Aims To describe the prevalence and associations of epiretinal membrane (ERM) in an elderly urban Chinese population in China.
Methods A population-based cross-sectional study was conducted using a cluster random sample of residents aged 50 years or older living in the Jiangning Road Sub-district, Shanghai, China. All participants underwent a standardised interview and comprehensive eye examinations, including digital retinal photography and spectral domain optical coherence tomography (SD-OCT) examinations of both eyes between November 2012 and February 2013. ERM was identified and classified as either cellophane macular reflex (CMR) or preretinal macular fibrosis (PMF) based on the fundus photography and OCT features.
Results Of the 2044 subjects who participated (82.5% response rate), 2005 had fundus photographs and OCT results of sufficient quality for grading of ERM signs. ERM was present in 8.4% of participants, including 5.0% with CMR and 3.4% with PMF. After age and gender standardisation to the 2010 Chinese census population, the prevalence rate of ERM in mainland Chinese individuals of 50 years of age or older was estimated to be 7.3%. After adjusting for age and/or gender, idiopathic ERM was positively associated with age (OR, 1.06; 95% CI 1.04 to 1.08), female gender (OR 1.67, 95% CI 1.16 to 2.40), myopia (OR 2.21, 95% CI 1.51 to 3.22) and hyperlipaemia (OR 1.80, 95% CI 1.04 to 3.12).
Conclusions The prevalence of ERM in elderly urban Chinese was similar to that in Caucasians. Risk factors for idiopathic ERM were older age, female gender, myopia and hyperlipaemia.
Statistics from Altmetric.com
Epiretinal membrane (ERM) is a condition characterised by the proliferation of abnormal tissues on the inner retinal surface.1 Studies have reported a great discrepancy in the prevalence of ERM, ranging from 1.0% to 28.9% among different ethnic groups.2–16 In relation to the prevalence of ERM in the Chinese population, three studies have provided population-based prevalence data on ERM, ranging from 1.0% to 3.4%.10 ,11 ,15 The results suggested that the prevalence of ERM is much lower in the Chinese population than in the Caucasian population. However, in another multi-ethnic population study, American of Chinese origin was identified as a risk factor for ERM.13 Additional population-based epidemiology studies should be conducted to better understand the epidemiology of ERM in China.
Studies on ERM have traditionally relied on fundus photography for identifying and characterising ERM. Recently, the development of spectral domain optical coherence tomography (SD-OCT) made it possible to obtain high-resolution near-histologic images of the retina and epiretinal structure, and this has significantly increased the diagnostic capabilities, with some studies detecting ERM in up to 90% of cases.17 ,18 The purpose of the current study was to investigate the prevalence and factors associated with ERM in an elderly urban Chinese population, based on the fundus photography and SD-OCT.
The Jiangning Eye Study, a population-based cross-sectional study of Chinese urban elders aged 50 years and older, living in the Jiangning Road Sub-district, Shanghai, was conducted to investigate the prevalence and risk factors of eye disease. The study design and the details of the population sampling for the Jiangning Eye Study have been described elsewhere.19 Informed written consent was obtained from all participants before enrolment. The study complied with the guidelines in the Declaration of Helsinki, and ethical approval was received from the Medical Ethics Committee of the Xinhua Hospital, Shanghai Jiao Tong University School of Medicine.
A detailed interviewer-administrated questionnaire was conducted to collect information about medical history, cigarette smoking, alcohol consumption, current medication use and socioeconomic status factors. The eye examinations were conducted according to a standardised protocol that included visual acuity measurement, autorefraction, non-contact tonometry, slit-lamp biomicroscopy, indirect ophthalmoscopy, axial length and SD-OCT (Topcon 3D OCT-2000; Topcon). Fundus photography was undertaken with a 45° 16.2-megapixel digital non-mydriatic fundus camera (integrated high-resolution fundus camera of Topcon 3D OCT-2000; Topcon) in a darkened room.
Definition and classification of ERM
ERM was diagnosed based on the fundus photography and OCT images. For the photographic grading, ERM was classified into two forms.2 ,11 Earlier stage of ERM was defined as the presence of cellophane macular reflex (CMR) only, characterised by a patch or patches of irregular increased reflection from the inner surface of the retina. The later stage of ERM was defined as the presence of preretinal macular fibrosis (PMF), characterised by a thickening and contraction of the membrane, with opaque or grey superficial retinal folds or traction lines. On the OCT images, ERM was defined as hyper-reflective signals at the inner retinal surface and evidence of contractility, including any distortion, corrugation or flattening of the inner retina.11 ,20 Any ERM was defined as the presence of either CMR or PMF in one or both eyes. Participants who had both CMR and PMF were classified as having PMF.
