Meibomian gland dysfunction and its determinants in Iranian adults: A population-based study

Meibomian gland dysfunction and its determinants in Iranian adults: A population-based study

Contact Lens and Anterior Eye xxx (xxxx) xxx–xxx Contents lists available at ScienceDirect Contact Lens and Anterior Eye journal homepage: www.elsev...

167KB Sizes 0 Downloads 31 Views

Contact Lens and Anterior Eye xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Contact Lens and Anterior Eye journal homepage: www.elsevier.com/locate/clae

Full length paper

Meibomian gland dysfunction and its determinants in Iranian adults: A population-based study Hassan Hashemia, Hadith Rastadb, Mohammad Hassan Emamianc, Akbar Fotouhid,



a

Noor Ophthalmology Research Center, Noor Eye Hospital, Tehran, Iran Noor Research Center for Ophthalmic Epidemiology, Noor Eye Hospital, Tehran, Iran c Ophthalmic Epidemiology Research Center, Shahroud University of Medical Sciences, Shahroud, Iran d Department of Epidemiology and Biostatics, School of Public Health, Tehran University of Medical Sciences, Tehran, Iran b

A R T I C L E I N F O

A B S T R A C T

Keywords: Meibomian gland dysfunction Iran Population-based study Prevalence

Purpose: To estimate the prevalence of Meibomian gland dysfunction (MGD) and determine the associated factors in the general population in Iran. Methods: This cross-sectional study is based on the data from the second phase of the Shahroud Eye Cohort Study conducted in 2014. Of the 4737 participants of the second phase, data was available for 4700 people; their mean age was 55.9 ± 6.2 years and 2768 (58.9%) were women. Diagnosis of MGD was made based on the classification of the International Workshop on MGD as judged by the examining ophthalmologist. The prevalence of MGD was summarized as percentage and 95% confidence intervals (CI), and related factors were studied through simple and multiple logistic regressions. Results: The diagnosis of MGD in at least one eye was recorded for 1235 (26.3%) participants; the prevalence of unilateral and bilateral MGD was 26.3% (95% CI: 24.5-28.1) and 26.1% (95% CI: 24.3–27.9), respectively. In the multiple logistic regression analyses, MGD significantly correlated with pinguecula [odds ratio (OR) = 1.3, 95% CI: 1.12–1.50], hypertension (OR = 1.34, 95% CI: 1.11–1.61), high-density lipoprotein (HDL) level (OR = 0.0992, 95% CI: 0.986–0.999), diabetes mellitus (OR = 0.83, 95% CI: 0.71–0.97), and years of education (OR = 0.98, 95% CI: 0.96–0.99). Conclusions: MGD prevalence in this study was lower than the rates reported in other studies on Asian populations. Besides HDL level, MGD is associated with another ocular surface disorder, namely pinguecula, as well as certain systemic diseases such as hypertension and diabetes mellitus. These associations should be taken into consideration when diagnosing MGD.

1. Introduction Meibomian gland dysfunction (MGD) is one of the most common disorders of the eyelids which is often overlooked [1]. The pathogenesis is not completely understood, but in general, it appears that changes in the quality and quantity of Meibomian gland secretions (Meibum) and the resulting changes in the tear lipid composition increases tear evaporation and osmolality, and hence, the ocular surface becomes more prone to many diseases [1–3]. MGD has been shown to be the leading cause of evaporative dry eye [4] which is commonly detected in patients with aqueous-deficient dry eye [4]. MGD, with or without inflammation, can cause discomfort and visual symptoms due to tear film instability. Prolonged inflammation or obstruction of meibomian glands may lead to their atrophy or loss of functioning glands [5]. It is estimated that the prevalence of MGD in the general population



is high (range: 39–50%) [6] and is associated with different systemic diseases such as diabetes mellitus, hypertension, hypercholesterolemia, atopy, and atopic dermatitis [6–11]. The few population-based studies available on this topic have reported MGD prevalence in association with dry eye which ranges from 3.5% in the Salisbury Eye Evaluation Study [12] to 68.3% in the Beijing Eye Study in China [13]. Some fewer studies have investigated potential risk factors related to MGD, although they have mostly studied patients who have MGD and dry eye [14]. In this population-based study, the prevalence of MGD and its association with other ophthalmic and systemic diseases is investigated in an adult population of Shahroud, Iran 2. Methods In this study, data were used from the second phase of the Shahroud

Corresponding author at: Department of Epidemiology and Biostatics, School of Public Health, Tehran University of Medical Sciences, 14155-6446, Tehran, Iran. E-mail address: [email protected] (A. Fotouhi).

http://dx.doi.org/10.1016/j.clae.2017.05.003 Received 12 August 2016; Received in revised form 20 April 2017; Accepted 5 May 2017 1367-0484/ © 2017 Published by Elsevier Ltd on behalf of British Contact Lens Association.

