Ethan Frome - University of Manchester



Lid wiper epitheliopathyNathan Efrona,1*, Noel A Brennan b,1, Philip B Morgan c,1, Tawnya Wilson b,1 aInstitute of Health and Biomedical Innovation, and School of Optometry and Vision Science, Queensland University of Technology, 60 Musk Avenue, Kelvin Grove, Queensland 4059, AustraliabJohnson & Johnson Vision Care, Inc., 7500 Centurion Parkway, Jacksonville FL 32256, USAcEurolens Research, The University of Manchester, Dover Street, Manchester M13 9PL, UK*Corresponding author. Tel: +617 3138 6401E-mail address: n.efron@qut.edu.au (N. Efron)1Percentage of work contributed by each author in the production of the manuscript is as follows: Efron, 50; Brennan, 25; Morgan, 15; Wilson, 10. A B S T R A C TSome recent research has resulted in a hypothesis that there is a common 'lid wiper' region that is apposite to the ocular surface or anterior lens surface (where contact lenses are worn), responsible for spreading tears during blinking. In the upper eyelid, it extends about 0.6 mm from the crest of the sharp posterior (inner) lid border (i.e. the mucocutaneous junction, or line of Marx) to the subtarsal fold superiorly and from the medial upper punctum to the lateral canthus horizontally. Histologically, it is seen as an epithelial elevation comprising of stratified epithelium with a transitional conjunctival structure of (moving posteriorly) squamous cells then cuboidal cells, with some parakeratinised cells and goblet cells. Lid wiper epitheliopathy (LWE) denotes staining of the lid wiper after instillation of dyes such as fluorescein, rose bengal or lissamine green. There have been some reports of higher rates of LWE in dry eye patients and contact lens wearers, but others have failed to find such associations. The primary cause of LWE is thought to be increased friction between the lid wiper and ocular or anterior contact lens surface due to inadequate lubrication, which could be caused by dry eye and may be exacerbated by factors such as abnormal blinking patterns, poor contact lens surface lubricity and adverse environmental influences. Recent evidence suggests that LWE is associated with sub-clinical inflammation. LWE has the potential to provide the missing mechanistic link between clinical observation and symptoms associated with dry eye and contact lens wear. Clinical and fundamental research into LWE is still in its infancy and in many instances equivocal; however, it is an idea that provides a potentially important new avenue for further investigation of anterior eye discomfort associated with ocular dryness and contact lens wear. Keywords: Contact lens; Dry eye; Eye lids; Lid wiper epitheliopathy. Contents1. Introduction2. Terminology3. Literature search strategy4. Structure of the eyelids and lid wiper4.1. Gross anatomy4.2. Histology4.2.1. Mucocutaneous junction (line of Marx)4.2.2. Lid wiper4.3. Neuroanatomy5. Examination of the lid wiper with staining agents5.1. Staining techniques5.2. Staining patterns observed5.3. Grading the severity of staining5.3.1. Subjective grading5.3.1. Automated grading6. LWE in presenting populations7.LWE associated with dry eye8.LWE associated with contact lens wear8.1. Publications demonstrating a link between LWE and contact lens associated dryness/discomfort8.2. Publications failing to demonstrate a link between LWE and contact lens associated dryness/discomfort8.3. Meta-analysis8.4Summary9. Pathology9.1. Immunocytochemistry9.2. Structural changes9.2.1. Lid wiper epithelial cell morphology9.2.2. Fractal dimensions of the lid wiper microvascular network9.3. Functional changes10. Aetiology10.1. Friction and contact lens wear10.2. Tribology10.3. Tear Film10.3.1. Overall structure10.3.2. Mucins10.3.3. Osmolarity10.4. Biotribology of blinking10.5. Saccadic eye movements10.6. Inflammation10.6.1. Cellular response10.6.2. Vascular response10.6.3. Temperature response10.7. Lid wiper elasticity and eyelid pressure10.8. Upper versus lower LWE10.9. Age10.10. Sex10.11. Humidity10.12. Psychological factors10.13 Multiple lid wiper epitheliopathies?11. Treatment11.1. Rebamipide11.2. Corticosteroids11.3. Lubricant eye drops11.4. Basic fibroblast growth factor11.5. LipiFlow Thermal Pulsation System11.6. Punctal plugs11.7. Fit contact lenses of high surface lubricity11.8. Alter lens wearing modalities11.9. Improve blinking behaviour12. Differential diagnosis12.1. Line of Marx12.2. Lid imbrication syndrome12.3. Blepharitis12.4. Papillary conjunctivitis12.5. Demodicosis12.6. Meibomian gland dysfunction12.7. Iatrogenic lid wiper staining13. Significance14. Conclusions and future directionsConflict of interestsAcknowledgementsReferencesList of abbreviationLWELid wiper epitheliopathy1. IntroductionThe contemporary ophthalmic literature contains few examples of reports of new anatomical structures and conditions that rapidly gain traction and enter the common lexicon. One example in the field of anterior eye is a structure known as the ‘lid wiper’ and a proposed condition brought about by disturbance to this structure referred to as ‘lid wiper epitheliopathy’ (LWE). These concepts were first described only 14 years ago by Korb et al. (2002), but they have attracted considerable attention as evidenced by a growing number of publications and abstracts specifically referring to them. Certainly, the terms lid wiper and LWE have are becoming increasingly common when referring to dry eye aetiology, diagnosis and management in contact lens wearers and non-lens wearers (see, for example, Bron et al., 2011; Bron, 2015; Caparas, 2015; Efron et al., 2013; Jones et al., 2013a).The lid wiper is proposed to be that portion of the marginal conjunctiva of the upper eyelid that acts as a wiping mechanism for spreading the tear film over the ocular surface, or the surface of a contact lens. There is also a lid wiper in the lower lid, but its role in spreading tears is less certain, as discussed later. LWE is the term used to describe disruption to the surface epithelium of the lid wiper, typically observed using vital dyes (Knop et al., 2010). Keen interest in the concept of the lid wiper and LWE possibly stems, at least in part, from the apparent difficulty encountered by clinicians and scientists in diagnosing, reconciling signs and symptoms and treating dry eye syndrome (Begley et al., 2003; Bron et al., 2007; Bron et al., 2014; Nichols et al., 2004; Pflugfelder, 2007; Pult et al., 2011). As well, little progress has been made in understanding the root cause of contact lens discomfort (Korb et al., 2002; Jones et al., 2013a), which in turn is a major reason for discontinuation from contact lens wear (Dumbleton, 2013), causing great inconvenience to lens wearers (Pesudovs et al., 2006) and missed economic opportunity for contact lens practitioners and the contact lens industry (Rumpakis, 2010). This paper will review the current literature on LWE, with the aim of identifying the strengths and weaknesses of the science underpinning this emerging concept. 2. TerminologyBroadly speaking, the lid margin is divided into three distinct anatomical sub-zones: (1) the skin epidermis extending over the meibomian gland orifices; (2) the mucocutaneous junction, which is the transition region between the skin epidermis and the conjunctival tissue – the surface that represents the line of Marx; and (3) the lid wiper (Fig. 1) (Knop et al., 2011b).Interestingly, the term ‘lid wiper’ is protected as a registered trademark, in respect of “performing diagnosis of diseases, namely, diagnosis of disease of the eye, namely, disease of the portion of the upper eyelid marginal conjunctiva of the eye that wipes the ocular or contact lens surface during blinking”. The trademark is owned by Ocular Research of Boston, Inc. (Swider, 2009). The owners of the trademark have indicated that while general academic use of the term ‘lid wiper’ is acceptable, they may seek to protect this term and claim exclusive right to its use in respect of commercial goods and services relating to examination of this tissue structure (Donald Korb, personal communication). Numerous studies have investigated various aspects of the lid margin, but there is some difficulty in interpreting the literature because the region of the lid margin that is being examined is not always clear. In terms of phraseology, Ehlers (1965a) appears to have first connected the action of the lid to that of the windscreen wiper of a car. He did not specifically provide terminology for the region, but when referring to the part of the lid that rubs against the bulbus during blinking, he stated “This soft (unkeratinised) bead gliding over the cornea must be assumed to be a perfect ‘wind screen wiper’”. Fatt (1992) later referred to the ‘wiper blade’ action of the upper lid in spreading tears over the ocular surface.In their initial paper on this topic, Korb et al. (2002) carefully defined the lid wiper region and precisely mapped out its location in relation to adjacent commonly recognised anatomical landmarks. Accordingly, the location of the lid wiper is clear and unambiguous in all subsequent papers by Korb and other researchers who have coined this specific term. However, some have adopted different terminology. For example, Munger and Halata (1984) referred to the ‘angle’. McGowan et al. (1994) used the term ‘marginal zone’ and ‘marginal angle’. Navascues-Cornago et al. (2015c) were clearly examining the lid wiper, but used the terms ‘lid margin’ and ‘marginal conjunctiva’, both of which were apparently considered to be synonymous with ‘lid wiper’. Golebiowski et al. (2012) also used the term ‘lid margin’ to indicate the lid wiper region for staining purposes, but measured ‘lid margin sensitivity’ at the ‘occlusal surface’.In respect of the term lid wiper epitheliopathy, it is perhaps worth considering the appropriateness of the suffix ‘-pathy’, which indicates a ‘morbid condition or disease; generally used to designate a non-inflammatory condition’ (Miller et al., 2005). Two questions arise: (1) To what extent is LWE truly a ‘disease’? and (2) Is there evidence that it has an inflammatory component (see Section 10.6)? If the answer to the latter question is in the affirmative, then the term “lid wiper epitheliitis” may be more appropriate. These nuances of etymology are discussed further in Section 13.In conclusion, there is no ambiguity when the term ‘lid wiper’ is used. However, studies investigating the ‘lid margin’, eyelid margin’, ‘marginal conjunctiva’, ‘marginal zone’ or ‘occlusal surface’ may or may not relate to the lid wiper, either exclusively or inclusively. Such ambiguity may be resolved in instances where the authors provide specific details of the precise location, anatomical features and adjacent structures of the tissue area they are investigating. The term lid wiper epitheliopathy implies a disease process with a non-inflammatory basis, which may or may not be entirely appropriate, as will be explored in this review. 3. Literature search strategyWe searched the ophthalmic literature on March 26, 2016 for scientific papers and abstracts relating to LWE published since 2002, the year Korb et al. (2002) first described this condition. Specifically, we searched PubMed (National Center for Biotechnology Information, USA) for the terms ‘lid wiper’ (yielding 33 references since 2002) and ‘lid margin’ (yielding 275 references since 2002), and manually scanned through these listings to find relevant references. The bibliography of all relevant papers was also checked for references that we may have missed employing our primary search strategy, and pertinent papers were accessed. Because the topic of LWE is relatively new, we also searched for abstracts, containing the search terms noted above, presented at the annual meetings of the Association for Research in Vision and Ophthalmology and the American Academy of Optometry. There are of course other meetings but we felt that we would identify most of the key work in this area by concentrating on the annual meetings of these two major academic and clinical ophthalmic organisations. As a result of the latter search strategy in particular, this review comes with an undoubtedly familiar, but nevertheless important caveat: abstracts are necessarily incomplete scientific accounts of emerging research that have not been subjected to a full peer review process. Except in a few instances, we will not issue this caution each time we report information derived from a specific abstract; therefore, when reading this review, due weight should be accorded to information derived from published papers versus abstracts. 4. Structure of the eyelid marginThe anatomy and histology of the eyelid margin has been described in a small number of papers throughout the 20th century, although the structure of the lid wiper as a defined anatomical region has only been investigated fairly recently. Here we shall (a) review pertinent information describing the gross anatomy of the lid wiper as well as its detailed cellular structure, and (b) attempt to reconcile historical and contemporary accounts. 4.1. Gross anatomyOf particular interest in the context of the lid wiper is the relationship of areas of contact and noncontact between the upper eyelid – which typically travels about 10 to 12 mm in the vertical direction during each complete blink (Sun et al., 1997) – and the ocular surface. According to Korb et al (2002), the anatomical area of the lid wiper region is the portion of palpebral marginal conjunctiva in the upper and lower lids that is in contact with the globe. In the upper lid, it extends from the crest of the sharp posterior (inner) lid border (i.e. the mucocutaneous junction, or line of Marx) to the subtarsal fold superiorly and from the medial upper punctum to the lateral canthus horizontally. This configuration is similar in the lower lid. Over a century ago it was proposed by Parsons (1904) that the ‘sharp’ inner lid border lies in close contact with the globe and contributes to the distribution of tears. A thickened epithelium in this area had previously been described by Sattler (1877). Ehlers (1965a) further advanced this thinking with the suggestion that this inner lid border region acted like a ‘wind-screen wiper’ in spreading tears across the ocular surface. Up until this time, it may have been assumed that other parts of the inner eyelid, such as the tarsal conjunctiva, also contacted the underlying ocular surface and contributed to the tear spreading function.The idea that the lid wiper is essentially the only part of the upper lid wiping tears across the ocular or contact lens surface assumes that the tarsal conjunctiva does not contact the opposing ocular or contact lens surface. Part of this reasoning – which may or may not be correct (see Section 4.2.2) – is that the lid wiper is the only part of the eyelid to exhibit non-keratinised stratified squamous epithelium, a characteristic feature of other body tissues that experience frequent rubbing (Ehlers, 1965a). Ehlers (1965a) made the further observation that blinking would be impeded if the tarsal region of the lid was ‘sucked’ closely against the eye.Kessing (1967) investigated the notion that there is a gap between the tarsal conjunctiva and ocular surface. A thick, curved needle was used to inject 0.2 to 0.3 ml barium sulphate contrast medium into the superior and inferior conjunctival fornices of the anaesthetised left eye of a 33 year old male experimental subject who was lying on his right side. About 1 min after injection of the contrast medium, tomography of the left orbit was performed in the antero-posterior plane, affording a sagittal section midway through the eyeball and conjunctival sac. Using a tomographic technique, the eyelid and associated anterior ocular structures were projected onto an X-ray film in normal size and qualitatively assessed.The images presented in the paper of Kessing (1967) are of low resolution and difficult to interpret; however, Kessing concluded “It is apparent from the studies, then, that in the upper eyelid only the marginal area, while in the lower eyelid the entire tarsal area, is in close contact with the eyeball.” The presumed gap between the ocular surface and the posterior surface of the eyelid, which extends vertically from the lid wiper to a region approaching the upper fornix, has become popularly known as ‘Kessing’s Space’ (Korb et al., 2002).Contemporary ocular imaging techniques such as high-resolution microscopy coil magnetic resonance imaging (HR-MRI) may hold promise in developing a more thorough understanding of the apposition of the eyelids with the ocular surface. Although not specifically investigating this issue, Georgouli et al. (2008) used HR-MRI to image anterior ocular structures of subjects with normal eyes and vision. It is not possible from inspection of the images presented in this paper to identify Kessing’s space; however, the precise anatomical location of the published image sections was not defined. Even if HR-MRI could be undertaken with an optimised imaging protocol aimed at identifying the relationship between the lids and ocular surface, the current resolution of this technique (250 ?m2) may be insufficient to reveal Kessing’s space, should it exist.The word ‘space’ in the term ‘Kessing’s space’ infers an empty area; however, this may not be the case. Using electron microscopy, Doughty (1997) observed a very thick, fixed gel layer across the palpebral conjunctiva in rabbits. He suggests that this mucus layer is important in providing a robust lubrication system to facilitate regular eye-blink activity. Doughty (1997) goes on to propose that the mucus-based material likely fills Kessing’s space, although this has yet to be confirmed in humans.Knop et al. (2012) measured the upper lid wiper in human cadaver eyes (see below) and noted that it is of variable width, ranging from 0.3 to 0.5 mm, and forms a slope. To investigate the contact region between the upper eyelid and the ocular surface – which should correspond to the width of the lid wiper if it is true that this is the only region of contact between the upper lid and globe during blinking – Shaw et al. (2010) used pressure sensitive paper which was adhered to the anterior surface of a rigid contact lens. The eyelid of 11 subjects was placed over the pressure-sensitive paper for 10s, and after removal the width of the pressure area was determined to be 0.60 ± 0.16 mm (range, 0.33 to 0.84 mm). These results essentially indicate the width of the region of the upper eyelid that contacts the globe, which is presumed to be the lid wiper. This finding of Shaw et al. (2010) is broadly consistent with the estimated width of the lid wiper of 0.3 to 0.5 mm, as reported by Knop et al. (2012). It is possible that the Shaw et al. (2010) overestimateed the width of the lid wiper as a result of small involutary eye movements during to 10s application period of the pressure-sensitive paper, which would have resulted in a slightly wider pressure marking than would have been the case if the eye was completely still. Nevertheless, these observations have been considered to be in line with the recent proposal that the lid wiper is the primary contact region (Korb et al., 2002). In the same paper, Shaw et al. (2010) measured the width of the line of Marx by photographing and measuring the width of this anatomical feature, which was determined to be 0.09 ± 0.02 mm (Fig. 2). 4.2. HistologyThere are two regions of interest here; namely, the mucocutaneous junction and the newly proposed 'lid wiper' region (Knop et al., 2012). These two adjacent structures shall be considered separately.4.2.1 Mucocutaneous junction (line of Marx)The first detailed account of the mucocutaneous junctions was provided by Marx (1924); hence the eponymous name of this anatomical feature as the ‘line of Marx’. In the Marx paper, which has recently been translated into English from the original German text (Pult et al., 2010), Marx discusses in detail his research into revealing the mucocutaneous junction using a variety of dyes.Marx noted that the mucocutaneous junctions “is often situated on a higher level relative to the epithelium of the epidermis of the eyelid” (Pult et al., 2010). This observation has not been confirmed in the histological analysis of Pult et al (2011b); they reported that the lid wiper formed “an elevation of multiple cell layers (typically 8–12, occasionally up to 15 layers)”, but made no mention of elevation of the mucocutaneous junctions. In fact, in Figure 3A of the paper of Pult et al (2011b), the mucocutaneous junctions appears depressed compared to the adjacent lid wiper, although this is diffucult to judge because the tissue is curved at this point and there is no obvious point of reference for assessing ‘elevation’.The histology of the various tissue structures that comprise the lid margin has been studied in detail by Knop et al. (2011b). These authors examined serial sections of the upper and lower lid margin of whole-mount specimens from 10 human body donors and compared their findings to images from 8 eyelids of living volunteers obtained by performing in vivo laser scanning confocal microscopy with a Heidelberg retina-tomograph (HRT II) and attached Rostock cornea module. Knop et al. (2011b) propose that the mucocutaneous junction is comprised of two distinct zones. The cornified and granular epidermis of the skin of the eyelid stops abruptly, just behind the posterior margin of the meibomian orifices. Knop et al. (2011b) define this point as the beginning of the mucocutaneous junction, which has a total width of 150 to 350?m ?m. The tissue structure then becomes a continuous surface layer of parakeratinised cells for the first 150 to 200 ?m (the continuous para-keratinuised zone), followed by discontinuous parakeratinised cells interspersed with ordinary squamous cells for the remaining 100 to 150 ?m (the squamous trannsition zone). The tissue in both zones is clearly stratified. Thus, according to this description provided by Knop et al (2011b), the mucocutaneous junction (line of Marx) can be characterised as a transitional zone of stratified epithelium populated at the surface by adjacent zones of continuous and discontinuous parakeratinised and ordinary squamous cells.Impression cytology has provided different insights to the histology of the line of Marx. Doughty (2013) obtained impression cytology specimens using a Millicell-CM filter unit placed across the line of Marx of the lower eyelid of 10 healthy male adults aged 18 to 57 years. In vivo staining with rose bengal was used in some subjects to highlight the line. The filters were stained with Giemsa. Cell and nucleus areas and dimensions were measured. The impression cytological specimens included three to eight lines of squamous-appearing cells. The cells had average areas ranging from 702 to 1,119 mm2 (group mean and SD of 894 ± 136 mm2). Most cells (average 93 ± 4 %) contained a nucleus but with sizes from 3 to 124 mm2 including a proportion (40 %) of pyknotic nuclei. The average nuclear areas ranged from 13.7 to 57.9 mm2, with a group mean area of 37.0 ± 15.0 mm2. Doughty (2013) concluded that cells along the line of Marx are moderate-sized squamous cells with nuclei smaller than in normal bulbar conjunctival cells. The cells may be pyknotic (shrunken) and/or anucleate.Jalbert et al. (2012) used impression cytology to examine the lid margins of a mixed group of contact lens wearers and non-lens wearers. They fixed and stained samples with haematoxylin and eosin and periodic acid-Schiff for cell morphology analysis. At the line of Marx, epithelial cells displayed parakeratinised features, with dense cytoplasm but more regular cell size, shape and nuclei. Thus, while Knop et al (2011b) and Doughty et al (2013) generally characterised the mucocutaneous junction as being composed, at least in part, of squamous cells, the description provided by Jalbert et al (2012) is more consistent with non-squamous cells. The discrepancy between the observations of Doughty et al (2013) and Jalbert et al. (2012) may relate to a number of factors, including the use of different staining techniques; assessment of the line of Marx in the upper lid (Jalbert et al., 2012) versus lower lid (Doughty, 2013); and the observation of non-contact lens wearers only (Doughty, 2013) versus a mixed cohort of lens wearers and non-lens wearers (Jalbert et al., 2012).The width of the line of Marx was measured by Knop et al. (2012) to be 0.2 to 0.3 mm wide. Doughty et al (2013) determined the line of Marx to be up to 0.3 mm wide, but noted that in some subject, when stained with lissamine green, it could be close to 0.5 mm wide. These values are somewhat greater than the estimates of Donald et al. (2003) (0.10 ± 0.09 mm) and Shaw et al. (2010) (0.09 ± 0.02 mm). Differences in accounts of the width of mucocutaneous junction may be attributed to differences in in-vitro versus in-vivo measurement.4.2.2. Lid wiperConvention suggests that surfaces exposed to a certain degree of mechanical friction are typically composed of squamous epithelium, such as the cornea, oral epithelium, or esophagus (Gray et al., 2005). The expectation would therefore be that the lid wiper – as a surface that would experience extensive frictional forces as a result of blinking – would largely be comprised of squamous epithelium. Parsons (1904) offered the what is believed to be the first description of the region of the lid margin, which he apparently believed to be the wiping surface: “The intermarginal zone of the lid between the anterior and posterior borders is covered with stratified epithelium, and this passes upwards for a short distance on to the posterior surface of the lid. This part of the lid is in closest apposition with the globe, and mutual pressure of the two may perhaps be the cause of the flattening of the superficial cells. It ceases in a line parallel with the posterior border of the lid, which shows a slight depression and is called the sulcus subtarsalis.” Thus, it was the view of Parsons (1904) that the lid wiper is comprised of squamous epithelium.Descriptions of the anatomical region of the lid margin provided by Ehlers (1965a) suggest that he may have believed that the mucocutaneous junction (line of Marx), or part thereof, was the primary lid wiping surface. This is because his histological description of the wiping surface as being comprised of non-keratinised stratified squamous epithelium is consistent with the histology of the squamous transition zone of the mucocutaneous junction as defined by Knop et al. (2011b). Ehlers then states that the squamous epithelium extends away from the lid margin on to the conjunctival side of the lid “ ... for some distance, until, rather abruptly, it continues in a single- or multi-layered, almost cubical epithelium with goblet cells.” This account is broadly consistent with the description provided by Knop et al (2011b) (see below) of the change in tissue structure from the squamous transition zone of the mucocutaneous junction into the lid wiper. It is difficult to completely reconcile these accounts as it is likely that Ehlers (1965a) and Knop et al (2012) differ in their interpretations as to where the mucocutaneous junction ends and the lid wiper begins.Knop et al. (2011b) described the lid wiper as an epithelial elevation that typically has an initial thickness of about 100 ?m. They described the lid wiper as consisting of stratified epithelium with a conjunctival structure of cuboidal cells, some parakeratinised cells and goblet cells, with a superficial flattened polygonal mesh of small vessels. The goblet cells observed in the lid wiper occur either singly or in clusters. Specifically, they are located in both superficial and deeper layers of the epithelium, and along epithelial crypts that are connected to the surface (Fig. 3) (Knop et al., 2012). These goblet cells secrete soluble mucins onto the lid wiper surface. Along with the tear film, the mucins form a hydrated gel between the lid wiper and the ocular surface to provide lubrication, so that the lid wiper and ocular surface are separated by a fluid film consisting of mucin and aqueous components.Knop et al (2011b) note that, although the lid wiper shows a typical conjunctival structure, the change of the epithelial surface morphology at its beginning is not sharply defined. They maintain that, as the initial part of the lid wiper can still show considerable numbers of interspersed parakeratinised cells or patches of one very flat cell layer (squamous cells) on top of cuboidal cells, the lid wiper in this respect can still have characteristics of a transition zone. Therefore the earlier descriptions of Parsons (1904) and Ehlers (1965a) of the lid wiper as having squamous cells at the surface are not entirely inconsistent with the contemporary account of Knop et al (2011b) of the epithelial structure of the lid wiper, especially given the uncertainty of where the mucocutaneous junction ends and the lid wiper begins. In relation to the expectation that a squamous epithelium would be more suited to the function of a lid wiper, Knop et al (2012) argue that the structure they described – a stratified cuboidal surface that contains goblet cells and is hence covered by secreted mucins, forming a hydrodynamic fluid layer between the lid wiper and the bulbar surface – may be just as suitable for the task of distributing precorneal tears