Dry eye, a common condition often manifested with visual

Transcription

1 Cornea Screening for Meibomian Gland Disease: Its Relation to Dry Eye Subtypes and Symptoms in a Tertiary Referral Clinic in Singapore Louis Tong, 1,2,3 Shyam S. Chaurasia, 2 Jodhbir S. Mehta, 1,2,3 and Roger W. Beuerman 2,4,5 From the 1 Singapore National Eye Center, Singapore; the 2 Singapore Eye Research Institute, Singapore; the 3 Office of Clinical Sciences, Duke-NUS (National University of Singapore) Graduate Medical School, and the 5 Duke-NUS SRP (Signature Research Program) Neuroscience and Behavioral Disorders, Singapore; and the 4 Department of Ophthalmology, Yong Loo Lin School of Medicine, National University Hospital, Singapore. Supported by National Medical Research Council individual research grants NMRC/1206/2009, IBG 2008, NMRC/CSA/013/2009 and a National University of Singapore-Exxon Mobil Award for Clinician Researchers. Submitted for publication August 7, 2009; revised November 9, 2009, and February 2, 2010; accepted February 7, Disclosure: L. Tong, None; S.S. Chaurasia, None; J.S. Mehta, None; R.W. Buerman, None Corresponding author: Louis Tong, Singapore National Eye Center, 11, Third Hospital Avenue, Singapore ; louis.tong.h.t@snec.com.sg. PURPOSE. To study screening methods and associated factors of Meibomian gland disease and dry eye subtypes in a specialized eye clinic in Singapore. METHODS. This cross-sectional study involved 200 patients in a dry eye clinic. The outcome measures evaluated were dysfunctional tear syndrome (DTS) level, meibomian gland disease grade, Schirmer test (ST) result, fluorescein tear break-up time (TBUT), corneal fluorescein staining grade, and irritative eye symptoms. RESULTS. The meibomian gland screening grade was associated with TBUT (P 0.007), especially in the upper eyelid and correlated with reading difficulty (P 0.007) and reversibility of symptomatic blurring with lubricants (P 0.006). An abnormal ST result was associated with early morning discomfort (P 0.001), and reduced TBUT was linked to discomfort in windy conditions (P 0.001). In all patients examined, evaporative dry eye (58%) was the most common type, followed by the mixed evaporative and aqueous tear deficiency (30.5%) types. Fluorescein staining in the central and inferior cornea was most severe in the mixed type. CONCLUSIONS. In dry eye patients, screening for meibomian gland disease based on anteriorization of Marx s line may predict a decrease in TBUT and difficulty in performing certain visual-function related activities. These findings facilitate better understanding of the meibomian gland s contribution to multifactorial dry eye syndrome, apart from routine conventional tests performed in clinics. (Invest Ophthalmol Vis Sci. 2010;51: ) DOI: /iovs Dry eye, a common condition often manifested with visual disturbance and irritative symptoms, 1,2 may be classified as evaporative or hyposecretory, on the basis of the underlying pathologic processes. This classification often neglects patients with concurrent hyperevaporation and hyposecretion. In addition, it is unclear whether patient demographics are unique. There is no definitive single diagnostic test for dry eye, nevertheless, the clinical signs and symptoms of the disorder are important in the diagnosis, assessment of disease progression, and monitoring of treatment outcomes. 3 Currently, it is unclear whether clinical symptoms and signs of dry eye help in evaluating the different aspects of this multifactorial disease. 4 Objective tests for dry eye include the Schirmer test (ST), which measures tear production; tear break-up time (TBUT), which is an indicator of tear film stability; and dye staining tests, which assess ocular surface epithelial damage. Population-based 5,6 and other studies 7 9 have shown that the presence of dry eye symptoms and objective tests demonstrate poor agreement. However, 96% of symptomatic individuals exhibited one abnormal clinical sign. 10 Clearly, more studies are needed to define how symptoms and signs relate to each other. 11 Meibomian gland disease (MGD) is a common condition that is prevalent throughout Asia, (e.g., Taiwan, 57.8% 62.5%, 10 and Thailand, 50%). 12 The Salisbury Eye Study found the prevalence of MGD (based on meibomian gland plugging or collarettes) to be only 3.5% (95% confidence interval [CI]: ). 