easilydistinguishable by clinical exams.287 High resolution OpticalCoherence Tomography (OCT) offers another promisingapproach.288
A recent study evaluated the ability of a composite ofocular symptoms, lid-margin abnormalities, and meibomianchanges (measured by meibography) to differentiate be-tween MGD and ADDE. Results demonstrated that evenwhen all 3 parameters were used to make the diagnosis ofMGD, sensitivity and specificity were 68% and 80%, respec-tively, suggesting that additional parameters, including theSchirmer test, may be necessary to differentiate between dis-ease subtypes.180 Evaluating meibomian loss in both upperand lower lids has been shown to improve the ability to pre-dict DED.289
G. Patient-Reported Outcomes (PROs)Because in milder forms of DED, clinical symptoms are
not well correlated with objective signs (with the possibleexception of the tear osmolarity test), clinical signs ofDED often underestimate the severity of the condition,even with new clinical evaluations being developed andimplemented, it is important to also assess the health-related quality of life (HRQoL) of patients as a diagnosticand treatment criterion.290 Although DED is frequently asymptom-driven disease, many patients with DED diag-nosed on the basis of objective tests don’t report symp-toms.238,291 A number of DED symptom questionnairesand vision-related quality-of-life questionnaires exist, asrecently reviewed comprehensively by Guillemin.290
The DEQ292,293 and the McMonnies instruments294 arethe most widely used symptom questionnaires for patientswith DED, but these do not capture HRQoL.290 TheOSDI295 is the DED-specific questionnaire that has beenaccepted by the FDA for use in clinical trials. It assessesboth symptoms and HRQoL, although the lack of contentvalidity may limit future use according to recent FDAPRO guidance.290,296 The National Eye Institute VisualFunction Questionnaire-25 (NEI-VFQ25) has also beenused in trials, but is used for a number of ophthalmic con-ditions and places emphasis on visual function, which isonly 1 component of DED.297 The Impact of Dry Eye inEveryday Life (IDEEL) was developed in accordance withthe FDA PRO guidance298 and is a good candidate forPRO assessment in the DED clinical trial setting, althoughit has experienced limited use thus far.290 The OSDI, NEI-VFQ25, and IDEEL have undergone standard linguistic vali-dation in different languages, which makes them useful forinternational clinical trials.290
Summary: New Tests on the Horizond The more recently described MMP-9 in-office mea-surement, which is a marker for inflammatory disease,although not specific for DED, should provide the
THE OCULAR SURFACE / APRIL 2014, VOL.
opportunity to identify patients with significant surfaceinflammation.
d Other diagnostic tests under development includeimproved interferometry to measure the thickness ofthe lipid layer and newer measures of visual function.-Clinical Conclusion
d The evidence published over the past 5 years suggeststhat a rethinking of traditional concepts of DED isneeded.
d DED is best viewed as a disease of the lacrimal func-tional unit leading to a loss of homeostatic protectionof the ocular surface from environmental stress. It isnot simply an inconvenience of advancing age butrather a bonafide disease with important clinicalimplications.
d Tear instability and hyperosmolarity are core mecha-nisms of the disease. Inflammation is an importantconsequent feature related to observed tissue damage.
d DED has two major subtypes, ADDE and EDE, themost common form of which is MGD.
d Although most patients present with symptoms, a sig-nificant number of patients with objective evidence ofdisease are asymptomatic. Diagnosis by symptomsalone is insufficient.
d Tear osmolarity is the best single metric to diagnoseDED and is linearly related to increasing severity ofdisease. It is recommended to test both eyes, take thehigher of the two values as indicative of the disease ef-fects, and note the inter-eye variability. A thresholdvalue of 308 mOsm/L appears to be the most sensitivevalue, and inter-eye variability greater than 8 mOsm/Lis indicative of DED-related instability of the tear film.Other clinical exam tests differentiate which subtype ofdisease is present.
d All objective tests for DED are variable in patients withDED. This variability is not seen in healthy individ-uals. With effective treatment, tear osmolarity hasbeen reported to return to normal, and the variabilityis reduced.
d Inflammatory cytokines and other proinflammatoryproteins appear in DED and are most prominent inmoderate to severe disease.
d Many new therapeutic approaches are under develop-ment, including anti-inflammatory agents, secretorystimulants, tear film stabilizers, and others. Withincorporation of improved endpoints for clinical trials,it is likely that a variety of therapeutic agents willemerge in the foreseeable future.
d Accurate recognition of disease is achievable, and suc-cessful management of DED is within our grasp for themajority of our patients.
ACKNOWLEDGEMENTSThis article was developed from a roundtable meeting held on
December 1-2, 2012 and organized by MediTech Media. All authors(with the exception of Professors Baudouin and Nichols, attended theroundtable meeting. The meeting was supported by an unrestricted
educational grant from TearLab, Inc, who had no input into the meeting orcontent of this article. All the authors contributed to the drafting of thismanuscript, and it represents a consensus of their opinions gathered atthe roundtable meeting and clarified in subsequent communications. Edito-rial assistance provided by Beth Burke, PhD, Daniel Sinsimer, PhD, andJoelle Suchy, PhD through MediTech Media was funded by the educationalgrant from TearLab Corp.
Additional faculty disclosures appear below:Anthony Bron has previously been a consultant for Acucela Inc., SAR-
code Bioscience Inc., and Santen Pharmaceutical Company and has receivedpayment for the development of educational presentations from MedEdicusLLC; has stock in TearLab Corporation.
Alan Tomlinson was, unfortunately, unable to provide any additionaldisclosures for medical reasons.
Gary Foulks is a consultant for Bausch & Lomb Inc., TearLab Corpo-ration, Eleven Biotherapeutics, R-Tech Ueno Ltd., Insight Pharmaceuticals,Parion Sciences Inc., Santen Pharmaceutical Company, Lexitas Pharma,Rigel Pharmaceuticals Inc.; has stock in TearLab Corporation.
Jay Pepose has received grants or has grants pending from AllerganInc., and SARcode Bioscience Inc.; has received consulting fees/honorariafrom TearLab Corporation; received payment for lectures including serviceon speakers’ bureaus from TearScience; and has stock/stock options inMimetogen Pharmaceuticals and in TearLab Corporation.
Christophe Baudouin is a board member and consultant for Alcon,Allergan, Thea Pharma, and Santen Pharmaceutical Company and is aconsultant for TearLab Corporation; has received payment for lecturesincluding service on speakers bureaus from Allergan Inc., Alcon and SantenPharmaceutical Company. The Quinze-Vingts National OphthalmologyHospital, and Vision Institute, University Paris 6, Paris, France has receivedgrants or has grants pending from Allergan Inc., Thea Pharma, and SantenPharmaceutical Company for activities conducted by CB.
Gerd Geerling serves as a board member and/or consultant for Aller-gan Inc., Alcon, Thea Pharma, TearScience Inc., Novagali/Santen, Bausch& Lomb Inc., has received payment for lectures from Alcon, Allergan Inc.and Thea Pharma and has stock/stock options in TearLab Corporation.The University of Duesseldorf has received grants from TearScience Inc.
Kelly Nichols serves as a consultant for TearLab Corporation; is aconsultant for Alcon, Allergan Inc., Bausch & Lomb Inc., Insite Pharmaceu-ticals, Nicox, SARcode Bioscience Inc., and Eleven Biotherapeutics; receivedpayment for lectures including service on speakers bureaus from AllerganInc., and Bausch & Lomb Inc.; has stock/stock options in SARcode Biosci-ence Inc. and TearLab Corporation (non-exercised). The Foundation forEducation and Research in Vision, University of Houston, College ofOptometry, Houston, Texas, USA has received grants or has grants pendingfrom Allergan Inc. and SARcode for activities conducted by KN.
Michael Lemp is a consultant and serves as Chief Medical Officer atTearLab and receives fees for participation in review activities such asdata monitoring boards, statistical analysis, end point committees fromTearScience, Novagali Pharma, and Merck & Co., Inc.; is a consultantfor Novagali/Santen Pharma, Merck & Co., Inc., and SARcode BioscienceInc.; has stock/stock options in TearLab Corporation and NovagaliPharma.