into a thin film during blinking to form an optically perfect tissue–air interface without resulting in trauma to the ocular surface epithelia. Jalbert et al. (2012) appears to offer a view of lid wiper histology that agrees more with the earlier descriptions of Parsons (2004) and Ehlers (1965a), although their decriptions are ambiguous. They used impression cytology to examine the lid margin and described epithelial cells collected “around meibomian gland openings” as being flat and polygonal shaped, with dense (keratinised) cytoplasm and small or no nuclei. This was characterised as a squamous epithelium. It is unclear whether some of these cells were also collected more posteriorly, from the lid wiper region as defined by Korb et al. (2002). Figure 1(C) of the Jalbert et al (2012) seems to incorrectly indicate that lid margin epithelium is found posterior to the mucocutaneous junctions, whereas Figure 1(C) appears to be correctly labelled. Muntz et al. (2015) also performed impression cytology of the lid wiper. Specifically, they examined the lid wiper of 5 subjects (n = 10 eyes) and observed the presence of goblet cells, mucins, cell nuclei and various degrees of pre- and parakeratinisation. Using Calcein AM, Ethidium and Annexin V, the authors were able to demonstrate metabolic activity, compromised cell membranes, nucleic acids and apoptosis in cells.It is difficult to reconcile the above accounts for a number of reasons. The earlier reports of Parsons (1904) and Ehlers (1965a), and the contemporary report of Jalbert et al (2012), suffer from a lack of clarity as to the precise anatomical locations to which histological descriptions pertain. Knop et al (2011b) examined tissue from cold stored cadavers with an average age 77 years. A primary application of lid wiper theory is contact lens wear (see section 8), and according to one large international demographic survey, the average age of contact lens wearers is between 26 and 31 years, depending on the country. This raises the question of the relevance of the observations of Knop et al (2011b), especially in view of the notion that pressure between the lid wiper and ocular surface may be the cause of the flattening of the superficial cells (Parsons, 2004), which would be exacerbated in the elderly if this is a cumulative effect.In relation to assessment of the lid wiper using impression cytology (Jalbert et al, 2012; Muntz et al, 2015), and notwithstanding the ambiguity of the descriptions of Jalbert et al (2012), it is difficult to deduce undisturbed three-dimensional epithelial morphology from flat-mounted impression cytology samples that in the first place were collected by pressing a substrate against the target tissue; that is, it is not possible to tell with certainty if cells are squamous, cuboidal or columnar, or the extent to which the tissue is stratified, if at all. As well, it is not possible to determine from the work of Jalbert et al (2012) the extent to which the observed cell morphology relates to contact lens wearers versus non-lens wearers.4.3. NeuroanatomyUnlike the neural architecture of the cornea, which has been extensively investigated (Oliveira-Soto and Efron, 2015), there is minimal information available in the literature on the neuroanatomy of the lid wiper. This is despite the potential importance of lid wiper sensation as an ocular defense mechanism. Munger and Halata (1984) published what appears to be the only histological investigation of the neural tissue of this region. They observed a few special nerve endings consisting of scattered simple corpuscles, as well as scant Meissner corpuscles. The simple corpuscles are often associated with presumptive intraepithelial free nerve endings. Numerous papers make comment about the neuroanatomy of the lid margin by reference to Munger and Halata (1984), but these comments generally refer to the whole lid margin area. When referring to lid wiper innervation, the limited information provided by Munger and Halata (1984), on the region they referred to as the ‘angle between occlusal and conjunctival surfaces’, should be noted. Further aspects of lid wiper neuroanatomy can be inferred from measures of tactile sensitivity of this tissue. While a number of authors claim to have measured the sensitivity of the lid margin, only two appear to have genuinely measured the sensitivity of the lid wiper. McGowan et al. (1984) used a Cochet-Bonnet aesthesiometer to measure touch thresholds of three regions of the eyelid: the middle of the tarsus, the lid wiper (referred to as the ‘marginal angle’) and the non-lid wiper region of the lid margin (referred to as the ‘occlusal surface’). The lid wiper was the most sensitive of the three regions. Navascues-Cornago et al. (2015c) also used the Cochet-Bonnet instrument to measure touch sensitivity of the cornea, bulbar conjunctiva, lid wiper and tarsal conjunctiva. They found that the cornea was the most sensitive of these tissues, but that the lid wiper was more sensitive than other regions of the conjunctiva. The implication from these works is that the lid margin is more richly innervated than surrounding conjunctival and lid tissue, but is less richly innervated than the cornea..5. Examination of the lid wiper with staining agentsWith the upper eyelid everted, the lid wiper can be seen in white light as a long, thin pale band of tissue beneath the eye lashes extending laterally from inner to outer canthus (Fig. 4). Disturbance to the superficial epithelium of the lid wiper is not visible in white light but can be observed with the aid of vital dyes. 5.1. Staining techniquesIn their original paper, Korb et al. (2002) described the application of fluorescein followed by rose bengal (which we shall term ‘Korb Protocol A’) to examine the lid wiper (Fig. 5). Other combinations of dyes were subsequently evaluated in various concentrations for staining the lid wiper – namely, fluorescein, rose bengal and lissamine green. Korb et al. (2008) later concluded that a mixture of 2% fluorescein and 1% lissamine green (‘Korb Protocol B’) offers optimal visualisation of lid wiper staining, and advocated the following procedure.A 40-mL drop of an unpreserved mixture of equal volumes of 2% fluorescein and 1% lissamine green is instilled by pipette onto the inferior palpebral conjunctiva of the eye, which is exposed by gentle depression of the lower eyelid. This procedure is repeated after 5 min. This ‘sequential staining’ approach is recommended to ensure an adequate volume and concentration of dye to reveal the full extent of staining of the lid wiper epithelium (Korb et al., 2008). One minute after the instillation of the second drop of the combination dye, the upper eyelid is everted. Korb et al. (2002) recommend grasping the eyelashes or external eyelid surface, but not any part of the lid margin posterior to the lashes, to prevent finger contact with the transitional or lid wiper epithelium and the possibility of iatrogenic staining (Varikooty et al., 2015). The area of the lid wiper is examined for staining with a slit lamp biomicroscope using 16X magnification, a cobalt blue filter, and a slit beam approximately 5 mm in width and 10 mm in height. A cobalt filter, with the highest level of illumination and transmitting 330 to 400 nm of light, is used in the illumination system to enhance fluorescein staining. The horizontal length of the lid wiper, extending from the superior punctum to the lateral canthus, and the sagittal width of the lid wiper, extending from just proximal to the line of Marx to the subtarsal fold, can then be examined. The lid wiper is re-examined using white light and a slightly lower illumination level in order to reveal lissamine green staining. Korb et al. (2002) explains that the reason for using two stains is based upon his initial observation that of all subjects exhibiting with staining of the lid wiper with fluorescein and rose bengal, 23.5% stained with fluorescein and not with rose bengal; 20.5% stained with rose bengal and not with fluorescein; and 56% stained with both dyes.Different research groups have used a variety of staining approaches when examining the lid wiper, including the use of fluorescein only, lissamine green only, fluorescein and rose bengal (referred to as Korb Protocol A) (Korb et al., 2002), fluorescein and lissamine green (referred to as Korb Protocol B) (Korb et al., 2008) and a triple combination of fluorescein, lissamine green and rose bengal. As can be seen from Table 1, which is a chronological listing of all studies that have used dyes to examine the lid wiper, there has been a general trend over the past three years to depart from using either of the Korb protocols, in favour of using lissamine green only. Varikooty et al. (2012) have described an optimised technique for staining the lid wiper. The authors assessed the impact of concentration, volume and effect of repeat applications of fluorescein and lissamine green. The best outcome of each phase was used in each subsequent phase. The optimal procedure, using paper strips impregnated with either 1 mg fluorescein or 1 mg lissamine green, was determined to be as follows: moisten two strips, held together, with 50μl of saline and apply to the eye. Repeat this procedure 1 minute later. Evert the eyelid three minutes after the second instillation to visualize the lid wiper (Varikooty, 2015). Visualization of LWE with fluorescein is enhanced by a Kodak Wratten 12 barrier yellow filter (transmitting above 495nm) in the slit lamp observation system. The beam of the slit lamp is set to maximum width and 10 mm height, and the potentiometer is set to provide maximum illumination through a Wratten 47 or 47A cobalt blue exciter filter located in the illumination system.5.2. Staining patterns observedIn assessing the lid wiper following contact lens wear, Korb et al. (2002, 2005) simply considered the lateral extent of the band of staining as the basis for grading LWE. However, complex patterns of staining have been reported.Varikooty et al. (2008) noted an atypical manifestation of upper lid wiper staining that occurred in 4 out of 38 adapted silicone hydrogel contact lens wearers who complained of ocular surface dryness. These patients showed fimbriated or ‘feathery’ extensions from the superior margin of the subtarsal fold onto the upper tarsal plate. The extent of these feathery extensions varied between subjects, with a mean length of 2.0 ± 0.8 mm. In all cases, a broad band of staining with both fluorescein and lissamine green was demonstrated, which extended along the entire length of the lid margin. The authors suggested that this appearance could be attributed to a direct mechanical effect of the lid-lens interactions or more complex tissue-lens interactions. Furthermore, they suggested that the fact that this occurred with wearers of silicone hydrogel materials may indicate that the staining is related to the higher modulus or altered surface properties of these materials compared with HEMA-based lenses, although they did not test the latter lens type in their study.More recently, using Korb Protocol B, Varikooty et al. (2015) further investigated staining patterns observed in people wearing silicone hydrogel contact lenses. Sometimes no staining was observed. Staining that was present was categor into six patterns: vertical streaks, short horizontal band, broad horizontal band, speckled appearance, comb-shaped and atypical appearance (Fig. 6). Little difference was observed between the patterns of stains reveled by fluorescein and lissamine green, suggesting that the dual-staining approach originally described by Korb et al. (2002) and subsequently adopted by many investigators, may not be according any additional diagnostic capability compared to the use of a single dye.5.3. Grading the severity of stainingVarious approaches have been adopted for grading the severity of LWE. Initially this involved subjective grading based on the appearance of the lid wiper following instillation of various dyes. With growing interest in LWE, attention has turned to the development of automated grading systems. These two approaches are reviewed here.5.3.1. Subjective gradingAs described above, Korb et al. (2002) advocated a grading technique that accounted for staining observed with both fluorescein and rose bengal (Korb Protocol A). Specifically, fluorescein staining of the lid wiper was graded from 0 to 3 for each of 2 characteristics: the linear area of involvement (Table 2) and the severity of the staining (Table 3). The total grade for fluorescein staining of the lid wiper was the average of the grades for the linear area and severity of staining. This procedure was repeated for rose bengal. The total grades for fluorescein and rose bengal staining were averaged to produce a single final score, whereby 0.5 to 1.0 = grade 1 LWE (mild); 1.25 to 2.0 = grade 2 LWE (moderate); and 2.25 to 3.0 = grade 3 LWE (severe).Korb et al. (2002) make the point that it is very important to differentiate any staining across the proposed lid wiper zone/region from that staining seen at the line of Marx, which invariably stains brightly as a clearly delineated thin line at the distal border of the lid wiper with all three staining agents. Others have noted that grade 0 LWE shows only staining of the line of Marx but grade 1 LWE, for example, shows a slight staining of the lid wiper lesion in addition to the staining of the line of Marx (Yamamoto et al., 2016). In attempting to objectively assess overall staining across the marginal zone, the line of Marx may also be included in the measurements and still be referred to as the lid wiper (Navascues-Cornago et al., 2015a).Guillon and Maissa (2008) judged the extent of lissamine green staining of the lid wiper in terms of staining type (0 = none, 1 = broken line, 2 = thin line, 3 = thick line/patch) and staining severity (0 = none, 1 = slight, 2 = mild, 3 = moderate, 4 = severe). Korb et al. (2010) subsequently modified their earlier grading scheme (Korb et al., 2002) when using a combination of fluorescein and lissamine green staining (Korb Protocol B). The higher of the final fluorescein or lissamine green staining grades were used as the LWE severity grade (rather than the average of the two).A number of other grading systems have been applied, but detailed descriptions are lacking. For example, Willis et al. (2011) graded LWE on a 0-4 scale, although details were not presented in their abstract. Jalbert and Rejab (2015) used lissamine green to grade the severity of LWE in 0.5 steps using a 4-point simplified pictorial severity grading scale where 0 = none and 3 = severe, but they did not publish the grading pictures.Most papers and abstracts describing LWE research to date have used the Korb Protocol B, and data reported from different studies using this procedure can be validly compared. However, caution needs to be exercised when comparing data from studies that have used different staining protocols, because the extent of staining revealed may be procedure- and dye-dependent (Korb et al., 2008).5.3.2. Automated gradingThere have been two documented attempts at automating grading of lid wiper staining, both of which involve digital image capture of the stained lid wiper and subsequent image analysis.Varikooty et al. (2013) stained the upper lid margin of 22 subjects using lissamine green and applied an optim procedure to detect LWE. Three images – focused on the central, temporal and nasal regions of each upper lid – were captured at 12x magnification using a digital camera attached to a slit-lamp biomicroscope. Using custom built software, metrics characterizing the stained area were computed for each image. LWE was also graded by an investigator using Korb Protocol B. A number of image processing metrics were found by Varikooty et al. (2013) to be significantly associated with LWE grade: these included area, convexity and staining region thickness (Spearman rho, all at least 0.61, all p < 0.05). When LWE grades were binarised into approximately ‘moderate’ and ‘severe’ grades, separation of these diagnostic groups was most effective in respect of the parameters ‘area’, ‘convexity’ and ‘region thickness’ (receiver-operator characteristic area under curve = 0.96, 0.93 and 0.88, respectively). The authors concluded that their automated grading system facilitated repeatable, objective measurements of LWE. Kunnen et al. (2014) digitally photographed the upper eyelid margins of 27 participants after staining with lissamine green. Images were processed using custom designed software developed in MATLAB. After manual delineation of the lid margin, images were transformed into a hue/saturation/value format. Spurious reflections were automatically eliminated prior to contrast enhancement, thresholding and binarisation. The algorithm automatically extracted the following variables: proportion of staining relative to area of the eyelid; median intensity of staining; median intensity of staining in red, green and blue channels; relative greenness staining; green-red difference; green-blue difference; and median intensity of hue value. For the proportion of lissamine green staining relative to area of the eyelid, the mean difference between replicates was -0.06% with 95% limits of agreement being 1.07 and -1.13% (Kunnen et al., 2014). Correlation analysis showed significant associations between the mean grades by human observers and many parameters (p < 0.05). The proportion of lissamine green staining relative to area of the lid wiper was revealed to have the highest linear correlation with subjective grading (R = 0.63). Kunnen et al. (2014) concluded that the system they developed enabled objective, repeatable measures of lissamine green staining.6. LWE in presenting populationsGuillon and Maissa (2008) evaluated the incidence and type of lid margin staining in 75 consecutive patients (30 Male, 45 Female) aged between 18 and 62 years (mean 33.1± 11.5 years) presenting to an eye clinic; the population included non-contact lens wearers and contact lens wearers. The upper and lower lid margin staining was evaluated with lissamine green and was judged using the grading system described above (Guillon and Maissa, 2008) (see Section 5.3). In the majority of cases, staining was rated as mild or moderate for both the upper (mild = 39.6%; moderate = 37.6%) and lower (mild = 38.9%; moderate = 38.3%) eyelid margins. The staining observed was most commonly of the ‘thin line’ type (upper lid = 65.3%; lower lid = 63.3%). Clinically significant ‘thick line/patch’ staining was observed in about 1 in 4 eyelids (upper = 23.3%; lower = 25.3%). The results were highly correlated between the upper and lower lid margins (r = 0.895 to 0.918, p < 0.001) and between the right and left eyes (r = 0.820 to 0.987, p < 0.001). The authors concluded that LWE is present in 25% of a presenting population to an eye clinic.The prevalence of upper and lower LWE was examined using Korb Protocol B in 405 eyes of 208 soft contact lens wearers, 135 eyes of 71 rigid contact lens wearers and 443 eyes of 229 non-lens wearers by Shiraishi et al. (2014). Tear break-up time, and corneal and conjunctival staining was also assessed. The prevalence/mean grade of lower LWE (39.5%/0.79) was significantly higher than that of upper LWE (12.5%/0.21) in non-lens wearers (p < 0.001). The prevalence/mean grade of upper-LWE was significantly higher in rigid contact lens wearers (84.4%/1.3) than soft lens wearers (58.1%/2.1) (p < 0.001). There was no difference in prevalence/mean grade of lower-LWE between rigid and soft lens wearers (Shiraishi et al., 2014).The prevalence of both upper-LWE and lower-LWE were significantly correlated with age and corneal and conjunctival staining, but not sex and tear break-up time, in non-lens wearers. The prevalence of both upper-LWE and lower-LWE was significantly higher in younger subjects (p < 0.001). Upper-LWE and lower-LWE were detected in a higher percentage of contact lens wearers than in non-lens wearers (p < 0.001) (Shiraishi et al., 2014).Navascues-Cornago et al. (2015a) captured digital images of upper lid wiper staining in the right eye of 35 healthy non-contact lens wearers. The area of lid wiper staining was measured from the digital images using ImageJ (V.1.46; National Institutes of Health, Bethesda, MD, ). It is important to note that Navascues-Cornago et al. (2015a) included the area of the stained line of Marx in the measurement of lid wiper staining area. The area of upper lid wiper staining was found to be 2.7 ± 2.0 mm2 (mean ± standard deviation), with confidence intervals of 2.0 to 3.4 mm2 and range 0.4 to 7.7 mm2. The authors also measured three dimensions of the tarsal plate in these subjects (tarsal length, height and area), but found none of these measures correlated with area of lid wiper staining (all p > 0.05). Alghamdi et al. (2016) examined the lid wiper in 60 current contact lens wearers, 20 previous contact lens wearers and 20 non-lens wearers. They found no difference in the severity of LWE between the three groups. 7. LWE associated with dry eyeAlthough the original motivation of Korb et al. (2002) for elucidating the concept of LWE was to explain contact lens associated discomfort and dryness, there is also the possibility that an aberrant interaction between the lid wiper and ocular surface could have a direct role in the aetiology of dry eye in the absence of contact lenses. Indeed, Korb et al. (2002) reported anecdotally that they had observed lid wiper staining in patients presenting with dry-eye symptoms and disorders but who had never worn contact lenses.To this end, Korb et al. (2002) followed up their assessment in contact lens wearers with a study of non-contact lens wearers having symptoms but not signs of dry-eye (Korb et al., 2005). The presence or absence of dry eye symptoms was determined with the Standard Patient Evaluation of Eye Dryness (SPEED) questionnaire (Ngo et al., 2013). Recruitment continued until there were 50 patients in each of symptomatic and asymptomatic groups. Patients with ‘ambiguous’ scores (that is, from 2 to 9 on the SPEED questionnaire) were excluded to ensure that only those patients with a clear presence or absence of symptoms were studied. Criteria for admission to the study also included tear break-up time of 10 s or more, Schirmer test value of 10 mm or more, and absence of fluorescein corneal staining. Patients who had used any ocular ointment in the previous 3 days or contact lenses within the previous 6 months were excluded. Korb Protocol A was used to assess LWE.The two groups were well matched for age (44.3 years for the symptomatic patients and 42.8 years for the asymptomatic patients). Of a possible 24 points, where 0 = no symptoms and 24 = severe symptoms, the mean (± standard error) scores for the SPEED questionnaire were 0.46 ± 0.07 points for the asymptomatic group and 16.8 ± 0.55 points for the symptomatic group. Of the 50 symptomatic patients, 76% had LWE versus 12% of the 50 asymptomatic patients. These proportional differences were significant (p = 0.0001, χ2 = 24.47, 95% confidence interval: 56.3% to 71.7%). Korb et al. (2005) noted that LWE was present six times more frequently with symptomatic than with asymptomatic patients (p < 0.0001), and that grades 2 and 3 LWE were eight times more common with symptomatic than with asymptomatic patients (p = 0.0005). This is the first publication suggesting that LWE may have utility in diagnosing dry eye in a group of non-contact lens wearers.Korb et al. (2010) followed their earlier study with an investigation of LWE in a population who would be considered as having dry eye disease by all commonly accepted criteria, including positive signs as well as symptoms. They consecutively recruited non-contact lens wearing patients presenting for an ocular examination at two practices in Massachusetts, USA. Recruitment continued until there were 50 patients in a dry eye group and 50 in a non-dry eye group. Inclusion criteria for the dry eye group were: a score greater than 10 with the SPEED questionnaire; fluorescein break-up time ≤ 5 s; corneal and conjunctival staining with fluorescein; lissamine green grade 1 or greater (scale 0 to 3); and Schirmer test with anesthesia ≤ 5 mm. For the asymptomatic group, inclusion criteria were: a score of 0 or 1 point on the SPEED questionnaire; fluorescein break-up time ≥ 10 s; no corneal or conjunctival staining; and Schirmer test ≥ 10 mm. Korb Protocol B was used to test for the presence of LWE.The authors found that, in symptomatic patients, 88% had LWE, of which 22% was grade 1, 46% grade 2 and 20% grade 3. In asymptomatic patients, 16% had LWE, of which 14% was grade 1, 2% grade 2 and 0% grade 3. Although not supplied by Korb et al. (2010), we calculated the odds ratio (± 95% CIs) for symptomatic patients having any LWE compared to the control group to be 38.5 (± 12.3 – 120.4), and for grade 2 or greater to be 95.1 (± 12.1 – 749.7), using the formulae provided by Altman (1991). These odds ratios are very high as a general rule for diagnostic signs and are highly statistically significant (p < 0.0001) for both criteria. Korb et al. (2010) proposed that their data establish LWE as a diagnostic sign of dry eye disease.Yeniad et al. (2010) assessed LWE using fluorescein, rose bengal and lissamine green in 46 patients with dry eye and 40 control subjects. Dry eye was assessed in all participants using a subjective questionnaire and three objective tests – Schirmer I test, fluorescein break up time and fluorescein corneal staining. The prevalence of staining with the three dyes in the dry eye/control groups were: fluorescein 48%/8% (p = 0.001); rose bengal 43%/11% (p = 0.008); and lissamine green 39%/9% (p = 0.006). The authors were unable to demonstrate a correlation between LWE and fluorescein break-up time or Schirmer I test results. It was concluded that LWE should be investigated in patients with symptoms characteristic of dry eye but with normal dry-eye tests.Pult et al. (2011) used Korb Protocol B to examine the capability of LWE and a battery of objective dry eye tests, alone or in combination, to predict the development of dry eye symptoms on a cohort of 47 healthy, non-lens wearers (17 males, 30 females, median age 35 years, range 19 to 70 years). The other dry eye tests investigated were lid-parallel conjunctival folds, non-invasive tear break-up time, tear meniscus height, phenol red tear test, ocular hyperaemia and corneal and conjunctival staining. Symptoms were evaluated using the Ocular Surface Disease Index questionnaire (Walt et al., 1997).LWE was found to correlate with ocular surface disease index scores (r = 0.453, p < 0.001) as well as three of the objective tests: non-invasive break-up time (r = 0.419); and temporal (r = 0.537) and nasal (r = 0.607) lid-parallel conjunctival folds objective tests. LWE was also shown to have acceptable capability (area under the receiver-operator characteristic curve = 0.749) for discriminating between subjects with Ocular Surface Disease Index scores above and below 15. Pult et al. (2011) concluded that the best method for predicting dry eye symptoms was to apply a combination of non-invasive tear break-up time and nasal lid-parallel conjunctival folds. The importance of these findings in diagnosing and monitoring dry eye should be kept in context, because the subjects in this sample constituted a cohort of non-contact lens wearers presenting with ocular signs within a relatively normal range.Sonomura et al. (2014) examined 76 eyes of 76 dry eye patients (mean age: 62.2 years) who were divided into two groups in relation to their Schirmer Ι test values: Group A – Schirmer strip wetting ≤ 5 mm (28 eyes of 28 patients); and Group B – Schirmer strip wetting ≥ 6 mm (48 eyes of 48 patients). Upper and lower LWE was compared between the two groups. LWE scores were significantly greater (all p < 0.009) in Group A than in Group B at the upper lid wiper (Group A: 0.9 ± 1.0, Group B: 0.3 ± 0.6), lower lid wiper (Group A: 0.9 ± 1.0, Group B: 0.3 ± 0.5) and total score for upper and lower lid wiper (Group A: 1.7 ± 2.0, Group A: 0.5 ± 0.9). The authors concluded that LWE scores increase in relation to the tear deficiency.Research to date has centred principally on non-Sj?gren’s dry eye but there is no reason to believe that the findings would not be relevant to Sj?gren’s patients. Of course, there are other signs of this disease and the severity of dry eye is likely to render LWE as a less important sign.All of the studies to date have found an increased incidence of LWE in patients with dry eye symptoms. These results are valuable not just in the context of diagnosis but also with respect to pathogenesis, as will be discussed below.8. LWE associated with contact lens wearA number of studies have explored the relationship between LWE and symptoms of dryness or discomfort in contact lens wearers. This question has generally been approached by assessing the incidence and/or severity of LWE among symptomatic versus asymptomatic contact lens wearers. Here we consider separately studies that have and have not been able to demonstrate a link between LWE and symptoms of dryness and discomfort among contact lens wearers. As mentioned previously, this is a new and emerging area of research, and much of the literature pertaining to clinical aspect of lid wiper epitheliopathy has only appeared as conference