13 Although MGD is classified as a non dry eye disease in a recent publication of ocular surface disease classification, 3 it has been linked to evaporative dry eye. 14 There have been discrepancies in associating MGD with specific clinical features of dry eye. Results in a Chinese and Taiwanese study showed no correlation of MGD with dry eye symptoms 15 or ST, 16 whereas in the Salisbury Eye Study, MGD was associated with dry eye symptoms. 13 Moreover, several population-based studies reporting dry eye 5,17 19 did not report its relationship with MGD. Recently, a simple method of assessment for MGD based on the position of Marx s line has been proposed. 20 The correlation of this test with dry eye signs and severity, however, remains to be evaluated. We wanted to achieve the following in an Asian dry eye clinic: characterize patients with aqueous deficiency of evaporative or mixed etiologies, analyze the relationship between symptoms and signs, and evaluate a screening test for MGD and its associations with dry eye. METHODS A hospital clinic based, cross-sectional assessment of dry eye patients was conducted in Singapore National Eye Center from 2006 to The study protocol formed part of a routine, ongoing arm of a longterm clinical audit and was approved by the Institutional Review Board of the Singapore Eye Research Institute. All study procedures were performed as part of standard clinical care and complied with the tenets of the Declaration of Helsinki regarding human research, and as all procedures performed were essential for standard clinical care of these patients, written consent was not required, but consent was obtained by assent. The patients were aware of the privacy policy of the hospital that states that information released for publication would not include patient identifiers. Investigative Ophthalmology & Visual Science, July 2010, Vol. 51, No. 7 Copyright Association for Research in Vision and Ophthalmology 3449

2 3450 Tong et al. IOVS, July 2010, Vol. 51, No. 7 There were no uniform criteria for referral to this specialized eye clinic within the Singapore National Eye Center, as individual ophthalmologists exercise their clinical judgment concerning a diagnosis of dry eye. However, the assessment at the clinic was performed by a single ophthalmologist (LT). Data from the first consecutive 200 patients in the clinic with a diagnosis of dry eye were analyzed. Assessment of Predisposing Factors The assessment included a history of dry mouth, previous or current contact lens wear, and known history of diabetes mellitus, rheumatoid arthritis, thyroid disease, or ocular surgery (LASIK or cataract surgery). Use of the following systemic medications was determined by a clinician-administered standardized form: oral contraceptives and hormonal preparations, antihistamines, antispasmodics, and drugs for hypertension, depression, and Parkinsonism. Exposure keratopathy and lid deformity were clinically assessed in addition to an ocular surface examination. Assessment of Dry Eye All patients underwent a TBUT test, grading of corneal fluorescein staining, and ST in that order. TBUT was evaluated by placing a single fluorescein strip over the inferior tear meniscus after shaking off a small drop of normal saline. According to standard instructions, time between eye opening and the first appearance of a dry spot on the cornea was recorded as the TBUT. ST (without anesthesia) was performed by placing a standard strip (Sno Strips; Chauvin, Montpelier, France) over the midlateral one third of the lower lid margin and noting the extent of wetness on the strip after 5 minutes. Spots of corneal fluorescein staining were recorded over the central zone and the superior, inferior, nasal, and temporal quadrants, as described elsewhere. 3 The questionnaire used (Appendix) was adapted from the Ocular Surface Disease Index. 