Please cite this article as: Hashemi, H., Contact Lens and Anterior Eye (2017), http://dx.doi.org/10.1016/j.clae.2017.05.003

Contact Lens and Anterior Eye xxx (xxxx) xxx–xxx

H. Hashemi et al.

Eye Cohort Study conducted in 2014. The Shahroud Eye Cohort Study is an epidemiological study on a random sample of 40–64 year old adult residents of Shahroud, which aims to identify the causes of eye disease and visual disorders. The details of the methodology and population sampling are described elsewhere [15]. In brief, from the total population of 40 to 64 year old adults of about 28,000 people, 6311 individuals were selected through random cluster sampling, and 5190 participated in phase I of the study in 2009 (response rate: 82.2%). In the second phase of the study, all baseline participants (5190 people) were invited to have follow-up examinations and 4737 participated (response rate: 91.3%). After completing the interviews, all participants had comprehensive eye examinations including near and distance vision testing, refraction, slit lamp exam, and direct and indirect ophthalmoscopy. For each individual the necessary information in relation to anthropometric measures, blood pressure, lipid profile, fasting blood sugar, and hemoglobin A1c level was recorded. The diagnosis of MGD was based on the clinical judgment of the study ophthalmologist. According to the international workshop on Meibomian gland dysfunction: Report of the diagnosis subcommittee [16], MGD can be clinically categorized into: 1. Asymptomatic MGD, 2. MGD with ocular surface damage, 3. MGD-related evaporative dry eye, and 4. MGD associated with other ocular disorders. In this study all subtypes of above definition were considered as MGD. MGD cases included both hypersecretory and hyposecretory types and the cases with Meibomian gland diseases excluded from MGD definition. This study was approved by the Ethics Committee of Shahroud University of Medical Sciences, and all study procedures adhered to the tenets of the Declaration of Helsinki for human research. Informed written consents were obtained from all participants.

Table 1 Prevalence of Meibomian gland dysfunction in at least one eye by gender, age, and other variables in the Shahroud, Iran, 2014.

2.1. Statistical analysis In this study, the prevalence of MGD and 95% confidence intervals (CI) are provided for the total population as well as study subgroups. In order to investigate the relation between MGD with potential risk factors, simple and multiple logistic regression analyses were used. In the multiple logistic regression analysis model, the forward likelihood ratio method was used to remove non-significant variables. Thus, the final model with the remaining variables provided the adjusted odds ratio and 95% CI. In all analyses, the design effect was calculated to correct for the effect of cluster sampling on calculating the confidence intervals. In all statistical tests, a significance level of 0.05 was considered.

Independent Variables

Number of participants

Prevalence% (95% Confidence Intervals)

Totals

4700

26.3 (24.5–28.1)

Gender

Male Female

1931 2769

27.8 (25.4–30.2) 25.2 (23.2–27.2)

Age group

45–49 50–54 55–59 60–64 65–69

870 1262 1179 857 531

25.4 24.6 25.2 39.9 27.6

(22.2–28.5) (21.8–27.5) (22.6–28.0) (26.4–33.5) (24.0–31.3)

Education level

illiterate primary secondary high school college

525 1483 712 1454 522

29.5 26.9 27.1 24.7 24.2

(25.2–33.4) (24.0–29.8) (23.6–30.6) (21.9–27.5) (20.4–28.0)

BMI group

< 25 ≥25

992 3736

28.6 (25.3–31.8) 25.6 (23.8–27.5)

Current Smoker

no yes

4023 669

26.0 (24.2–27.8) 28.0 (24.4–31.5)

Diabetes

no yes

3547 1115

26.8 (24.9–28.6) 24.5 (21.5–27.4)

Hypertension

no yes

1803 2896

23.1 (20.8–25.5) 28.2 (26.1–30.3)