abstracts. As a rapid guide as to the source of research we discuss in this review, Table 1 provides a listing of all references cited that describe accounts of clinical studies of the lid wiper using staining agents, and each is identified as either a paper (P) or abstract (A).8.1. Publications demonstrating a link between LWE and contact lens associated dryness/discomfort The original multi-centre LWE study of Korb et al. (2002) prospectively recruited 105 soft contact lens wearers who were assigned into two groups, according to whether they were asymptomatic or symptomatic based on the presence or absence of dry eye symptoms. The prerequisite for admission of patients to the symptomatic group was the presence of ocular symptoms of dryness, grittiness, scratchiness, soreness, irritation, burning or watering occurring within the first 4 hours of wearing their best-fit soft contact lenses. The prerequisite admission of patients to the asymptomatic group was a daily wearing period of 12 hours or more, without any of the dry-eye symptoms cited above.All patients were examined using Korb Protocol A. The mean score for symptoms from the questionnaire was 0.3 points for the asymptomatic group and 7.2 points for the symptomatic group, out of a possible 12 points. Twenty-four (80%) of the 30 symptomatic subjects displayed LWE as compared to 10 (13%) of 75 asymptomatic subjects; these proportional differences were highly statistically significant (p = 0.0001; 95% confidence intervals 51% to 83%). Based on these results, the authors suggested that LWE is associated with symptoms of dry eye in contact lens wearers and went to the extent of claiming that a diagnosis of LWE may explain the conundrum of symptomatic patients with no other clinically objective findings.Berry et al. (2008) randomly recruited 50 experienced hydrogel contact lens wearers from a single clinic. The subjects were grouped into symptomatic (n =19) and asymptomatic (n = 31) patients according to their response to a contact lens dry eye questionnaire (CLDEQ) published by Nichols et al. (2002), and the upper and lower lid wiper region was stained and graded according to Korb Protocol B. The authors found that upper LWE scores were significantly greater in symptomatic versus asymptomatic contact lens wearers (p < 0.035), although there was no such association demonstrated for lower LWE.Pult et al. (2008) investigated the capability of a variety of tests to predict symptoms of dryness during contact lens wear. They examined the right eye of 61 experienced contact lens wearers who were randomly selected from contact lens patients attending a clinic in Germany. The subjects were grouped into symptomatic and asymptomatic patients according to their response to the Contact Lens Dry Eye Questionnaire (Nichols et al., 2002), and the following parameters were assessed: pre-lens tear break-up time, limbal and bulbar hyperaemia, corneal staining, LWE (using Korb Protocol B) and lid parallel conjunctival folds.The authors reported that 38 subjects were classified as asymptomatic and 23 as symptomatic. Severity scores for LWE and lid parallel conjunctival folds were significantly greater in symptomatic versus asymptomatic patients (U-test, p < 0.03), while no significant differences were found between groups for pre-lens tear break-up time, corneal staining or limbal or bulbar hyperaemia. Significant positive correlations were found between scores for LWE and lid parallel conjunctival folds (temporal p <0.001; nasal p < 0.001), and between LWE and bulbar (p < 0.001) and limbal (p < 0.001) hyperaemia. The ‘positive predictive value/negative predictive value/cut-off value’ statistics were recorded as 53.1%/81.1%/≥1 for LWE and 79.8%/86.5%/≥2 for lid parallel conjunctival folds (sum value of nasal and temporal scores). The areas under the receiver-operator characteristic curve were 0.746 for lid parallel conjunctival folds (sum value) and 0.654 for LWE.Pult et al. (2008) concluded that lid parallel conjunctival folds (sum value) is an improved test to predict contact lens-induced dry eye. They noted that the predictive values derived from the receiver-operator characteristic curve indicate that LWE serves better as a test for excluding dry eye symptoms.Yan et al. (2008) assessed LWE using Korb Protocol B and three other objective tests of dry eye – tear break-up time, Schirmer I test and corneal fluorescein staining – on three groups of subjects: Group A: 60 patients (60 eyes) with no dry eye symptoms and normal objective test results; Group B: 51 patients (51 eyes) with dry eye symptoms and normal objective test results; and Group C: 30 patients (30 eyes) with dry eye symptoms and abnormal objective test results. The patients in group A and group B were further divided into groups of those not wearing contact lenses (Groups A1 and B1) and those wearing contact lenses (Groups A2 and B2).The prevalence of LWE was found to be 18.3% (11 patients), 86.3% (44 patients) and 100.0% (30patients) in group A, B and C, respectively (χ2 = 78.26, p < 0.01). The prevalence of LWE was 13.3% (4 patients), 23.3% (7 patients), 81.0% (17 patients) and 90.0% (27 cases) in Groups A1, A2, B1 and B2, respectively. There were no difference in the prevalence of LWE between Groups A1 and A2 (χ2 = 1.00, p = 0.253) or between B1 and B2 (χ2 = 0.85, p = 0.301). These findings suggests an association between LWE and dry eye irrespective of whether or not contact lenses are worn; however, it is not clear if the experiment was sufficiently powered to detect a difference in LWE between those who were and were not wearing contact lenses.Pult et al. (2009) explored the utility of combining tests for LWE (using Korb Protocol B) and lid parallel conjunctival folds plus other tests conducted before contact lens fitting – tear meniscus height, non-invasive break-up time, ocular hyperaemia, phenol red thread test, corneal and conjunctival staining and subjective evaluation – to predict dry eye symptoms during contact lens wear. All of these tests were performed on 33 subjects prior to being fitted with contact lenses for the first time. Subjects were grouped into those who were symptomatic (n = 20) and asymptomatic (n = 13) during the study. LWE was significantly increased when wearing contact lenses (visit 1 versus visit 3; Wilcoxon test; p < 0.016). At the enrolment visit, the subjects who became symptomatic exhibited significantly decreased non-invasive tear break-up time and increased lid parallel conjunctival folds and ocular surface disease index. LWE score, tear meniscus height, ocular hyperaemia and phenol red thread test results were found to be of little value in predicting contact lens induced dry eye.Varsani and Wong (2009) examined the upper and lower lid wiper in 242 patients (105 males and 137 females) aged 18 to 79 years who presented consecutively to a clinic for routine examination. The population included non-contact lens wearers and contact lens wearers, who were characterised as being asymptomatic or symptomatic according to the Ocular Surface Disease Index questionnaire (Walt et al., 1997). Upper and lower LWE was evaluated with lissamine green and graded using the system described by Guillon and Maissa (2008).The authors reported that the type of staining observed was similar (upper lid, p = 0.399; lower lid, p = 0.897) for the non-contact lens wearing and contact lens wearing groups. In the majority of cases, staining was of the ‘thin line’ type (upper lid: non-contact lens wearers, 64%; contact lens wearers, 64%; lower lid: non-contact lens wearers, 54%; contact lens wearers, 66 %). The severity of staining was also similar (upper lid, p = 0.568; lower lid, p = 0.130) for the two groups, with staining most commonly rated as mild or moderate. The type of staining for the population overall and for non-contact lens wearers and contact lens wearers was similar for the asymptomatic and symptomatic sub-groups (p = 0.121 to 338) and ‘thin line’ staining was observed in the majority of cases. The severity of staining was greater for the symptomatic group than the asymptomatic groups overall for both lids (upper lid, p = 0.021; lower lid, p = 0.015), for the upper lid in non-contact lens wearers (p = 0.022), and for the lower lid in contact lens wearers (p = 0.048). Staining type and severity were highly correlated between the upper and lower lid margins (r = 0.812 to 0.867). It was concluded that lid margin staining was similar for contact lens wearers and non-wearers, but more severe for symptomatic than asymptomatic patients in both groups.Yeniad et al. (2010) assessed LWE using fluorescein, rose bengal and lissamine green in 42 symptomatic and 27 asymptomatic contact lens wearers. Dry eye was assessed in all participants using a subjective questionnaire and three objective tests – Schirmer I test, fluorescein break up time and fluorescein corneal staining. The prevalence of staining with the three dyes in the symptomatic and asymptomatic contact lens wearers were: fluorescein 67%/32% (p = 0.001); rose bengal 59%/41% (no statistics provided); and lissamine green 61%/39% (no statistics provided). The authors were unable to demonstrate a correlation between LWE and fluorescein break-up time or Schirmer I test results. The authors concluded that LWE should be investigated in symptomatic contact lens wearers who yield normal results with dry-eye tests.8.2. Publications failing to demonstrate a link between LWE and contact lens associated dryness/discomfortA number of studies have failed to find an association between contact lens associated dryness or discomfort and LWE. Curiously, and most likely coincidentally, these studies all occurred after those studies which have found a difference in LWE between symptomatic contact lens wearers and asymptomatic wearers or non-contact lens wearers.Best et al. (2013) fitted monthly replacement silicone hydrogel contact lenses to 60 subjects who had never worn contact lenses previously. The following objective dry eye tests were performed at baseline (prior to lens wear): non-invasive break-up time (Keratograph and Tearscope Plus), fluorescein break-up time, tear meniscus height, bulbar and limbal hyperaemia, lid parallel conjunctival folds, osmolarity, phenol red test, corneal and conjunctival staining, LWE (using Korb Protocol B) and symptoms using the Ocular Surface Disease Index questionnaire (Walt et al., 1997). These tests were repeated on the 33 subjects who returned after having worn the lenses on a full-time basis for 6 months. At baseline, LWE did not correlate with ocular surface disease index scores, and correlated with only one of the subjective tests – the phenol red test (r = 0.257, p = 0.045). Lid wiper staining scores increased from 0.3 ± 0.7 at baseline to 1.5 ± 1.2 at 6 months (p = 0.002). LWE was unable to discriminate between those who dropped out of lens wear versus those who completed the 6 months lens wearing trial (area under the receiver-operator characteristic curve = 0.511). Read et al. (2014) assessed lissamine green staining of the upper and lower lid wiper in 10 symptomatic lens wearers, 10 asymptomatic lens wearers and 10 non-lens wearers, at afternoon visits. Lens wearers wore their habitual contact lenses and reported 0 to 100 comfort scores. The authors found that lissamine green staining of the upper lid wiper was greater for the lens wearers vs. non-wearers (p = 0.03); however, there was no difference between symptomatic and asymptomatic wearers (p = 0.27). The authors concluded that inspection of the lid wiper region with lissamine green staining does not appear to differentiate between symptomatic and asymptomatic contact lens wearers.Navascues-Cornago et al. (2015b) measured lid wiper staining (lissamine green) using image analysis in 35 soft contact lens wearers and 35 non-wearers. Measurements were repeated at 12 hours in 11 non-wearers and 10 symptomatic lens wearers. Lower lid wiper staining was greater than upper lid wiper staining in both groups (all p < 0.05), confirming the previous observations of Golebiowski et al. (2012). Among lens wearers, a significant increase in lower lid wiper staining was found at 12 hours compared with the morning (p = 0.04). Lid margin staining was not significantly different between symptomatic and asymptomatic lens wearers (Mann–Whitney U test, all p > 0.05).In a large multi-centre study, Schulze et al. (2015) assessed subjective comfort in 246 participants wearing their habitual soft contact lenses. After lens removal, the lid wiper was graded using Korb Protocol B. The authors were unable to demonstrate a relationship between comfort scores and the severity of LWE. 8.3. Meta-analysis We conducted a meta-analysis of the relationship between LWE and contact lens comfort, using data extracted from six studies relating to different aspects of soft contact lens wear. A total of 587 subjects were enrolled in these studies, comprising 180 males and 407 females, age 28.5 ± 7.8 years (range 18 to 62 years) with an ethnicity distribution as follows: White 72%, Asian 18%, Other 10%. LWE was assessed according to Korb Protocol B, and comfort was assessed using a validated patient reported outcomes instrument known as the ‘contact lens user experience’ (CLUE) metric (Wirth and Riley, 2011). Scores with this instrument follow a normal distribution with a population average score of 60 ± 20, where higher scores indicate a more favorable/positive response and lower scores a less favourable/negative response.Only data from the first period of each study were used in the analysis. The Mean LWE grade of the two eyes was used as the outcome so that there was one grade per subject, aligning with the structure of the CLUE scores (one score per subject). A linear mixed model with both fixed effects (age, race, sex, and treatment) and random effects was used.The mean ± standard deviation (minimum - maximum) metrics recorded were: CLUE 56.3 ± 7.5 (13.1 - 97.9); and LWE 0.72 ± 0.11 (0.00 - 1.20). We were unable to demonstrate a significant relationship between grade of LWE and CLUE score (F = 1.21, p = 0.298) (Fig. 7). 8.4. SummaryThere are roughly equal numbers of papers that have found LWE to be greater in contact lens wearers as those that have not. However, it should be kept in mind that reports of the absence of a statistically significant difference could be due to methodological limitations such as the experiment being insufficiently powered (e.g. sample size too small) or one or more of the techniques used to assess LWE or comfort/dryness being too insensitive to detect subtle differences. In addition, it is unclear whether LWE is an acute or chronic condition, or whether there is a predictable diurnal variation in this condition. Accordingly, differences between studies in respect of the length of time that lenses have been worn, or the time of day measurements are recorded, may have confounded the capacity to detect differences or associations.9. PathologyCritical to the development of a management strategy for any adverse condition is gaining an understanding of the pathological changes occurring in the affected tissue, and any adverse functional correlates. Preliminary studies employing immunocytochemistry, impression cytology, functional slit lamp biomicroscopy and corneal confocal microscopy have shed some light on the likely tissue pathology underlying LWE, and aesthesiometry of the lid wiper margin has highlighted possible functional deficits associated with this condition. 9.1. ImmunocytochemistryJalbert et al. (2012) performed immunocytochemistry for the keratinization-related proteins filaggrin, transglutaminase1 and cytokeratin 1/10 on impression cytology samples collected from a broad region of the upper lid margin. The authors demonstrated expression of all of these proteins from lid margin epithelium. Filaggrin is known to be upregulated in superficial and intermediate layers of keratinised epithelium of diseased bulbar conjunctiva and does not appear to be expressed in normal bulbar conjunctiva (Nakamura et al. 2001). Similarly, transglutaminase1 gene and protein expression are reported to accompany pathological keratinization of conjunctival epithelial cells (located in or above the supra-basal region) in severe Stevens Johnson syndrome. This protein is absent, or only slightly expressed, in normal ocular surface epithelia (Nakamura et al., 2001). Cytokeratins are expressed in the superficial and intermediate layers of diseased conjunctiva epithelia. It is not found in any layers of the normal conjunctival epithelium (Tanioka et al., 2010). Expression of these proteins in impression cytology samples taken from the lid wiper region therefore suggest the possibility of pathological processes occurring in superficial epithelial cells of the lid wiper. 9.2. Structural changesA sensible starting point for considering changes in the lid wiper epithelium in patients with LWE is interpretation of the staining observed with different dyes. However, such considerations are limited because the meaning of corneal staining in adjacent ophthalmic tissues such as the cornea and conjunctiva is far from complete. Here we shall review what is known about the relationship between the appearance of staining with different dyes and cellular changes in the cornea and conjunctiva, and will try to extrapolate this to explain what might be happening in the lid wiper. Recent evidence from studies using invasive techniques (impression cytology) and non-invasive approaches (laser scanning confocal microscopy and functional slit lamp biomicroscopy) will also be considered.9.2.1. Lid wiper epithelial cell morphologyAs outlined in detail above, clinical examination of the lid wiper has been largely restricted to observations of patterns of staining following instillation of fluorescein, lissamine green, rose bengal and various combinations of these staining agents. Interpretation of these staining patterns is confounded as a result of significant gaps in our understanding of the relationship between staining patterns observed and the specific nature of the tissue abnormality (Morgan and Maldonado-Codina, 2009).Much attention has been directed in recent times to our understanding of tissue changes associated with staining of the cornea that has been subjected to tonicity, chemical or mechanical insult. A further challenge is the translation of research into mechanisms of corneal staining to what might be happening in the conjunctiva.In the cornea, healthy cells can be observed to take up fluorescein when viewed with specific excitation and emission filter sets rather than the insensitive cobalt blue filter often used at the slit lamp biomicroscope (Bandamwar et al., 2014; Thinda et al., 2010). Punctate spots have been shown to be individual hyperfluorescent epithelial cells, which predominate in the superficial epithelial layers (Mokhtarzadeh et al., 2011). Using high magnification photography through a slit lamp biomicroscope, Maldonado-Codina et al. (2014) examined eyes with solution-induced corneal epitheliopathy and noted that the size of the small punctate dots (22 ?m in diameter) induced by the solution was broadly consistent with the known diameter of superficial human corneal epithelial cells, which range in size from 20 to 37 ?m (Romano et al., 2003). The same region of the cornea was also imaged after instillation of fluorescein, rose bengal and lissamine green. Identical patterns of cells could be observed irrespective of the dye used, indicating that all three dyes identify the same damaged cells.Fluorescein staining properties of corneal epithelial cells under normal and stressed conditions were studied by Bandamwar et al. (2014) using human corneal limbal epithelial cell cultures and rabbit organ culture models. The cultures were stressed by exposing them to hypotonicity, hypertonicity, preservatives, and scratch and alkaline wounding. In addition to fluorescein, cells were stained with Hoechst-33342, Propidium-iodide, and Annexin-V to identify live, dead and apoptotic cells, respectively. Observations with the clinical slit lamp biomicroscopy and fluorescence confocal microscopy facilitated identification of the association between patterns of staining and corresponded cellular damage.The authors found that micropunctate fluorescein staining identifies individual cells that are undergoing apoptosis, whereby the cells have internal fluorescein and are hyper-fluorescent relative to neighbouring cells. Normal, healthy cells also take up fluorescein, but fluoresce less brightly than apoptotic cells. Dead cells do not fluoresce appreciably. It was concluded that fluorescein can be considered as a clinical probe, capable of indicating cells in the corneal epithelium that are undergoing apoptosis (Bandamwar et al., 2014).Rose bengal is thought to stain dead as well as ‘devitalised’ cells, i.e. cells which have lost their protective mucin coating (Feenstra and Tseng, 1992). It is widely accepted that lissamine green and rose bengal stain the same cells (Efron, 2013; Machado et al., 2009). It has also been suggested that disruption of the mucin layer can lead to fluorescein staining in corneas which have undergone cell damage after impression cytology (Thinda et al., 2010).Doughty and Hagan (2013) assessed the staining of human (Chang) conjunctival epithelial cells in culture with rose bengal and lissamine green. They observed that the conjunctival cultures, which were indicative of flattened epithelial cells, stained very intensely with rose bengal, with the nucleus of the cells showing the most notable dye uptake. Cultures showing such intensive staining with rose bengal were consistently non-staining with lissamine green. Although it is only possible to surmise how these observations translate to the lid wiper epithelium in the living human eye, the notion that conjunctival cells in vitro stain more intensely with rose bengal than lissamine green suggests that these dyes may be revealing different aspects of conjunctival pathophysiology.Extrapolating these findings to explain staining of the lid wiper is difficult because the epithelium of the cornea and lid wiper have distinct differences, despite having the same embryonic origin (surface ectoderm). The corneal epithelium is stratified, squamous and non-keratinised and consists of 5 or 6 layers of cells, whereas the lid wiper epithelium has a transitional, stratified structure with some squamous cells at the beginning, a conjunctival structure of cuboidal cells, some parakeratinised cells and goblet cells, and consists of 8 to 12 layers (see Section 4.2.2). Nevertheless, the established principle in relation to the corneal epithelium – that cells uptake some fluorescein and hyperfluorescence cells are undergoing apoptosis – may also apply to the lid wiper epithelium. There have been no reports in the literature of light microscopy evaluations of excised tissue samples of the lid wiper in human patients with LWE. The only detailed examination of the lid wiper is that undertaken by Knop et al. (2011b) on 14 eyes harvested from 10 cold stored human body donors with a macroscopically normal ocular surface. The authors noted that, in selected cases, cells with atypical keratinization (parakeratinization) increase in number and extend from the natural stainable line of Marx, where they physiologically occur, over the surface of the lid wiper epithelium. This is a possible indication of the type of pathological change that might occur in LWE.Certainly, the detailed observations of Knop et al. (2011b) on normal lid wiper tissue will constitute a useful reference against which tissue morphology in LWE can be evaluated, at least in elderly eyes. Perhaps samples of the lid wiper of younger patients – say, aged 20 to 30 years, in accordance with the age demographic of contact lens wearers (Morgan et al., 2010) – suffering from LWE could be obtained using micro-biopsy techniques such as those developed for examining corneal tissue (Kompa et al., 1999).An analogy for changes in the human lid wiper perhaps can be drawn from studies of the mouse palpebral conjunctival epithelium, which according to Henriksson et al. (2013) is structurally similar to that in humans. Henriksson et al. (2013) observed that, following exposure to experimental ocular surface desiccating stress to mouse eyes, the palpebral conjunctival epithelium decreased in thickness and goblet cell access to the surface seemed to be inhibited by surrounding epithelial cells, potentially slowing down their migration to the surface. Differential staining with periodic acid-Schiff and Alcian blue suggests that there may be different subtypes of conjunctival goblet cells. It remains to be demonstrated whether such changes also occur in LWE.Jalbert et al. (2015) performed impression cytology on the lid margin, including the lid wiper, in 19 soft contact lens wearers and 21 non-lens wearers, and observed lissamine green staining in 17% of subjects. There was no difference between contact lens and non-contact lens wearers for lid margin staining or Nelson grade (p = 0.4, Fisher’s exact test). The authors did not report any differences between samples obtained from those who did and did not display lissamine green staining of the lid wiper region in respect of cell morphology, histochemistry or immunocytochemistry.Knop et al. (2011b) conducted a detailed analysis of the lid wiper of healthy eyes of four volunteers using laser scanning confocal microscopy and was able to clearly image a variety of normal cellular structures in this tissue. Clearly, this approach could be applied to examine the lid wiper in vivo in those with LWE.9.2.2. Fractal dimensions of the lid wiper microvascular networkWang et al. (2015) conducted a preliminary quantitative analysis of the microvascular network of the lid wiper area imaged using functional slit lamp biomicroscopy (Jiang et al., 2014). This technique involves capturing high resolution digital images of the tissue of interest (in this case, the lid wiper) and using custom software for automated segmentation of the microvascular network. The analysis determines the ‘fractal dimension’ D0, which in essence defines the degree of ‘density and complexity’ of the vascular tissue in the selected area of the image.The authors used this technique to capture two images from one eye of each of three habitual contact lens wearers and three non-lens wearing control participants. They recorded lid wiper D0 values of 1.358 ± 0.213 in contact lens wearers and 1.667 ± 0.116 in non-lens wearers; the difference was weakly significant (p = 0.057). The authors noted that the vascular network of the lid wiper in contact lens wearers appeared to be loose compared to that in healthy control subjects. Wang et al. (2015) suggest that their finding of reduced fractal dimensions in the lid wiper of lens wearers may be due to lens-induced surface damage of the lid wiper. This result of Wang et al. (2015) is counter to expectations based on the original observation of Korb et al. (2002) who noted that vascular injection (hyperaemia), observed with the slit lamp biomicroscope, might be suggestive of LWE. Korb et al. (2002) stated, however, that vascular changes were usually subtle and difficult to observe, except when the condition is severe.The approach of Wang et al. (2015) has promise for assessing the impact of contact lens wear on the lid wiper, but further studies need to be conducted on larger cohorts to properly validate the technique and to more accurately define the relationship between fractal dimensions of the lid wiper microvascular network and LWE. 9.3. Functional changesPresumed tissue structure compromise indicated by staining of the lid wiper may be associated with functional alterations. A functional attribute of the anterior ocular structures that can be readily investigated is sensitivity to touch. Navascues-Cornago et al. (2015c) measured mechanical sensitivity at 8 locations on the ocular surface and adnexa (cornea, upper and lower lid wiper and bulbar and tarsal conjunctivae) of 35 soft contact lens wearers and 35 non-wearers using a Cochet–Bonnet aesthesiometer. The lid wiper was found to be the most sensitive conjunctival region (all p < 0.001), which is consistent with a previous report (McGowan et al., 1994). Previous morphological studies investigating the sensory innervation of the human eyelid have revealed a greater density of nerve endings present in the lid margin area (Munger and Halata, 1984), which may explain the increased sensitivity of the lid wiper compared with that of the tarsal conjunctiva.Measurements were repeated by Navascues-Cornago et al. (2015c) at 12 hours in 11 non-wearers and 10 symptomatic lens wearers. At this time, the sensitivity of the lower lid wiper, but not the upper lid wiper, was found to be reduced in lens wearers compared with non-wearers (all p < 0.05). The upper lid primarily travels in a vertical direction during blinking and the vertical distance traversed by the lid margin is much greater than that of the lower lid; however, the lower lid translates in short horizontal movements during blinking (Doane, 1980). This lower lid horizontal shift may result in a different physical interaction with the anterior cornea or contact lens surface than occurs with predominantly large vertical sweeps of the upper lid, possibly differentially impacting lid wiper sensitivity.This finding of