21 All symptoms were graded as 0 (never present) or 1, 2, 3, or 4 (always present-at least once a day in the past week). There were two questions on whether activities (driving a car and computer use) precipitated or aggravated symptoms that were graded 0 (never) to 6 (always aggravated). There were two additional questions on the physical factors (wind or drafts and air conditioning) that may aggravate the symptoms that were graded 0 (never) to 4 (always aggravated). Dry eye was classified according to Delphi Consensus Dysfunctional Tear Syndrome (DTS) levels 22 : DTS I (mildest), II, III, and IV (most severe). Briefly, cases with no corneal fluorescein staining at all, but with symptoms of dry eye, were designated as DTS I; cases with staining but sparing the central zone were designated as DTS II; cases with filaments or staining involving the central zone were designated DTS III; and severe cases with signs, such as opacified corneas or symblepharon, were designated DTS IV. All features in the DTS classification, 22 except conjunctival staining, were considered. The severity of MGD was screened as previously described by using a series of tests: Yamaguchi grading, 20 microscopic signs of MGD, and expressivity of the meibomian glands. Briefly, we designated Yamaguchi grade 0 (least severe) to indicate a posterior location of the mucocutaneous junction (Marx s line), indicated by fluorescein dye staining; grade 3 (most severe) to indicate an anterior position of the Marx s line relative to the meibomian gland (MG) orifices; and grades 1 and 2 to indicate intermediate positions of Marx s line. Two-sectors per eyelid grading, which requires a total of eight assessments of the two eyes of a patient, instead of three-sector grading was adopted to save time in busy ophthalmic clinics. Microscopic signs of MGD include the presence of misdirected lashes, fragility of lashes, scurf formation, irregularity of the MG orifice, loss of expressibility of the MG, and formation of plaques. Expressibility of the MG was assessed by gently squeezing the lower eyelids, and the consistency of the expressed secretion was graded as watery or viscous. Data Analysis The ST results were dichotomized into normal or abnormal ( 8mmin either eye), and the TBUT into normal or abnormal ( 5 seconds), by using previous 4 and other mean-value based thresholds. Evaporative dry eye was defined as abnormal TBUT but normal ST result. Hyposecretory dry eye was defined as normal TBUT and abnormal ST result, and mixed dry eye was defined as abnormal TBUT and ST result. The Yamaguchi score was analyzed for each of the lid sectors, and the total score for each eye amounted to the sum of the individual grades of the four sectors in that eye. Data analysis for categorical and continuous variables was performed with the 2 test and either the t-test or the analysis of variance (ANOVA), respectively (SPSS Inc., Chicago, IL). For significant changes in the ANOVA, post hoc tests with Bonferroni correction were applied. When using ANOVA to analyze the age and the extent of central and inferior corneal fluorescein staining in different dry eye categories, multiple paired comparisons were used in the statistical software. A separate Bonferroni correction was applied, dividing 0.05 by the number of remaining statistical comparisons to obtain a corrected. Because the analyses for each of the three goals of the study (characterization of dry eye subtypes, comparison of symptoms and signs of dry eye, and characterization of MGD) differ in the number of statistical comparisons (3, 26, and 7, respectively), the corrected values were different: 0.05/3 or 0.017, 0.05/26 or and 0.05/7 or , respectively. For logistic regression models with a binary outcome variable, such as a specific dry eye symptom present in 40% of 200 participants, the power of detecting an odds ratio of 1.5 at 0.05 for a single independent variable is ( eugened/ power-samplesize.php). Unless otherwise stated, only one eye (right) of each patient was used in the analyses presented in the results, although we also performed the same analyses to check for consistency with the left eye data. RESULTS Demographics of Dry Eye Clinic Patients The mean age of the patients was years, and 62.5% were women. Eighty seven percent were Chinese and 13% of minority ethnicity (3.5% Asian Indians, 2% Caucasians, 1.5% Malay, and 6% other). Table 1 shows classification into DTS levels. Being female (P 0.006), but not age (P 0.496), was associated with a higher DTS level. Age correlated directly with the total Yamaguchi score (r 0.44, P 0.001) and inversely but weakly correlated with TBUT (r 0.19, P 0.007) and ST result (r 0.16, P 0.026). A history of dry mouth was found in 19.5% of the patients, 11.5% were wearing contact lens currently (at least once in the past week), and 