Losartan Use

no yes

2578 2114

24.6 (22.5–26.7) 28.2 (25.9–30.5)

Pinguecula

no yes

3509 1193

24.7 (22.9–26.5) 30.7 (27.7–33.7)

Pterygium

no yes

4312 390

26.3 (24.5–28.1) 24.9 (20.1–29.6)

and subgroups. All variables of gender, age, years of education, BMI, and diabetes, as well as hypertension, Losartan use, HDL, and pinguecula were entered in simple logistic regression models. The first 5 variables showed a p < 0.20 and were fit in the multiple logistic regression model. The odds ratios obtained from univariate and multivariable logistic regression models are presented in Table 2. Variables which maintained their significance in the final model, in the order added to the model, were pinguecula (OR = 1.30, 95% CI: 1.12–1.50), hypertension (OR = 1.34, 95% CI: 1.11–1.61), HDL level (OR = 0.992, 95% CI: 0.986–0.999), diabetes mellitus (OR = 0.83, 95% CI: 0.71–0.97), and years of education (OR = 0.98, 95% CI: 0.96–0.99).

3. Results Of the 4737 participants, 37 cases had missing data and the final analyses were therefore performed with data from 4700 people with a mean age ( ± SD) of 55.9 ± 6.2, and 2769 (58.9%) of them were women. In total, 1235 participants (26.3%, 95% CI: 24.5–28.1) were diagnosed with MGD in at least one eye, and in more than 99% of these cases, MGD was bilateral. Therefore, the prevalence rates of unilateral and bilateral MGD were 26.3% (95% CI: 24.5–28.1) and 26.1% (95% CI: 24.3–27.9), respectively. Considering MGD in at least one eye as the outcome measure, MGD prevalence was higher in men (27.8%, 95% CI: 25.4–30.2) than woman (25.2%, 95% CI: 23.2–27.2). Among the studied age groups, the 60–64 year age group (39.9%, 95% CI: 26.4–5.4) and among education categories, illiterate people (29.5%, 95% CI: 25.2–33.4) had the highest MGD prevalence rates. The prevalence of MGD was lower in people with body mass index (BMI) > 25 (25.6%, 95% CI: 23.8–27.5) compare to those with BMI < 25 (28.6%, 95% CI: 25.3–31.8). MGD was more common among smokers than non-smokers (28.0% vs. 26.0%). Table 1 shows the prevalence of MGD in at least one eye in the total study population

4. Discussion The MGD prevalence rates reported in the available literature broadly vary, but they generally point to higher rates in Asian populations compared to predominantly Caucasian populations [14]. In the present study, the prevalence of MGD in the 45–69 year old population was 26.3% (95% CI: 24.5–28.1) which is lower compared to other studies conducted on Asian populations (from 46.2% in the Bangkok Study to 69.3% in the Beijing Eye Study) [13,17–19], but higher than population-based studies on Caucasians (from 3.5% in the Salisbury Eye Evaluation Study [12] to 19.9% in the Melbourne Visual Impairment Project [20]). Genetic predisposition of Asians or racial differences may be the cause of the higher prevalence of this disorder in the Asian race. However, due to differences in the study populations and definitions, it is difficult to estimate the exact MGD prevalence rate in the general population, and drawing any comparison between studies or an overall conclusion must be done with caution. 2