Navascues-Cornago et al. (2015c) is contrary to previous studies – albeit on a different but adjacent tissue (the cornea) – that have shown an increase in sensitivity throughout the day (Du Toit et al., 2003; Millodot, 1972). Methodological differences may account for this discrepancy; Navascues-Cornago et al. (2015c) measured lid sensitivity 2 hours after eye opening whereas Du Toit et al. (2003) and Millodot (1972) measured corneal sensitivity immediately after awakening, at which time there may have been a greater impact on sensitivity.Navascues-Cornago et al. (2015c) reported that the wear of contact lenses over a 12-hour period hadno significant impact on mechanical thresholds of the surface of the lid wiper in symptomatic subjects. Although reduced sensitivity of the lid margin with rigid gas-permeable, polymethyl methacrylate and conventional hydrogel contact lenses has been reported by Lowther and Hill (1968), it is unclear whether the specific region of tissue assessed included the lid wiper in those studies.The cause of changes in ocular surface sensitivity during contact lens wear has been attributed to a variety of possible factors, such as metabolic disorders caused by lens-induced hypoxia (Millodot et al., 1979; Velasko et al., 1994), neural adaptation to mechanical stimulation (Lum et al., 2013; Millodot et al., 1979;) and neural sensitization in response to inflammation or hyperosmolarity (Situ et al., 2010). Either or both of the neural factors may play a role in altering lid wiper sensitivity but not ‘lens-induced hypoxia’, because the eyelids do not become hypoxic during contact lens wear.10. AetiologyThe primary hypothesis for the aetiology of LWE can be summar as increased friction between the lid wiper and ocular or anterior contact lens surface due to inadequate lubrication. Numerous theories have been advanced in relation to the mechanism; some are based on solid evidence, some on preliminary evidence and some are conjecture. These theories relate to either the association between LWE and dry eye or between LWE and contact lens induced discomfort/dry eye and some relate to both. Here we shall critically review the relevant literature, identifying proposals which are evidence-based and those which are conjecture.In their original paper, Korb et al. (2002) hypothes that LWE with soft contact lens wear is “most likely a product of the characteristics of the tear film between the lens surface and the lid wiper”. The presence of the lens alters the normal structure of the tear film, which in turn could result in “physical trauma and mechanical abrasion to the epithelial cells of the lid wiper, disclosed only by staining of those cells”. This theory was later general to non-contact lens wearers (Korb et al., 2005), whereby it was suggested that an optimal lid wiper arrangement would require an adequate “lubricating layer at the interface between the opposing tissues to dispel the shear and frictional forces”, thereby preserving tissue integrity. Such conjecture leads to discussion of evidence for the role of friction in LWE, the structure of the tear film under the lid wiper and the tribology of blinking and eye movements.10.1. Friction and contact lens wearWhile the finding of LWE described in the papers of Korb et al. (2002, 2005) was novel, the previous literature was not entirely silent on the mechanical nature of blinking. The potential importance of friction in contact lens wear was recogn as early as 1936 by Feinbloom (1938). Ehlers (1965a) certainly alluded to ‘friction during blinking” in his seminal work. Nairn and Jiang (1995) anticipated that frictional forces during blinking with a contact lens on the eye might impact comfort and went on to make measures of the coefficient of friction of contact lenses. Numerous others have since measured contact lens surface friction. Nonetheless, the practical implications of these measures have not been quantified until recently. It is not practical to measure the frictional forces at play under the human eyelid during blinking in either dry eye or normals. Somewhat serendipitously, insight is provided by studies of comfort during wear of contact lenses with different surface coefficients of friction.Although peer-reviewed articles on the topic are yet to appear, four separate conference abstracts provide consistent evidence of a relation between contact lens surface friction and wearing comfort. Brennan (2009) reported a negative correlation (r = -0.90, p < 0.01) between end-of-day comfort scores obtained with visual analog scales after one month of wear of five silicone-hydrogel contact lenses; the coefficient of friction measures of these lenses were obtained from a conference poster presentation by Ross et al. (2005). The presentation of this data in combination with the LWE results from Korb et al. (2002) has provided strong impetus for further research investigating the potential role of contact lens surface friction and the lid wiper in explaining contact lens discomfort and dryness symptoms (for example, see Coles and Brennan, 2012; Jones et al., 2013a; Kern et al., 2013; Pult et al., 2015; Samsom et al., 2014; Wilson et al., 2015).Coles and Brennan (2012) expanded on the earlier presentation of Brennan (2009) by using different coefficient of friction data (Roba et al., 2011) which increased the number of lenses with both friction and comfort data available for analysis to 6 silicone-hydrogel and 3 hydrogel materials. End-of-day comfort data from over 700 one-month wearing trials were available for analysis. Again, a strong correlation between comfort and coefficient of friction was evident (p< 0.01) (Fig. 8). Regression analyses excluded oxygen transmissibility, modulus and water content as predictors of comfort, leaving only coefficient of friction as a significant factor.Kern et al. (2013) similarly conducted a meta-analysis to examine the association between subjective comfort and contact lens coefficient of friction, albeit on fewer lenses with a smaller sample size. They used yet another set of friction data, measured with the inclined plane method of Tucker et al. (2012), replicating the finding of a highly statistically significant correlation between coefficient of friction and a number of different comfort measures.Brennan and Coles (2013) used comfort data available from the staining grid website of Andrasko (2013) and correlated these data with the friction data from Roba et al. (2011). The staining grid presents corneal staining measurements with 9 lens brands combined with 14 care solution brands with a sample size of 29 to 30 in each block of the grid. Fortuitously, Andrasko (2013) also measured comfort after 2 hours of wear with these lens-solution combinations. Coefficient of friction was found to be highly correlated with the mean comfort measures (r2 = 0.91, p = 0.0002)Although not published in peer-reviewed journals, these data provide considerable support for the role of friction in ocular comfort. In total, there are three separate independent combinations of friction and comfort data – Brennan (2009)/Ross et al., (2005); Kern et al. (2013)/ Tucker et al. (2012); and Andrasko (2013)/ Roba et al. (2011) – with each combination being derived from different methodologies for both friction and comfort measurement. Such variations in technique and data source might be expected to provide differing results, yet the data convey a consistent pattern. It should be borne in mind that the correlations are not necessarily causal and a separate common factor (for example, lens modulus or edge design) may also be determinants of comfortable contact lens wear.10.2. TribologyCharacterising the role of the lid wiper and the aetiology of LWE requires consideration of the principles of tribology, which is the science and engineering of interacting surfaces in relative motion. It includes the study and application of the principles of friction, lubrication and wear. In classical tribology, the relationship between coefficient of friction, load, relative speed and lubricious fluid properties – in particular viscosity – has been described in the well-known Stribeck curve (Jacobson, 2003). In this model, three different regimes were identified: boundary lubrication, dominated by the close contact of the solid surfaces; the mixed regime, where occasional contact between the solid surfaces occurs; and the hydrodynamic regime, where a full lubricant film is present between the two surfaces moving relative to each other. Boundary friction appears principally to be a function of the local surface microarchitecture, whereas in hydrodynamic friction, both surfaces are fully separated and friction depends on the viscosity of the fluids between the surfaces. Ehlers (1965b) recogn that the epithelial cells of the eyelid and the globe are subject to shear forces (friction) during blinking and discussed classical lubrication theory when describing this interaction. He concluded that boundary lubrication must apply to the lid margin zone – specifically, that of the mucocutaneous junction rather than the currently proposed lid wiper region. He noted “The movement is here a mutual rubbing between the cornea and the squamous-epithelium-lined part of the lid in the immediate neighbourhood of the limbus”. Korb et al. (2005) also recogn the application of lubrication theory to the lid wiper phenomenon, stating “… it would be expected that hydrodynamic lubrication is the mechanism operating in the healthy eye, whereas boundary lubrication would manifest in dry eye conditions and other ocular disease states in which alterations of the normal surface cytoarchitecture unfavorably change the relationship between the volume of fluid available for lubrication and the tissue surface requiring lubrication.”These simple approaches to understanding lubrication of the lid wiper do not adequately communicate the complexity of the system. Numerous authors have described models of the tear layer at the lid wiper-cornea interface based on theoretical considerations (Cher, 2008; Dunn et al., 2013; Jones et al., 2008; Pult et al., 2015; Radke and Chauhan, 2008). A further construct – the elastohydrodynamic domain – is described as lying between the mixed and hydrodynamic regimes and has been applied to the study of the normal tear film and the tear film in contact lens wear (Dunn et al., 2013; Jones et al., 2008). A number of important features impact our ability to fully model the friction at the lid wiper, including lack of knowledge of the components and structure of the tear film at the lid wiper-cornea interface, variation in the pressure of the upper eyelid during the blink, the presence of glycoprotein surface brushes arising from transmembrane mucins, and variable viscosity of the tear fluid arising from the interaction of different mucins and the aqueous. To further explore the biotribology at the lid wiper, a discussion of the normal composition and structure of the tear film is presented below.10.3. Tear filmIt is beyond the scope of this paper to fully detail all that is known about the tear film and its dynamics. This discussion concentrates on factors that are relevant to friction between the lid wiper and ocular or anterior contact lens surface due to inadequate lubrication. 10.3.1. Overall structureThe classical view of preocular tear film structure is of a multi-layered structure – a thin, visco-elastic lipid layer at the tear-air interface, a thicker aqueous layer and a mucin layer overlying the ocular surface epithelium. More recently, researchers have proposed that the mucin and aqueous components, in the absence of a distinct interface between them, co-exist in a mucoaqueous layer with a gradient concentration of mucin decreasing away from the epithelial surface (Cher, 2000). In the conventional understanding of the lipid layer, McCulley and Shine (1997) reported that it was composed of two phases – a thin polar phase abutting the junction with the aqueous-mucin phase and a thicker non-polar phase between the polar phase and the air interface. They associated low levels of certain polar phospholipids with evaporative dry eye. The international workshop on meibomian gland dysfunction agreed that the lipid layer functions to slow evaporation of the aqueous component (Nichols et al., 2011). However, the microstructure, composition and role of the lipid layer remains controversial. In a critical review, Millar and Schuett (2015) concluded that, while a normal tear film resists evaporation, an emerging idea from the literature is that the main purpose of the lipid layer is “ ... to allow the spread of the tear film and to prevent its collapse onto the ocular surface, rather than to be an evaporative blanket.” Further, the role of meibomian gland phospholipids in creating an interphase between the outer nonpolar lipid layer and inner aqueous layers remains controversial (Geogiev, 2015; Pucker and Haworth, 2015).In their nanoscale study of the tear film, Khanal and Millar (2010) observed markedly different flow dynamics in the various corneal layers. Lipophilic quantum dots moved in a separate manner to hydrophilic quantum dots consistent with separation of lipid and aqueous layers. The lipophilic quantum dots dispersed quickly but did not drain from the ocular puncta. Rather, they appeared on the skin and eyelashes, suggesting to the authors that lipids are likely removed from the eye via the eyelashes and the inner canthus where they are periodically wiped away as we rub our eyes. Numerous authors have described models of the tear layer at the lid wiper-cornea interface based on theoretical considerations (Cher, 2008; Jones et al., 2008; Pult, et al. 2015; Radke and Chauhan, 2008). These authors assume that the lipid layer does not extend beneath the eyelid. Given these models, the data from Khanal and Millar (2010), and an absence of data suggesting lipid contamination of the tear film at the lid wiper-cornea interface, no further discussion of the lipid layer is warranted with respect to the aetiology of LWE – with the exception of its possible role in the development of evaporative dry eye (Bron, 2015).The aqueous component of the tears is largely derived from the exocrine lacrimal glands, the major secretory source of the fluids, proteins and salts that bathe and protect the cornea and ocular surface. Lacrimal secretions are mixed with the preocular film during the blink and drained via the puncta. Some water is also lost by evaporation. Reduced production of aqueous by the lacrimal gland is considered one of the major causes of dry eye (Bron, 2015). 10.3.2. MucinsMucins have taken longer to characterize than other tear proteins because of their heavy glycosylation. As well as having other functions, mucins are thought to be involved in hydration and lubrication of the ocular surface. They are very large glycoproteins generally classified as either transmembrane or secreted. Three transmembrane mucins, MUC1, MUC4, and MUC16, at the ocular surface have been well character (Gipson, 2004; Hori et al., 2004;). They are found at the anterior-most side of the ocular surface epithelium, extending from cell surface microplicae as stiff rods to form the glycocalyx. Additional transmembrane mucins found at the conjunctival surface – MUC13, MUC15, and MUC17 – are less well studied. Secreted mucins can be further subcategor as gel-forming or soluble. MUC5AC, a gel-forming mucin, is produced by the goblet cells of the conjunctiva. MUC7, a soluble mucin, is synthes by the lacrimal gland and by the stratified epithelium of the conjunctiva (Gipson, 2004). Because of its solubility, it can largely be considered to be dispersed evenly within the aqueous of the tear film (Radke and Chauhan, 2008). It was thought that goblet cell derived mucins were the primary source of ocular mucins but more recent data suggests that the signal for mucin gene expression in human conjunctival epithelium is stronger overall for membrane associated mucin (Woodward and Argueso, 2014).The transmembrane and gel-forming mucins form a complex dynamic hydrated layer. MUC5AC is thought to be connected largely by weak non-covalent interactions to the surface glycocalyx, which is comprised of transmembrane mucins. The gel likely disintegrates and is restored during each blink (Gipson, 2004; Yokoi et al., 2014). The result is likely a dynamic gradient of mucin concentration from the epithelial surface through the aqueous to the lipid, with variability of the gradient dependent on the blink. The mucin molecules, with their characteristic bottlebrush structure, immobilize large amounts of water within the contact region, while the backbone provides interconnections to the other mucins or to the corneal surface. Such ‘hairy’ polymers impart a remarkable lubricating effect with the resulting fluid-like cushioning layer capable of withstanding substantial externally applied pressure, thereby lowering frictional forces. This behavior is ascribed to interchain repulsion which leads to the incorporation of large quantities of aqueous rather than being an intrinsic property of either component (Lee and Spencer, 2014).Khanal and Millar (2010) offered a practical demonstration of different flow dynamics in the mucin layer compared to the aqueous. They applied hydrophilic quantum dots to the tear layer and some of these seemed to adhere to the mucin layer in a mesh like pattern. These dots did not mobilize rapidly and multiple blinks were required to clear them. On the contrary, other dots flowed immediately on the blink to the tear prisms and remained there until drained through the puncta. The authors took this to indicate the presence of a thick glycocalyx at the epithelial surface. The structure of the tear film between the lid wiper and cornea is much less certain, not least because of difficulties in imaging this region in vivo. The findings of Khanal and Millar (2010), described above, are consistent with the restriction of lipids to the interpalpebral aperture. The structure of the tear film is thus likely to consist of a mucin layer at each surface with a certain amount of aqueous in between, when there is no contact lens in place. Unknown variables within this structure include the composition of the mucins in the layer, the local aqueous content, the thickness of the layer and the viscosity gradient in the region. When a contact lens is in place, there is also uncertainty as to the type and amount of mucin that may be found on the surface of the lens in the lid wiper region.Berry et al. (2008) proposed that – given the assumption that LWE is associated with frictional and mechanical forces during blinking – insufficient mucins, or an altered composition of the resident mucins at the ocular surface, may result in increased friction between the lid wiper and ocular or contact lens surface. This in turn could lead to reduced comfort.To test this hypothesis, Berry et al. (2008) assessed comfort levels using the Contact Lens Dry Eye Questionnaire (Nichols et al., 2002) and LWE (Korb Protocol B) in the right eyes of 50 experienced lens wearers (19 men, 31 women; mean age 32.1 ± 11.4 years). Thirty-one subjects were classified as asymptomatic and 19 as symptomatic by the questionnaire. LWE was significantly higher (p ≤ 0.035) and MUC5AC reactivity was significantly lower (p < 0.050) in the symptomatic group. MUC4 correlated to LWE (r = 0.46; p < 0.01). The authors concluded that increased friction might follow from insufficient mucins, or an altered composition of the resident mucins at the ocular surface.10.3.3. OsmolarityMcMonnies (2015) proposed that the upper and lower conjunctival sacs, upper and lower menisci, and exposed and over-exposed regions of the ocular surface can be considered as compartments with corresponding differences in osmolarity. In normal eyes, there appears to be a progressive increase in osmolarity from the isotonic freshly produced tears in the upper conjunctival sac, to the upper meniscus, the upper exposed ocular surface, the lower over-exposed ocular surface, the lower meniscus and the lower conjunctival sac, with the last having the highest osmolarity. McMonnies (2015) suggests that these compartmental differences may be exaggerated in dry eye disease, especially when hyperosmotic tears are delivered to the upper conjunctival sac. Stahl et al. (2011) explored the relationship between osmolality, comfort and LWE (assessed with lissamine green) in 20 contact lens wearers in a randomised, masked study. Subjects each wore two unidentified lenses (lenses ‘A’ and ‘B’). Comfort (Lens A, 77 ± 24; Lens B, 79 ± 23; p > 0.05), tear osmolarity (Lens A, 300 ± 15; Lens B, 293 ± 10 mmol/L; p > 0.05) and contact lens osmolality (Lens A, 380 ± 90; Lens B, 351 ±58 mmol/kg; p > 0.05) were not affected by lens type. A relationship between tear or lens osmolality, comfort and LWE at the upper lid margins could not be demonstrated.Although friction-related damage appears to be the primary basis for upper LWE, tear hyperosmolarity in the exposed areas may also have a role. A positive correlation was observed by Golebiowski et al. (2012) between upper LWE and tear osmolarity measured in the inferior meniscus (r = 0.41, p = 0.004). Blink-related excursions for the lower lid wiper are much less than for the upper lid wiper. McMonnies (2015) argues that, as consequence, there is less opportunity in the lower lid wiper for friction-related damage, and any epitheliopathy observed in this tissue structure is more likely to be due to hyperosmotic insult. This theory conflicts with the suggestion of Shiraishi et al. (2014) that lower LWE is due to small repeated lateral lower lid excursions (see above) and is inconsistent with the observations of Stahl et al. (2011).10.4. Biotribology of blinkingCher (2003) discussed the concept of blink-related microtrauma, which he defined as a pathological process that results mechanically from blinking under prolonged unphysiologic conditions. Although Cher (2003) introduced this concept to explain various ocular conditions such as superior limbic keratoconjunctivitis, filamentary keratitis, blepharospasm, severe ptosis, canthal/palpebral froth, affections from disordered eyelid lining and contact lens related damage, he discusses extensively many of the principles introduced by Korb et al. (2002), such as interfacial friction between opposing mucosal ocular surface structures, the movement of the eyelids over the ocular surface during blinking, eyelid pressure, frictional effects and lubricity; however, he did not use the term ‘lid wiper’ in his review. The initial paper by Korb et al. (2002) was published in April 2002, and Cher (2003) published his review 11 months later, in March 2003. Furthermore, Cher briefly introduced his idea of blink-related microtrauma (although he did not specifically use this term) three years earlier (Cher, 2000), so it is clear that Cher conceived the concept of blink-related microtrauma prior to the first publication describing LWE of Korb et al. (2002). The apparently independent thinking of Cher (2000, 2003) and Korb et al. (2002) at about the same point in time, introducing new concepts with a common basis of frictional effects with blinking, is remarkable. Pult et al. (2015) have applied these tribological principles to model the dynamic interaction between the lid wiper and the ocular or contact lens surface during normal and abnormal blinking and dry eye situations. They begin by pointing out that the classic form of the Stribeck curve is not strictly applicable to the process of blinking, because of the presence – in a healthy tear film – of the glycocalyx and the gel-forming mucin brush-like structure on the ocular surface and lid wiper. The expected high coefficient of friction at low blink speeds, caused by the solid contacts, does not occur because of the presence of the brush. In addition to the hydrophilic polymer brushes formed by the mucins, the tribological model of the eye may be influenced by: (a) physical parameters, such as lid pressure (see below), blink speed, surface roughness and texture; (b) material parameters, such as the elastic modulus of the tissue or contact lenses; and (c) tear film composition, which ultimately has an influence on tear film viscosity and lubricant thickness. Tear film viscosity is also influenced by temperature (Pult et al., 2015).In healthy eyes, friction between the sliding partners – the cornea and lid wiper or contact lenses and lid wiper – is independent of the surface of the partners when moving at high velocity, since full fluid film lubrication is operating. Contact lenses with higher intrinsic lubricity may not induce wear on the lid wiper in high sliding velocity either, because of the presence of full fluid film lubrication (Pult et al., 2015). In contrast, at low velocity, well-formed brushes are vital to reduce the coefficient of friction, particularly at the beginning and end of blinks, as well as at the reversal points of eyelid movement, and perhaps during other low-velocity eye movements. In dry-eye patients, the collapsed brushes appear to be the dominant factor in determining friction between the sliding partners. An additional component of shear stress in blinks may be a delayed transition to full fluid film lubrication and viscous forces that are too great at high sliding velocities. Failure of a proper transition may be due to the partner moduli, geometry of the lid wiper, blink speed, collapsed brushes, improper surface roughness, tear film viscosity, low film thickness, or evaporation (Pult et al., 2015) (Fig. 9).The markedly increased viscosity of the tear film in dry eye patients will result in much higher viscous fluid shear stress between contacting partners than in healthy patients. Pult et al. (2015) suggest that this may explain why the stained area of the lid wiper in LWE is much larger than the assumed touching zone. Composition of the tear film appears to be vital and can also affect the non-Newtonian properties of the tear film. The effect of the increased viscous shear on the ocular surface and lids may be compounded by the increased blink rate in dry eye patients. The coefficient of friction of the back surface of the contact lens may play a dominant role, since the friction that occurs during small movements and low velocities happens in the brush-to-brush regime. However, in contrast to symptomatic contact lens wearers, asymptomatic wearers may adapt to this wear, or more likely, well-formed brushes at the cornea and limbal region may reduce mechanical forces between the ocular surface and the back surface of contact lenses to a minimum (Pult et al., 2015).In respect of contact lens wear, two key parameters that impact coefficient of friction can be controlled or modified – one by practitioners and the other by the contact lens manufacturer. Practitioners can encourage patients to periodically instill various forms of in-eye lubricants, which can modify, at least temporarily, the volume and viscosity of the tear film, in turn impacting coefficient of friction (see Treatment).Contact lens manufacturers can alter the chemistry and thus surface properties of contact lenses in order to optimise lubricity and reduce the coefficient of friction during blinking. The coefficient of friction of the first-generation soft silicone-hydrogel contact lenses was substantially higher than that of the cornea (Dunn et al., 2008). Subsequently, new materials have been developed with a lower coefficient of friction; these lenses either have high water content surfaces and/or incorporate wetting agents such as polyvinyl pyrrolidone and polyvinyl alcohol. In vitro measurement of the coefficient of friction of soft contact lenses depends on many variables (Roba et al., 2011). The in vitro coefficient of friction during lubrication is potentially influenced by sliding speed (velocity), normal force, and solution viscosity, so a comparison between measurements can only be drawn when evaluated in the same experiment. Such an experiment was undertaken by Roba et al. (2011), who demonstrated significant differences between currently available contact lenses.McMonnies (2007) has proposed that LWE is related to abnormal blinking activity. The starting point for his proposal is the notion that adequate lubrication at the lid wiper-ocular surface interface with a concomitant increase in the frictional coefficient and an increase in the potential damage to either or both of these ocular surfaces, is the primary cause of LWE. Exposure keratopathy may result in increased friction between an abnormal epithelium of the lid wiper and the desiccated inferior cornea. In respect of dry eye disease, McMonnies (2007) proposes that this frictional increase may be greatest for the complete blink that follows an incomplete blink, because the tear layer over the desiccated exposed cornea may be thinnest, and its lubricant properties most deficient, at the end of the prolonged interblink interval that is associated with an incomplete blink. He suggests that the risk of mechanical trauma to the lid wiper epithelium would be significantly higher following an incomplete blink and proposes that improved blink efficiency, achieved through a reduction in the rate of incomplete blinking, may be an important component of management. McMonnies (2007) goes on to suggest that the prime sensory mechanism for lens awareness or discomfort associated with an