REFERENCES1. The definition and classification of dry eye disease: Report of the defi-
nition and classification subcommittee of the international dry eyeWorkShop (2007). Ocul Surf 2007;5(2):75-92
2. Research in dry eye: Report of the research subcommittee of the inter-national dry eye WorkShop (2007). Ocul Surf 2007;5(2):179-93
3. Design and conduct of clinical trials: Report of the clinical trials sub-committee of the international dry eye WorkShop (2007). Ocul Surf2007;5(2):153-62
S24 THE OCULAR SURFACE / APRIL 2014, V
4. Methodologies to diagnose and monitor dry eye disease: Report of thediagnostic methodology subcommittee of the international dry eyeWorkShop (2007). Ocul Surf 2007;5(2):108-52
5. The epidemiology of dry eye disease: Report of the epidemiology sub-committee of the international dry eye WorkShop (2007). Ocul Surf2007;5(2):93-107
6. Management and therapy of dry eye disease: Report of the manage-ment and therapy subcommittee of the international dry eye Work-Shop (2007). Ocul Surf 2007;5(2):163-78
7. Stern ME, Gao J, Siemasko KF, et al. The role of the lacrimal func-tional unit in the pathophysiology of dry eye. Exp Eye Res 2004;78:409-16
8. Bron AJ. The doyne lecture: reflections on the tears. Eye (Lond)1997;11(Pt 5):583-602
9. Mishima S, Gasset A, Klyce Jr SD, Baum JL. Determination of tear vol-ume and tear flow. Invest Ophthalmol 1966;5(3):264-76
10. Dartt DA. Regulation of mucin and fluid secretion by conjunctivalepithelial cells. Prog Retin Eye Res 2002;21(6):555-76
11. Gaffney EA, Tiffany JM, Yokoi N, Bron AJ. A mass and solute balancemodel for tear volume and osmolarity in the normal and the dry eye.Prog Retin Eye Res 2010;29(1):59-78
12. Bron AJ, Tiffany JM, Gouveia SM, Yokoi N, Voon LW. Functional as-pects of the tear film lipid layer. Exp Eye Res 2004;78(3):347-60
13. Knop E, Knop N, Millar T, Obata H, Sullivan DA. The internationalworkshop on meibomian gland dysfunction: Report of the subcom-mittee on anatomy, physiology, and pathophysiology of the meibo-mian gland. Invest Ophthalmol Vis Sci 2011;52(4):1938-78
14. Blackie CA, Korb DR. The diurnal secretory characteristics of individ-ual meibomian glands. Cornea 2010;29(1):34-8
15. Chew CK, Hykin PG, Jansweijer C, Dikstein S, Tiffany JM, Bron AJ. Thecasual level ofmeibomian lipids in humans.Curr EyeRes 1993;12(3):255-9
16. Chew CK, Jansweijer C, Tiffany JM, Dikstein S, Bron AJ. An instru-ment for quantifying meibomian lipid on the lid margin: The meibom-eter. Curr Eye Res 1993;12(3):247-54
17. McDonald JE. Surface phenomena of the tear film. Am J Ophthalmol1969;67(1):56-64
18. Korb DR, Greiner JV, Herman JP, et al. Lid-wiper epitheliopathy anddry-eye symptoms in contact lens wearers. CLAO J 2002;28(4):211-6
19. King-Smith PE, Fink BA, Hill RM, Koelling KW, Tiffany JM. Thethickness of the tear film. Curr Eye Res 2004;29(4-5):357-68
20. King-Smith PE, Fink BA, Fogt N, Nichols KK, Hill RM, Wilson GS.The thickness of the human precorneal tear film: Evidence from reflec-tion spectra. Invest Ophthalmol Vis Sci 2000;41(11):3348-59
21. Holly FJ, Lemp MA. Wettability and wetting of corneal epithelium.Exp Eye Res 1971;11(2):239-50
22. Spurr-Michaud S, Argueso P, Gipson I. Assay of mucins in humantear fluid. Exp Eye Res 2007;84(5):939-50
23. Gipson IK, Hori Y, Argueso P. Character of ocular surface mucins andtheir alteration in dry eye disease. Ocul Surf 2004;2(2):131-48
24. King-Smith PE, Hinel EA, Nichols JJ. Application of a novel interfer-ometric method to investigate the relation between lipid layer thick-ness and tear film thinning. Invest Ophthalmol Vis Sci 2010;51(5):2418-23
25. Tiffany JM. The lipid secretion of the meibomian glands. Adv LipidRes 1987;22:1-62
26. Butovich IA. The meibomian puzzle: Combining pieces together. ProgRetin Eye Res 2009;28(6):483-98
27. Butovich IA, Millar TJ, Ham BM. Understanding and analyzing mei-bomian lipidsea review. Curr Eye Res 2008;33(5):405-20
28. Butovich IA. On the presence of (O-acyl)-omega-hydroxy fatty acidsand of their esters in human meibomian gland secretions. Invest Oph-thalmol Vis Sci 2011;52(1):639-41
29. Millar TJ, Tragoulias ST, Anderton PJ, et al. The surface activity of pu-rified ocular mucin at the air-liquid interface and interactions withmeibomian lipids. Cornea 2006;25(1):91-100
30. Green-Church KB, Butovich I, Willcox M, et al. The internationalworkshop on meibomian gland dysfunction: Report of the subcom-mittee on tear film lipids and lipid-protein interactions in healthand disease. Invest Ophthalmol Vis Sci 2011;52(4):1979-93
31. Garreis F, Gottschalt M, Schlorf T, et al. Expression and regulation ofantimicrobial peptide psoriasin (S100A7) at the ocular surface and inthe lacrimal apparatus. Invest Ophthalmol Vis Sci 2011;52(7):4914-22
32. Garreis F, Schlorf T, Worlitzsch D, et al. Roles of human beta-defensins in innate immune defense at the ocular surface: Armingand alarming corneal and conjunctival epithelial cells. HistochemCell Biol 2010;134(1):59-73
33. Garreis F, Gottschalt M, Paulsen FP. Antimicrobial peptides as a majorpart of the innate immune defense at the ocular surface. Dev Ophthal-mol 2010;45:16-22
34. Paulsen FP, Berry MS. Mucins and TFF peptides of the tear film andlacrimal apparatus. Prog Histochem Cytochem 2006;41(1):1-53
35. Ni M, Tam C, Verma A, et al. Expression of surfactant protein D inhuman corneal epithelial cells is upregulated by pseudomonas aerugi-nosa. FEMS Immunol Med Microbiol 2008;54(2):177-84
36. Van Haeringen NJ. Clinical biochemistry of tears. Surv Ophthalmol1981;26(2):84-96
37. Norn MS. Dead, degenerate, and living cells in conjunctival fluid andmucous thread. Acta Ophthalmol (Copenh) 1969;47(5):1102-15
38. Bron AJ, Yokoi N, Gaffney EA, Tiffany JM. A solute gradient in thetear meniscus. I. A hypothesis to explain marx’s line. Ocul Surf2011;9(2):70-91
39. Bron AJ, Yokoi N, Gaffney EA, Tiffany JM. A solute gradient in thetear meniscus. II. implications for lid margin disease, including meibo-mian gland dysfunction. Ocul Surf 2011;9(2):92-7
40. Yokoi N, Kinoshita S, Bron AJ, Tiffany JM, Sugita J, Inatomi T. Tearmeniscus changes during cotton thread and schirmer testing. InvestOphthalmol Vis Sci 2000;41(12):3748-53
41. Yokoi N, Bron AJ, Tiffany JM, Maruyama K, Komuro A, Kinoshita S.Relationship between tear volume and tear meniscus curvature. ArchOphthalmol 2004;122(9):1265-9
42. Mainstone JC, Bruce AS, Golding TR. Tear meniscus measurement inthe diagnosis of dry eye. Curr Eye Res 1996;15(6):653-61
43. Yokoi N, Bron A, Tiffany J, Brown N, Hsuan J, Fowler C. Reflectivemeniscometry: A non-invasive method to measure tear meniscus cur-vature. Br J Ophthalmol 1999;83(1):92-7
44. Yokoi N, Bron AJ, Tiffany JM, Kinoshita S. Reflective meniscometry:A new field of dry eye assessment. Cornea 2000;19(3 Suppl):S37-43
45. Shen M, Li J, Wang J, et al. Upper and lower tear menisci in the diag-nosis of dry eye. Invest Ophthalmol Vis Sci 2009;50(6):2722-6
46. Yokoi N, Komuro A. Non-invasive methods of assessing the tear film.Exp Eye Res 2004;78(3):399-407
47. Naase T, Doughty MJ, Button NF. An assessment of the pattern ofspontaneous eyeblink activity under the influence of topical ocularanaesthesia. Graefes Arch Clin Exp Ophthalmol 2005;243(4):306-12
48. DoughtyMJ. Consideration of three types of spontaneous eyeblink activityin normal humans: During reading and video display terminal use, in pri-mary gaze, and while in conversation. Optom Vis Sci 2001;78(10):712-25
49. Collins M, Seeto R, Campbell L, Ross M. Blinking and corneal sensi-tivity. Acta Ophthalmol (Copenh) 1989;67(5):525-31
50. Tsubota K. Tear dynamics and dry eye. Prog Retin Eye Res 1998;17(4):565-96
51. Glasgow BJ, Marshall G, Gasymov OK, Abduragimov AR, Yusifov TN,Knobler CM. Tear lipocalins: Potential lipid scavengers for the cornealsurface. Invest Ophthalmol Vis Sci 1999;40(13):3100-7
THE OCULAR SURFACE / APRIL 2014, VOL.