Contact Lens and Anterior Eye xxx (xxxx) xxx–xxx

H. Hashemi et al.

asymptomatic cases. In this regard, the population-based study by Siak et al. [27] did not find a significant relation between the male gender and MGD either. In the case-control study by Pinna et al. [8], gender showed no significant effect. Although the exact pathogenesis of the disease is not clear, multiple factors are known to lead to MGD [30,31]. Among possible etiologies, there is evidence confirming the role of sex steroid hormones in causing MGD [1,30,32–36]. Laboratory studies have shown that androgens and estrogens can regulate the metabolism of Meibomian glands and control the gene expression and lipid production in these glands [37,38]. Hormonal change scan lead to different types of MGD, and this usually happens after a course of estrogen replacement therapy or anti-androgen therapy [32,38]. In regards to the role of gender and the direction of it correlation with MGD, inter-study differences may be due to the type of MGD assessed in these studies. The precise mechanism through which sex hormones can cause MGD is still unclear. Despite the general agreement regarding the correlation between MGD and age [4,22,36,39–41], previous studies attribute the impact of aging to age-related atrophy of Meibomian gland acini and decreased lipid expression [2,42–44]. Other observed changes include focal hyper keratinization of the ductal epithelium, cystic dilation, and lipogranulomatoses, though their exact relation with age is not clear [45]. In present study, limiting the sample to the middle-aged (45–69 years old) can be the reason for contradictory results. Based on the findings, a higher level of education was associated with reduced odds of MGD. In the study by Nemet et al. [46], the prevalence of different types of blepharitis including posterior blepharitis, which occurs following Meibomian gland inflammation and dysfunction, was higher in populations with lower socioeconomic status and in urban areas with high population density. In this study, the relation of other socioeconomic factors with different types of blepharitis did not investigated. The role of microbial factors and eyelid health in certain cases of MGD is well known [31]. Hence, with regards to the role of these factors, a lower level of health in individuals with lower social economic status may explain the observed relationship. Unlike pterygium, the association between pinguecula and MGD maintained statistical significance which is in line with the findings of other studies [1,27,47]. Pinguecula can disrupt tear film coverage over the ocular surface, and therefore, may lead to the deterioration of MGD symptoms such as dry eye. A common factor may be causing both diseases. In the present study, no significant correlation between MGD and smoking was found which was in line with Siak et al. [1], but in the study by Viso et al. [27], a significant inverse relation was reported. A more detailed assessment of this variable in terms of the average number of cigarettes per day, number of years of smoking, etc. can help with a better assessment of the impact of smoking on MGD. It was also found a significant inverse relation between HDL levels and MGD. Bukhari et al. [48] demonstrated that dyslipidemia correlated with the degree of MGD rather than its presence. In present study, the degree of MGD was not recorded. The relationship between hypercholesterolemia and MGD has been reported by others [7,8]. Dao et al. [7] studied the association of serum lipid profile with MGD and observed that patients with moderate to severe MGD have a higher total cholesterol level compared to the general population, and the higher level of HDL was the main cause of the higher level of total cholesterol. In other words, unlike the findings in the present study, a higher serum HDL level was a risk factor for MGD with an unknown mechanism. In line with other studies [1,8], in present study a significant relation between BMI and MGD was not found.

Table 2 The effect of different independent variables on Meibomian gland dysfunction (in at least one eye) in simple and multiple logistic regression models, Shahroud, Iran, 2014. P value

Independent variables

Crude Odds Ratio (95% Confidence Interval)

P value

Adjusted Odds Ratio (95% Confidence Interval)

Sex (Male/ Female)

1.14 (1.01–1.30)

0.038



Reference 0.96 (0.79–1.16) 0.99 (0.81–1.2) 1.26 (1.01–1.56) 1.12 (0.89–1.40) 0.86 (0.73–1.01)

– 0.680 0.983 0.038 0.309 0.075

Reference – – – – –

– – – – – –

0.98 (0.97–0.99) 1.10 (0.93–1.30) 1.31 (1.14–1.51) 1.20 (1.05–1.38) 0.88 (0.76–1.03) 1.001 (1.000–1.002) 1.0001 (1.000–1.008) 0.99 (0.986–0.998) 1.35 (1.18–1.55) 0.93 (0.72–1.19)

0.019 0.26 < 0.001 0.007 0.122 0.348

0.98 (0.96–0.99) – 1.34 (1.11–1.61)

0.023 – 0.002

0.83 (0.71–0.97) –

0.021 –

0.771





0.009

0.992 (0.986–0.999) 1.30 (1.12–1.50) –

0.025

Age (Year) 45–49 50–54 55–59 60–64 65–69 Body Mass Index (≥25/ < 25) Education (year) Current Smoker Hypertension Losartan Use Diabetes Total Cholesterol Triglyceride HDL Pinguecula Pterygium