3.5% had worn them in the past. A known history of diabetes mellitus was found in 1.5% of the patients, 3% had rheumatoid arthritis, 5.5% had thyroid disease, and 1.0% had undergone ocular surgery (LASIK or cataract). Seventeen percent of the patients had used one or more systemic medications within the past week. There were no patients with significant exposure keratopathy or lid deformity. Subtypes of Dry Eye Clinic Patients Thirty-four percent and 88.5% of participants had abnormal ST ( 8 mm) and TBUT ( 5 seconds), respectively. Based on these tests, the patients were classified into four categories (Table 2, Fig. 1). Age (P 0.001), but not sex (P 0.36), was associated with the dry eye category (Table 2): The patients who had normal TBUT and ST result had a mean age of years, younger than patients with evaporative dry eye, whose

3 IOVS, July 2010, Vol. 51, No. 7 Dry Eye and Meibomitis 3451 TABLE 1. Dry Eye by Severity Levels DTS 1 DTS 2 DTS 3 Total n (%) 28 (14) 72 (36) 100 (50) 200 (100) Age, mean y SD Females % TBUT, mean s SD ST, mean mm SD Central corneal staining % (n) 0.0 (0) 0.0 (0) 70.0 (70) 35.0 (70) Inferior corneal staining, % (n) 0.0 (0) 72.2 (52) 72.0 (72) 62.0 (124) Symptom score, mean SD* Yamaguchi score, mean SD Total Upper temporal Upper nasal Lower temporal Lower nasal * Calculated by adding the scores for all 12 questions related to dry eye symptoms. The sum of the individual grades of the four sectors in that eye. mean age was years (P 0.006), or mixed dry eye, with a mean age of years (P 0.001). The presence of central cornea fluorescein staining was different between the subtypes of dry eye (P 0.004). The mixed type had more staining (P 0.042) than did the evaporative type, with a mean of spots compared with spots. Inferior corneal staining also varied between dry eye types (P 0.001), with the mixed type showing more staining than the evaporative type (P 0.016) or those with normal ST result and TBUT (P 0.006). The dry eye subtypes had similar individual or total symptom scores (all P 0.05). Relationship between Clinical Features and Dry Eye Severity More severe dry eye (DTS level 3) was associated with a lower ST outcome (P 0.019) and a lower TBUT (P 0.026). The mean ST result in those with and without severe dry eye was and mm, respectively. The mean TBUT in the patients with and without severe dry eye was and seconds, respectively. In clinical terms, these differences were not significant. Table 3 shows that TBUT (using threshold: 3.2 seconds) and inferior corneal staining were the only significant predictors of DTS level 3. The symptom scores (individual or total) were similar between DTS levels (P 0.05). Relationship between Clinical Signs and Symptoms of Dry Eye Pain in the night or on waking up were more frequent in those with abnormal ST findings (P 0.001). Reduced TBUT was linked to discomfort in windy conditions (P 0.001). The total symptom scores were not different between the ST and TBUT categories. The number of staining spots in the inferior cornea was associated with abnormal ST result (P 0.001) but not TBUT (P 0.227). Staining in the central cornea was not associated with ST result (P 0.596) or TBUT (P 0.776). The presence of any corneal fluorescein staining was also not associated with abnormal ST result (P 0.034) or TBUT (P 0.249). The ST result did not correlate with TBUT (r 0.123, P 0.082). Relationship between MGD and Clinical Features of Dry Eye The total Yamaguchi score was associated with decreased TBUT (P 0.007), but not abnormal ST results (P 0.376). When each eyelid segment was analyzed individually, the Yamaguchi score in the upper nasal (P 0.007), but not in the upper temporal, lower temporal, or lower nasal lid segments, was associated with abnormal TBUT. TABLE 2. Dry Eye Subtypes Evaporative Mixed Aqueous Tear Deficiency ST and TBUT Normal P n (%) 116 (58) 61 (30.5) 7 (3.5) 16 (8) Age, mean y SD * Females % TBUT, mean s SD ST, mean mm SD Central corneal staining, % (n) 32.8 (38) 44.3 (27) 42.9 (3) 12.5 (2) Inferior corneal staining, % (n) 62.1 (72) 67.2 (41) 71.4 (5) 37.5 (6) Symptom score, mean SD Yamaguchi score, mean SD Total Upper temporal Upper nasal Lower temporal Lower nasal * P Calculated by adding the scores for all 12 questions related to dry eye symptoms. The sum of the individual grades of the four sectors in that eye.