52. Yokoi N, Yamada H, Mizukusa Y, et al. Rheology of tear film lipidlayer spread in normal and aqueous tear-deficient dry eyes. InvestOphthalmol Vis Sci 2008;49(12):5319-24
53. Goto E, Tseng SC. Differentiation of lipid tear deficiency dry eye bykinetic analysis of tear interference images. Arch Ophthalmol2003;121(2):173-80
54. Owens H, Phillips J. Spreading of the tears after a blink: Velocity andstabilization time in healthy eyes. Cornea 2001;20(5):484-7
55. Brown SI, Dervichian DG. Hydrodynamics of blinking. in vitro studyof the interaction of the superficial oily layer and the tears. Arch Oph-thalmol 1969;82(4):541-7
57. Tsubota K, Nakamori K. Effects of ocular surface area and blink rateon tear dynamics. Arch Ophthalmol 1995;113(2):155-8
58. Tsubota K, Hata S, Okusawa Y, Egami F, Ohtsuki T, Nakamori K.Quantitative videographic analysis of blinking in normal subjectsand patients with dry eye. Arch Ophthalmol 1996;114(6):715-20
59. Himebaugh NL, Begley CG, Bradley A, Wilkinson JA. Blinking and tearbreak-up during four visual tasks. Optom Vis Sci 2009;86(2):E106-14
60. Montes-Mico R, Caliz A, Alio JL. Wavefront analysis of higher orderaberrations in dry eye patients. J Refract Surg 2004;20(3):243-7
61. Montes-Mico R, Alio JL, Charman WN. Dynamic changes in the tear
film in dry eyes. Invest Ophthalmol Vis Sci 2005;46(5):1615-962. Ferrer-Blasco T, Garcia-Lazaro S, Montes-Mico R, Cervino A, Gonza-
lez-Meijome JM. Dynamic changes in the air-tear film interface mod-ulation transfer function. Graefes Arch Clin Exp Ophthalmol2010;248(1):127-32
63. Denoyer A, Rabut G, Baudouin C. Tear film aberration dynamics andvision-related quality of life in patients with dry eye disease. Ophthal-mology 2012;119(9):1811-8
64. Benito A, Perez GM, Mirabet S, et al. Objective optical assessment oftear-film quality dynamics in normal and mildly symptomatic dryeyes. J Cataract Refract Surg 2011;37(8):1481-7
65. Tomlinson A, Khanal S, Ramaesh K, Diaper C, McFadyen A. Tear filmosmolarity: Determination of a referent for dry eye diagnosis. InvestOphthalmol Vis Sci 2006;47(10):4309-15
66. Lemp MA, Bron AJ, Baudouin C, et al. Tear osmolarity in the diag-nosis and management of dry eye disease. Am J Ophthalmol2011;151(5):792-798.e1
67. Lemp MA, Hamill Jr JR. Factors affecting tear film breakup in normaleyes. Arch Ophthalmol 1973;89(2):103-5
68. Bron AJ, Tiffany JM, Yokoi N, Gouveia SM. Using osmolarity to diag-nose dry eye: A compartmental hypothesis and review of our assump-tions. Adv Exp Med Biol 2002;506(Pt B):1087-95
69. Nichols JJ, Mitchell GL, King-Smith PE. Thinning rate of the precornealand prelens tear films. Invest Ophthalmol Vis Sci 2005;46(7):2353-61
70. Benedetto DA, Clinch TE, Laibson PR. In vivo observation of tear dy-namics using fluorophotometry. Arch Ophthalmol 1984;102(3):410-2
71. Craig JP, Blades K, Patel S. Tear lipid layer structure and stabilityfollowing expression of the meibomian glands. Ophthalmic PhysiolOpt 1995;15(6):569-74
72. Arciniega JC, Wojtowicz JC, Mohamed EM, McCulley JP. Changes inthe evaporation rate of tear film after digital expression of meibomianglands in patients with and without dry eye. Cornea 2011;30(8):843-7
73. Rantamaki AH, Javanainen M, Vattulainen I, Holopainen JM. Dolipids retard the evaporation of the tear fluid? Invest Ophthalmol VisSci 2012;53(10):6442-7
74. Craig JP, Tomlinson A. Importance of the lipid layer in human tearfilm stability and evaporation. Optom Vis Sci 1997;74(1):8-13
75. Goto E, Endo K, Suzuki A, Fujikura Y, Matsumoto Y, Tsubota K. Tearevaporation dynamics in normal subjects and subjects with obstructive
meibomian gland dysfunction. Invest Ophthalmol Vis Sci 2003;44(2):533-9
76. Mengher LS, Bron AJ, Tonge SR, Gilbert DJ. Effect of fluoresceininstillation on the pre-corneal tear film stability. Curr Eye Res1985;4(1):9-12
77. de Paiva CS, Lindsey JL, Pflugfelder SC. Assessing the severity of kera-titis sicca with videokeratoscopic indices. Ophthalmology 2003;110(6):1102-9
78. Gumus K, Crockett CH, Rao K, et al. Noninvasive assessment of tearstability with the tear stability analysis system in tear dysfunction pa-tients. Invest Ophthalmol Vis Sci 2011;52(1):456-61
79. Patel S, Murray D, McKenzie A, Shearer DS, McGrath BD. Effects offluorescein on tear breakup time and on tear thinning time. Am JOptom Physiol Opt 1985;62(3):188-90
80. Mengher LS, Pandher KS, Bron AJ. Non-invasive tear film break-uptime: Sensitivity and specificity. Acta Ophthalmol (Copenh)1986;64(4):441-4
81. Lemp MA. Report of the national eye Institute/Industry workshop onclinical trials in dry eyes. CLAO J 1995;21(4):221-32
82. Abelson MB, Ousler 3rd GW, Nally LA, Welch D, Krenzer K. Alter-native reference values for tear film break up time in normal anddry eye populations. Adv Exp Med Biol 2002;506(Pt B):1121-5
84. Peng CC, Cerretani C, Braun RJ, Radke CJ. Evaporation-driven insta-bility of the precorneal tear film. Manuscript Submitted. 2013.