< 0.001 0.549

< 0.001 –

A number of clinic-based studies with smaller sample sizes have been conducted as well [21–25]. Since there are only few populationbased studies on this subject, the information in these clinic-based studies (though limited) can be used in regard to MGD prevalence and its related factors, although the accuracy and generalizability of their results may be limited. Since these studies are conducted on highly selective and specific groups of patients, comparing their findings with those from population-based studies can be difficult. In this study, it was found a significant positive correlation between MGD and hypertension, which is in line with other reports addressing this relationship. [1,26] This relation maybe an indirect inflammatory or hormonal effect, but a direct effect is possible as well. In present study, in contrast to the study by Siaket al. [1], the impact of Losartan did not maintain a statistical significance in the multivariable model; however, in Siak et al. [1], study the impact of Losartanon on MGD was investigated in a multivariable model including diastolic blood pressure only without taking account of systolic blood pressure. In contrast to the study by Pathan [6], it was found a significant inverse relationship between MGD and diabetes mellitus. The eyes, in general, and particularly Meibomian glands, are rich in arteries, and diseases affecting blood vessels can impair blood delivery to these glands and lead to MGD [27]. The direct impact of diabetes on MGD in the study by Viso et al. [27] was observed only in symptomatic cases, and Siak et al. [1] stated that the association was not statistically significant. Lack of symptoms in certain diabetic cases has been attributed to probable sensory neuropathy. In the present study, the observed protective effect of diabetes may be due to the lack of MGD symptoms in diabetic patients. On the other hand, the possible effect of diabetes medications on lipid concentrations (serum lipids and lipid profile) may explain the absence of this expected relation in present study [28,29]. Further studies are warranted to examine this association. In the present study, significant association between MGD and gender was not found. In the population-based study by Viso et al. [27], a significant relation with the male gender was only observed for

5. Conclusion The prevalence of MGD in present study is lower than the rates reported in studies of other Asian populations but higher compared to the reported rates in predominantly Caucasian populations. The 3