ERM was also classified as primary (idiopathic) or secondary. Secondary ERM was defined as any ERM present in the eyes with other ocular conditions associated with ERM, including retinal vein occlusion, diabetic retinopathy, late age-related macular degeneration, retinal detachment or a post-cataract surgery state.2 ,4
Two experienced graders initially assessed the fundus photographs and OCT images for ERM independently in a masked fashion. All eyes recorded as having ERM were verified by a third experienced grader. Any discrepancies were discussed and adjudicated by two retinal specialists.
The overall age-specific and gender-specific prevalence (%) of ERM was calculated. The age-standardised and gender-standardised prevalence was estimated using direct standardisation of the study samples to the Chinese population of the 2010 Chinese census.21 Logistic regression analyses were conducted to estimate the ORs and the 95% CIs, adjusting for age and gender. Statistical analysis was performed using the Statistical Package for Social Science (SPSS V.15.0, SPSS, Chicago, Illinois, USA). All p values were two-sided and were considered statistically significant when the values were <0.05.
Of 2478 eligible subjects identified for this study, 2044 (82.5% response rate) underwent an eye examination between November 2012 and February 2013. Of these individuals, 2005 had fundus photographs and OCT results of sufficient quality for grading of ERM signs. Before the grading was initiated for all subjects, intergrader agreements were assessed using the κ statistic on a random subset of 60 eyes. The κ for the presence of CMR and PMF was 0.9 and 1.0, respectively.
The number of cases and individual eyes diagnosed by fundus photography only and by both fundus photography and OCT images were shown in table 1. When diagnosed by both fundus photography and OCT, ERM was observed in 207 eyes of 169 participants and was present bilaterally in 38 cases (22.5%). The prevalence of ERM was similar in the right (4.9%, 99/2005) and left eyes (5.4%, 108/2005, p=0.568).
In the present study, CMR, PMF and any ERM were found in 5.0% (101/2005), 3.4% (68/2005) and 8.4% (169/2005) participants, respectively. The age-standardised and gender-standardised prevalence of CMR, PMF and any ERM in the 2010 Chinese census population aged 50 years or older was 4.3% (95% CI 3.4% to 5.2%), 3.0% (95% CI 2.3% to3.8%) and 7.3% (95% CI 6.2% to 8.5%), respectively (table 2). The prevalence of any ERM, as well as of CMR and PMF when considered separately, increased significantly with increasing age (p for trend <0.01 for all).
Of the 169 persons with ERM, 26 (15.4%) cases were secondary ERM. The most frequent cause of secondary ERM was a history of cataract surgery (65.4%, 17/26).
Logistic regression analyses were performed to assess factors significantly associated with idiopathic ERM, after removing participants with any known secondary cause of ERM (table 3). Older age was associated with a higher likelihood of ERM, with increased odds of approximately 6% per year of age (OR, 1.06; 95% CI 1.04 to 1.08). Female gender was associated with ERM (OR 1.67, 95% CI 1.16 to 2.40). After adjustment for age and gender, axial myopia (axial length ≥25 mm; OR 2.21, 95% CI 1.51 to 3.22) and hyperlipaemia (OR 1.80, 95% CI 1.04 to 3.12) were associated with ERM. Other characteristics assessed included a history of arthritis, stroke, hypertension and diabetes, body mass index, smoking (current vs past or non-smoker), alcohol intake, intraocular pressure and level of education. After adjustment for age and gender, none of these previously suggested risk factors was significantly associated with ERM in the current study (OR: 1.00–1.47, all p>0.05).
This population-based study reports on the prevalence of ERM in an elderly Chinese population in an urban setting in China and documents an overall ERM prevalence of 8.4%, made up of 5.0% with CMR and 3.4% with PMF. After age and gender standardisation to the 2010 Chinese census population, the prevalence rate of ERM in mainland Chinese individuals of 50 years of age or older was estimated to be 7.3%.
The prevalence rate in our study was markedly higher than the rates in the Beijing Eye Study (2.2%)10 and the Beixinjing Block Study (1.0%).15 Although the reasons for the discrepancies remain unclear, potential sources of errors could have led to underestimation of ERM in both the previous studies. First, ERM was diagnosed using only non-stereoscopic photography in the Beijing and Beixinjing studies, which may have missed subtle early macular changes, especially CMR. It has been reported that more than one-third of ERM could have been missed with a strictly photograph-based approach.20 ,22 Second, ERM rates may appear lower in eyes with other ocular diseases, such as cataract and myopia, owing to the increased difficulty of detecting this lesion. Notably, myopia is prevalent in Chinese.23 However, it is difficult to identify ERM in eyes with myopic degenerative retinal changes with fundus photographs only.11 The current study differed from these two studies in that SD-OCT was performed for all participants in the current study, so that ERM was diagnosed with both fundus photography and SD-OCT images. SD-OCT can provide useful data on the anatomical features of the macula in some cases where no macular details can be seen on the fundus photography, and this significantly increased the ERM rate detected in our study. The Handan Eye Study also detected ERM from both fundus photography and OCT in a large rural Chinese population and reported a slightly lower ERM rate (6.7%)11 than that of our study after age standardisation for participants of 50 years of age or older (table 4). This difference in the prevalence of ERM between our study and the Handan Eye Study can be explained by urban–rural and regional differences.