4 3452 Tong et al. IOVS, July 2010, Vol. 51, No. 7 TABLE 4. Logistic Regression Models with Transient Visual Blurring and Nocturnal/Early Morning Symptoms as Outcome Variables Crude OR Adjusted OR* FIGURE 1. The four subtypes of dry eye patients studied. Top left quadrant: evaporative dry eye; bottom left quadrant: mixed evaporative and hyposecretory dry eye; bottom right quadrant: hyposecretory dry eye; top right quadrant: normal ST result and TBUT. The horizontal and vertical lines indicate conventional thresholds for ST and TBUT, respectively. The percentages indicate the proportion of participants in each group (classification based on right eyes). The total Yamaguchi score correlated weakly with the total symptom score (r 0.151, P 0.032), and correlated only to improvement of vision with artificial tears (r 0.195, P 0.006) and discomfort with reading or driving a car (r 0.191, P 0.007), among individual symptoms. Table 4 shows that having a higher score in the upper eyelids increased the risk of having blurred vision that was correctable with lubricants. Table 5 shows that that sort of score was also associated with difficulty in reading or driving a car. The grades in the lower eyelid segments (data not shown) were not associated with these activities. Microscopic signs of MGD (such as MG plugging) were not associated with ST result or TBUT or with symptoms or staining. DISCUSSION This clinic-based study showed that a simple MGD screening test based on the position of Marx s line in the eyelid was correlated to TBUT and visual functional difficulties. Abnormal Dependent Variable: Blurring of Vision Improving with Artificial Tears Abnormal ST, 8 mm ( ) ( ) Abnormal ST, 13 mm ( ) ( ) Abnormal TBUT, 5 s 0.9 ( ) ( ) Abnormal TBUT, 3 s ( ) ( ) Inferior corneal staining ( ) ( ) Central corneal staining ( ) 1.63 ( ) Yamaguchi score 1.51 ( ) ( ) Right upper temporal Right upper nasal ( ) ( ) Dependent Variable: Burning in Middle of Night or on Waking Abnormal ST, 8 mm ( ) ( ) Abnormal ST, 13 mm ( ) ( ) Abnormal TBUT, 5 s ( ) ( ) Abnormal TBUT, 3 s ( ) ( ) Inferior corneal staining ( ) ( ) Central corneal staining ( ) ( ) Yamaguchi score Right upper temporal ( ) ( ) Right upper nasal 0.96 ( ) ( ) * Adjusted by age, sex. TBUT was associated with discomfort in windy conditions, whereas abnormal ST was associated with early morning or nocturnal discomfort. Fluorescein staining in the central and inferior cornea was most severe in mixed evaporative and hyposecretory dry eye. The sex of the patient, but not the age, was associated with DTS score and age, but not sex, was associated with dry eye category, perhaps because the DTS category is predominantly influenced by the corneal staining, whereas the dry eye category is influenced by ST result and TBUT. Comparison with Previous Studies The advancement of Marx s line observed in the study and its relation to TBUT was consistent with that in previous studies. The MG secretes the lipid layer in human tears 23 and lack of lipid confluence increases evaporation fourfold. 24 Previously, TABLE 3. Multiple Logistic Regressions with Severe Dry Eye (DTS at Least 3) as the Dependent Variable Crude OR Model 1 OR Model 2 OR Abnormal ST, 8 mm ( ) ( ) ( ) Abnormal ST, 13 mm ( ) ( ) ( ) Abnormal TBUT, 5 s ( ) ( ) ( ) Abnormal TBUT, 3 s ( ) ( ) ( ) Inferior corneal staining ( ) ( ) ( ) ( ) Yamaguchi score Upper temporal ( ) ( ) ( ) ( ) Upper nasal ( ) ( ) ( ) ( ) Model 1 was adjusted for age and sex. Model 2 was adjusted for the covariates age, sex, Yamaguchi upper temporal and upper nasal, and inferior corneal fluorescein staining spots, positive TBUT, positive ST result, and inferior corneal staining. Inferior corneal staining and Yamaguchi scores have two odds ratios in Model 2: top row, ST ( 8 mm) and TBUT ( 5 s) as covariates; bottom row: ST ( 13 mm) and TBUT ( 3 s) as covariates. OR, odds ratio.