85. Tsubota K, Yamada M. Tear evaporation from the ocular surface.Invest Ophthalmol Vis Sci 1992;33(10):2942-50
86. Uchiyama E, Aronowicz JD, Butovich IA, McCulley JP. Increasedevaporative rates in laboratory testing conditions simulating airplanecabin relative humidity: An important factor for dry eye syndrome.Eye Contact Lens 2007;33(4):174-6
87. Rieger G. The importance of the precorneal tear film for the quality ofoptical imaging. Br J Ophthalmol 1992;76(3):157-8
88. Tong L, Waduthantri S, Wong TY, et al. Impact of symptomatic dryeye on vision-related daily activities: The singapore malay eye study.Eye (Lond) 2010;24(9):1486-91
89. Liu H, Thibos L, Begley CG, Bradley A. Measurement of the timecourse of optical quality and visual deterioration during tear break-up. Invest Ophthalmol Vis Sci 2010;51(6):3318-26
90. Tutt R, Bradley A, Begley C, Thibos LN. Optical and visual impact oftear break-up in human eyes. Invest Ophthalmol Vis Sci 2000;41(13):4117-23
91. Kaido M, Dogru M, Ishida R, Tsubota K. Concept of functional visualacuity and its applications. Cornea 2007;26(9 Suppl 1):S29-35
92. Kaido M, Matsumoto Y, Shigeno Y, Ishida R, Dogru M, Tsubota K.Corneal fluorescein staining correlates with visual function in dryeye patients. Invest Ophthalmol Vis Sci 2011;52(13):9516-22
93. Kaido M, Ishida R, Dogru M, Tamaoki T, Tsubota K. Efficacy of punc-tum plug treatment in short break-up time dry eye. Optom Vis Sci2008;85(8):758-63
94. Rolando M, Iester M, Macri A, Calabria G. Low spatial-contrast sensi-tivity in dry eyes. Cornea 1998;17(4):376-9
95. Puell MC, Benitez-del-Castillo JM, Martinez-de-la-Casa J, et al.Contrast sensitivity and disability glare in patients with dry eye.Acta Ophthalmol Scand 2006;84(4):527-31
97. Ridder 3rd WH, LaMotte J, Hall Jr JQ, Sinn R, Nguyen AL, Abufarie L.Contrast sensitivity and tear layer aberrometry in dry eye patients.Optom Vis Sci 2009;86(9):E1059-68
S26 THE OCULAR SURFACE / APRIL 2014, V
98. Diaz-Valle D, Arriola-Villalobos P, Garcia-Vidal SE, et al. Effect oflubricating eyedrops on ocular light scattering as a measure of visionquality in patients with dry eye. J Cataract Refract Surg 2012;38(7):1192-7
99. Ridder 3rd WH, Tomlinson A, Huang JF, Li J. Impaired visual perfor-mance in patients with dry eye. Ocul Surf 2011;9(1):42-55
100. Walker PM, Lane KJ, Ousler 3rd GW, Abelson MB. Diurnal variationof visual function and the signs and symptoms of dry eye. Cornea2010;29(6):607-12
101. Tong BC, Harpole Jr DH. Molecular markers for incidence, prognosis,and response to therapy. Surg Oncol Clin N Am 2012;21(1):161-75
102. Tsubota K, Miyake M, Matsumoto Y, Shintani M. Visual protectivesheet can increase blink rate while playing a hand-held video game.Am J Ophthalmol 2002;133(5):704-5
103. Tsubota K, Nakamori K. Dry eyes and video display terminals. N EnglJ Med 1993;328(8):584
104. Jansen ME, Begley CG, Himebaugh NH, Port NL. Effect of contactlens wear and a near task on tear film break-up. Optom Vis Sci2010;87(5):350-7
105. Ford JG, Davis RM, Reed JW, Weaver RG, Craven TE, Tyler ME.Bilateral monocular diplopia associated with lid position during nearwork. Cornea 1997;16(5):525-30
106. Shaw AJ, Collins MJ, Davis BA, Carney LG. Corneal refractive changesdue to short-term eyelid pressure in downward gaze. J Cataract RefractSurg 2008;34(9):1546-53
107. Buehren T, Collins MJ, Carney L. Corneal aberrations and reading.Optom Vis Sci 2003;80(2):159-66
108. Lemp MA, Crews LA, Bron AJ, Foulks GN, Sullivan BD. Distributionof aqueous-deficient and evaporative dry eye in a clinic-based patientcohort: A retrospective study. Cornea 2012;31(5):472-8
109. Schaumberg DA, Dana R, Buring JE, Sullivan DA. Prevalence of dryeye disease among US men: Estimates from the physicians’ healthstudies. Arch Ophthalmol 2009;127(6):763-8
110. Schaumberg DA, Nichols JJ, Papas EB, Tong L, Uchino M, Nichols KK.The international workshop on meibomian gland dysfunction: Reportof the subcommittee on the epidemiology of, and associated risk factorsfor, MGD. Invest Ophthalmol Vis Sci 2011;52(4):1994-2005
111. Gayton JL. Etiology, prevalence, and treatment of dry eye disease. ClinOphthalmol 2009;3:405-12
112. Moss SE, Klein R, Klein BE. Incidence of dry eye in an older popula-tion. Arch Ophthalmol 2004;122(3):369-73
113. Moss SE, Klein R, Klein BE. Long-term incidence of dry eye in anolder population. Optom Vis Sci 2008;85(8):668-74
114. Zhang Y, Chen H, Wu X. Prevalence and risk factors associated withdry eye syndrome among senior high school students in a county ofshandong province, china. Ophthalmic Epidemiol 2012;19(4):226-30
115. Uchino M, Nishiwaki Y, Michikawa T, et al. Prevalence and risk fac-tors of dry eye disease in japan: Koumi study. Ophthalmology2011;118(12):2361-7
116. Lin PY, Tsai SY, Cheng CY, Liu JH, Chou P, Hsu WM. Prevalence ofdry eye among an elderly chinese population in Taiwan: The Shihpaieye study. Ophthalmology 2003;110(6):1096-101
121. Fraunfelder FT, Sciubba JJ, Mathers WD. The role of medications incausing dry eye. J Ophthalmol 2012;2012:285851
122. Guillon M, Maissa C. Dry eye symptomatology of soft contact lenswearers and nonwearers. Optom Vis Sci 2005;82(9):829-34
123. Khanal S, Tomlinson A. Tear physiology in dry eye associated withchronic GVHD. Bone Marrow Transplant 2012;47(1):115-9
124. Akpek EK, Klimava A, Thorne JE, Martin D, Lekhanont K,Ostrovsky A. Evaluation of patients with dry eye for presence of un-derlying sjogren syndrome. Cornea 2009;28(5):493-7
125. Abusharha AA, Pearce EI. The effect of low humidity on the humantear film. Cornea 2013;32(4):429-34
126. Schaumberg DA, Sullivan DA, Buring JE, Dana MR. Prevalence of dryeye syndrome among US women.Am J Ophthalmol 2003;136(2):318-26
127. Galor A, Feuer W, Lee DJ, et al. Prevalence and risk factors of dry eyesyndrome in a united states veterans affairs population. Am J Ophthal-mol 2011;152(3):377-384.e2
128. Moss SE, Klein R, Klein BE. Prevalence of and risk factors for dry eyesyndrome. Arch Ophthalmol 2000;118(9):1264-8
129. Schaumberg DA, Buring JE, Sullivan DA, Dana MR. Hormone replace-ment therapy and dry eye syndrome. JAMA 2001;286(17):2114-9
130. Bron AJ, Yokoi N, Gafney E, Tiffany JM. Predicted phenotypes of dryeye: Proposed consequences of its natural history. Ocul Surf 2009;7(2):78-92
131. Sullivan BD, Evans JE, Dana MR, Sullivan DA. Influence of aging onthe polar and neutral lipid profiles in human meibomian gland secre-tions. Arch Ophthalmol 2006;124(9):1286-92
132. Smith JA, Vitale S, Reed GF, et al. Dry eye signs and symptoms inwomen with premature ovarian failure. Arch Ophthalmol2004;122(2):151-6
133. Nakaishi H, Yamada Y. Abnormal tear dynamics and symptoms ofeyestrain in operators of visual display terminals. Occup EnvironMed 1999;56(1):6-9
134. Nakamura S, Kinoshita S, Yokoi N, et al. Lacrimal hypofunction as anew mechanism of dry eye in visual display terminal users. PLoS One2010;5(6):e11119
135. Madden LC, Tomlinson A, Simmons PA. Effect of humidity variationsin a controlled environment chamber on tear evaporation after dry eyetherapy. Eye Contact Lens 2013;39(2):169-74
136. Lee S, Lam S, Luk F. Investigation of cabin air quality in commercialaircrafts. Proc Healthy Build 2000;1:471-6
137. Wolkoff P, Nojgaard JK, Troiano P, Piccoli B. Eye complaints in theoffice environment: Precorneal tear film integrity influenced by eyeblinking efficiency. Occup Environ Med 2005;62(1):4-12
138. Gilbard JP, Gray KL, Rossi SR. A proposed mechanism for increasedtear-film osmolarity in contact lens wearers. Am J Ophthalmol1986;102(4):505-7
139. Stahl U, Willcox MD, Naduvilath T, Stapleton F. Influence of tear filmand contact lens osmolality on ocular comfort in contact lens wear.Optom Vis Sci 2009;86(7):857-67
140. Begley CG, Caffery B, Nichols KK, Chalmers R. Responses of contactlens wearers to a dry eye survey. Optom Vis Sci 2000;77(1):40-6
141. Doughty MJ, Fonn D, Richter D, Simpson T, Caffery B, Gordon K.A patient questionnaire approach to estimating the prevalence ofdry eye symptoms in patients presenting to optometric practicesacross canada. Optom Vis Sci 1997;74(8):624-31
142. Dumbleton KA, Woods CA, Jones LW, Fonn D. The relationship be-tween compliance with lens replacement and contact lens-relatedproblems in silicone hydrogel wearers. Cont Lens Anterior Eye2011;34(5):216-22
143. Dumbleton K, Woods CA, Jones LW, Fonn D. The impact of contem-porary contact lenses on contact lens discontinuation. Eye ContactLens 2013;39(1):93-9
THE OCULAR SURFACE / APRIL 2014, VOL.