The prevalence of ERM appears to vary with ethnicity.13 In previous population-based studies in Chinese, the ERM prevalence was lower than that in mainly Caucasian populations.10 ,11 ,15 However, for participants aged >50 years, the ERM prevalence rate (7.6%) in the Chinese population of our study was similar to that in Caucasian populations (Melbourne Visual Impairment Project (VIP) Study, 7.4%,7 and Melbourne Collaborative Cohort Study, 7.8%16) and higher than that of the Blue Mountains Eye Study (5.5%)3 after age standardisation to the WHO world population.24 This rate was lower than that in an Indian population (Singapore Indian Eye Study, 12.1%),14 Malay population (Singapore Malay Eye Study, 13.4%)9 and Hispanic population (Los Angeles Latino Eye Study, 24.0%)6 (table 4).
As in previous population-based studies,2 ,5 ,7 ,9 ,10 ,12–14 ,16 increasing age was significantly associated with idiopathic ERM. Female gender was associated with ERM after adjusting for age.6 ,9 In contrast to reports from the Beijing Eye Study,10 the Singapore Malay Eye Study9 and the Blue Mountains Eye Study,3 our data indicated that axial myopia was significantly associated with idiopathic ERM after adjusting for age and gender. This result was also reported by the Handan Eye Study,11 the Singapore Indian Eye Study14 and the Melbourne VIP Study.7 Possible reasons for the positive association between myopia and ERM include increased posterior vitreous detachment and vitreoretinal traction in myopic eyes, which are both known risk factors for ERM.25 As mentioned above, the increased detection of ERM by SD-OCT in participants with myopia may have helped to detect this association. This association between myopia and ERM could become increasingly important in Chinese as the prevalence and severity of myopia increases.14 In addition, we found that hyperlipaemia was related to ERM after adjusting for age and gender.5 ,13 However, we could not confirm previous associations of idiopathic ERM with diabetes,3 ,12 ,13 smoking7 ,9 ,11 or vascular risk factors9 after age and gender adjustment.
The strengths of our study include its large sample size and the high response rate in a population-based sample, and we performed SD-OCT on all participants to detect possible ERM. However, this study does have several limitations. First, fundus photography was performed using a non-mydriatic non-stereo camera, and this could have missed subtle ERM and other subtle retinal abnormalities even though SD-OCT was adopted as the diagnostic tool of choice. Second, several potential risk factors reported previously could not be assessed in current study. Several characteristics were assessed only on the basis of participants’ self-reported data. For example, diabetes tended to be underestimated in our study, as this information was based on the self-reported history only, without plasma glucose testing. Third, our study is subject to the limitations of a cross-sectional design, and further prospective studies are needed to confirm the incidence and risk factors of ERM.
In summary, we described the age-specific and gender-specific prevalence of ERM lesions in an elderly urban Chinese population in mainland China. We found an ERM prevalence similar to that reported for Caucasian populations. In addition to age, female gender, myopia and hyperlipaemia were significantly associated with idiopathic ERM in this older Chinese population.
The authors thank Jian Zhang (Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou, China) for providing suggestions on the study sampling and statistical analysis.
HY and QZ contributed equally.
Contributors Conception and design of the study (HY, QZ, PZ); analysis and interpretation (HY, QZ, XL, BY, PZ); writing of the article (HY, QZ); critical revision of the article (XL, PZ); final approval of the article (HY, QZ, XL, XC, WY, SY, TW, WL, XL, YH, BY, PZ); data collection (HY, XL, XC, WY, SY, TW, WL, XL, YH, BY); obtaining funding (HY, PZ); literature search (HY, XL, XC); administrative, technical or logistic support (QZ, XL, PZ).
Funding This study was supported by the National Natural Science Foundation Project of China (81200682, 81271045, 81470642), the Shanghai Outstanding Young Scientist Foundation from the Shanghai Municipal Health Bureau (XBR2011060) and the Outstanding Scientist Foundation from the Science and Technology Commission of Shanghai Municipality (15XD1502800).
Competing interests None declared.
Patient consent Obtained.
Ethics approval The Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.