5 IOVS, July 2010, Vol. 51, No. 7 Dry Eye and Meibomitis 3453 TABLE 5. Logistic Regression Models Showing Reading and Watching Television/Computer Use as Aggravating Factors Crude OR Adjusted OR* Dependent Variable: Irritation with Reading or Driving Abnormal ST, 8 mm ( ) ( ) Abnormal ST, 13 mm ( ) ( ) Abnormal TBUT, 5 s ( ) 2.02 ( ) Abnormal TBUT, 3 s ( ) ( ) Inferior corneal staining ( ) ( ) Central corneal staining 1.07 ( ) 1.03 ( ) Yamaguchi score Right upper temporal ( ) ( ) Right upper nasal ( ) ( ) Dependent Variable: Irritation with Watching Television/Computer Use Abnormal ST, 8 mm ( ) ( ) Abnormal ST, 13 mm ( ) 1.67 ( ) Abnormal TBUT, 5 s ( ) ( ) Abnormal TBUT, 3 s ( ) ( ) Inferior corneal staining ( ) ( ) Central corneal staining ( ) ( ) Yamaguchi Right upper temporal ( ) ( ) Right upper nasal ( ) ( ) * Adjusted by age, sex. the presence of at least one of six dry eye symptoms, but not ST or TBUT, was associated with MGD. 12 In our study, almost all patients had abnormal TBUT but fewer than half had a reduced ST result. A Chinese study 10 found that TBUT was reduced ( 10 seconds) in 79.3% (95% CI, ) whereas only 58.4% (95% CI, ) had a decreased ST result ( 5 mm). In an Australian study, 5 however, abnormal TBUT ( 8 seconds, 8.6%) was less prevalent than abnormal ST ( 8 mm, 16.3%). A Thai study showed similar prevalence of abnormal TBUT ( 10 seconds, 54.7%) and abnormal ST-I ( 10 mm, 50%). 12 Contrary to a previous study, 16 we did not find correlation between the ST and TBUT. These discrepancies between studies originate from differences in thresholds, TBUT and ST techniques used, 3 or from participant differences from population-based studies. Strengths The strength of this study is the thorough assessment of the meibomian glands and dry eye status by a single examiner in a large sample of patients. Greater uniformity and robustness was achieved by prospective data collection. The clinical MGD screening method was based on anatomic changes in MG. 20 Limitations Our data were not compared to normal and MGD cases without dry eye. Therefore, we did not calculate the efficiency of various tests as screening modalities. This deficiency could result in a selection bias. For practical reasons, we did not perform some other tests such as the ST-II (Schirmer s test performed after topical anesthesia application), tear osmolarity, meibography, and meibometry. However, there is no current consensus on the most appropriate diagnostic criteria or combination of criteria. 2,25 Clinical Relevance In clinics with dry eye patients, information on aggravation of symptoms by television viewing or computer use may be used to supplement a diagnosis of evaporative dry eye and MGD. In addition, the Yamaguchi grading may be used to support these diagnoses if slit lamp microscopy is available. These patients may be offered MGD-targeted therapy 14 (reduction of meibomitis-related inflammation), lipid-containing eye drops, education on lid hygiene, warm compresses, doxycycline, and stabilization of tears. 26,27 MGD may be a contributor to dry eye, especially in the evaporative subgroup. A similar opinion has also been sounded in the literature. 22 Since TBUT is abnormal in most of the patients in our study, it is a better single screening tool than ST for detecting dry eye. However, we recommend performing other tests such as ST and Yamaguchi grading to classify dry eye and predict MGD, respectively. MGD assessment may distinguish the predominantly evaporative type of dry eye from the predominantly aqueous-deficient dry eye. It is important in future studies to determine whether the Yamaguchi score correlates with the degree of observed inflammation of the margin of the eyelid as well as the character of the meibum and degree of meibomian gland plugging. CONCLUSIONS In this clinic-based study conducted in Singapore, an MGD screening test based on anteriorization of Marx s line provides clinical information in addition to conventional dry eye tests (i.e., ST result, TBUT, and fluorescein staining) and is associated with reversibility of visual blurring and difficulty in reading. More studies are needed to evaluate MGD screening in people without dry eye. References 1. Miljanovic B, Dana R, Sullivan DA, Schaumberg DA. Impact of dry eye syndrome on vision-related quality of life. Am J Ophthalmol. 