144. Richdale K, Sinnott LT, Skadahl E, Nichols JJ. Frequency of and fac-tors associated with contact lens dissatisfaction and discontinuation.Cornea 2007;26(2):168-74
145. Nichols JJ, Jones L, Nelson JD, et al. The TFOS international workshopon contact lens discomfort: Introduction. Invest Ophthalmol Vis Sci2013;54(11). TFOS1-6
146. Campbell R, Kame G, Leach N, Paul M, White E, Zigler L. Clinicalbenefits of a new multipurpose disinfecting solution in silicone hydro-gel and soft contact lens users. Eye Contact Lens 2012;38(2):93-101
147. Dumbleton K, Woods C, Fonn D. An investigation of the efficacy of anovel ocular lubricant. Eye Contact Lens 2009;35(3):149-55
148. Chalmers R, Long B, Dillehay S, Begley C. Improving contact-lensrelated dryness symptoms with silicone hydrogel lenses. Optom VisSci 2008;85(8):778-84
149. Johnson ME, Murphy PJ, Boulton M. Carbomer and sodium hyalur-onate eyedrops for moderate dry eye treatment. Optom Vis Sci2008;85(8):750-7
150. Stahl U, Willcox M, Stapleton F. Role of hypo-osmotic saline drops inocular comfort during contact lens wear. Cont Lens Anterior Eye2010;33(2):68-75
151. McDonald M, D’Aversa G, Perry HD, Wittpenn JR, Donnenfeld ED,Nelinson DS. Hydroxypropyl cellulose ophthalmic inserts (lacrisert)reduce the signs and symptoms of dry eye syndrome and improve pa-tient quality of life. Trans Am Ophthalmol Soc 2009;107:214-21
152. Iskeleli G, Karakoç Y, Aydin O, Yetik H, Uslu H, Kizilkaya M. Com-parison of tear-film osmolarity in different types of contact lenses.CLAO J 2002;28(4):174-6
153. Martin DK. Osmolality of the tear fluid in the contralateral eye duringmonocular contact lens wear. Acta Ophthalmol (Copenh) 1987;65(5):551-5
154. Sarac O, Gurdal C, Bostanci-Ceran B, Can I. Comparison of tear os-molarity and ocular comfort between daily disposable contact lenses:Hilafilcon B hydrogel versus narafilcon A silicone hydrogel. Int Oph-thalmol 2012;32(3):229-33
155. Farris RL. Tear osmolarityda new gold standard? Adv Exp Med Biol1994;350:495-503
156. Miller WL, Doughty MJ, Narayanan S, et al. A comparison of tear vol-ume (by tear meniscus height and phenol red thread test) and tearfluid osmolality measures in non-lens wearers and in contact lenswearers. Eye Contact Lens 2004;30(3):132-7
157. Vishnubhatla S, Borchman D, Foulks GN. Contact lenses and the rateof evaporation measured in vitro; the influence of wear, squalene andwax. Cont Lens Anterior Eye 2012;35(6):277-81
158. Lee SY, Petznick A, Tong L. Associations of systemic diseases, smok-ing and contact lens wear with severity of dry eye. Ophthalmic PhysiolOpt 2012;32(6):518-26
159. Nichols JJ, Bickle KM, Zink RC, Schiewe MD, Haque RM, Nichols KK.Safety and efficacy of topical azithromycin ophthalmic solution 1.0%in the treatment of contact lens-related dry eye. Eye Contact Lens2012;38(2):73-9
160. Forsblad-d’Elia H, Carlsten H, Labrie F, Konttinen YT, Ohlsson C.Low serum levels of sex steroids are associated with disease character-istics in primary sjogren’s syndrome; supplementation with dehydro-epiandrosterone restores the concentrations. J Clin Endocrinol Metab2009;94(6):2044-51
161. Oncel BA, Pinarci E, Akova YA. Tear osmolarity in unilateral pseu-doexfoliation syndrome. Clin Exp Optom 2012;95(5):506-9
162. Rolando M, Refojo MF, Kenyon KR. Increased tear evaporation ineyes with keratoconjunctivitis sicca. Arch Ophthalmol 1983;101(4):557-8
163. Khanal S, Tomlinson A, Diaper CJ. Tear physiology of aqueous defi-ciency and evaporative dry eye. Optom Vis Sci 2009;86(11):1235-40
164. Mathers WD, Daley TE. Tear flow and evaporation in patients withand without dry eye. Ophthalmology 1996;103(4):664-9
165. Hamano H, Mitsunaga M. Application of an evaporimeter to the fieldof ophthalmology. J Japan Contact Lens Soc 1980;22:101-7
166. Shimazaki J, Goto E, Ono M, Shimmura S, Tsubota K. Meibomiangland dysfunction in patients with sjogren syndrome. Ophthalmology1998;105(8):1485-8
167. Mathers WD, Choi D. Cluster analysis of patients with ocular surfacedisease, blepharitis, and dry eye. Arch Ophthalmol 2004;122(11):1700-4
168. Tomlinson A, Doane MG, McFadyen A. Inputs and outputs of thelacrimal system: Review of production and evaporative loss. OculSurf 2009;7(4):186-98
169. Sommer A, Emran N. Tear production in vitamin A-responsivexerophthalmia. Am J Ophthalmol 1982;93(1):84-7
170. Li DQ, Chen Z, Song XJ, Luo L, Pflugfelder SC. Stimulation of ma-trix metalloproteinases by hyperosmolarity via a JNK pathway in hu-man corneal epithelial cells. Invest Ophthalmol Vis Sci 2004;45(12):4302-11
171. Li DQ, Luo L, Chen Z, Kim HS, Song XJ, Pflugfelder SC. JNK and ERKMAP kinases mediate induction of IL-1beta, TNF-alpha and IL-8following hyperosmolar stress in human limbal epithelial cells. ExpEye Res 2006;82(4):588-96
172. Lee JH, Kim M, Im YS, Choi W, Byeon SH, Lee HK. NFAT5 inductionand its role in hyperosmolar stressed human limbal epithelial cells.Invest Ophthalmol Vis Sci 2008;49(5):1827-35
173. Luo L, Li DQ, Corrales RM, Pflugfelder SC. Hyperosmolar saline is aproinflammatory stress on the mouse ocular surface. Eye Contact Lens2005;31(5):186-93
174. Luo L, Li DQ, Pflugfelder SC. Hyperosmolarity-induced apoptosis inhuman corneal epithelial cells is mediated by cytochrome c andMAPK pathways. Cornea 2007;26(4):452-60
175. Pflugfelder SC, Tseng SC, Yoshino K, Monroy D, Felix C, Reis BL.Correlation of goblet cell density and mucosal epithelial membranemucin expression with rose bengal staining in patients with ocular irri-tation. Ophthalmology 1997;104(2):223-35
176. Beuerman RW, Stern ME. Neurogenic inflammation: A first line of de-fense for the ocular surface. Ocul Surf 2005;3(4 Suppl):S203-6
177. Li N, Deng XG, He MF. Comparison of the schirmer I test with andwithout topical anesthesia for diagnosing dry eye. Int J Ophthalmol2012;5(4):478-81
178. Shimazaki J, Sakata M, Tsubota K. Ocular surface changes anddiscomfort in patients with meibomian gland dysfunction. Arch Oph-
thalmol 1995;113(10):1266-70179. Arita R, Itoh K, Maeda S, Maeda K, Tomidokoro A, Amano S. Efficacy
of diagnostic criteria for the differential diagnosis between obstructivemeibomian gland dysfunction and aqueous deficiency dry eye. Jpn JOphthalmol 2010;54(5):387-91