undamaged and well-fitted soft contact lens appears to be the blink related action of the lid wiper over the lens surface. Lack of wetness of the soft lens front surface and associated symptoms of dryness may be an important factor in increasing or decreasing the overall blink rate as well as the incomplete blink rate. He notes that contact lens discomfort has been reported to have the effect of reducing the completeness and rate of blinking, whereby awareness of lens dryness and associated discomfort detected by the lid wiper during a blink may be a stimulus to reduced blink rate and/or incomplete blinks. Alternatively, awareness of lens dryness by the lid wiper may be a stimulus to reflex tearing and blinking. Any associated increased reflex aqueous production and blinking may provide relief of dryness symptoms.Kato et al. (2015) investigated a broad range of dry eye measures (tear meniscus radius; tear film spread rate; lipid layer thickness; fluorescein break-up time; upper lid wiper staining grade; and Schirmer 1 test) and blinking activity (blink rate; palpebral aperture height; and upper lid ascending/descending distance and maximum velocity) on 45 eyes of 45 female dry-eye patients (mean age 57.6 years). Significant correlations were found between LWE and tear film spread rate (R=0.75, p < 0.0001) and fluorescein break-up time (R = -0.79, p < 0.0001). Maximum upper eyelid descending velocity was also a determinant of LWE grade. The authors suggested that their results provide evidence to support the notion that LWE is caused by sheer friction.10.5. Saccadic eye movementsThe focus of attention in most tribological considerations of the lid wiper has been the frictional activity of the blink (Cher et al., 2008; Jones et al., 2008; Korb et al., 2002; Korb et al., 2005; Pult et al., 2015). This may be a false lead. Dunn et al. (2013) modelled thickness of the tear film at the lid wiper, finding that it increases from approximately 0.1 μm during the interblink period to around 3 μm during the downward motion of the upper lid. At this thickness and with a lid speed to approximately 10-30 cm/sec, hydrodynamic lubrication can be assumed for the majority of the lid movement. Thus, despite the 10,000 or more blinks which take place during the course of a day, it is only the motion at the beginning of the blink that may be in the boundary lubricating regime.Perhaps more important are saccadic eye movements – the small jerky ocular rotations that occur between two periods of visual fixation. Depending on the task, from reading a book to scanning a scene, the distance of movement of the cornea relative to the eyelid will normally be from 0.2 mm to 0.8 mm. Saccadic eye movements occur about 10% of the time during the waking hours. In total, an eye will likely make more than 150,000 such movements during the waking hours (eye movement characteristics are different during sleep but these will not be considered here as this discussion is limited to open eye lens wear). The sliding speed during these movements is generally below 1 cm/s and the lubricity in this regime is likely to be the determining factor for lubricity-related comfort. Moreover, these saccades are probably occurring in the boundary regime, with the thickness of the lubricating film in the order of 0.1 μm thick. Because the majority of saccadic eye movements are horizontal, different considerations are required for modelling frictional dynamics at the lid wiper. In vertical movement of the eyelid, a leading edge of contact with the cornea is apparent, but that is no longer the case for horizontal eye movements. The greater lid wiper staining observed on the lower lid compared to the upper lid (Golebiowski et al., 2012; Navascues-Cornago et al., 2015c) in part may be explained by differences in friction and lubrication with saccadic eye movements versus blinks.10.6. InflammationThe notion that uncomplicated contact lens wear is intrinsically inflammatory was originally proposed by Efron in 1985 and re-articulated by the same author more recently (Efron, 2012c, 2016). This constant sub-clinical upregulation of the immune system in the ocular tissues essentially represents a state of readiness, whereby the tissues are primed to deal with any mechanical, toxic, infectious or adverse environmental challenge (Efron, 2016).Korb et al. (2005) proposed that LWE in contact lens wearers represents friction-induced microtrauma as a result of movement of the anterior lens surface across the lid wiper epithelium; they suggest that this in turn initiates a classic inflammatory cascade as seen in dermal tissue. One way of testing this hypothesis is to examine the lid wiper in lens wearers with dry eye symptoms for evidence of inflammation. In this regard, various approaches have been applied, including non-invasive assessment of cells in the lid wiper known to modulate inflammation, measuring vascular changes in the lid wiper, and monitoring the temperature of the lid wiper.10.6.1. Cellular responseVarious studies have used a laser scanning confocal microscope to quantify the number of inflammatory cells – which are presumed to be Langerhans cells – in the cornea (Alzahrani et al. 2016b; Sindt et al., 2012; Su et al., 2006; Zhivov et al., 2007) and conjunctiva (Alzahrani et al. 2016b; Efron et al., 2010) in response to contact lens wear. Furthermore, Knop et al. (2011b) have identified the presence of bright spots in the lid wiper which they presumed to be some form of inflammatory cell, possibly leukocytes. Morgan et al. (2013) used a laser scanning confocal microscope to examine the lid wiper of 10 non-contact lens wearers and 10 contact lens wearers. Five of the contact lens wearers wore low-surface friction lenses and 5 wore high-surface friction lenses. All participants were examined early in the morning (before lens wear in the lens wearing group) and then late in the afternoon (after lens wear in the lens wearing group).Morgan et al. (2013) observed small bright spots – similar to those previously described in non-lens wearers as ‘inflammatory cells’ by Knop et al. (2011b) – at the proximal edge of the lid margin in some subjects at the afternoon examination. The spots were assessed in a masked manner, using a 0 to 4 grading scale. There was an average increase from baseline for bright spots of 0.3, 0.6 and 2.0 for the non-wearing controls, low surface friction lens wearers and high surface friction lens wearers, respectively (chi-square = 9.4, p = 0.009). Thus, the greatest change was observed in the high surface friction lens wearers.The authors suggested that the observed changes in bright spots in the lid margin – presumed to be inflammatory cells – after a few hours of contact lens wear, may be related to the frictional properties of the lens surface, although they noted that other factors may also be important (Morgan et al., 2013).Alzahrani et al. (2016a) set out to determine if Langerhans cells in the lid wiper are upregulated in contact lens-induced dry eye. The lid wiper of one eye of each of 46 contact lens wearers was examined using a Heidelberg laser scanning confocal microscope (Fig. 10) following six months wear of daily disposable hydrogel contact lenses (Biomedics 1 day Extra). Seventeen participants were assessed as having contact lens-induced dry eye and 29 participants were assessed as not having contact lens-induced dry eye, using a subjective test (Contact Lens Dry Eye Questionairre-8 (Chalmers et al., 2012)) and objective tests (non-invasive tear break-up time, phenol red cotton thread tear test and fluorescein corneal staining). Twenty non-contact lens-wearing controls were also examined. Langerhans cell density in each participant was taken as the mean cell count calculated manually from 6 clear, randomly-selected images of known dimensions (Fig. 11).There were significant overall differences in Langerhans cell density in the lid wiper among the three groups (p < 0.001) (Fig. 12). Langerhans cell density was significantly greater in the lid wiper in those with contact lens-induced dry eye (17 ± 10 cells/mm2) compared to controls (8 ± 4 cells/mm2) (p < 0.001); however, there was no difference in Langerhans cell density between those without contact lens-induced dry eye (10 ± 5 cells/mm2) and controls (p = 0.489). Langerhans cell density was significantly greater in those with contact lens-induced dry eye than in those without contact lens-induced dry eye (p = 0.002). Alzahrani et al. (2016a) concluded that Langerhans cells in the lid wiper are upregulated in contact lens-induced dry eye, suggesting an inflammatory component in the aetiology of this condition.10.6.2. Vascular responseThe lid wiper is highly vascularised, so it would be expected that any inflammation of this tissue structure would be associated with hyperaemia. Certainly, Korb et al. (2002) made the anecdotal observation that severe cases of LWE were associated with lid wiper hyperaemia. However, in a pilot study, Wang et al. (2015) found reduced vascular fractal dimensions in the lid wiper (less hyperaemia) and suggested that this indicated lens-induced surface damage of the lid wiper. It is noteworthy that the subjects in the study were long term contact lens wearers, and the observation may reflect tissue which has adapted to constant mechanical effects on the lid wiper. Read et al. (2014) assessed changes in redness of the lid wiper (as an analogue of vascular response) for contact lens wearers and non-lens wearers. Ten symptomatic lens wearers, 10 asymptomatic lens wearers and 10 non-lens wearers underwent imaging of the upper and lower lid wiper at a morning and afternoon visit (≥ 6 h between visits). Imaging was performed using a slit lamp biomicroscope-based system comprising of a fixed exposure red-green-blue camera, macro optics and diffuse illumination. At the afternoon visit, the lid wiper was also assessed with lissamine green. Lens wearers wore their habitual contact lenses and reported 0 to 100 comfort scores. Image analysis (with calibration) was performed in a consistent region of the lid margin. A vertical red-green-blue intensity profile was also evaluated across the tarsal conjunctiva of the upper lid.The authors found no significant change in lid wiper redness between the morning and afternoon visits (p = 0.84) or between the three subject groups (p = 0.69). No significant association was observed between lid wiper redness and subjective comfort (p = 0.60) or Contact Lens Dry Eye Questionairre-8 (CLDEQ8) score (Chalmers et al., 2012) (p = 0.60). The mean red-green-blue intensity profiles for the three study groups did not differ significantly between visits. The association between lissamine green staining and lid wiper redness approached statistical significance (p = 0.07). Read et al. (2014) concluded that lid wiper redness did not alter significantly through the day or following contact lens wear. No association was observed with lid wiper redness or subjective comfort (either CLDEQ8 or 0-100 comfort score). Inspection of the lid wiper region with respect to either redness or lissamine green staining did not appear to differentiate between symptomatic and asymptomatic contact lens wearers. This finding does not support the notion that LWE is inflammatory, in the context of hyperaemia being one of the five cardinal signs of inflammation as observed clinically. 10.6.3. Temperature responseIncreased temperature of an affected tissue can be a sign of inflammation. Nepp (2013) measured the temperature of the lid wiper in the ‘worse eye’ of 61 patients (it is unclear from this abstract what was meant by ‘worse eye’ or whether the patients were wearing contact lenses). LWE was graded following staining with lissamine green. The patients were divided into those with and without signs of inflammation. Nepp (2013) claimed that there was more lid wiper staining and redness in the inflamed eyes, and the temperature of the lid wiper was higher in those with inflamed eyes (34.5°C) than those without inflamed eyes (33.7°C). He also suggested that there was a positive correlation between temperature of the lid wiper and signs of inflammation. Statistical validation of these results was not provided in the abstract.10.7. Lid wiper elasticity and eyelid pressureAs discussed previously, frictional forces between the lid wiper and ocular surface or contact lens anterior surface are influenced by upper eyelid pressure against the globe (Pult et al., 2015). Jones et al. (2008) developed an elastohydrodynamic model of the lid wiper. Their model allowed predictions of tear film flux from under the upper eyelid, as well as normal and shear stresses acting on the ocular surface. These factors are important in relation to dry eye syndrome. It was found that the pressure and shear stress under the eyelid act across a width of approximately 0.1 mm, which the authors claimed “is consistent with clinical observations”, but is considerably narrower than current estimates of the width of the lid wiper as being 0.3 to 0.6 mm wide (Knop et al., 2012; Shaw et al., 2010). It order to achieve a flow of tears from under the upper eyelid during a blink, the model requires that the normal force the eyelid applied to the ocular surface during the closing phase of the blink is significantly higher than during the opening phase of the blink.Shaw et al. (2010) measured static upper eyelid pressure for 11 subjects using a piezoresistive pressure sensor attached to that anterior surface of a rigid contact lens. Measures of eyelid pressure were derived from an active pressure cell (1.14 mm2) beneath the central upper eyelid margin. To investigate the contact region between the upper eyelid and the ocular surface, the authors used pressure-sensitive paper adhered to the front surface of a rigid contact lens. These measures, combined with the pressure sensor readings, were used to derive estimates of eyelid pressure. The authors concluded that a mean pressure of 8.0 ± 3.4 mmHg is the most reliable estimate of static upper eyelid pressure against the globe.The extent to which eyelid pressure is involved in the development of LWE was investigated directly in two experiments by Yamamoto et al. (2015). In the first experiment, eyelid pressure was measured with a blepharo-tensiometer, and the degree of upper and lower LWE was assessed using lissamine green staining in 79 eyes of 43 non-contact lens wearers. Upper LWE was detected in 24 of 79 eyes (30.4%) and no significant difference was detected in the eyelid pressure between any grade of upper LWE. Lower LWE was detected in 41 of 79 eyes (51.9%) and the eyelid pressure (27.9 ± 2.8 mmHg) in eyes with grade 3 LWE was significantly higher than that in eyes with grade 0LWE (19.7 ± 1.3 mmHg; p < 0.05).In the second experiment of Yamamoto et al. (2015), movements of the eyelids and displacement of the eyes during spontaneous blinking were photographed with a high-speed camera. Theeyelid pressure was also measured in 34 normal eyes of 19 non-contact lens wearers. The authors observed that lower eyelid pressure was significantly correlated with the length of the horizontal movement of the lower eyelids (p < 0.05) and also with the degree of posterior movement of the globe (p < 0.05). The authors concluded that higher pressure from the eyelid may be one of the causes for the development of lower LWE.10.8. Upper versus lower LWEShiraishi et al. (2014) have offered an interesting hypothesis to explain the observation of greater staining of the lower versus upper lid wiper (Fig. 13), which on initial consideration seems counterintuitive. They note that, during a blink, the upper eyelid has a large vertical movement while the lower lid has a shorter horizontal nasalward movement, as revealed by high speed photography of the eye during blinking (Doane et al., 1980). The large excursion self-evident in the upper eyelid has perhaps captured the attention of most researchers attempting to investigate and explain LWE. Because the lower eyelid moves horizontally, the lower eyelid margin rubs over the same restricted region of the contact lens or ocular surface. As a result, frictional effects, and therefore LWE, may be greater in the lower eyelid than the upper eyelid – the latter gliding over a vast expanse of generally well-lubricated ocular surface (in a non-dry eye situation). McMonnies (2015) has proposed an alternative theory to explain lower LWE; he proposes that blink-related excursions for the lower lid wiper are much less than for the upper lid wiper. As a consequence, this tissue structure has less opportunity to incur friction-related damage, and any epitheliopathy observed may more likely be due to hyperosmotic insult (see Section 10.3.3.). 10.9. AgeWhen examining the line of Marx using rose bengal, Norn (1970) noted anecdotally that the prevalence of staining 'was perhaps a little lower' in those under 30 years of age. Contrary to this, Doughty et al. (2004) observed that the intensity of lissamine green staining was qualitatively similar in younger and older, white Northern European subjects.Pult et al. (2011) reported a significant positive correlation between LWE and age (r = 0.340, p < 0.034), in a cohort of 17 males and 30 females aged 19 to 70 years. This indicates that LWE is more likely in older patients. The find of Pult et al. (2011) contradicts that of Shiraishi et al. (2014), who assessed LWE in 443 eyes of 229 non-contact lens wearers (100 men and 129 women, age 3 to 94 years; mean ± standard deviation 52.1 ± 24.0 years). The prevalence and average lid wiper staining scores of the eyes with LWE were examined after dividing subjects into 20-year age groups. Both the prevalence and average lid wiper grading scores decreased significantly with increasing age (p < 0.0001), and the most significant difference was detected between the 0 to 19 year and 20 to 39 year groups for upper LWE and between the 20 to 39 year and 40 to 59 year groups for lower LWE. When the subjects were divided into younger (≤ 39 years) and older (≤ 40 years) groups, both the prevalence and average lid wiper scores were significantly higher in the younger groups for both upper and lower LWE.10.10. SexTwo studies have failed to demonstrate an association between sex and LWE. Pult et al. (2011) found no relationship between sex and LWE ((U-test, p > 0.05). These authors indicated that this lack of association between sex and LWE is unexpected, since the prevalence of dry eye is known to be increased in elderly women (Albietz, 2000; Moss et al., 2000). Pult et al. (2011) suggested that their inability to demonstrate a relation between sex and LWE may be due to the mild nature of the dry eye symptoms reported in their subject cohort. Shiraishi et al. (2014) also could not detect a significant difference in sex between those with and without LWE. Using lissamine green staining, Doughty et al. (2004) observed that staining of the line of Marx was quantitatively similar in men and women.10.11. HumidityThe impact of a low humidity environment on the status of LWE was investigated by Jones et al. (2013b). Ten symptomatic contact lens wearers were random to contralateral lens wear with narafilcon A and etafilcon A lenses. Lens wear was discontinued 48 h prior to assessments and Refresh Plus artificial tears were instilled three times per day in both eyes. Upper and lower LWE were assessed using Korb protocol B at baseline, prior to lens insertion and entry into a low humidity environmental exposure chamber (temperature 22 ± 3°C, relative humidity 10 ±3 % and air velocity 5ft/sec) for 180 min. Upon exit, lenses were removed and artificial tears were instilled every 15 min for 120 min. LWE was evaluated immediately after post-chamber exposure and at 30min, 90min and 120 min. After 180 min in the chamber, upper LWE increased from baseline to post-chamber assessment; from 1.25 to 2.23 for narafilcon A and from 1.18 to 1.93 for etafilcon A. Lower LWE changed from 1.00 to 2.48 for narafilcon A and from 0.90 to 2.03 for etafilcon A (all p < 0.05). The mean upper and lower LWE grades did not return to baseline levels at 120 min after leaving the chamber and were 2.47 and 2.11 for narafilcon A and 2.42 and 1.89 for etafilcon A, respectively. These LWE responses were similar for the two lenses, despite their markedly different material properties.These results suggest that low humidity environments can have an adverse impact on upper and lower LWE during contact lens wear, presumably as a result of degradation of the pre-lens tear film and exacerbation of frictional contact between the lid wiper and anterior lens surface. 10.12. Psychological factorsMcMonnies (2013) suggests that there may be an important psychological component to the development of LWE. He proposes that higher expectations of comfort may be established for soft lens wearers based on their initial experience with diagnostic lenses. These expectations are usually reinforced by the high comfort levels experienced on daily insertion of fully hydrated lenses and before significant lens front surface drying occurs after many hours of wear, resulting in greater lid wiper friction and typical ‘end-of-day symptoms’. McMonnies (2013) proposes that lens awareness can increase under provocative conditions for front surface drying, such as prolonged reading and/or computer use, air conditioning, air movement and incomplete and/or infrequent blinking. Such adverse environmental conditions could compromise the tear film and exacerbate frictional effects between the lid wiper and lens surface. Failure of lens wearers to appreciate the potential significance of these factors may contribute to discomfort experiences; conversely, understanding the potential significance of these factors may help patients to reduce exposure to, or even to avoid, some of these adverse influences (McMonnies, 2013). 10.13. Multiple lid wiper epitheliopathies?The term ‘lid wiper epitheliopathy’ implies that this is a unitary condition, but the discussion above of the different forms of clinical presentation and various possible aetiological factors contributing to this condition suggest that there may be many different forms of LWE. An analogy would be corneal staining – an over-arching descriptor of a range of conditions with different aetiologies, such as the following: solution-induced corneal staining, which is a phenomenon of uncertain aetiology related to the impact of contact lens solution preservatives on the cornea (Efron, 2013); 3 and 9 o’clock corneal staining, possibly caused by poor wetting or the nasal and temporal cornea due to the lid wiper in that region being physically separated from the cornea by a rigid contact lens (van der Worp et al., 2003); desiccation corneal staining, due to accelerated pervaporation through thin, high water content hydrogel contact lenses that draws water out of the cornea (Orsborn and Zantos, 1988); superior epithelial arcuate lesions, produced by mechanical chaffing at the peripheral cornea (Holden et al., 2001; and peroxide burn, a toxic reaction to non-neutralised hydrogel peroxide lens care solution entering the eye (Paugh et al., 1988). Although all of these conditions can be included under the over-arching descriptor ‘corneal staining’, they represent a wide range of clinical presentations, with different causes, that require different treatments.As discussed throughout Section 10 of this paper, the aetiology of LWE is likely to be multifactorial in the same way that corneal staining is multifactorial. Consider the following examples. The cause of contact lens-induced LWE (frictional effects between the lid wiper and anterior lens surface) may be different from dry eye-induced LWE (deficient tear film and shifts in tear osmolarity). Upper LWE may have a different aetiology to the apparently more common lower LWE (long vertical sweeps with blinking versus short lateral movements associated with eye saccades). Lid imbrication syndrome –an abnormality of lid apposition in which the upper lid overlies the lower lid – may induce a unique form of LWE, different from that seen in contact lens and dry eye induced forms of the condition. Perhaps therefore, as in the case of corneal staining, LWE should be thought of as a range of conditions with disparate underlying causes. Accordingly, it might be more appropriate to think in terms of there being a range of ‘lid wiper epitheliopathies’.11. TreatmentThe choice of strategies for the treatment of LWE is confounded by uncertainty as to the link between LWE and ocular comfort, and lack of a complete understanding of the aetiology of the condition, which may be multifactorial and vary among individuals. Despite these uncertainties, a number of reports have advocated various management strategies alleviating LWE. Some authors have directly assessed the benefits of certain therapies on LWE whereas others have demonstrated an impact of therapies on LWE in the course of alleviating dry eye or contact lens discomfort. In the latter case, improvements in LWE essentially served as a surrogate marker for alleviation for dry eye.Because LWE is a relatively new concept, many accounts of treatment of this condition are at present just starting to appear; accordingly, a number of these reports are in the form of conference abstracts. To assist readers in assigning due weight to full papers versus abstract, Table 1 can be consulted for a summary of all clinical reports disscussed in this section, and indeed elsewhere in this paper. 