Prevalence of Glaucoma in North China: The Beijing Eye Study


To assess the prevalence of glaucoma in adult Chinese.


Population-based study.


The Beijing Eye Study in 2001 included 4439 subjects with an age of 40+ years. Glaucoma was determined using the ISGEO (International Society of Geographical and Epidemiological Ophthalmology) classification scheme.


Glaucoma was detected in 158 subjects (3.7%, 95% confidence interval [CI] 3.1%–4.2%), in which open-angle glaucoma (OAG), primary angle-closure glaucoma (PACG), and secondary glaucoma (SG) accounted for 2.6% (95% CI 2.1%–3.0%), 1.0% (95% CI 0.7%–1.3%), and 0.07% (95% CI 0%–0.1%), respectively. The overall glaucoma prevalence for the age groups of 40 to 49 years, 50 to 59 years, 60 to 69 years, and 70+ years was 2.2%, 2.5%, 4.4%, and 9.5%, respectively. Glaucoma prevalence increased significantly with age ( P = .001), myopic refractive error ( P < .001), and intraocular pressure ( P < .001). The age-standardized prevalence of OAG, PACG, and SG was 2.5%, 1.0%, and 0.1%, respectively. Prevalence of glaucoma-related bilateral blindness or unilateral blindness was significantly ( P = .02 and P = .03) higher in PACG than in OAG. Frequency of glaucoma-associated blindness and low vision was significantly higher in the rural area than in the urban region (6/62 vs 2/96, P = .04).


In the adult population of Greater Beijing, glaucoma prevalence was 3.6% and increased with age, myopic refractive error, and intraocular pressure. Glaucoma prevalence of 3.6% was comparable with figures from Caucasian populations. The ratio of OAG to PACG of 2.6:1 agrees with recent other studies from East Asia. Glaucoma-related blindness and low vision occurred significantly more often in PACG than in OAG.

Glaucoma has been estimated to affect about 66 million people worldwide, with Asians accounting for almost half of the world’s glaucoma population. Recent studies on the prevalence of glaucoma revealed that primary angle-closure glaucoma was more common in Sino-Mongolian peoples than in Caucasians. Some of these studies suggested that primary angle-closure glaucoma, compared with chronic open-angle glaucoma, was the major form of glaucoma in Asia. In 1989, Hu and associates examined residents of Shunyi County, Beijing, and found a glaucoma prevalence of 1.4% in the population with an age of 40 or more years. Primary angle-closure glaucoma was by far the most common type of glaucoma in that study. In the more recent Liwan Eye Study in Guangzhou, South China, He and associates applied the criteria of the International Society of Geographical and Epidemiological Ophthalmology (ISGEO) for the uniform definition of glaucoma. The crude glaucoma prevalence was 3.8% (95% confidence interval [CI] 2.8%–4.8%). Open-angle glaucoma was found in 2.1% and primary angle-closure glaucoma in 1.5% of the study participants, with a ratio of open-angle glaucoma to primary angle-closure glaucoma of 1.4:1. The age-adjusted rate of open-angle glaucoma was similar to that found in European-derived populations, while the rate of primary angle-closure glaucoma was higher in the Chinese population.

Since there have been no other recent glaucoma-related population-based studies in mainland China, particularly not in North China or for the rural population, and since the number of glaucoma patients in China represents a major portion of the global glaucoma population, we conducted the present investigation in rural and urban regions of Greater Beijing to examine the prevalence of glaucoma overall, the ratio of open-angle glaucoma to primary angle-closure glaucoma, differences in the prevalence of both glaucoma types between the rural region and the urban region, and the influence of parameters such as age, gender, and refractive error on the prevalence of glaucoma.