Contact Lens and Anterior Eye xxx (xxxx) xxx–xxx

H. Hashemi et al.

[21] B. Ong, J. Larke, Meibomian gland dysfunction: some clinical, biochemical and physical observations, Ophthalmic. Physiol. Opt. 10 (1990) 144–148. [22] M.M. Hom, J.R. Martinson, L.L. Knapp, J.R. Paugh, Prevalence of Meibomian gland dysfunction, Opto. Vis. Sci 67 (1990) 710–712. [23] M. Zhang, J.Q. Chen, Z.G. Liu, L.H. Lou, Q.G. Xiao, Y. Yao, et al., Clinical characteristics of patients with dry eye syndrome, Zhonghua. Yan. Ke. Za. Zhi. 39 (2003) 5–9. [24] J. Shimazaki, E. Goto, M. Ono, S. Shimmura, K. Tsubota, Meibomian gland dysfunction in patients with Sjögren syndrome, Ophthalmology 105 (1998) 1485–1488. [25] B.-L. Ong, Relation between contact lens wear and Meibomian gland dysfunction, Opto. Vis. Sci. 73 (1996) 208–210. [26] K.K. Nichols, G.N. Foulks, A.J. Bron, B.J. Glasgow, M. Dogru, K. Tsubota, et al., The international workshop on meibomian gland dysfunction: executive summary, Invest. Ophthalmol. Vis. Sci. 52 (2011) 1922–1929. [27] E. Viso, M.T. Rodríguez-Ares, D. Abelenda, B. Oubiña, F. Gude, Prevalence of asymptomatic and symptomatic meibomian gland dysfunction in the general population of Spain, Invest. Ophthalmol. Vis. Sci. 53 (2012) 2601–2606. [28] A.C. Robinson, J. Burke, S. Robinson, D.G. Johnston, R.S. Elkeles, The effects of metformin onglycemic control and serum lipids in insulin-treated NIDDM patients with suboptimal metabolic control, Diab. Care. 21 (1998) 701–705. [29] B. Keidan, J. Hsia, R. Katz, Plasma lipids and antidiabetic agents: a brief overview, Br. J. Diab. Vasc. Dis. 2 (2002) 40–43. [30] O.G. Sahin, E. Kartal, N. Taheri, Meibomian gland dysfunction: endocrine aspects, ISRN Ophthalmol. 2011 (2011) 1–6. [31] C. Baudouin, E.M. Messmer, P. Aragona, G. Geerling, Y.A. Akova, J. Benítez-DelCastillo, et al., Revisiting the vicious circle of dry eye disease: a focus on the pathophysiology of meibomian gland dysfunction, Br. J. Ophthalmol. 100 (2016) 300–306. [32] D.A. Sullivan, B.D. Sullivan, J.E. Evans, F. Schirra, H. Yamagami, M. Liu, et al., Androgen deficiency meibomian gland dysfunction, and evaporative dry eye, Ann. N. Y. Acad. Sci. 966 (2002) 211–222. [33] D.A. Sullivan, B.D. Sullivan, M.D. Ullman, E.M. Rocha, K.L. Krenzer, J.M. Cermak, et al., Androgen influence on the meibomian gland, Invest. Ophthalmol. Vis. Sci. 41 (2000) 3732–3742. [34] D.A. Sullivan, H. Yamagami, M. Liu, R.J. Steagall, F. Schirra, T. Suzuki, et al., Sex steroids, the meibomian gland and evaporative dry eye, Adv. Exp. Med. Biol. 506 (2001) 389–399. [35] F. Schirra, T. Suzuki, D.P. Dickinson, D.J. Townsend, I.K. Gipson, D.A. Sullivan, Identification of steroidogenic enzyme mRNAs in the human lacrimal gland, meibomian gland, cornea, and conjunctiva, Cornea 25 (2006) 438–442. [36] S. Den, K. Shimizu, T. Ikeda, K. Tsubota, S. Shimmura, J. Shimazaki, Association between meibomian gland changes and aging sex, or tear function, Cornea 25 (2006) 651–655. [37] F. Schirra, T. Suzuki, S.M. Richards, R.V. Jensen, M. Liu, M.J. Lombardi, et al., Androgen control of gene expression in the mouse meibomian gland, Invest. Ophthalmol. Vis. Sci. 46 (2005) 3666–3675. [38] T. Suzuki, F. Schirra, S.M. Richards, R.V. Jensen, D.A. Sullivan, Estrogen and progesterone control of gene expression in the mouse meibomian gland, Invest, Ophthalmol. Vis. Sci. 49 (2008) 1797–1808. [39] R. Arita, K. Itoh, K. Inoue, S. Amano, Noncontact infrared meibography to document age-related changes of themeibomian glands in a normal population, Ophthalmology 115 (2008) 911–915. [40] M. Guillon, C. Maïssa, Tear film evaporation—effect of age and gender, Cont. Lens. Anterior. Eye. 33 (2010) 171–175. [41] C. Maïssa, M. Guillon, Tear film dynamics and lipidlayer characteristics—effect of age and gender, Cont. Lens. Anterior. Eye. 33 (2010) 176–182. [42] H. Obata, Anatomy and histopathology of human meibomian gland, Cornea 21 (2002) S70–S74. [43] J.D. Nelson, J. Shimazaki, J.M. Benitez-del-Castillo, J.P. Craig, J.P. McCully, S. Den, et al., Theinternational workshop on meibomian gland dysfunction: report of the definition and classification subcommittee, Invest. Ophthalmol. Vis. Sci. 52 (2011) 1930–1937. [44] P. Hykin, A. Bron, Age-related morphological changes in lid margin and meibomian gland anatomy, Cornea 11 (1992) 334–342. [45] V.J. Gutgesell, G.A. Stern, C.I. Hood, Histopathology of meibomian gland dysfunction, Am. J. Ophthalmol. 94 (1982) 383–387. [46] A.Y. Nemet, S. Vinker, I. Kaiserman, Associated morbidity of blepharitis, Ophthalmology 118 (2011) 1062–1068. [47] E. Viso, F. Gude, M. Rodríguez-Ares, Prevalence of pinguecula and pterygium in a general population in Spain, Eye 25 (2011) 350–357. [48] A.A. Bukhari, Associations between the grade of meibomian gland dysfunction and dyslipidemia, Ophthal. Plast. Reconstr. Surg. 29 (2013) 101–103.