11.1. RebamipideItakura et al. (2013) presented two case reports of patients who were treated with topical rebamipide to treat LWE. Rebamipide is an amino acid derivative of 2-(1H)-quinolinone and is used for mucosal protection, healing of gastroduodenal ulcers and treatment of gastritis. It works by enhancing mucosal defense, scavenging free radicals and temporarily activating genes encoding cyclooxygenase-2. It is also thought to have anti-inflammatory properties (Kashima et al., 2014). Administration of topical rebamipide increases secretion of mucins from goblet cells and improves the ocular surface in the ‘short break-up time’ type of dry eye (Kinoshita et al., 2012; Koh et al., 2013).One of the cases reported by Itakura et al. (2013), the patient had previously been treated with sodium hyaluronate ophthalmic solution and diquafosol sodium eye drops by other doctors for several weeks prior to presentation, but this treatment regime was deemed to have been ineffective in alleviating dry eye. The other case was not previously treated. In both cases, fluorescein staining of the cornea and lid margin was ‘remarkably improved’, ocular symptoms decreased, and tear film break-up times increased with rebamipide eye drops administered four times daily for 2–3 weeks. The authors stated that this drug may provide a novel approach to the treatment of LWE, but acknowledged that they were only reporting anecdotal observations of two patients and there was no masking or control group.Rebamipide is used in a number of Asian countries including Japan, South Korea, China and India, but it is not approved by the Food and Drug Administration for use in the United States.11.2. CorticosteroidsThe efficacy of a corticosteroid, Lotemax (Bausch & Lomb), for the treatment of LWE was investigated by El-Rayess et al. (2009). LWE was diagnosed using Korb Protocol B. Symptoms were evaluated with the Standard Patient Evaluation of Eye Dryness questionnaire (Ngo et al., 2013). Forty subjects were treated with Lotemax (Bausch & Lomb) and 30 served as untreated controls. After a one month treatment phase with patients using 1 drop twice daily, there was an improvement in signs (p < 0.001) and symptoms (p < 0.001) in the treatment groups.The authors claimed that their study demonstrates effectiveness of a corticosteroid for the initial treatment of LWE. The corticosteroid was claimed to be effective in minimizing or eliminating LWE signs and symptoms (El-Rayess et al., 2009).11.3. Lubricant eye dropsEl-Rayess et al. (2009) examined the efficacy of an oil-in-water emulsion lubricant for the treatment of LWE. Korb Protocol B was used for the diagnosis of LWE and symptoms were evaluated using the Standard Patient Evaluation of Eye Dryness questionnaire (Ngo et al., 2013). Forty subjects were treated with Soothe Emollient Eye Drops (Bausch & Lomb) and 30 served as untreated controls. After a treatment phase (1 drop twice daily) of at least one month, followed by a one-year maintenance phase, there was an improvement in signs (p < 0.001) and symptoms (p < 0.001) in the treatment groups. Of subjects completing the maintenance phase, 85 % demonstrated sustained improvement (p < 0.01). According to the authors, the oil-in-water emulsion eye drop was efficacious for both the initial treatment of LWE as well as maintaining long-term improvement of the lid wiper epithelium. Guthrie et al. (2015) examined the clinical impact of using Systane Balance Lubricant Eye Drops (Alcon, Fort Worth, TX), an oil-in-water emulsion, as a rewetting eye drop in symptomatic contact lens wearers. One hundred and six subjects who had previously experienced contact lens discomfort, with a mean lens wearing history of 18.6 ± 12.8 years, were randomly assigned to use a test drop (Systane Balance Lubricant Eye Drops; n = 76) or control drop (habitual non-lipid contact lens rewetting eye drop; n =30) over their contact lenses within 5 min of lens insertion. They were advised to continue using their drops at 2 h intervals up to a maximum of 4 drops per eye daily for a one month period. Assessments of subjective comfort, comfortable wearing time, LWE, and corneal staining were conducted at baseline and after 1 month, following 6 h of lens wear on each measurement occasion. The authors reported that comfort, wearing time, LWE, and corneal staining all showed statistically significant improvements in the test group using Systane Balance Lubricant Eye Drops at the 1-month visit compared with baseline data (all p < 0.01) and compared with the control group at the 1-month visit (p < 0.01, p = 0.01, p < 0.01, and p ≤ 0.03, respectively). They concluded that the use of Systane Balance Lubricant Eye Drops in a group of wearers who experienced symptoms of contact lens discomfort improved subjective comfort scores, increased comfortable wearing time, and reduced signs of LWE and corneal staining, when compared with the use of non–lipid-containing contact lens rewetting eye drops.Srinivasan et al. (2015) investigated the combined effect of TheraTears Lubricant Eye Drops, TheraTears SteriLid Eyelid Cleanser, and TheraTears Nutrition on dry eye signs and symptoms, in a prospective, random, controlled, single-masked study. Thirty three dry eye participants, who were already on a dry eye regimen that consisted of using artificial tears at least three times a week, were enrolled. Participants were seen at baseline and were random into either the treatment group or the control group. In the treatment group, participants were instructed to use TheraTears Lubricant Eye Drops one to two times a day (or when necessary), TheraTears SteriLid Eyelid Cleanser one to two times a day (or when necessary), and TheraTears Nutrition 3 capsules once a day, as per product label. Participants in the control group remained on their own treatment regimen for the rest of the study. Each participant was followed up at one and three months, and 18 dry eye tests were performed, including staining for LWE. Seven of the measures showed improvement, but not LWE. Therefore, this treatment regimen was deemed to be inefficacious for alleviating LWE.11.4. Basic fibroblast growth factorThe treatment of LWE with basic fibroblast growth factor was investigated by Zhang and Mo (2011). Eighty-six patients were divided into a treatment group, who were administered preservative-free artificial tears combined with basic fibroblast growth factor, and a control group, who were administered preservative-free artificial tears. After 3 weeks of treatment, all patients were assessed for LWE, using Korb Protocol B, and a questionnaire survey was used to evaluate dry eye. The therapeutic effect of LWE in the treated group was deemed to be ‘excellent’ in 74 eyes (86%) and ‘effective’ in 12 eyes (14%), with a ‘cure rate’ of 86%. The therapeutic effect in the control group was deemed to be ‘excellent’ in 51 eyes (59%), ‘effective’ in 34 eyes (40%) and ‘ineffective’ in 1 eye (1%), with a cure rate was 59%. There was a statistically significant difference between the two groups (p < 0.05). The authors concluded that artificial tears combined with basic fibroblast growth factor were effective in treating LWE (Zhang and Mo, 2011).11.5. LipiFlow Thermal Pulsation SystemSatjawatcharaphong et al. (2015) sought to identify patient characteristics – including LWE – at a baseline ocular surface evaluation that correlated with improvement in dry eye symptoms at a follow-up visit after treatment with the LipiFlow Thermal Pulsation System (Lane et al., 2012). Only initial symptoms and sex (and not LWE) were found to be predictive of improved post-treatment dry eye symptoms.11.6. Punctal plugsThe benefits of punctal plug insertion in alleviating symptoms and signs of aqueous tear deficient dry eye, including LWE, were investigated in 13 subjects by Yokoi et al. (2015). Symptoms and signs before and at more than one month post-punctal occlusion were assessed. The authors reported that a range of symptoms, including dryness, difficulty in opening the eye, foreign body sensation, pain, redness, blurred vision, sensitivity to light, heavy eyelid, and eye fatigue were significantly improved (all p < 0.01), as well as tear meniscus radius, interference grade/spread grade of the tear-film lipid layer, non-invasive breakup time, ocular surface epithelial damage score (both corneal and conjunctival), corneal filament grade and LWE grade (all p < 0.02). Punctal plug insertion was therefore deemed efficacious in alleviating LWE (Yokoi et al., 2015).11.7. Fit contact lenses of high surface lubricityA strategy that may alleviate LWE associated with contact lens wear is to fit a lens with high surface lubricity. Roba et al. (2011) have demonstrated that contact lenses have a wide range of surface lubricity values. Further to this, Coles and Brennan (2012) have shown that contact lenses with higher lubricity values tend to be more comfortable. No studies to date, however, have demonstrated a relationship between contact lens surface lubricity and LWE. Such an experiment would be difficult to conduct, as lenses would have to be custom manufactured for such an experiment, whereby a range of lenses would need to be fabricated that are identical in every respect except for surface lubricity.11.8. Alter lens wearing modalitiesIt also remains to be seen whether alterations to lens wearing modalities could alleviate LWE. Such strategies might include reducing lens wearing time, increasing lens replacement frequency, changing lens material properties (modulus of elasticity, silicone/water content, surface-active agents or packaging wetting agents) (Lin and Yeh, 2013), changing lens dimensions (thickness, diameter, radius of curvature) or altering lens fit. 11.9. Improve blinking behaviourMcMonnies (2007) has proposed that improved blink efficiency, achieved through a reduction in the rate of incomplete blinking, may be an important component of management of LWE related to dry eye in contact lens wearers and non-lens wearers. Early anecdotal reports proposed a variety of strategies for enhancing blinking activity. These ranged from simple instructions and reminders (Korb and Korb, 1974), to the employment of a small buzzer that sounded every 10 seconds, which acted as a prompt to execute a full blink (Jenkins et al., 1978). Although it was realised that such strategies were only stimulating reflex rather than spontaneous blinks, the underlying assumption was that spontaneous blinking activity could be learned via training using reflex stimulation techniques.Collins et al. (1987) tested the hypothesis that blinking could be trained by subjecting a group of unsuspecting contact lens-wearing volunteers to blink exercises (the volunteers were told that the purpose of the exercises was to improve vision). The exercise consisted of placing the index finger of each hand just lateral to the outer canthus to hold the lids taught whilst performing 10 complete forced blinks. This exercise was repeated three times daily for two weeks. Blinking exercises resulted in an increased frequency of complete blinks and a decreased frequency of incomplete and twitch blinks. This approach could be applied to patients with LWE in whom blinking anomalies are deemed to be the cause.12. Differential diagnosisIt is important to be able to differentiate LWE from other physiological and pathological eyelid conditions that may take on a similar appearance. Although there is no clear evidence that the lid wiper increases in redness in cases of LWE, other pathological conditions in adjacent anatomical structures of the eyelid may be associated with lid wiper staining; clinicians may need to differentially diagnosis such conditions from LWE.12.1. Line of MarxKorb et al. (2002) emphasise the importance of differentiating between LWE and staining of the line of Marx (see Section 4.2.1). With the patient looking upwards, it is possible to view the line of Marx without lid eversion; however, observation of the lid wiper requires lid eversion (Korb and Blackie, 2010).12.2. Lid imbrication syndromeHinkle (2006) suggested that LWE needs to be differentiated from lid imbrication syndrome – a condition in which patients have a lax upper eyelid that does not easily evert and a normal tarsal plate (Donnenfeld et al., 1993; Karesh et al., 1993). Korb (2006) noted the following points of differentiation between LWE and lid imbrication syndrome, with reference to his own previous observations of LWE (Korb et al., 2005). The mean age of the population with LWE was 44.3 years (range 24 to 64 years), versus 71 years (range, 54 to 93 years) and 64.5 years (range, 21 to 86 years) for those with lid imbrication syndrome in the studies of Karesh et al. (1993) and Donnenfeld et al. (1994), respectively. Thus, the typical population with lid imbrication syndrome is much older than that with LWE. Karesh et al. (1993) reported that all patients with lid imbrication syndrome exhibited overriding of the upper lid on the lower lid, a feature not observed in the study of Korb et al. (2005). Karesh et al. reported that of 18 patients observed with lid imbrication syndrome, 13 required surgery and five were successfully treated with lubrication. LWE does not require surgery. LWE is a very common feature of contact lens wear, and cessation of contact lens wear usually leads to the resolution of most cases of LWE, indicating that LWE may be induced by contact lenses. 12.3. BlepharitisBlepharitis is typically classified as being either anterior or posterior and may be related to contact lens wear (Jackson, 1993). The condition is sometimes called ‘marginal blepharitis’ because it is observed along the lid margins. Anterior blepharitis is directly related to infections of the base of the eyelashes and manifests in two forms – staphylococcal blepharitis and seborrhoeic blepharitis (Efron, 2012a). Staphylococcal blepharitis is associated with the presence of hyperaemia, telangiectasis and scaling of the anterior lid margins. In seborrhoeic anterior blepharitis, the anterior lid margin displays a shiny, waxy appearance with mild erythema and telangiectasis. Soft, yellow greasy scales are observed along the lid margin and the eyelashes may also become greasy and stuck together.Posterior blepharitis describes inflammatory conditions of the posterior lid margin, of which meibomian gland dysfunction is a possible cause. The resulting lid margin redness may result in a similar appearance to LWE, so the two conditions need to be differentiated. Although the symptoms of blepharitis and LWE may be similar in mild cases, all form of blepharitis involve tissue anterior to the line of Marx, and in this way can be differentiated from LWE, which only involved the lid wiper, which is posterior to the line of Marx (Efron, 2012a).12.4. Papillary conjunctivitisPapillary conjunctivitis is common in contact lens wearers (Forister et al., 2009). Upon lid eversion, numerous papillae can be observed on the superior tarsal conjunctiva, which also appears hyperaemic. This condition is also characterised by excess mucus production, seen as white strands on the contact lens or corneal surface, and excessive lens movement. In the early stages of this condition, patients may complain of lens discomfort and itching; these symptoms are similar to those reported in association with LWE. In severe cases of papillary conjunctivitis, the discomfort worsens and can progress to a burning sensation, leading to an urge to remove lenses (Efron, 2012e). Differentiation of LWE from papillary conjunctivitis can be effected by lid eversion and careful inspection of the underlying palpebral tissue to see which is affected. Instillation of fluorescein will reveal papillae on the tarsal conjunctiva in the case of papillary conjunctivitis and staining of the lid wiper in LWE, or both if the two conditions co-exist. 12.5. DemodicosisMite infestation in the eyelashes is ubiquitous and generally sub-clinical, but if present in excessive numbers, adverse signs and symptoms may develop (Efron, 2012a). The mode of transmission of mites between humans is not clear but may arise from intimate contact. Mites are also more abundant in diabetic individuals, those with autoimmune deficiency syndrome, and in patients on long-term corticosteroid therapy, with involvement of the latter two conditions suggesting that compromised immunity may also influence mite infestation (Edmondson and Christenson, 1992).Two species of mite (Demodex) are found in the human pilosebaceous gland complex. Demodex folliculorum is a cigar shaped mite with four evenly spaced stubby legs on the upper third of its body. It prefers to live in the space between the eyelash and the follicle wall, above the level of the gland of Zeis. Demodex brevis is found in human skin rich in sebaceous glands and sebum production. It prefers to infest the gland of Zeis, and can get to this gland because of its very small size (Edmondson and Christenson, 1992).Signs of demodicosis include erythema of the lid margins, lid hyperplasia and madarosis (all of which give the impression of blepharitis), conjunctival injection, cuffing around lashes, follicular distension, and meibomian gland blockage. Typical symptoms of demodicosis are pruritus, burning, crusting, itching, swelling of the lid margins and loss of lashes. The itching often parallels the 10 day reproductive cycle of mites (Edmondson and Christenson, 1992). Although the symptoms of demodicosis may be similar to those in LWE, careful inspection of the lashes for the signs of mites or collarettes will assist differential diagnosis. In a group of 40 female Asian subjects, Jalbert and Rejab (2015) were unable to detect significant differences in lid wiper staining between Demodex-infested subjects and those without Demodex regardless of whether they were contact lens wearers or not.12.6. Meibomian gland dysfunctionMeibomian gland dysfunction is defined as a chronic, diffuse abnormality of the meibomian glands, commonly character by terminal duct obstruction and/or qualitative/quantitative changes in glandular secretion. This may result in alteration of the tear film, symptoms of eye irritation, clinically apparent inflammation and ocular surface disease (Nelson et al., 2011). Hypersecretory meibomian gland dysfunction is character by the release of a large volume of meibomian lipid at the lid margin in response to pressure on the tarsus. This appearance is accompanied by symptoms of smeary vision, greasy contact lenses, dry eyes and reduced tolerance to lens wear. In severe cases where the meibomian orifices are blocked, there may be an absence of gland secretion. Long standing cases of this condition may be associated with additional signs such as irregularity, distortion and thickening of eyelid margins, slight distension of glands, mild to moderate papillary hypertrophy, vascular changes (Efron, 2012d).The absence of oil globules at the meibomian gland orifices could indicate normality, hyposecretion or obstructive meibomian gland dysfunction. If oil globules are not observed on the lid margins of a symptomatic contact lens wearer, they patient may have a condition which Blackie et al. (2010) refer to as ‘non-obvious meibomian gland dysfunction’. Such cases may appear similar to LWE and it may be necessary to conduct a provocative test in order establish the state of health of the meibomian glands by manual expression of the glands, so that the nature of the expressed oils can be assessed. Abnormal secretions will assist differentiation of meibomian gland dysfunction, especially the ‘non-obvious’ form, from LWE.Willis et al. (2011) evaluated the grade of LWE in individuals symptomatic for dry eye meibomian gland dysfunction. Nineteen female and 5 male non-contact-lens-wearing patients (mean age 45.2 ± 14.3 years) presenting for routine eye examinations reported their symptoms using the Standard Patient Evaluation of Eye Dryness questionnaire (Ngo et al., 2013). Patients with a score of ≥ 6 (maximum score = 28) were considered symptomatic for dry eye. Meibomian gland expression was performed using a standard instrument which exerts a pressure of about 0.3 pounds per square inch over the lower lid meibomian glands. The total number of meibomian glands yielding liquid secretion (indicating functional glands) was recorded and only those with ≤ 4 functional glands were admitted to the study. Korb Protocol B was used to assess LWE. The mean number of functional glands was found to be 2.1 ± 1.3. The mean symptom score was 12.3 ± 4.3. The mean grade of LWE was 2.0 ± 0.5. The authors concluded that patients reporting dry eye symptoms are very likely to display meibomian gland dysfunction and LWE as comorbidities. In cases where LWE and meibomian gland dysfunction co-exist, clinicians should remain aware that the precise location of the line of Marx – a key anatomical landmark the assessment of LWE – varies in location relative to the meibomian gland orifices, tending to run on the eyelid-margin side of gland orifices in more severe cases (Yamaguchi et al., 2006).Alghamdi et al. (2016) assessed the position of the line of Marx among contact lens wearers and non-lens wearers and noted that this anatomical landmark was displaced more anteriorly (i.e. more towards the skin side of the meibomian gland orifices) among contact lens wearers compared to non-lens wearers, and this trend was greater among long-term contact lens than previous wearers.The implication of the above observations is that an awareness or current or previous meibomian gland dysfunction and/or contact lens wear will assist practitioners in interpreting the appearance of the various substructures of the lid margin when examining patients with LWE.12.7. Iatrogenic lid wiper staining In the course of research experiments characterising the various appearances of lid wiper staining, Varikooty et al. (2015) occasionally observed a false positive artifact on the lid wiper if a finger of the examiner touched the lid margin region before instillation of fluorescein or lissamine green (Fig. 14). Care must therefore be taken when manipulating the lids during eversion, and while maintaining lid eversion, to avoid the occurrence of iatrogenic lid wiper staining. 13. SignificanceWe noted in the introduction that the term lid wiper had become part of the mainstream terminology when discussing ocular discomfort associated with dry eye or contact lens wear. Despite this, it is more often referred to in a supplementary capacity, in the form of an interesting observation or perhaps a diagnostic sign, in major reviews of the tear film, rather than being a core component of the mechanism. Contact lens wearers most often cite dryness as their principal symptom (Brennan and Efron, 1989). The finding of LWE in both symptomatic contact lens wearers and dry eye sufferers leads to the intriguing possibility of a unifying model to explain the mechanism of the dryness sensation and ties together the similarity of symptoms experienced by these groups. It would explain the disconnect, as mentioned in our introduction, between signs and symptoms of dry eye and the inability hitherto of researchers to identify the basis for discomfort during contact lens wear (Nichols et al., 2013).The Dry Eye Workshop defined dry eye as a “multifactorial disease of the tears and ocular surface that results in symptoms of discomfort, visual disturbance, and tear film instability, with potential damage to the ocular surface” (Bron, 2007). The definition also states that increased tear film osmolarity and inflammation of the ocular surface are of aetiological significance in the disease. Dry eye is a thus a condition that is principally symptom based. It may occur without damage to the ocular surface. There is an inherent but unproven assumption that change in tear film osmolarity is, in part, responsible for the symptoms of dry eye. At least in the short term, major changes in solution tonicity are needed to bring about awareness (Fletcher and Brennan, 1993). Studies have not found a significant correlation between tear osmolarity and ocular symptoms, nor change in tear osmolarity with change in symptoms upon treatment (Caffery et al., 2014; Amparo et al., 2014).While it has been stated that dry eye is an inflammatory disease (Hessen and Akpek, 2014; Stern et al., 2013; Wei and Asbell, 2014), the role of inflammation in dry eye is yet to be fully elucidated. Aqueous insufficiency may result from inflammation of the lacrimal gland; however, meibomian gland dysfunction – apparently the principal cause of dry eye disease – is not inherently an inflammatory process (Knop et al., 2011a). The ocular surface may show an influx of inflammatory cells and release of inflammatory biomarkers during dry eye development, but the role of inflammation in the development of symptoms is not clear.If LWE is the basis of symptoms in dry eye – and given that dry eye is first and foremost a symptom-based disease – then understanding of LWE could provide clinicians and researchers with insights into the disease process. It may become the preferred sign for diagnosis and monitoring of treatment efficacy. Although Pult et al. (2011) found a combination of non-invasive tear break-up time and nasal lid-parallel conjunctival folds to be a better predictor of dry eye symptoms than LWE, their sample group consisted of normal healthy volunteers. In other studies of dry eye disease, LWE has been shown to be diagnostic. In those cases where symptoms are present but no staining is evident, the mismatch may result from methodological issues, such as time of day at which the staining was performed. Further development of tests for LWE in addition to staining, such as vascular image analysis, confocal microscopy and impression cytology, may improve diagnosis as well as understanding of the disease process. It should however be recogn that LWE is neither the origin of dry eye disease nor the endpoint, which is frank ocular surface disease.In Fig. 15, we present a simple pathway for the process of dry eye disease that incorporates the pathological origins of the disease through to the endpoint of ocular surface damage, which highlights the potential role of LWE and tear film hyperosmolarity. The connection between tear film hyperosmolarity and symptoms is shown as being uncertain, as is the possible role of LWE in ocular surface damage.With respect to contact lens wear, researchers have been searching without reward for many decades for the mechanism of lens-associated discomfort (Nichols et al., 2013). The demonstrated association between contact lens surface lubricity and comfort has focused the attention of researchers to try and discover which particular tissue sitting in apposition with a contact lens is the source of this comfort. Although the lid wiper – as the tissue part that moves extensively across the contact lens surface – is a strong candidate in this regard, the equivocal nature of research attempting to draw a direct association between contact lens discomfort and disruption of the lid wiper epithelium undermines the clinical utility of LWE. Ongoing consideration of other tissue parts, such as the cornea, bulbar and tarsal conjunctiva, should continue in parallel with efforts to more fully understand the role of frictional damage to ocular tissue as a determinant of contact lens comfort.A consistent theme among wearers is the report of end-of-day dryness despite, in many cases, satisfactory lens comfort earlier in the day (Jones et al., 2013a). The interaction between the lens and the lid wiper may provide an explanation for this problem. A contact lens surface which is inadequate to provide appropriate lubrication may cause mechanical trauma to the lid wiper as a result of blinking and eye movements during the day. During the course of the day, friction at the lens surface may also increase. The trauma may initially be insufficient to cause awareness of the lens but the cumulative trauma may ultimately lead to sufficient surface damage at the lid wiper to produce both staining and symptoms. Upon lens removal and after sleeping overnight, the lid wiper region may undergo repair such that, on reinsertion of a contact lens the next day, there is initially no awareness or discomfort from the lens and the cycle repeats. This process of “wear and tear and then repair” can explain why greater variations in comfort occur on an intra-day basis (Papas et al., 2015; Santodomingo-Rubido et al., 2010) than, say, between days over the lifetime of a reusable contact lens (Dumbleton et al., 2008).Two recent studies are consistent with this process. In these studies replacement of a contact lens during the middle of the day had minimal impact on the end-of-day comfort outcome (Navascues-Cornago et al., 2015b; Papas et al., 2015). If the discomfort was caused by a deterioration of the lens surface as a result of wear, then replacement of the lenses should have restored comfort. The observed effect is consistent with a degradation of some aspect of the ocular surface – such as the lid wiper – which recovers on a daily basis.14. Conclusions and future directionsIn any new and emerging field, there are bound to be more questions than answers. This is certainly the case in respect of LWE. At the time of writing this review, 14 years has elapsed since LWE was first described as a clinical entity. Research into this condition has proliferated at considerable pace, but the research findings published so far have opened up gaps in our understanding of LWE. Here we shall review what we consider the more significant knowledge gaps relating to this condition, and will try to identify future research directions that may pave the way towards developing a fuller understanding of LWE.The question as to the histology of the lid wiper needs to be resolved. The classical literature and contemporary impression cytology studies suggests that the lid wiper is a stratified squamous epithelium. The comprehensive study of Knop et al demonstrated that the lid wiper of 77 year old eyes from cold-stored cadavers is best characterised as a stratified epithelium with a conjunctival structure of cuboidal cells. While the observations of Knop et al (2011b) are undoubtedly accurate, the extent to which these findings relate to young adults aged 25 to 35 (the demographic of contact lens wearers, which is a key application of lid wiper theory) (Morgan et al., 2010) remains uncertain. As well, Parsons (1904) noted, albeit anecdotally, “The epithelium varies ... at different ages.” Further histological studies of fresh cadaver tissue from young adults could resolve this issue.There needs to be a consensus on terminology. The term ‘lid wiper’ appears to have been well accepted and it is now generally understood what this term means, assisted by a clear initial statement of this concept by Korb et al. (2002), followed by careful anatomical and histological documentation of this region of eyelid anatomy by Knop et al. (2012) and others.The rules of etymology dictate that phrases ending in the suffix ‘-pathy’ indicate a ‘morbid condition or disease; generally used to designate a non-inflammatory condition’ (Miller et al., 2005). This raises two key questions in respect of the term lid wiper epitheliopathy. First, we need to consider whether LWE truly is a disease. Certainly, lid wiper staining is frequently observed in healthy non-contact lens wearing subjects (see Section 6), those with dry eye (see Section 7) and contact lens wearers (see Section 8). As to whether this indicates true pathology, versus a mechanical frictional-based tissue disruption of no clinical consequence, is yet to be fully answered.Second, preliminary evidence suggests that there may be an inflammatory component to LWE, in which case the word ‘epitheliopathy’ might not be entirely appropriate. If LWE was clearly inflammatory, then the suffix ‘-itis’ would be more appropriate (Miller et al., 2005). However, evidence of overt inflammation is lacking; for example, Read et al. (2014) were unable to demonstrate a link between LWE and lid wiper hyperaemia. Alzahrani et al. (2016a) demonstrated an upregulation of antigen-presenting Langerhans cells in the lid wiper in patients with contact lens-associated dry eye, compared to lens wearers without dry eye; however, (a) Alzahrani et al. (2016a) did not specifically examine subjects for LWE, and (b) the response they observed is ‘subclinical’, and is perhaps best categor as ‘para-inflammation’ – which Medzhitov (2008) describes as an adaptive response to tissue stress, that offers protection without causing further tissue damage. With the benefit of hindsight, a more prudent approach may have been to simply use the term ‘lid wiper staining’; nevertheless, as noted above, LWE is now a generally accepted term. Perhaps LWE should be taken to mean ‘staining observed on the lid wiper’.The revised twin-dye staining protocol described by Korb et al. (2008) – which we have referred to as ‘Korb Protocol B’ throughout this review – was initially adopted by many research groups, although not universally, but over the past three years appears to have been used less, in favour staining with lissamine green alone (see Table 1). Certainly, Varikooty et al. (2015) have suggested that the twin-dye staining protocol may not be according any additional diagnostic capability compared to the use of a single dye. Possible