181. Caffery B, Simpson T, Wang S, et al. Rose bengal staining of the tem-poral conjunctiva differentiates sjogren’s syndrome from keratocon-junctivitis sicca. Invest Ophthalmol Vis Sci 2010;51(5):2381-7
182. Guillon M, Maissa C. Contact lens wear affects tear film evaporation.Eye Contact Lens 2008;34(6):326-30
183. Ogawa Y, Kuwana M. Dry eye as a major complication associated withchronic graft-versus-host disease after hematopoietic stem cell trans-plantation. Cornea 2003;22(7 Suppl):S19-27
184. De Paiva CS, Villarreal AL, Corrales RM, et al. Dry eye-inducedconjunctival epithelial squamous metaplasia is modulated by inter-feron-gamma. Invest Ophthalmol Vis Sci 2007;48(6):2553-60
S28 THE OCULAR SURFACE / APRIL 2014, V
185. Stern ME, Gao J, Schwalb TA, et al. Conjunctival T-cell subpopula-tions in sjogren’s and non-sjogren’s patients with dry eye. Invest Oph-thalmol Vis Sci 2002;43(8):2609-14
186. El Annan J, Chauhan SK, Ecoiffier T, Zhang Q, Saban DR, Dana R.Characterization of effector T cells in dry eye disease. Invest Ophthal-mol Vis Sci 2009;50(8):3802-7
187. Brignole F, Pisella PJ, Goldschild M, De Saint Jean M, Goguel A,Baudouin C. Flow cytometric analysis of inflammatory markers inconjunctival epithelial cells of patients with dry eyes. Invest Ophthal-mol Vis Sci 2000;41(6):1356-63
188. Zhang X, Chen W, De Paiva CS, et al. Desiccating stress inducesCD4þ T-cell-mediated sjogren’s syndrome-like corneal epithelialapoptosis via activation of the extrinsic apoptotic pathway by inter-feron-gamma. Am J Pathol 2011;179(4):1807-14
189. Yoon KC, De Paiva CS, Qi H, et al. Expression of th-1 chemokines andchemokine receptors on the ocular surface of C57BL/6 mice: Effects ofdesiccating stress. Invest Ophthalmol Vis Sci 2007;48(6):2561-9
190. Zhang X, Chen W, De Paiva CS, et al. Interferon-gamma exacerbatesdry eye-induced apoptosis in conjunctiva through dual apoptoticpathways. Invest Ophthalmol Vis Sci 2011;52(9):6279-85
191. Chen Y, Chauhan SK, Saban DR, Sadrai Z, Okanobo A, Dana R. Inter-feron-gamma-secreting NK cells promote induction of dry eye disease.J Leukoc Biol 2011;89(6):965-72
192. Chauhan SK, El Annan J, Ecoiffier T, et al. Autoimmunity in dry eye isdue to resistance of Th17 to treg suppression. J Immunol 2009;182(3):1247-52
193. Zhang X, Volpe EA, Gandhi NB, et al. NK cells promote th-17 medi-ated corneal barrier disruption in dry eye. PLoS One 2012;7(5):e36822
194. De Paiva CS, Chotikavanich S, Pangelinan SB, et al. IL-17 disruptscorneal barrier following desiccating stress. Mucosal Immunol2009;2(3):243-53
195. Schaumburg CS, Siemasko KF, De Paiva CS, et al. Ocular surfaceAPCs are necessary for autoreactive T cell-mediated experimentalautoimmune lacrimal keratoconjunctivitis. J Immunol 2011;187(7):3653-62
196. Zheng X, de Paiva CS, Li DQ, Farley WJ, Pflugfelder SC. Desiccatingstress promotion of Th17 differentiation by ocular surface tissuesthrough a dendritic cell-mediated pathway. Invest Ophthalmol VisSci 2010;51(6):3083-91
197. Redfern RL, McDermott AM. Toll-like receptors in ocular surface dis-ease. Exp Eye Res 2010;90(6):679-87
198. Chotikavanich S, de Paiva CS, Li de Q, et al. Production and activityof matrix metalloproteinase-9 on the ocular surface increase indysfunctional tear syndrome. Invest Ophthalmol Vis Sci 2009;50(7):3203-9
199. Meloni M, De Servi B, Marasco D, Del Prete S. Molecular mechanismof ocular surface damage: Application to an in vitro dry eye model onhuman corneal epithelium. Mol Vis 2011;17:113-26
200. Smith VA, Rishmawi H, Hussein H, Easty DL. Tear film MMP accu-mulation and corneal disease. Br J Ophthalmol 2001;85(2):147-53
201. Pflugfelder SC, Farley W, Luo L, et al. Matrix metalloproteinase-9knockout confers resistance to corneal epithelial barrier disruptionin experimental dry eye. Am J Pathol 2005;166(1):61-71
202. Afonso AA, Sobrin L, Monroy DC, Selzer M, Lokeshwar B,Pflugfelder SC. Tear fluid gelatinase B activity correlates with IL-1alpha concentration and fluorescein clearance in ocular rosacea.Invest Ophthalmol Vis Sci 1999;40(11):2506-12
in dry eye patients and in vitro models. Invest Ophthalmol Vis Sci2011;52(8):5488-96
205. Brignole F, Pisella PJ, De Saint Jean M, Goldschild M, Goguel A,Baudouin C. Flow cytometric analysis of inflammatory markers inKCS: 6-month treatment with topical cyclosporin A. Invest Ophthal-mol Vis Sci 2001;42(1):90-5
206. Baudouin C, Brignole F, Becquet F, Pisella PJ, Goguel A. Flow cytom-etry in impression cytology specimens. A new method for evaluationof conjunctival inflammation. Invest Ophthalmol Vis Sci 1997;38(7):1458-64
207. Baudouin C, Haouat N, Brignole F, Bayle J, Gastaud P. Immunopath-ological findings in conjunctival cells using immunofluorescence stain-ing of impression cytology specimens. Br J Ophthalmol 1992;76(9):545-9
208. Enriquez-de-Salamanca A, Castellanos E, Stern ME, et al. Tear cyto-kine and chemokine analysis and clinical correlations inevaporative-type dry eye disease. Mol Vis 2010;16:862-73
209. VanDerMeid KR, Su SP, Krenzer KL, Ward KW, Zhang JZ. A methodto extract cytokines and matrix metalloproteinases from schirmerstrips and analyze using luminex. Mol Vis 2011;17:1056-63
210. VanDerMeid KR, Su SP, Ward KW, Zhang JZ. Correlation of tear in-flammatory cytokines and matrix metalloproteinases with four dry eyediagnostic tests. Invest Ophthalmol Vis Sci 2012;53(3):1512-8
211. Yoon KC, Jeong IY, Park YG, Yang SY. Interleukin-6 and tumor ne-crosis factor-alpha levels in tears of patients with dry eye syndrome.Cornea 2007;26(4):431-7
212. Boehm N, Riechardt AI, Wiegand M, Pfeiffer N, Grus FH. Proinflam-matory cytokine profiling of tears from dry eye patients by means ofantibody microarrays. Invest Ophthalmol Vis Sci 2011;52(10):7725-30
213. Na KS, Mok JW, Kim JY, Rho CR, Joo CK. Correlations between tearcytokines, chemokines, and soluble receptors and clinical severity ofdry eye disease. Invest Ophthalmol Vis Sci 2012;53(9):5443-50
214. Food and Drug Administration. Priority NDA and BLA approvals in2002. http://www.fda.gov/Drugs/DevelopmentApprovalProcess/HowDrugsareDevelopedandApproved/DrugandBiologicApprovalReports/PriorityNDAandBLAApprovals/ucm051251.htm. Updated 2002.Accessed 12/6, 2012.