2007;143: Pflugfelder SC, Solomon A, Stern ME. The diagnosis and management of dry eye: a twenty-five-year review. Cornea. 2000;19: Methodologies to diagnose and monitor dry eye disease: report of the Diagnostic Methodology Subcommittee of the International Dry Eye WorkShop. Ocul Surf. 2007;5: The definition and classification of dry eye disease: report of the Definition and Classification Subcommittee of the International Dry Eye WorkShop. Ocul Surf. 2007;5: McCarty CA, Bansal AK, Livingston PM, Stanislavsky YL, Taylor HR. The epidemiology of dry eye in Melbourne, Australia. Ophthalmology. 1998;105: Schein OD, Tielsch JM, Munoz B, Bandeen-Roche K, West S. Relation between signs and symptoms of dry eye in the elderly: a population-based perspective. Ophthalmology. 1997;104: Begley CG, Chalmers RL, Abetz L, et al. The relationship between habitual patient-reported symptoms and clinical signs among patients with dry eye of varying severity. Invest Ophthalmol Vis Sci. 2003;44: Chalmers RL, Begley CG, Edrington T, et al. The agreement between self-assessment and clinician assessment of dry eye severity. Cornea. 2005;24: Nichols KK, Nichols JJ, Mitchell GL. The lack of association between signs and symptoms in patients with dry eye disease. Cornea. 2004;23: Lin PY, Tsai SY, Cheng CY, Liu JH, Chou P, Hsu WM. Prevalence of dry eye among an elderly Chinese population in Taiwan: the Shihpai Eye Study. Ophthalmology. 2003;110: Nichols KK. Patient-reported symptoms in dry dye disease. Ocul Surf. 2006;4: Lekhanont K, Rojanaporn D, Chuck RS, Vongthongsri A. Prevalence of dry eye in Bangkok, Thailand. Cornea. 2006;25:

6 3454 Tong et al. IOVS, July 2010, Vol. 51, No Schein OD, Munoz B, Tielsch JM, Bandeen-Roche K, West S. Prevalence of dry eye among the elderly. Am J Ophthalmol. 1997;124: Bron A, Tiffany JM. Contribution of Meibomian gland disease to dry eye. Ocul Surf. 2004;2: Jie Y, Xu L, Wu YY, Jonas JB. Prevalence of dry eye among adult Chinese in the Beijing Eye Study. Eye. 2009;23(3): Lin PY, Cheng CY, Hsu WM, et al. Association between symptoms and signs of dry eye among an elderly Chinese population in Taiwan: the Shihpai Eye Study. Invest Ophthalmol Vis Sci. 2005; 46: Caffery BE, Richter D, Simpson T, Fonn D, Doughty M, Gordon K. CANDEES: The Canadian Dry Eye Epidemiology Study. Adv Exp Med Biol. 1998;438: Moss SE, Klein R, Klein BE. Prevalence of and risk factors for dry eye syndrome. Arch Ophthalmol. 2000;118: Schaumberg DA, Sullivan DA, Buring JE, Dana MR. Prevalence of dry eye syndrome among US women. Am J Ophthalmol. 2003; 136: Yamaguchi M, Kutsuna M, Uno T, Zheng X, Kodama T, Ohashi Y. Marx line: fluorescein staining line on the inner lid as indicator of meibomian gland function. Am J Ophthalmol. 2006; 141: Schiffman RM, Christianson MD, Jacobsen G, Hirsch JD, Reis BL. Reliability and validity of the Ocular Surface Disease Index. Arch Ophthalmol. 2000;118: Behrens A, Doyle JJ, Stern L, et al. Dysfunctional tear syndrome: a Delphi approach to treatment recommendations. Cornea. 2006; 25: Mishima S, Maurice DM. The oily layer of the tear film and evaporation from the corneal surface. Exp Eye Res. 1961;1: Craig JP, Tomlinson A. Importance of the lipid layer in human tear film stability and evaporation. Optom Vis Sci. 1997;74: Lemp MA. Report of the National Eye Institute/Industry workshop on Clinical Trials in Dry Eyes. CLAO J. 1995;21: Pflugfelder SC. Antiinflammatory therapy for dry eye. Am J Ophthalmol. 2004;137: Sheppard JD. Dry eye moves beyond palliative therapy. Manag Care. 2003;12:6 8. APPENDIX Dry eye questionnaire used in the study: 1. Are your eyes sensitive to light? 2. Do you have a gritty or scratchy sensation in your eyes? 3. Do you have a burning or stinging sensation in your eyes? 4. Do you have blurred vision? 5. Do you have vision fluctuating with blinking? 6. Do you have vision fluctuating with artificial tears? 7. Do you have spontaneous tearing? 8. Do you have pain or burning in the eyes in the middle of the night or on awakening in the morning?