reasons why the twin-dye approach is losing favour are (a) it is cumbersome and time consuming to perform, (b) it has not been fully validated, in terms of repeatability and accuracy, and (c) a number of authors have demonstrated that staining with a single agent (generally lissamine green) is sufficient to identify LWE. The requirement to use two stains (fluorescein and lissamine green according to Korb Protocol B) is not well justified; the fact that different amounts of staining are observed in different subjects is not a rational justification for using both stains. Such an outcome, for example, would be equally likely to occur when observing the corneal or bulbar conjunctival with two different stains, yet no-one has advocated double-staining for assessing these tissues. Possible sources of error in using two dyes to observe and grade the lid wiper are (a) increased possibility of an iatrogenic toxic effect on the lid wiper epithelium, and (b) revealing of the maximum possible staining on each grading occasion; both of these factors would lead to an overestimation of the overall extent of staining. The above confounding factors are likely to reduce the sensitivity of the procedure. For example, estimates of the prevalence LWE as high as 85% in general presenting populations of lens wearers and non-lens wearers to eye clinics may be an over-estimation as a result of the double staining and grading procedure adopted the researchers who generated this data (see Section 6). As well, these confounding factors may have contributed to the failure of a number of researchers (Best et al., 2013; Navascues-Cornago et al., 2015c; Read et al., 2014; Schulze et al., 2015), and our meta-analysis (see Section 8.3), to demonstrate a relationship between LWE and discomfort or dry eye symptoms. In view of the above, we recommend the application of lissamine green alone for assessing LWE. Efron (2012b) has argued that if grading scales are to be generally accepted by eye care practitioners for routine clinical use, they should be straight forward and easy to apply. For example, a simple 5-step pictorial grading scale, depicting images of lid wiper staining of grade 0 (normal), grade 1 (slight), grade 2 (mild), grade 3 (moderate) and grade 4 (severe) – such as those used by Willis et al. (2011) and Jalbert and Rejab (2015) – is likely to be used more often and with greater confidence by eye care practitioners, leading to more widespread, perhaps universal use. At the time of writing, six research papers (Berry et al., 2008; Korb et al., 2002, 2005; Pult et al., 2008, 2009; Yeniad et al., 2010) and two abstracts (Varsani and Wong, 2009; Yan et al., 2008) have been published which claim to have demonstrated an association between LWE and contact lens-associated dryness/discomfort. Putting aside the abstracts, which provide an incomplete account of the scientific methods employed, all studies were prospective, but only the study of Yeniad et al. (2010) contained a true control group; the other 5 studies involved comparison of groups, such as symptomatic versus asymptomatic lens wearers. Only one study (Pult et al., 2009) reported observer masking, but even these workers noted a risk of observer bias because all measurements were taken by a single researcher. Furthermore, Pult et al. (2009) explained that full masking was not possible in their study comparing LWE in response to the wearing of two lens types because markings on the lenses revealed to the observer which lens type was being worn.It is also perhaps worth noting that five of the six published clinical trials referred to above were essentially conducted by two research groups – the Korb group (Korb et al., 2002, 2005) and the Pult group (Berry et al., 2008; Pult et al., 2008, 2009). Given the understandable enthusiasm of these two groups for the concept of LWE, and absence of effective controls or observer masking in these studies, the possibility of inadvertent experimenter bias in favour of the stated hypotheses cannot be discounted. Clearly, more clinical studies by independent groups of researchers, adhering to high standards of scientific rigor – i.e., sufficiently powered experiments employing appropriate controls, with randomization and double-masking, employing rigorous statistical analysis and published in refereed journals – are required before LWE can be fully accepted as a distinct, validated clinical entity. However, we feel that there is a reasonable prima facie case for continuing to explore this condition and its relationship to contact lens-associated dryness and discomfort.The causes of LWE is poorly understood and the tear layer structure and dynamics in the lid wiper region are not well defined in normals, dry eye disease, or asymptomatic and symptomatic contact lens wear. Modelling has provided some insights, but the gathering of further input data to these models will improve our understanding of the frictional forces to which the lid wiper is subjected, and will assist us in determining whether these forces are responsible for LWE. Based on the variety of clinical appearances of LWE, and our demonstration that the aetiology of LWE is multifactorial, we have entertained the idea of there being a range of different ‘lid wiper epitheliopathies’. This way of thinking may serve to support different approaches to the management or treatment of excessive lid wiper staining. Further work is also needed to ascertain whether LWE is the source of symptomatology in dry eye disease and contact lens wear, and whether common symptoms of dryness arise from identical or disparate mechanisms. Additionally, it would be useful to determine whether development of LWE early in contact lens wear is prognostic of contact lens intolerance.The time course of development and resolution of LWE is unclear. The lid wiper epithelium shares some common characteristics with the cornea, being a richly innervated, largely unkeratinised, stratified epithelium. The corneal epithelium is known to recover quickly from minor surface damage (Dua et al., 1994). Neighbouring epithelial cells will migrate to cover a modest breach within a matter of hours. There may well be a diurnal pattern of lid wiper damage in both dry eye patients and contact lens wearers, both of whom may experience symptoms more acutely later in the day. The lid wiper may suffer cumulative damage leading to such symptoms, but recover overnight so that on awakening the next morning the individual is symptom free. Then the pattern repeats. Close observation over a period of days will determine this diurnal rhythm.Light microscopy evaluations of excised tissue samples of the lid wiper in patients with LWE have yet to be undertaken. Such studies may reveal chronic changes to the lid wiper epithelium; for example, persistent and excessive frictional rubbing on the lid wiper in a lubricant deficient environment may lead to changes in the morphology of the epithelium, whereby those parts of the lid wiper that are of a stratified cuboidal form change to a more flattened, squamous form. The anatomical features of LWE need to be elucidated in the human eye at a microscopic level, especially in view of the observations of Jalbert et al. (2012) suggesting a pathological basis of LWE (discussed further below).There have only been two studies of longitudinal changes in LWE. Pult et al. (2009) demonstrated that LWE increased significantly three weeks after commencing contact lens wear, and Best et al. (2013) reported that lid wiper staining scores increased from 0.3 ± 0.7 at baseline to 1.5 ± 1.2 at 6 months (p = 0.002). Further longitudinal research is required to determine how long it takes for LWE to become manifest after commencing contact lens wear, as well as the pattern of changes over the course of a day and over longer periods of time. The time course of resolution of LWE after ceasing lens wear, or commencing some form of treatment, also needs to be determined. These time-related aspects of presentation of LWE could be discerned by conducting randomised, controlled longitudinal studies. For example, it would be instructive to monitor changes in the lid wiper, and other associated signs and symptoms, in a cohort of subjects fitted with contact lenses for the first time. Subjects could be followed for a period of lens wear, followed by a period after cessation of lens wear. It is known that adaptation occurs in terms of comfort with wearers of rigid contact lenses (Carracedo et al., 2016). If the lid wiper is a primary source of discomfort during rigid lens wear, then some form of adaptation may be taking place in the lid wiper to enable this effect. Investigation of such adaptive mechanisms in both rigid and soft lenses would provide useful insights into the pathophysiology of LWE.Perhaps more compelling than the conduct of clinical trials, in terms of understanding LWE, are the more fundamental research studies that have been performed to date. Here are three examples that were considered in this review. The work by Jalbert et al. (2012) studying the keratinization-related proteins filaggrin, transglutaminase1 and cytokeratin 1/10 extracted from impression cytology samples, collected from a broad region of the upper lid margin, provides valuable insights into the keratinised status of the lid wiper epithelium. These data also point towards a possible pathological basis for LWE. Such techniques may be able to be applied to study the impact of different contact lens types on the lid wiper.Mucins provide critical lubrication at the interface of the lid wiper and ocular or contact lens surface, so the study of Berry et al. (2008) revealing changes in the composition of these mucin components during symptomatic and asymptomatic contact lens wear may assist our understanding of the role of mucins in LWE. Such analyses could also assist tribologists in modelling frictional effects – and on this note, the engineering science of tribology has the potential to lead us towards novel engineering solutions for optimising the lubricity of contact lens surfaces and therefore minimising LWE. Tracking the sub-clinical inflammatory status of the lid wiper by measuring changes in Langerhans cell density with non-invasive laser scanning confocal microscopy is another promising technique. The findings of Alzahrani et al. (2016a) of an upregulated Langerhans cell response in the lid wiper of those with and without contact lens induced dry eye points towards an inflammatory contribution to the aetiology of this condition. Such an approach could be employed in future work to assess the impact of different lens types on subclinical lid wiper inflammation, and guide researchers towards different strategies for alleviating LWE.Studies of the efficacy of various treatments for LWE are in their infancy, as indicated by the fact that most of the reports to date are in the form of conference abstracts. An increasing number of studies on therapies for dry eye are employing the strategy of monitoring the lid wiper, as one of a battery of tests for assessing dry eye. The flaw in this approach at this point in time is our lack of understanding as to whether or not LWE is truly associated with other dry eye signs and symptoms. Nevertheless, the weight of evidence of such research when published as full refereed papers will undoubtedly further our understanding of LWE and of ways of minimising lid wiper staining.The eyelid is composed of various anatomical and histological substructures and can exhibit many forms of pathology. We have considered a number of conditions that may need to be differentially diagnosed from LWE, but in general, this ought not be difficult in view of the detailed anatomical accounts of the lid wiper (Knop et al., 2011b) and valuable published descriptions and clinical pictures by many authors of the distinct appearance of LWE when viewed in association with appropriate staining agents.AcknowledgementsWe wish to thank Ryan Butterworth for performing the statistical analysis of our meta-analysis (Section 8.3) as well as Chantal Coles-Brennan and Kristy Canavan for study procedural reviews. N.E. and P.B.M. served as consultants to Johnson and Johnson Vision Care, Inc. in the execution of this work. ReferencesAlbietz, J.M., 2000. Prevalence of dry eye subtypes in clinical optometry practice. Optom. Vis. Sci. 77, 357-363.Alghamdi, W., Markoulli, M., Holden, B.A., Papas, E.B., 2016. Impact of duration of contact lens wear on the structure and function of the meibomian glands. Ophthalmic Physiol. Opt. 36, 120-131.Altman, D.G., 1991. Practical Statistics for Medical Research. Chapman and Hall, London.Alzahrani, Y., Colorado, L.H., Pritchard, N., Efron, N., 2016a. Inflammatory cell upregulation of the lid wiper in contact lens dry eye. Optom. Vis. Sci. in press.Alzahrani, Y., Colorado, L.H., Pritchard, N., Efron, N., 2016b. Longitudinal changes in Langerhans cell density in the cornea and conjunctiva in contact lens induced dry eye. Clin. Exp. Optom. in press.Amparo, F., Jin, Y., Hamrah, P., Schaumberg, D.A. and Dana, R., 2014. What is the value of incorporating tear osmolarity measurement in assessing patient response to therapy in dry eye disease? Am. J. Ophthalmol. 157, 69-77.Andrasko, G., 2013. Andrasko corneal staining grid. The Staining Grid Center. . (Last accessed August 31, 2013).Bandamwar, K.L., Papas, E.B., Garrett, Q., 2014. Fluorescein staining and physiological state of corneal epithelial cells. Contact Lens Ant. Eye 37, 213-223.Begley, C.G., Chalmers, R.L., Abetz, L., Venkataraman, K., Mertzanis, P., Caffery, B.A., Snyder, C., Edrington, T., Nelson, D. and Simpson, T., 2003. The relationship between habitual patient-reported symptoms and clinical signs among patients with dry eye of varying severity. Invest. Ophthalmol. Vis. Sci. 44, 4753-4761.Berry, M., Pult, H., Purslow, C., Murphy, P.J., 2008. Mucins and ocular signs in symptomatic and asymptomatic contact lens wear. Optom. Vis. Sci. 85, E930-938. Best, N., Drury, L., Wolffsohn, J.S., 2013. Predicting success with silicone-hydrogel contact lenses in new wearers. Cont. Lens Anterior Eye 36, 232-237. Brennan, N.A., 2009. Contact lens based correlates of soft lens wearing comfort. Optom. Vis. Sci. 86: E-abstract 90957.Brennan, N. A., Coles, C., 2013. Supportive data linking coefficient of friction and comfort. Contact Lens Ant. Eye, 36, e10.Brennan, N.A., Efron, N., 1989. Symptomatology of HEMA contact lens wear. Optom. Vis. Sci. 66, 834-838.Bron, A.J., 2015. The definition and classification of dry eye disease. In: Chan, C., (Ed.), Dry Eye. Springer Berlin, Heidelberg, pp. 1-19.Bron, A.J., Smith, J.A., Calonge, M., 2007. Methodologies to diagnose and monitor dry eye disease: report of the Diagnostic Methodology Subcommittee of the International Dry Eye WorkShop. Ocul. Surf. 5, 108-152. Bron, A.J., Tomlinson, A., Foulks, G.N., Pepose, J.S., Baudouin, C., Geerling, G., Nichols, K.K., Lemp, M.A., 2014. Rethinking dry eye disease: a perspective on clinical implications. Ocul. Surf. 12, S1-S31.Bron, A.J., Yokoi, N., Gaffney, E.A., Tiffany, J.M., 2011. A solute gradient in the tear meniscus. II. Implications for lid margin disease, including meibomian gland dysfunction. Ocul. Surf. 9, 92-97.Caffery, B., Chalmers, R.L., Marsden, H., Nixon, G., Watanabe, R., Harrison, W., Mitchell, G.L., 2014. Correlation of tear osmolarity and dry eye symptoms in convention attendees. Optom. Vis. Sci. 91, 142-149.Caparas, V.L., 2015. Medical management of dry eye. In: Chan, C., (Ed.), Dry Eye. Springer Berlin, Heidelberg, pp. 51-66.Carracedo, G., Martin-Gil, A., Peixoto-de-Matos, S.C., Abejón-Gil, P., Macedo-de-Araújo, R., González-Méijome, J.M., 2016. Symptoms and signs in rigid gas permeable lens wearers during adaptation period. Eye Contact Lens 42, 108-114.Chalmers, R.L., Begley, C.G., Moody, K., Hickson-Curran, S.B., 2012. Contact Lens Dry Eye Questionnaire-8 (CLDEQ-8) and opinion of contact lens performance. Optom. Vis. Sci. 89, 1435-1442.Cher I., 2000. Superior limbic keratoconjunctivitis: multifactorial mechanical pathogenesis. Clin. Experiment. Ophthalmol. 28, 181-184.Cher I., 2003. Blink-related microtrauma: when the ocular surface harms itself. Clin. Experiment. Ophthalmol. 31, 183-190.Cher, I., 2008. A new look at lubrication of the ocular surface: fluid mechanics behind the blinking eyelids. Ocul. Surf. 6, 79-86.Chim, K., Jalbert, I., 2011. The lid margin, its sensitivity and lid wiper epitheliopathy. Optom. Vis. Sci. 87: E-abstract 115115.Coles, C.M.L., Brennan, N.A., 2012. Coefficient of friction and soft contact lens comfort. Optom. Vis. Sci. 88: E-abstract 125603.Collins, M., Heron, H., Larsen, R., Lindner, R., 1987. Blinking patterns in soft contact lens wearers can be altered with training. Am. J. Optom. Physiol. Opt. 64, 100-103.Doane, M.G., 1980. Interactions of eyelids and tears in corneal wetting and the dynamics of the normal human eyeblink. Am. J. Ophthalmol. 89, 507-516.Donald, C., Hamilton, L., Doughty, M.J., 2003. A quantitative assessment of the location and width of Marx’s line along the marginal zone of the human eyelid. Optom. Vis. Sci. 80, 564-572.Donnenfeld, E.D., Perry, H.D., Schrier, A., Zagelbaum, B., Rodgers, R., 1994. Lid imbrication syndrome: Diagnosis with rose bengal staining. Ophthalmology 101, 763-766.Doughty, M.J., 1997. Scanning electron microscopy study of the tarsal and orbital conjunctival surfaces compared to peripheral corneal epithelium in pigmented rabbits. Doc. Ophthalmol. 93, 345-371.Doughty, M.J., 2013. Morphological features of cells along Marx’s line of the marginal conjunctiva of the human eyelid. Clin. Exp. Optom. 96, 76–84Doughty, M.J., Hagan, S., 2013. On the staining of human tissue-cultured (Chang) conjunctival cells with rose bengal and lissamine green. Cont. Lens Anterior Eye 36, 32-40. Doughty, M.J., Naase, T., Donald, C., Hamilton, L., Button, N.F., 2004. Visualisation of "Marx's line" along the marginal eyelid conjunctiva of human subjects with lissamine green dye. Ophthalmic Physiol. Opt. 24, 1-7.Dua, H.S., Gomes, J.A., Singh, A. 1994. Corneal epithelial wound healing. Br. J. Ophthalmol. 78, 401-408.Dumbleton, K.A., Woods, C.A., Jones, L.W., Fonn, D., 2008. Comfort and adaptation to silicone hydrogel lenses for daily wear. Eye Contact Lens 34, 215-223.Dumbleton, K.A., Woods, C.A., Jones, L.W., Fonn, D., 2013. The impact of contemporary contact lenses on contact lens discontinuation. Eye Contact Lens. 39, 93-99. Dunn, A.C., Cobb, J.A., Kantzios, A.N., Lee, S.J., Sarntinoranont, M., Tran-Son-Tyay, R., Sawyer, G., 2008. Friction coefficient measurement of hydrogel materials on living epithelial cells. Tribol. Lett. 30, 13-19.Dunn, A.C., Tichy, J.A., Urue?a, J.M., Sawyer, W.G., 2013. Lubrication regimes in contact lens wear during a blink. Tribol. Int. 63, 45-50.Du Toit, R., Vega, J.A., Fonn, D., Simpson, T., 2003. Diurnal variation of corneal sensitivity and thickness. Cornea 22, 205-209.Edmondson, W., Christenson, M.T., 1992. Lid parasites, in: Onefrey, B., (Ed.), Clinical Optometric Pharmacology and Therapeutics. Lippincott-Raven, Philadelphia.Efron, N., 1985. Is contact lens-induced corneal oedema inflammatory? Aust. J. Optom. 68, 167-172.Efron, N., 2012a. Eyelash disorders, in: Efron, N., (Au.), Contact Lens Complications, 3rd ed., Elsevier-Saunders, Oxford, pp. 67-75.Efron, N., 2012b. Grading scales, in: Efron, N., (Au.), Contact Lens Complications, 3rd ed., Elsevier-Saunders, Oxford, pp. 32-38.Efron, N., 2012c. Is contact lens wear inflammatory? Br. J. Ophthalmol. 96, 1447-1448.Efron, N., 2012d. Meibomian gland dysfunction, in: Efron, N., (Au.), Contact Lens Complications, 3rd ed., Elsevier-Saunders, Oxford, pp. 56-66.Efron, N., 2012e. Papillary conjunctivitis, in: Efron, N., (Au.), Contact Lens Complications, 3rd ed., Elsevier-Saunders, Oxford, pp. 122-132.Efron, N., 2013. Putting vital stains in context. Clin. Exp. Optom. 96, 400-421.Efron, N., 2016. Contact lens wear is inflammatory. Clin Exp. Optom. in pressEfron, N., Al-Dossari, M., Pritchard, N., 2010. Confocal microscopy of the bulbar conjunctiva in contact lens wear. Cornea 29, 43-52.Efron, N., Jones, L., Bron, A.J., Knop, E., Arita, R., Barabino, S., McDermott, A.M., Villani, E., Willcox, M.D., Markoulli, M., 2013. The TFOS International Workshop on Contact Lens Discomfort: Report of the Contact Lens Interactions with the Ocular Surface and Adnexa Subcommittee. Invest. Ophthalmol. Vis. Sci. 54, TFOS98-TFOS122.Ehlers, N., 1965a. On the lid margin, the meibomian glands and their secretion. Acta Ophthalmol. 43 (S81), 67-75.Ehlers, N., 1965b. The precorneal film. Biomicroscopical, histological and chemical investigations. Acta Ophthalmol. 43 (S81), 109-113.El-Rayess, H.M., Greiner, J.V., Herman, J.P., Korb, D.R., Kleiner-Goudy, S.J., 2009. Comparison of corticosteroid and an oil-in-water emulsion in the treatment of lid wiper epitheliopathy (LWE). Invest. Ophthalmol. Vis. Sci. 50:ARVO E-Abstract 546.Fatt, I., 1992. Architecture of the lid-cornea juncture. CLAO J. 18:, 187-192.Feenstra, R.P., Tseng, S.C., 1992. What is actually stained by rose bengal? Arch. Ophthalmol. 110, 984-993.Feinbloom, W., 1938. Contact lens. U.S. Patent 2,129,305, issued September 6, 1938. Washington, DC: U.S. Patent and Trademark Office.Forister, J.F., Forister, E.F., Yeung, K.K., Ye, P., Chung, M.Y., Tsui, A., Weissman, B.A., 2009. Prevalence of contact lens-related complications: UCLA contact lens study. Eye Contact Lens 35, 176-180.Georgiev, G.A., 2015. Controversies regarding the role of polar lipids in human and animal tear film lipid layer. Ocul. Surf. 13, 176-178.Georgouli, T., James, T., Tanner, S., Shelley, D., Nelson, N., Chang, B., Backhouse, O., McGonagle, D., 2008. High-resolution microscopy coil MR-eye. Eye 22, 994-996.Gipson, I.K., 2004. Distribution of mucins at the ocular surface. Exp. Eye Res. 78, 379-388.Golebiowski, B., Chim, K., So, J., Jalbert, I., 2012. Lid margins: sensitivity, staining, meibomian gland dysfunction, and symptoms. Optom. Vis. Sci. 89, 1443-1449.Gray H., Standring, S., Ellis, H., Berkovitz, B. K. B., 2005. Gray’s Anatomy: The Anatomical Basis of Clinical Practice. Churchill Livingstone, Edinburgh.Guillon, M., Maissa, C., 2008. Assessment of upper and lower lid margin with lissamine green. Optom. Vis. Sci. 84: E-abstract 80088.Guthrie, S.E., Jones, L., Blackie, C.A., Korb, D.R., 2015. A comparative study between an oil-in-water emulsion and nonlipid eye drops used for rewetting contact lenses. Eye Contact Lens 41, 373-377. Henriksson, J.T., De Paiva, C.S., Farley, W., Pflugfelder, S.C., Burns, A.R., Bergmanson, J.P., 2013. Morphologic alterations of the palpebral conjunctival epithelium in a dry eye model. Cornea 32, 483-90.Hessen, M., Akpek, E.K., 2014. Dry eye: an inflammatory ocular disease. J. Ophthal. Vis. Res. 9, 240.Hinkle, D.M., 2006. Lid wiper epitheliopathy and dry eye symptoms. Eye Contact Lens 32, 160.Holden, B.A., Stephenson, A., Stretton, S., Sankaridurg, P.R., O'Hare, N., Jalbert, I., Sweeney, D.F., 2001. Superior epithelial arcuate lesions with soft contact lens wear. Optom. Vis. Sci. 78, 912.Hori, Y., Spurr-Michaud, S., Russo, C.L., Argueso, P., Gipson, I.K., 2004. Differential regulation of membrane-associated mucins in the human ocular surface epithelium. Invest. Ophthalmol. Vis. Sci. 45, 114–122.Itakura, H., Kashima, T., Itakura, M., Akiyama, H., Kishi, S., 2013. Topical rebamipide improves lid wiper epitheliopathy. Clin. Ophthalmol. 7, 2137-2141. Jackson, W.B., 1993. Differentiating conjunctivitis of diverse origins. Surv. Ophthalmol. 38 Suppl., 91-104.Jacobson, B., 2003. The Stribeck memorial lecture. Tribol. Int. 36, 781-789.Jalbert, I., Madigan, M.C., Shao, M., Ng, J., Cheng, J., Wong, D., McMonnies, C., 2012. Assessing the human lid margin epithelium using impression cytology. Acta Ophthalmol 90, e547-552.Jalbert, I., Rejab, S., 2015. Increased numbers of Demodex in contact lens wearers. Optom. Vis. Sci. 92, 671-678. Jenkins, M.S., Rehkopf, P.G., Brown, S.I., 1978. A simple device to improve blinking. Am. J. Ophthalmol. 85, 869-872.Jiang, H., Zhong, J., DeBuc, D.C., Tao, A., Xu, Z., Lam, B.L., Liu, C., Wang, J., 2014. Functional slit lamp biomicroscopy for imaging bulbar conjunctival microvasculature in contact lens wearers. Microvasc. Res. 92, 62-71.Jones, L., Brennan, N.A., González-Méijome, J., Lally, J., Maldonado-Codina, C., Schmidt, T.A., Subbaraman, L., Young, G., Nichols, J.J., 2013a. The TFOS International Workshop on Contact Lens Discomfort: Report of the contact lens materials, design, and care subcommittee. Invest. Ophthalmol. Vis. Sci. 54, TFOS37-TFOS70.Jones, L., Varikooty, J., Kier, N., Soong, F., Patel, P., 2013b. The evaluation of lid wiper epitheliopathy in contact lens wearers in a controlled low humidity environmental exposure chamber. Invest. Ophthalmol. Vis. Sci. 54:ARVO E-Abstract 5475.Jones, M.B., Fulford, G.R., Please, C.P., McElwain, D.L.S., Collins, M.J., 2008. Elastohydrodynamics of the eyelid wiper. Bull. Math. Biol. 70, 323-343.Karesh, J.W., Nirankari, V.S., Hameroff, S.B., 1993. Eyelid imbrication: An unrecognized cause of chronic ocular irritation. Ophthalmology 100, 883-889.Kashima, T., Itakura, H., Akiyama, H., Kishi, S., 2014. Rebamipide ophthalmic suspension for the treatment of dry eye syndrome: a critical appraisal. Clin. Ophthalmol. 30, 1003-1010. Kato H., Yokoi, N., Niu, M., Sakai, R., Watanabe, A., Kinoshita, S., 2015. Relationship between lid-wiper epitheliopathy, tear abnormalities, and blink in dry-eye patients. Invest. Ophthalmol. Vis. Sci. 56:ARVO E-Abstract 4454.Kern, J., Rappon, J., Bauman, E., Vaughn, B., 2013. Assessment of the relationship between contact lens coefficient of friction and subject lens comfort. Invest. Ophthalmol. Vis. Sci. 54: ARVO E-abstract 494.Kessing, S.V., 1967. A new division of the conjunctiva on the basis of x-ray examination. Acta Ophthalmol. (Copenh) 45, 680–683.Khanal, S., Millar, T.J., 2010. Nanoscale phase dynamics of the normal tear film. Nanomedicine 6, 707-713.Kinoshita, S., Awamura, S., Oshiden, K., Nakamichi, N., Suzuki, H., Yokoi, N., 2012. Rebamipide Ophthalmic Suspension Phase II Study Group. Rebamipide (OPC-12759) in the treatment of dry eye: a randomized, double-masked, multicenter, placebo-controlled Phase II study. Ophthalmology 119, 2471–2478.Knop, E., Knop, N., Millar, T., Obata, H., Sullivan, D.A., 2011a. The international workshop on meibomian gland dysfunction: report of the subcommittee on anatomy, physiology, and pathophysiology of the meibomian gland. Invest. Ophthalmol. Vis. Sci. 52, 1938-1978.Knop, E., Knop, N., Zhivov, A., Kraak, R., Korb, D.R., Blackie, C., Greiner, J.V., Guthoff, R., 2011b. The lid wiper and mucocutaneous junction anatomy of the human eyelid margins: an in vivo confocal and histological study. J Anat. 218, 449-461. Knop, E.., Korb, D.R., Blackie, C.A., Knop, N., 2010. The lid margin is an underestimated structure for preservation of ocular surface health and development of dry eye disease. Dev. Ophthalmol. 45, 108-122. Knop, N., Korb, D.R., Blackie, C.A., Knop, E., 2012. The lid wiper contains goblet cells and goblet cell crypts for ocular surface lubrication during the blink. Cornea 31, 668-679. Koh, S., Inoue, Y., Sugmimoto, T., Maeda, N., Nishida, K., 2013. Effect of rebamipide ophthalmic suspension on optical quality in the short break-up time type of dry eye. Cornea 32, 1219–1223.Kompa, S., Langefeld, S., Kirchhof, B., Schrage, N., 1999. Corneal biopsy in keratitis performed with the microtrephine. Graefes Arch. Clin. Exp. Ophthalmol. 237, 915-919.Korb, D., 2006. Reply to Hinkle: Lid wiper epitheliopathy and dry eye symptoms. Eye Contact Lens 32, 160.Korb, D.R., Blackie, C.A, 2010. Marx’s line of the upper lid is visible in upgaze without lid eversion. Eye Contact Lens 36, 149-151.Korb, D.R., Greiner, J.V., Herman, J.P., Hebert, E., Finnemore, V.M., Exford, J.M., Glonek, T., Olson, M.C., 2002. Lid-wiper epitheliopathy and dry-eye symptoms in contact lens wearers. CLAO J., 28, 211-216.Korb, D.R., Herman, J.P., Blackie, C.A., Scaffidi, R.C., Greiner, J.V., Exford, J.M., Finnemore, V.M., 2010. Prevalence of lid wiper epitheliopathy in subjects with dry eye signs and symptoms. Cornea 29, 377-383. Korb, D.R., Herman, J.P., Finnemore, V.M., Exford, J.M., Blackie, C.A., 2008. An evaluation of the efficacy of fluorescein, rose bengal, lissamine green, and a new dye mixture for ocular surface staining. Eye Contact Lens 34, 61-64.Korb, D.R., Herman, J.P., Greiner, J.V., Scaffidi, R.C., Finnemore, V.M., Exford, J.M., Blackie, C.A., Douglass, T., 2005. Lid wiper epitheliopathy and dry eye symptoms. Eye Contact Lens 31, 2-8.Korb, D., Korb, J.E., 1974. Fitting to achieve normal blinking and lid action. Int. Contact Lens Clin. 1, 57-61.Kunnen, C.K., Lazon De La Jara, P., Holden, B.A., Papas, E.B., 2014. Automated assessment of lid margin lissamine green staining. Invest. Ophthalmol. Vis. Sci. 55: ARVO E-Abstract 1976.Lane, S.S., DuBiner, H.B., Epstein, R.J., Ernest, P.H, Greiner, J.V., Hardten, D.R., Holland, E.J., Lemp, M.A., McDonald, J.E. 2nd., Silbert, D.I., Blackie, C.A., Stevens, C.A., Bedi, R., 2012. A new system, the LipiFlow, for the treatment of meibomian gland dysfunction. Cornea 31, 396-404.Lee, S., Spencer, N.D., 2008. Materials science. Sweet, hairy, soft, and slippery. Science 319, 575-576.Lin, M.C., Yeh, T.N., 2013. Mechanical complications induced by silicone hydrogel contact lenses. Eye Contact Lens 39, 115-124. Lowther, G.E., Hill, R,M., 1968. Sensitivity threshold of the lower lid margin in the course of adaptation to contact lenses. Am. J. Optom. Arch. Am. Acad. Optom. 