215. Allergan.Restastis [package insert]. 2012:1e6.216. Stevenson D, Tauber J, Reis BL. Efficacy and safety of cyclosporin A
ophthalmic emulsion in the treatment of moderate-to-severe dry eyedisease: A dose-ranging, randomized trial. the cyclosporin A phase 2study group. Ophthalmology 2000;107(5):967-74
217. Sall K, Stevenson OD, Mundorf TK, Reis BL. Two multicenter, ran-domized studies of the efficacy and safety of cyclosporine ophthalmicemulsion in moderate to severe dry eye disease. CsA phase 3 studygroup. Ophthalmology 2000;107(4):631-9
218. Dastjerdi MH, Hamrah P, Dana R. High-frequency topical cyclo-sporine 0.05% in the treatment of severe dry eye refractory to twice-daily regimen. Cornea 2009;28(10):1091-6
219. Hong S, Kim T, Chung SH, Kim EK, Seo KY. Recurrence after topicalnonpreserved methylprednisolone therapy for keratoconjunctivitissicca in sjogren’s syndrome. J Ocul Pharmacol Ther 2007;23(1):78-82
220. Marsh P, Pflugfelder SC. Topical nonpreserved methylprednisolonetherapy for keratoconjunctivitis sicca in sjogren syndrome. Ophthal-mology 1999;106(4):811-6
221. De Paiva CS, Corrales RM, Villarreal AL, et al. Corticosteroid anddoxycycline suppress MMP-9 and inflammatory cytokine expression,MAPK activation in the corneal epithelium in experimental dry eye.Exp Eye Res 2006;83(3):526-35
222. Griffin MO, Fricovsky E, Ceballos G, Villarreal F. Tetracyclines: Apleitropic family of compounds with promising therapeutic properties.review of the literature. Am J Physiol Cell Physiol 2010;299(3):C539-48
THE OCULAR SURFACE / APRIL 2014, VOL.
223. Dartt DA, Hodges RR, Li D, Shatos MA, Lashkari K, Serhan CN.Conjunctival goblet cell secretion stimulated by leukotrienes isreduced by resolvins D1 and E1 to promote resolution of inflamma-tion. J Immunol 2011;186(7):4455-66
224. Li N, He J, Schwartz CE, Gjorstrup P, Bazan HE. Resolvin E1 improvestear production and decreases inflammation in a dry eye mousemodel. J Ocul Pharmacol Ther 2010;26(5):431-9
225. Brignole-Baudouin F, Baudouin C, Aragona P, et al. A multicentre,double-masked, randomized, controlled trial assessing the effect oforal supplementation of omega-3 and omega-6 fatty acids on aconjunctival inflammatory marker in dry eye patients. Acta Ophthal-mol 2011;89(7):e591-7
226. Rand AL, Asbell PA. Nutritional supplements for dry eye syndrome.Current opinion in ophthalmology JID - 9011108. 0902.
227. Wojtowicz JC, Butovich I, Uchiyama E, Aronowicz J, Agee S,McCulley JP. Pilot, prospective, randomized, double-masked, pla-cebo-controlled clinical trial of an omega-3 supplement for dry eye.Cornea 2011;30(3):308-14
228. Li J, Roubeix C, Wang Y, et al. Therapeutic efficacy of trehalose eyedrops for treatment of murine dry eye induced by an intelligentlycontrolled environmental system. Mol Vis 2012;18:317-29
229. Hayashi Y, Ishimaru N, Arakaki R, et al. Effective treatment of amouse model of sjogren’s syndrome with eyedrop administration ofanti-CD4 monoclonal antibody. Arthritis Rheum 2004;50(9):2903-10
230. Versura P, Profazio V, Campos EC. Performance of tear osmolaritycompared to previous diagnostic tests for dry eye diseases. Curr EyeRes 2010;35(7):553-64
231. Jacobi C, Jacobi A, Kruse FE, Cursiefen C. Tear film osmolarity mea-surements in dry eye disease using electrical impedance technology.Cornea 2011;30(12):1289-92
232. Tomlinson A, Bron AJ, Korb DR, et al. The international workshop onmeibomian gland dysfunction: Report of the diagnosis subcommittee.Invest Ophthalmol Vis Sci 2011;52(4):2006-49
233. Sullivan BD, Crews LA, Sonmez B, et al. Clinical utility of objectivetests for dry eye disease: Variability over time and implications forclinical trials and disease management. Cornea 2012;31(9):1000-8
234. McCann LC, Tomlinson A, Pearce EI, Papa V. Effectiveness of artifi-cial tears in the management of evaporative dry eye. Cornea2012;31(1):1-5
235. Gilbard JP, Rossi SR, Azar DT, Heyda KG. Effect of punctal occlusionby freeman silicone plug insertion on tear osmolarity in dry eye disor-ders. CLAO J 1989;15(3):216-8
236. Tomlinson A, Madden LC, Simmons PA. Effectiveness of dry eye ther-apy under conditions of environmental stress. Curr Eye Res2013;38(2):229-36
237. Sullivan BD, Whitmer D, Nichols KK, et al. An objective approach todry eye disease severity. Invest Ophthalmol Vis Sci 2010;51(12):6125-30
238. Sullivan BD, Crews LA, Messmer EM, et al. Correlations betweencommonly used objective signs and symptoms for the diagnosis ofdry eye disease: Clinical implications. Acta Ophthalmol. doi: 10.1111/aos.12012.
239. Johnson ME. The association between symptoms of discomfort andsigns in dry eye. Ocul Surf 2009;7(4):199-211
240. Benjamin WJ, Hill RM. Human tears: Osmotic characteristics. InvestOphthalmol Vis Sci 1983;24(12):1624-6
241. Lucca JA, Nunez JN, Farris RL. A comparison of diagnostic tests forkeratoconjunctivitis sicca: Lactoplate, schirmer, and tear osmolarity.CLAO J 1990;16(2):109-12
242. See C, Bilonick RA, Feuer W, Galor A. Comparison of two methodsfor composite score generation in dry eye syndrome. Invest Ophthal-mol Vis Sci 2013;54(9):6280-6
243. Amparo F, Jin Y, Hamrah P, Schaumberg DA, Dana R. What is thevalue of incorporating tear osmolarity measurement in assessing pa-tient response to therapy in dry eye disease? Am J Ophthalmol2014;157(1):69-77
244. Pepose JS, Sullivan BD, Foulks G, Lemp MA. The value of tear osmo-larity as a metric in evaluating the response to dry eye therapy in theclinic and in clinical trials. Am J Ophthalmology 2014;157(1):4-6[Editorial]
245. Sullivan BD. Challenges in using signs and symptoms to evaluate newbiomarkers of dry eye disease. Ocul Surf 2014;12(1):2-3 [Editorial]
246. Liu H, Begley C, Chen M, et al. A link between tear instability andhyperosmolarity in dry eye. Invest Ophthalmol Vis Sci 2009;50(8):3671-9
247. Begley CG, Himebaugh N, Renner D, et al. Tear breakup dynamics: Atechnique for quantifying tear film instability. Optom Vis Sci2006;83(1):15-21
248. Himebaugh NL, Nam J, Bradley A, Liu H, Thibos LN, Begley CG.Scale and spatial distribution of aberrations associated with tearbreakup. Optom Vis Sci 2012;89(11):1590-600
249. Montes-Mico R, Cervino A, Ferrer-Blasco T, Garcia-Lazaro S,Madrid-Costa D. The tear film and the optical quality of the eye.Ocul Surf 2010;8(4):185-92
250. Albietz JM, Lenton LM. Management of the ocular surface and tearfilm before, during, and after laser in situ keratomileusis. J RefractSurg 2004;20(1):62-71
251. Goto T, Zheng X, Klyce SD, et al. Evaluation of the tear film stabilityafter laser in situ keratomileusis using the tear film stability analysissystem. Am J Ophthalmol 2004;137(1):116-20
252. Konomi K, Chen LL, Tarko RS, et al. Preoperative characteristics and apotential mechanism of chronic dry eye after LASIK. Invest Ophthal-mol Vis Sci 2008;49(1):168-74
253. Dooley I, D’Arcy F, O’Keefe M. Comparison of dry-eye diseaseseverity after laser in situ keratomileusis and laser-assisted subepithe-lial keratectomy. J Cataract Refract Surg 2012;38(6):1058-64
254. Quinto GG, Camacho W, Behrens A. Postrefractive surgery dry eye.Curr Opin Ophthalmol 2008;19(4):335-41
255. Khanal S, Tomlinson A, Esakowitz L, et al. Changes in corneal sensi-tivity and tear physiology after phacoemulsification. Ophthalmic Phys-iol Opt 2008;28(2):127-34