evidence obtained from this study shows that besides HDL level, MGD is associated with another ocular surface condition, namely pinguecula, and also with some systemic diseases such as hypertension and diabetes mellitus. These associations should be considered in the diagnosis of MGD. 6. Conflict of interest The authors have no financial or proprietary interest in a product, method, or material described herein. 7. Financial support This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References [1] J.J. Siak, L. Tong, W.L. Wong, H. Cajucom-Uy, M. Rosman, S.M. Saw, et al., Prevalence and risk factors of meibomian gland dysfunction: the Singapore Malay eye study, Cornea 31 (2012) 1223–1228. [2] G.N. Foulks, A.J. Bron, Meibomian gland dysfunction: a clinical scheme for description, diagnosis, classification, and grading, Ocul. Surf. 1 (2003) 107–126. [3] V. Yaylali, C. Ozyurt, Comparison of tear function tests and impression cytology with the ocular findings in acne rosacea, Eur. J. Ophthalmol. 12 (2001) 11–17. [4] C.J. Nien, S. Massei, G. Lin, C. Nabavi, J. Tao, J.R. Paugh, et al., Effects of age and dysfunction on human meibomian glands, Arch. Ophthalmol. 129 (2011) 462–469. [5] V.J. Gutgesell, G.A. Stern, C.I. Hood, Histopathology of meibomian gland dysfunction, Am. J. Ophthalmol. 94 (1982) 383–387. [6] R. Pathan, Prevalence of meibomian gland disease in type II diabetic patients & its clinical presentations, J. Evid. Based. Med. 2 (2015) 346–353. [7] A.H. Dao, J.D. Spindle, B.A. Harp, A. Jacob, A.Z. Chuang, R.W. Yee, Association of dyslipidemia in moderate to severe meibomian gland dysfunction, Am. J. Ophthalmol. 150 (2010) 371–375. [8] A. Pinna, F. Blasetti, A. Zinellu, C. Carru, G. Solinas, Meibomian gland dysfunction and hypercholesterolemia, Ophthalmology 120 (2013) 2385–2389. [9] P. Ena, A. Pinna, F. Carta, Discoid lupus erythematosus of the eyelids associated with staphylococcal blepharitis and Meibomian gland dysfunction, Clin. Exp. Dermatol. 3 (2006) 77–79. [10] E.K. Akpek, A. Merchant, V. Pinar, C.S. Foster, Ocular rosacea: patient characteristics and follow-up, Ophthalmology 104 (1997) 1863–1867. [11] D.A. Schaumberg, R. Dana, J.E. Buring, D.A. Sullivan, Prevalence of dry eye disease among US men: estimates from the Physicians' Health Studies, Arch. Ophthalmol. 127 (2009) 763–768. [12] O.D. Schein, B. MUÑO, J.M. Tielsch, K. Bandeen-Roche, S. West, Prevalence of dry eye among the elderly, Am. J. Ophthalmol. 124 (1997) 723–728. [13] Y. Jie, L. Xu, Y. Wu, J. Jonas, Prevalence of dry eye among adult Chinese in the Beijing Eye Study, Eye 23 (2009) 688–693. [14] D.A. Schaumberg, J.J. Nichols, E.B. Papas, L. Tong, M. Uchino, K.K. Nichols, The international workshop on meibomian gland dysfunction: report of the subcommittee on the epidemiology of and associated risk factors for, MGD, Invest. Ophthalmol. Vis. Sci. 52 (2011) 1994–2005. [15] A. Fotouhi, H. Hashemi, M. Shariati, M.H. Emamian, K. Yazdani, Ebrahim jafarzadehpur, et al., Cohort Profile: shahroud eye cohort study, Int. J. Epidemiol. 42 (2013) 1300–1308. [16] A. Tomlinson, A.J. Bron, D.R. Korb, S. Amano, J.R. Paugh, E.I. Pearce, et al., The international workshop on meibomian gland dysfunction: report of the diagnosis subcommittee, Invest. Ophthalmol. Vis. Sci. 52 (2011) 2006–2049. [17] P.-Y. Lin, S.-Y. Tsai, C.-Y. Cheng, J.-H. Liu, P. Chou, W.-M. Hsu, Prevalence of dry eye among an elderly Chinese population in Taiwan: the Shihpai Eye Study, Ophthalmology 110 (2003) 1096–1101. [18] S.B. Han, J.Y. Hyon, S.J. Woo, J.J. Lee, T.H. Kim, K.W. Kim, Prevalence of dry eye disease in an elderly Korean population, Arch. Ophthalmol. 129 (2011) 633–638. [19] K. Lekhanont, D. Rojanaporn, R.S. Chuck, A. Vongthongsri, Prevalence of dry eye in Bangkok, Thailand, Cornea 25 (2006) 1162–1167. [20] C.A. McCarty, A.K. Bansal, P.M. Livingston, Y.L. Stanislavsky, H.R. Taylor, The epidemiology of dry eye in Melbourne, Australia, Ophthalmology 105 (1998) 1114–1119.

4