45, 587-594.Lum, E., Golebiowski, B., Gunn, R., Babhoota, M., Swarbrick, H., 2013. Corneal sensitivity with contact lenses of different mechanical properties. Optom. Vis. Sci. 90; 954-960.Machado, L.M., Castro, R.S., Fontes, B.M., 2009. Staining patterns in dry eye syndrome: rose bengal versus lissamine green. Cornea 28, 732-734.Maldonado-Codina, C., Read, M., Efron, N., Dobson, C.B., Morgan, P.B., 2013. Observation of solution-induced corneal staining with fluorescein, rose bengal and lissamine green. Contact Lens Ant. Eye 36, 267-270.Marx, E. 1924. ?ber vitale F?bungen am Auge und an den Lidern. I. ?ber Anatomie Physiologie und Pathologie des Augenlidrander und der Tr?nenpunkte. Graefes Arch. Ophthalmol. 114, 465–482.McCulley, J.P., Shine, W., 1997. A compositional based model for the tear film lipid layer. Trans. Am. Ophthalmol. Soc. 95, 79-88.McGowan, D.P., Lawrenson, J.G., Ruskell, G.L., 1994 Touch sensitivity of the eyelid margin and palpebral conjunctiva. Acta Ophthalmol. (Copenh.) 72, 57–60.McMonnies, C.W., 2007. Incomplete blinking: exposure keratopathy, lid wiper epitheliopathy, dry eye, refractive surgery, and dry contact lenses. Contact Lens Ant. Eye 30, 37-51. McMonnies, C.W., 2013. Psychological and other mechanisms for end-of-day soft lens symptoms. Optom. Vis. Sci. 2013, 90, e175-181. McMonnies, C.W., 2015. An examination of the relationship between ocular surface tear osmolarity compartments and epitheliopathy. Ocul. Surf. 13, 110-117.Medzhitov, R., 2008. Origin and physiological roles of inflammation. Nature 454 (7203), 428-435.Millar, T.J., Schuett, B.S., 2015. The real reason for having a meibomian lipid layer covering the outer surface of the tear film – A review. Exp. Eye Res. 137, 125-138.Miller, B.F., Keane, C.B., O’Toole, T., 2005. Miller-Keane Encyclopedia and Dictionary of Medicine, 7th Ed. Saunders-Elsevier, Oxford.Millodot, M., 1972. Diurnal variation of corneal sensitivity. Br. J. Ophthalmol. 56, 844-847.Millodot, M., Henson, D.B., O’Leary, D.J., 1979. Measurement of corneal sensitivity and thickness with PMMA and gas-permeable contact lenses. Am. J. Optom. Physiol. Opt. 56, 628-632.Mokhtarzadeh, M., Casey, R., Glasgow, B.J., 2011. Fluorescein punctate staining traced to superficial corneal epithelial cells by impression cytology and confocal microscopy. Invest. Ophthal. Vis. Sci. 52, 2127–2135.Morgan, P. B., Efron, N., Helland, M., Itoi, M., Jones, D., Nichols, J. J., van der Worp, E., Woods, C. A., 2010. Demographics of international contact lens prescribing. Contact Lens Ant. Eye 33, 27-29.Morgan, P. B., Maldonado-Codina, C., 2009. Corneal staining: do we really understand what we are seeing? Contact Lens Ant. Eye 32, 48–54.Morgan, P.B., Petropoulos, I., Read, M., Malik, R., Maldonado-Codina, C., 2013. Confocal microscopy of the lid margin area of contact lens wearers. Optom. Vis. Sci. 90: E-Abstract 130357.Moss, S.E., Klein, R., Klein, B.E.K., 2000. Prevalence of and risk factors for dry eye syndrome. Arch. Ophthalmol. 118, 1264-1268.Munger, B.L., Halata, Z., 1984. The sensorineural apparatus of the human eyelid. Am. J. Anat. 170, 181–204.Muntz, A., van Doorn, K., Subbaraman, L.N., Jones, L.W., 2015. Impression cytology of the lid wiper area. Invest. Ophthalmol. Vis. Sci. 56:ARVO E-Abstract 4432.Nairn, J.A., Jiang, T.B., 1995. Measurement of the friction and lubricity properties of contact lenses. Proc ANTEC Ann. Tech. Conf. 6, 1–5.Nakamura, T., Nishida, K., Dota, A., Matsuki, M., Yamanishi, K., Kinoshita, S., 2001. Elevated expression of transglutaminase 1 and 2 keratinization related proteins in conjunctiva in severe ocular surface disease. Invest. Ophthalmol. Vis. Sci. 42, 549-556.Navascues-Cornago, M., Maldonado-Codina, C., Gupta, R., Morgan, P.B., 2015a. Characterization of upper eyelid tarsus and lid wiper dimensions. Eye Contact Lens. in press.Navascues-Cornago, M., Morgan, P.B., Maldonado-Codina, C., 2015b. Effect of three interventions on contact lens comfort in symptomatic wearers: a randomized clinical trial. PLoS One 10(8), e0135323.Navascues-Cornago, M., Morgan, P.B., Maldonado-Codina, C., 2015c. Lid margin sensitivity and staining in contact lens wear versus no lens wear. Cornea 34, 808-816. Nelson, J.D., Shimazaki, J., Benitez-del-Castillo, J.M., Craig, J.P., McCulley, J.P., Den, S., Foulks, G.N., 2011. The international workshop on meibomian gland dysfunction: report of the definition and classification subcommittee. Invest. Ophthalmol. Vis. Sci. 52, 1930-1937.Nepp, J., 2011. Influence of alterations of the lid wiper on the temperature of the ocular surface. Invest. Ophthalmol. Vis. Sci. 52:ARVO E-Abstract 1923.Nichols, J.J., Mitchell, G.L., Nichols, K.K., Chalmers, R., Begley, C., 2002. The performance of the contact lens dry eye questionnaire as a screening survey for contact lens-related dry eye. Cornea 21, 469-475.Nichols, K.K., Foulks, G.N., Bron, A.J., Glasgow, B.J., Dogru, M., Tsubota, K., Lemp, M.A., Sullivan, D.A., 2011. The international workshop on meibomian gland dysfunction: executive summary. Invest. Ophthalmol. Vis. Sci., 52, 1922-1929.Nichols, K.K., Nichols, J.J., Mitchell, G.L, 2004. The lack of association between signs and symptoms in patients with dry eye disease. Cornea 23, 762-770Nichols, J.J., Willcox, M.D., Bron, A.J., Belmonte, C., Ciolino, J.B., Craig, J.P., Dogru, M., Foulks, G.N., Jones, L., Nelson, J.D., Nichols, K.K., Purslow, C., Schaumberg, D.A., Stapleton, F., Sullivan, D.A.; members of the TFOS International Workshop on Contact Lens Discomfort, 2013. The TFOS International Workshop on Contact Lens Discomfort: executive summary. Invest. Ophthalmol. Vis. Sci., 54, TFOS7-TFOS13.Ngo, W., Situ, P., Keir, N., Korb, D., Blackie, C., Simpson, T., 2013. Psychometric properties and validation of the Standard Patient Evaluation of Eye Dryness questionnaire. Cornea 32, 1204-1210.Norn, M.S., 1970. Rose bengal vital staining. Staining of cornea and conjunctiva by 10 prcent rose bengal, compared with 1 percent. Acta Ophthalmol (Copenh). 48, 546-559.Norn, M.S, 1973. Conjunctival sensitivity in normal eyes. Acta Ophthalmol. 51, 58-66.Orsborn, G.N., Zantos, S.G., 1988 Corneal desiccation staining with thin high water content contact lenses. CLAO J. 14, 81-85.Oliveira-Soto, L., Efron, N., 2015. Morphology of corneal nerves using confocal microscopy. Cornea 20, 374-384.Papas, E., Tilia, D., McNally, J., de la Jara, P.L., 2015. Ocular discomfort responses after short periods of contact lens wear. Optom. Vis. Sci. 92, 665-670.Parsons, J.H., 1904. The conjunctiva, In: Parsons, J. H., (Au), Pathology of the Eye, Vol 1. Histology. Holder and Stoughton, London, p. 34.Paugh, J.R., Brennan, N.A., Efron, N., 1988. Ocular response to hydrogen peroxide. Am. J. Optom. Physiol. Opt. 65, 91-98.Pesudovs, K., Garamendi, E., Elliott, D.B., 2006. A quality of life comparison of people wearing spectacles or contact lenses or having undergone refractive surgery. J Refract. Surg. 22, 19-27.Pflugfelder, S.C., 2007. Management and therapy of dry eye disease: report of the Management and Therapy Subcommittee of the International Dry Eye WorkShop. Ocul. Surf. 5, 163-178.Pucker, A.D., Haworth, K.M., 2015. The presence and significance of polar meibum and tear lipids. Ocul. Surf. 13, 26-42.Pult, H., Korb, D. K., Blackie, D. A., Knop, E. (2010). About vital staining of the eye and eyelids. I. The anatomy, physiology, and pathology of the eyelid margins and the lacrimal puncta by E. Marx. Optom. Vis. Sci. 2010, 87, 718-724.Pult, H., Murphy, P.J., Purslow, C., 2009. A novel method to predict the dry eye symptoms in new contact lens wearers. Optom. Vis. Sci. 86, E1042-1050. Pult, H., Purslow, C., Berry, M., Murphy, P.J., 2008. Clinical tests for successful contact lens wear: relationship and predictive potential. Optom. Vis. Sci. 85, E924-929. Pult, H., Purslow, C., Murphy, P.J., 2011. The relationship between clinical signs and dry eye symptoms. Eye (Lond) 25, 502-510. Pult, H., Tosatti, S.G., Spencer, N.D., Asfour, J.M., Ebenhoch, M., Murphy, P.J., 2015. Spontaneous blinking from a tribological viewpoint. Ocul. Surf. 13, 236-249. Radke, C.J., Chauhan, A., 2008. Comment on: a new look at lubrication of the ocular surface – fluid mechanics behind the blinking eyelids. Ocul. Surf. 6,152-153.Read, M., Maldonado-Codina, C., Morgan, P.B., Smith, S., 2014. Development of an imaging system to detect changes in redness of the eyelid margin. Optom. Vis. Sci. 90: E-abstract 140083.Roba, M., Duncan, E.G., Hill, G.A., Spencer, N.D., Tosatti, S.G.P., 2011. Friction measurements on contact lenses in their operating environment. Tribol. Lett. 44, 387-397.Romano, A.C., Espana, E.M., Yoo, S.H., Budak, M.T., Wolosin, J.M., Tseng, S.C., 2003 Different cell sizes in human limbal and central corneal basal epithelia measured by confocal microscopy and flow cytometry. Invest. Ophthalmol. Vis. Sci. 44, 5125-5129.Ross G, Nasso M, Franklin V, Lyndon F, Tighe B. Silicone hydrogels: trends in products and properties. Poster presented at: BCLA 29th Clinical Conference & Exhibition; June 3-5, 2005; Brighton, UK. Available at : . (Last accessed 31.08.13)Rumpakis, J., 2010. New data on contact lens dropouts: An international perspective. Rev. Optom. Available at: [Issue: January 15] (Accessed March 26, 2016).Samsom, M.L., Morrison, S., Masala, N., Sullivan, B.D., Sullivan, D.A., Sheardown, H., Schmidt, T.A., 2014. Characterization of full-length recombinant human proteoglycan 4 as an ocular surface boundary lubricant. Exp. Eye Res. 127, 14-19.Santodomingo-Rubido, J., Barrado-Navascués, E., Rubido-Crespo, M.J., 2010. Ocular surface comfort during the day assessed by instant reporting in different types of contact and non-contact lens wearers. Eye Contact Lens 36, 96-100.Satjawatcharaphong, P., Ge, S., Lin, M.C. Clinical outcomes associated with thermal pulsation system treatment. Optom. Vis. Sci. 92, e334-341.Sattler, H., 1877. Beitrag zur kenntnis der normalen bindehaut des menschen. Graefe Arch. Ophthalmol. 23, 1–28. Schulze, M-M.S., Srinivasan, S., Hickson-Curran, S.B., Toubouti, Y., Coz S., Mizra, A., Nichols, J.J., Morgan, P.B., Jones, L.W., 2015. Comparisons between age, gender, lens type and lid wiper epitheliopathy with soft contact lens comfort. Invest. Ophthalmol. Vis. Sci. 56:ARVO E-Abstract 6069.Shaw, A.J., Collins, M.J., Davis, B.A., Carney, L.G., 2010. Eyelid pressure and contact with the ocular surface. Invest. Ophthalmol. Vis. Sci. 51, 1911-1917.Shiraishi, A., Yamaguchi, M., Ohashi, Y., 2014. Prevalence of upper- and lower-lid-wiper epitheliopathy in contact lens wearers and non-wearers. Eye Contact Lens 40, 220-224. Situ, P., Simpson, T.L., Jones, L.W., Fonn, D., 2010. Effects of silicone hydrogel contact lens wear on ocular surface sensitivity to tactile, pneumatic mechanical, and chemical stimulation. Invest. Ophthalmol. Vis. Sci. 51, 6111-6117.Sindt, C.W., Grout, T.K., Critser, D.B., Kern, J.R., Meadows, D.L., 2012. Dendritic immune cell densities in the central cornea associated with soft contact lens types and lens care solution types: a pilot study. Clin. Ophthalmol. 6, 511-519.Sonomura, Y., Yokoi, N., Kato, H., Niu, M., Komuro, A., Kinoshita, S., 2014. The correlations between tear deficiency and conjunctival epithelial damage in dry eye. Invest. Ophthalmol. Vis. Sci. 55:ARVO E-Abstract 2007.Srinivasan, S., Ngo, W., Jones, L.W., 2015. The relief of dry eye signs and symptoms using a combination of lubricants, lid hygiene and ocular nutraceuticals. Invest. Ophthalmol. Vis. Sci. 56:ARVO E-Abstract 4465.Stahl, U.G., Delaveris, A., Madigan, M., Jalbert, I., 2011. Lid wiper epitheliopathy: exploring the links to comfort and osmolality in contact lens wear. Contact Lens Ant. Eye 34 (Supplement 1), S18.Stern, M.E., Schaumburg, C.S., Pflugfelder, S.C., 2013. Dry eye as a mucosal autoimmune disease. Int. Rev. Immunol., 32, 19-41.Su, P.Y., Hu, F.R., Chen, Y.M., Han, J.H., Chen, W.L., 2006 Dendritiform cells found in central cornea by in-vivo confocal microscopy in a patient with mixed bacterial keratitis. Ocul. Immunol. Inflamm. 14, 241-244.Sun, W.S., Baker, R.S., Chuke, J.C., Rouholiman, B.R., Hasan, S.A., Gaza, W., Stava, M.W., Porter, J.D., 1997. Age-related changes in human blinks. Passive and active changes in eyelid kinematics. Invest. Ophthalmol. Vis. Sci. 38, 92-99.Swider, N. D., 2009. LID WIPER Trademark of Ocular Research of Boston, Inc. Registration Number 3593279. Serial Number 78980898. Viewed at: . (Last accessed 03.01.16)Tanioka, H., Kawasaki, S., Sotozono, C., Nakamura, T., Inatomi, T., Kinoshita, S., 2010. The relationship between preoperative clinical scores and immunohistological evaluation of surgically resected tissues in chronic severe ocular surface diseases. Jpn. J. Ophthalmol. 54, 66-73. Thinda, S., Sikh, P.K., Hopp, L.M , Glasgow, B.J., 2010. Polycarbonate membrane impression cytology: evidence for fluorescein staining in normal and dry eye corneas. Br. J. Ophthalmol. 94, 406–409.Tucker, R.C., Quinter, B., Patel, D., Pruitt, J., Nelson, J., 2012. Qualitative and quantitative lubricity of experimental contact lenses. Invest. Ophthalmol. Vis. Sci. 53:ARVO E-Abstract 6093.van der Worp, E., De Brabander, J., Swarbrick, H., Nuijts, R., Hendrikse, F., (2003) Corneal desiccation in rigid contact lens wear: 3- and 9-o'clock staining. Optom. Vis. Sci. 80, 280-290.Varikooty, J., 2015. What is lid wiper epitheliopathy? Contact Lens Spectrum 30 (November), 36-38,40,41.Varikooty, J., Kier, N., Jones, L., 2012. Optimization of assessment and grading for lid wiper epitheliopathy. Optom. Vis. Sci. 88: E-abstract 120241.Varikooty, J., Lay, B., Kier, N., Burdin, H., Jones, L., Simpson, T., Lemp, J., 2013. The relationship between clinical grading and objective image analysis of lid wiper epitheliopathy. Invest. Ophthalmol. Vis. Sci. 54:ARVO E-Abstract 5460.Varikooty, J., Srinivasan, S., Jones, L., 2008. Atypical manifestation of upper lid margin staining in silicone hydrogel lens wearers with symptoms of dry eye. Contact Lens Ant. Eye 31:44-46.Varikooty, J., Srinivasan, S., Subbaraman, L., Woods, C.A., Fonn, D., Simpson, T.L., Jones, L.W., 2015. Variations in observable lid wiper epitheliopathy (LWE) staining patterns in wearers of silicone hydrogel lenses. Contact Lens Ant. Eye 38, 471-476. Varsani, S., Wong, S., 2009. Lid margins staining with lissamine green - contact lens wear and dry eye symptomatology. Optom. Vis. Sci. 85: E-abstract 95679. Velasco, M.J., Bermudez, F.J., Romero, J., Hita, E., 1994. Variations in corneal sensitivity with hydrogel contact lenses. Acta Ophthalmol (Copenh) 72, 53-56.Walt, J., Rowe, M., Stern, K., 1997. Evaluating the functional impact of dry eye: the Ocular Surface Disease Index (Abstract). Drug Inf. J. 31, 1436.Wang, J., Jiang, H., Mao, X., Yan, W., 2015. Quantitative analysis of microvascular network of the lid wiper area in contact lens wearers. Invest. Ophthalmol. Vis. Sci. 56:ARVO E-Abstract 6068.Willis, T., Blackie, C.A., Korb, D., 2011. Meibomian gland function, lid wiper epitheliopathy, and dry eye symptoms. Invest. Ophthalmol. Vis. Sci. 52:ARVO E-Abstract 3740.Wilson, T., Aeschlimann, R., Tosatti, S., Toubouti, Y., Kakkassery, J., Lorenz, K.O., 2015. Coefficient of friction of human corneal tissue. Cornea, 34,1179-1185.Wirth, R.J., Riley, C., 2011. Demonstrating content validity of CLUE: contact lens user experience. Optom. Vis. Sci. 87: E-abstract 110128.Woodward, A.M., Argüeso, P., 2014. Expression analysis of the transmembrane mucin MUC20 in human corneal and conjunctival epithelia. Invest. Ophthalmol. Vis. Sci. 55, 6132-6138.Yamaguchi, M., Kutsuna, M., Uno, T., Zheng, X., Kodama, T., Ohashi, Y., 2006. Marx line: fluorescein staining line on the inner lid as indicator of meibomian gland function. Am. J. Ophthalmol. 141, 669-675.Yamamoto, Y., Shiraishi, A., Sakane, Y., Ohta, K., Yamaguchi, M., Ohashi, Y., 2015. Involvement of eyelid pressure in lid-wiper epitheliopathy. Curr. Eye Res. 19, 1-8. Yan, X.M., Liu, S., Li, H.L., 2008. Preliminary observation the correlation between lid-wiper epitheliopathy and dry eye. Zhonghua Yan. Ke. Za. Zhi. 44, 436-441. Yeniad, B., Beginoglu, M., Bilgin, L.K., 2010. Lid-wiper epitheliopathy in contact lens users and patients with dry eye. Eye Contact Lens 36, 140-143. Yokoi, N., Bron, A.J., Georgiev, G.A., 2014. The precorneal tear film as a fluid shell: the effect of blinking and saccades on tear film distribution and dynamics. Ocul. Surf. 12, 252-266.Yokoi, N., Niu, M., Kato, H., Sakai, R., Komuro, A., Sonomura, Y., Kinoshita, S., 2015. Effect of punctal occlusion on aqueous tear deficient dry eye evaluated from tear film stability and blink-related friction. Invest. Ophthalmol. Vis. Sci. 56:ARVO E-Abstract 4453.Wei, Y., Asbell, P.A., 2014. The core mechanism of dry eye disease (DED) is inflammation. Eye Contact Lens, 40, 248.Zhang, C.W., Mo, J.S., 2011. Preliminary observation of the clinical therapeutic effect of artificial tears combined with bFGF on lid-wiper epitheliopathy. Int. J. Ophthalmol. 11, 1654-1655.Zhivov, A., Stave, J., Vollmar, B., Guthoff, R., 2007. In vivo confocal microscopic evaluation of Langerhans cell density and distribution in the corneal epithelium of healthy volunteers and contact lens wearers. Cornea 26, 47-54.Table 1. Staining agents used to evaluate LWE reported in papers and abstracts.AuthorsYearPaper (P)or abstract (A)Staining agents usedFluoresceinRose bengalLissaminegreenNot statedKorb et al.a2002P??Korb et al.a 2005P??Korb et al.b 2008P??Berry et al.b 2008P??Guillon and Maissa2008A?Pult et al.b 2008P??Yan et al.b 2008A??Varikooty et al.b 2008P??El-Rayess et al.b2009A??Pult et al.b 2009P??Shiraishi et al. 2009A?Varsani and Wong2009A?Korb et al.b2010P??Yeniad et al.2010P???Pult et al.b 2011P??Stahl et al.2011A?Willis et al.b2011A??Zhang and Mo2011A??Golebiowski et al.b2012P??Jalbert et al. 2012P?Varikooty et al.b 2012A??Best et al.b 2013P??Itakura et al. 2013P?Jones et al.b2013A??Morgan et al.2013A?Nepp2013A?Varikooty et al.2013A?Ianchenko et al.2014A?Kunnen et al.2014A?Read et al. 2014A?Shiraishi et al.b 2014P??Sonomura et al.2014A?Guthrie et al.b2015P??Jalbert and Rajab2015P?Kato et al.2015A?Navascues-Cornago et al.2015aP?Navascues-Cornago et al.2015bP?Satjawatcharaphong et al.2015P?Schulze et al.b2015A??Srinivasan et al.2015A?Yamamoto et al. 2015P?Yokoi et al.2015A?Varikooty et al.b 2015P??Alghamdi et al.2016P?Efron et al.b, c2016P??a. Korb protocol A b. Korb protocol B c. Meta-analysis reported in this paper (Section 8.3) Table 2Grading of linear area of staining of lid wiper [after Korb et al. (2002)]Linear Area of StainingGrade< 2 mm02 – 4 mm15 – 9 mm2≥ 10 mm3Table 3Grading of severity of staining of lid wiper [after Korb et al. (2002)]Severity of StainingGradeAbsent0Mild1Moderate2Severe3Figure CaptionsFig.1.Relationship of areas of contact and non-contact between the upper eyelid and ocular surface. The area of the lid wiper starts posterior to the meibomian glands, where the stratified squamous (sic) epithelium changes from keratinised to non-keratinised tissue, and extends superiorly to the subtarsal fold. In this unaltered reproduction of the original illustration of Korb et al (2002), the lid wiper is designated as ‘stratified squamous epithelium’. It should be recognised that the lid wiper has also been described as a stratified conjunctival structure of cuboidal cells (see Section 4.2.2). Reproduced from Korb et al. (2002) with permission of LIPPINCOTT WILLIAMS & WILKINS via Copyright Clearance Center. Fig. 2.Examples of the digital images (top) and analysis (middle and lower) from one subject of a contact imprint of the lid wiper (mean width 0.75 mm) and lissamine green staining of the line of Marx (mean width 0.09 mm). Both widths were averaged over approximately the central 6 mm of the eyelids. Reproduced from Shaw et al. (2010) with permission of ASSOCIATION FOR RESEARCH IN VISION AND OPHTHALMOLOGY via Copyright Clearance Center.Fig. 3.Overview of a lid wiper shows several goblet cells, some of them arranged in groups at the surface and others deep in the epithelium. Goblet cells are stained in serial sections either by Periodic Acid–Schiff (PAS) (A) and Alcian blue (B, C) or by combined staining (D, E). One isolated goblet cell deep inside the epithelium (C, Alcian blue) shows a narrow stained extension (double arrowhead) of the cytoplasmic content that points to the surface and conceivably indicates the delivery of mucus into a cryptal lumen or directly toward the surface. Staining with Alcian blue plus periodic acid–Schiff (D, E) shows that most goblet cells are double stained (dbl in E) and have an intermediate color. E, Some goblet cells stain preferably for either Alcian blue (ab) or periodic acid–Schiff (p). Stained mucus material from a goblet cell (gc) with basal flat nucleus (arrowhead) deep in the epithelium of the lid wiper is seen to be continuous (arrow) with mucus in a cryptal lumen (c), suggesting delivery of mucus into a luminal space. Immunohistochemistry for MUC5AC (F, G) revealed that some goblet cells are strongly positive for MUC5AC (arrowheads in G), whereas others are weakly stained (w) and yet others are negative (asterisks). In places, MUC5AC-positive material is seen deposited on the epithelial surface (F, arrowhead). Note that Alcian blue shows artificial stain concrements (B, C small arrowheads) on the section and on the glass slide. Paraffin histology: (A) periodic acid–Schiff; (B, C) Alcian blue; (D, E) Alcian blue plus periodic acid–Schiff; (F, G) MUC5AC. Size marker in A–G = 10 mm. Reproduced from Knop et al. (2012) with permission of LIPPINCOTT WILLIAMS & WILKINS via Copyright Clearance Center.Fig. 4.Normal everted eyelid, showing the location of the lid wiper and adjacent anatomical structures. Lissamine green stain highlights the mucocutaneous junction (line of Marx). Image courtesy of Maria Navascues Cornago.Fig. 5.(A) The everted upper eyelid with moderate LWE shown by fluorescein staining. The lid wiper was photographed with black light after two instillations of fluorescein dye and a single instillation of rose bengal dye. Rose bengal did not stain this lid wiper (grade 0). The horizontal length of fluorescein staining is more than 15 mm (grade 3), and the average sagittal width of the staining is approximately more than 75% of the lid wiper (grade 3). The fluorescein staining of the lid wiper is not continuous with the irregular border of the line of Marx, but separated by an area of nonstaining epithelium (arrows). Averaging grades 3 and 3 results in a score of 3.0 and a classification of severe LWE for fluorescein staining. The classification of the grade of LWE required averaging the grade of 3.0 for fluorescein staining and the grade of 0 for rose bengal staining, resulting in a final score of 1.5 and a classification of moderate LWE. (B) The everted upper eyelid with moderate LWE shown by rose bengal staining. The lid wiper was photographed with white light illumination after two instillations of fluorescein dye and a single instillation of rose bengal dye. Because fluorescein did not stain this lid wiper (grade 0), the ocular surfaces were irrigated with normal saline to remove the fluorescein and enhance the photograph. The horizontal length of rose bengal staining is approximately 15 mm (grade 3), and the average sagittal width of the staining is approximately 60% of the lid wiper (grade 2). The rose bengal staining of the lid wiper is continuous proximally with the irregular border of the line of Marx (arrows). Averaging grades 3 and 2 results in a score of 2.5 and a classification of severe LWE for rose bengal staining. The classification of the grade of LWE required averaging the grade of 2.5 for rose bengal staining and the grade of 0 for fluorescein staining, resulting in a final score of 1.25 and a classification of moderate LWE. (C) The everted upper eyelid with severe LWE shown by fluorescein and rose bengal staining. The lid wiper was photographed with black light illumination after two instillations of fluorescein dye and a single instillation of rose bengal dye, but without irrigation of the ocular surfaces. Fluorescein is visible as a film on the tarsal conjunctiva. The color of the rose bengal is darkened when viewed with black light. The horizontal length of rose bengal staining is approximately 15 mm (grade 3), and the average width of the staining is approximately 60% of the lid wiper (grade 2). The rose bengal and fluorescein staining of the lid wiper is continuous with the irregular border of the line of Marx (arrows). Averaging grades 3 and 2 results in a score of 2.5 and a classification of severe LWE for rose bengal staining. The horizontal length of fluorescein staining, partially obscured by the rose bengal staining, is more than 15 mm (grade 3), and the average width of the staining is more than 75% (grade 3), resulting in a classification of severe LWE. Averaging grade 2.5 for rose bengal staining and grade 3 for fluorescein staining results in a final score of 2.75 and a classification of severe LWE. Reproduced from Korb et al. (2005) with permission of LIPPINCOTT WILLIAMS & WILKINS via Copyright Clearance Center.Fig. 6.Various appearances of LWE revealed by fluorescein (left column) and lissamine green (right column): (A) vertical streaks, (B) short horizontal band, (C) speckled appearance, (D) comb appearance, (E) broad horizontal band. Reproduced from Varikooty et al. (2015) with permission of ELSEVIER via Copyright Clearance Center.Fig. 7.Contact lens user experience score versus lid wiper epithelial grade. There was no significant association between these two parameters (F = 1.21, p = 0.298).Fig. 8.Plot of median end-of-day comfort with contact lenses from Coles and Brennan (2012) versus contact lens coefficient of friction reported by Ross et al. (2005) (open circles, dashed line, scale above plot area) and Roba et al. (2011) (closed circles, unbroken line, scale below plot area). Reproduced from Jones et al. (2013) with permission of ASSOCIATION FOR RESEARCH IN VISION AND OPHTHALMOLOGY via Copyright Clearance Center.Fig. 9Coefficient of friction in contact lens wearers in low- and high-friction-coefficient contact lenses, and with healthy and collapsed brushes of the lid wiper, in relation to sliding velocity. A (red line): Highest coefficient of friction at low velocity in a dry eye patient (collapsed brush on the lid wiper) wearing high-coefficient-of-friction lenses (insufficient brush at contact lens surface). High shear forces observed at higher velocity due to increased tear film viscosity. B (yellow dashed line): Increased coefficient of friction at low velocity in normal tear film (sufficient brush of the lid wiper) and high-coefficient-of- friction lenses (insufficient brush at contact lens surface). Low shear forces at higher velocity due to low viscosity and non-Newtonian properties of the tear film. C (orange dashed line): Increased coefficient of friction at low velocity in a dry eye patient (collapsed brush on the lid wiper) wearing low-coefficient-of-friction lenses (sufficient brush at contact lens surface). High shear forces at increased velocity due to increased tear film viscosity. D (green line): Lowest coefficient of friction at low velocity in normal tear film (sufficient brush on the lid wiper) and low-coefficient-of-friction lenses (sufficient brush at contact-lens surface). Low shear forces at higher velocity due to low viscosity and non-Newtonian properties of the tear film. Reproduced from Pult et al. (2015) with permission of ETHIS COMMUNICATIONS INC. via Copyright Clearance Center.Fig. 10.View captured from side-mounted CCD camera of Heidelberg laser scanning confocal microscope while imaging the lid wiper. The upper section of the plano face of the Perspex Tomocap can be seen gently applanating the lid wiper. The Tomocap does not contact the exposed tarsal conjunctiva of the everted upper lid, eye lashes or brow. After Alzahrani et al. (in press). Fig. 11.Laser scanning confocal microscope image of Langerhans cells in the stroma of the lid wiper of a participant with contact lens-induced dry eye. Arrows indicate mature Langerhans cells with long dendrites. This image was captured at a depth of 46?m from the epithelial surface. After Alzahrani et al. (in press). Fig. 12.Langerhans cell density (LCD) in the lid wiper of participants with contact lens-induced dry eye (CLIDE), participants without contact lens-induced dry eye (NO-CLIDE) and non-lens wearing control participants. Box shows 25th and 75th percentiles; whiskers show data range; horizontal bar in box shows mean. After Alzahrani et al. (in press). Fig. 13.Staining of the lid wiper with lissamine green. (A) Moderate LWE of the upper eye lid. (B) Severe LWE of the lower eye lid. Images courtesy of Maria Navascues Cornago.Fig. 14.False positive lid wiper staining artifact. Short horizontal bands of staining seen on the lid wiper, caused by the finger touching the lid margin region, before instillation of (A) fluorescein and (B) lissamine green. Reproduced from Varikooty et al. (2015) with permission of ELSEVIER via Copyright Clearance Center.Fig. 15.Pathway for the process of dry eye disease that incorporates the pathological origins of the disease through to the endpoint of ocular surface damage, highlighting the roles of LWE and tear film hyperosmolarity. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download