257. Donnenfeld ED, Solomon R, Roberts CW, Wittpenn JR,McDonald MB, Perry HD. Cyclosporine 0.05% to improve visual out-comes after multifocal intraocular lens implantation. J CataractRefract Surg 2010;36(7):1095-100
258. Pflugfelder SC, Baudouin C. Challenges in the clinical measurement ofocular surface disease in glaucoma patients. Clin Ophthalmol 2011;5:1575-83
259. Stewart WC, Stewart JA, Nelson LA. Ocular surface disease in patientswith ocular hypertension and glaucoma. Curr Eye Res 2011;36(5):391-8
260. Baudouin C. Effects of topical drug preservatives on the tear film andocular surface. TFOS Abstracts 2010:48
261. Wong TT, Zhou L, Li J, et al. Proteomic profiling of inflammatorysignaling molecules in the tears of patients on chronic glaucoma medi-cation. Invest Ophthalmol Vis Sci 2011;52(10):7385-91
262. Labbe A, Terry O, Brasnu E, Van Went C, Baudouin C. Tear film os-molarity in patients treated for glaucoma or ocular hypertension.Cornea 2012;31(9):994-9
263. Neves Mendes CR, Hida RY, Kasahara N. Ocular surface changes ineyes with glaucoma filtering blebs. Curr Eye Res 2012;37(4):309-11
S30 THE OCULAR SURFACE / APRIL 2014, V
264. Yokoi N, Takehisa Y, Kinoshita S. Correlation of tear lipid layer inter-ference patterns with the diagnosis and severity of dry eye. Am J Oph-thalmol 1996;122(6):818-24
265. Danjo Y, Hamano T. Observation of precorneal tear film in patientswith sjogren’s syndrome. Acta Ophthalmol Scand 1995;73(6):501-5
266. Hosaka E, Kawamorita T, Ogasawara Y, et al. Interferometry in theevaluation of precorneal tear film thickness in dry eye. Am J Ophthal-mol 2011;151(1):18-23.e1
267. Ring MH, Rabensteiner DF, Horwath-Winter J, Boldin I, Horantner R,Haslwanter T. Introducing a new parameter for the assessment of thetear film lipid layer. Invest Ophthalmol Vis Sci 2012;53(10):6638-44
268. Nichols KK, Nichols JJ, Mitchell GL. The lack of association betweensigns and symptoms in patients with dry eye disease. Cornea 2004;23(8):762-70
269. Kaido M, Ishida R, Dogru M, Tsubota K. The relation of functionalvisual acuity measurement methodology to tear functions and ocularsurface status. Jpn J Ophthalmol 2011;55(5):451-9
270. Savini G, Barboni P, Zanini M. Tear meniscus evaluation by opticalcoherence tomography. Ophthalmic Surg Lasers Imaging 2006;37(2):112-8
271. Savini G, Goto E, Carbonelli M, Barboni P, Huang D. Agreement be-tween stratus and visante optical coherence tomography systems intear meniscus measurements. Cornea 2009;28(2):148-51
272. Uchida A, Uchino M, Goto E, et al. Noninvasive interference tearmeniscometry in dry eye patients with sjogren syndrome. Am J Oph-thalmol 2007;144(2):232-7
273. Nguyen P, Huang D, Li Y, et al. Correlation between optical coherencetomography-derived assessments of lower tear meniscus parametersand clinical features of dry eye disease. Cornea 2012;31(6):680-5
274. Solomon A, Dursun D, Liu Z, Xie Y, Macri A, Pflugfelder SC. Pro- andanti-inflammatory forms of interleukin-1 in the tear fluid and con-junctiva of patients with dry-eye disease. Invest Ophthalmol Vis Sci2001;42(10):2283-92
275. Srinivasan S, Thangavelu M, Zhang L, Green KB, Nichols KK. iTRAQquantitative proteomics in the analysis of tears in dry eye patients.Invest Ophthalmol Vis Sci 2012;53(8):5052-9
276. Kaufman HE. The practical detection of MMP-9 diagnoses ocular sur-face disease and may help prevent its complications. Cornea 2013;32(2):211-6
277. Sambursky R, Davitt 3rd WF, Latkany R, et al. Sensitivity and speci-ficity of a point-of-care matrix metalloproteinase 9 immunoassay fordiagnosing inflammation related to dry eye. JAMA Ophthalmol2013;131(1):24-8
278. Chiva A. Electrophoresis of tear proteins as a new diagnostic tool fortwo high risk groups for dry eye: Computer users and contact lenswearers. J Med Life 2011;4(3):228-33
280. Kijlstra A. The role of lactoferrin in the nonspecific immune responseon the ocular surface. Reg Immunol 1990;3(4):193-7
281. Goren MB, Goren SB. Diagnostic tests in patients with symptoms ofkeratoconjunctivitis sicca. Am J Ophthalmol 1988;106(5):570-4
282. Pult H, Nichols JJ. A review of meibography. Optom Vis Sci2012;89(5):E760-9
283. Srinivasan S, Menzies K, Sorbara L, Jones L. Infrared imaging of mei-bomian gland structure using a novel keratograph. Optom Vis Sci2012;89(5):788-94
284. Arita R. Validity of noninvasive meibography systems: Noncontactmeibography equipped with a slit-lamp and a mobile pen-shaped mei-bograph. Cornea 2013;(32 Suppl 1):S65-70
285. Kojima T, Matsumoto Y, Dogru M, Tsubota K. The application ofin vivo laser scanning confocal microscopy as a tool of conjunctival
in vivo cytology in the diagnosis of dry eye ocular surface disease. MolVis 2010;16:2457-64
286. Wakamatsu TH, Sato EA, Matsumoto Y, et al. Conjunctival in vivoconfocal scanning laser microscopy in patients with sjogren syndrome.Invest Ophthalmol Vis Sci 2010;51(1):144-50
287. Villani E, Beretta S, De Capitani M, Galimberti D, Viola F, Ratiglia R.In vivo confocal microscopy of meibomian glands in sjogren’s syn-drome. Invest Ophthalmol Vis Sci 2011;52(2):933-9
288. Hwang HS, Shin JG, Lee BH, Eom TJ, Joo CK. In vivo 3D meibogra-phy of the human eyelid using real time imaging fourier-domain OCT.PLoS One 2013;8(6):e67143
289. Pult H, Riede-Pult BH, Nichols JJ. Relation between upper and lowerlids’ meibomian gland morphology, tear film, and dry eye. Optom VisSci 2012;89(3):E310-5
290. Guillemin I, BegleyC, Chalmers R, BaudouinC, Arnould B. Appraisal ofpatient-reported outcome instruments available for randomized clinicaltrials in dry eye: Revisiting the standards. Ocul Surf 2012;10(2):84-99
291. Viso E, Rodriguez-Ares MT, Abelenda D, Oubina B, Gude F. Preva-lence of asymptomatic and symptomatic meibomian gland dysfunc-tion in the general population of spain. Invest Ophthalmol Vis Sci2012;53(6):2601-6
292. Begley CG, Caffery B, Chalmers RL, Mitchell GL. Dry Eye Investiga-tion (DREI) Study Group. Use of the dry eye questionnaire to measure
THE OCULAR SURFACE / APRIL 2014, VOL.
symptoms of ocular irritation in patients with aqueous tear deficientdry eye. Cornea 2002;21(7):664-70
293. Begley CG, Chalmers RL, Mitchell GL, et al. Characterization of ocularsurface symptoms from optometric practices in north america. Cornea2001;20(6):610-8
294. McMonnies CW, Ho A. Patient history in screening for dry eye con-ditions. J Am Optom Assoc 1987;58(4):296-301
295. Ozcura F, Aydin S, Helvaci MR. Ocular surface disease index for thediagnosis of dry eye syndrome. Ocul Immunol Inflamm 2007;15(5):389-93
296. U.S.Department of Health and Human Services, Center for Drug Eval-uation and Research, Center for Biologics Evaluation and Research,Center for Devices and Radiological Health. Guidance for industry.patient-reported outcome measures: Use in medical product develop-ment to support labeling claims. Available at: http://www.fda.gov/downloads/Drugs/Guidances/UCM193282.pdf. Accessed Jan 20, 2014.
297. Mangione CM, Lee PP, Gutierrez PR, et al. Development of the 25-item national eye institute visual function questionnaire. Arch Oph-thalmol 2001;119(7):1050-8
298. Rajagopalan K, Abetz L, Mertzanis P, et al. Comparing the discrimina-tive validity of two generic and one disease-specific health-relatedquality of life measures in a sample of patients with dry eye. ValueHealth 2005;8(2):168-74