The Beijing Eye Study is a population-based prospective cohort study in northern China that has been described in detail recently. It was carried out in 4 communities in the urban district of Haidian in the north of central Beijing and in 3 communities in the village area of Yufa of the Daxing District south of Beijing. Using this register as the sampling frame, all subjects living in the 7 communities and fulfilling the inclusion criterion of an age of 40+ years were eligible for the study. At the time of the survey in the year 2001, the 7 communities had a total population of 5324 individuals aged 40 years or older. In total, 4439 individuals (2505 women) participated in the eye examination, corresponding to an overall response rate of 83.4%. Mean age was 56.2 ± 10.6 years (range, 40–101 years). The study was divided into a rural part (1973 subjects [44.4%]; 1143 women; 3946 eyes) and an urban part (2466 subjects [55.6%]; 1362 women; 4932 eyes). The level of education was significantly ( P < .001) lower, and the reported family income was significantly ( P < .001) lower, in the rural region than in the urban region.

All examinations were carried out in the communities, either in schoolhouses or in community houses. After measurement of uncorrected visual acuity, automatic refractometry (Auto Refractometer AR-610; Nidek Co, Ltd, Tokyo, Japan) was performed, if uncorrected visual acuity was lower than 1.0. Visual field examinations were performed by frequency-doubling perimetry using the screening program C-20–1 (Zeiss-Humphrey, Dublin, California, USA). The rate of false-positive results had to be equal to or lower than 0.33, and the rate of fixation loss had to be equal to or lower than 0.33. The perimetric results were analyzed with reference to visual acuity and fundus appearance to ascertain the cause for a visual field defect. Intraocular pressure was measured using a noncontact pneumotonometer (CT–60 computerized tonometer; Topcon Ltd, Tokyo, Japan) by an experienced technician. Three measurements were taken, and the mean of the 3 measurements was taken for further statistical analysis. A slit-lamp examination was carried out by an ophthalmologist. During the slit-lamp examination, the anterior chamber depth was assessed using van Herick’s method. The pupil was dilated using tropicamide once or twice, until the pupil diameter was at least 6 mm. Photographs of the cornea and lens were taken (Neitz Instruments Co, Tokyo, Japan). The lens opacities were graded using the Age-Related Eye Disease Study (AREDS) system for classifying cataracts. Past history of eye diseases, eye trauma, diabetes mellitus, hypertension, and any ophthalmologic care the participant received were recorded. Additional information was obtained on family income.

Photographs (on film) of the macula and optic disc were taken using a fundus camera (Type CR6-45NM; Canon Inc, Tokyo, Japan). The optic disc photographs were digitized, and the width of the neuroretinal rim and the diameters of the optic cup and optic disc were measured in the vertical meridian of the optic disc. The vertical cup-to-disc diameter ratio (VCDR) was calculated. Glaucoma was defined according to the criteria of the ISGEO. In that definition, criteria for a category 1 diagnosis (structural and functional evidence) were a VCDR or an inter-eye asymmetry in the VCDR of ≥97.5th percentile for the normal population, or a neuroretinal rim width reduced to ≤0.1 VCDR (between 11 and 1 o’clock or 5 and 7 o’clock), in addition to a definite visual field defect consistent with glaucoma. Criteria for the category 2 diagnosis (advanced structural damage with unproven visual field loss) were a VCDR or a VCDR asymmetry ≥99.5th percentile for the normal population. Criteria for a category 3 diagnosis (for eyes the optic nerve head of which could not be examined or for which a visual field examination was not possible) were a visual acuity <3/60 combined with either an intraocular pressure >99.5th percentile or definite glaucoma medical records such as filtering surgery history.

The whole glaucoma group was differentiated into subjects with open-angle glaucoma (OAG), subjects with primary angle-closure glaucoma (PACG), and subjects with secondary glaucoma (SG). The differentiation between open-angle glaucoma and primary angle-closure glaucoma was performed in several steps. According to Goldmann gonioscopy, which was carried out for all glaucoma subjects and glaucoma suspects, open-angle glaucoma was characterized by an open anterior chamber angle, in addition to a normal depth of the anterior chamber as assessed by slit-lamp biomicroscopy. A glaucoma suspect was any person with an intraocular pressure higher than 21 mm Hg, a shallow anterior chamber, a narrow anterior chamber angle as assessed by slit-lamp biomicroscopy, anterior segment signs of primary angle-closure glaucoma such as glaukomflecken of the lens, or any glaucoma-like appearance of the optic nerve head as assessed during the examinations in the school or community houses. Gonioscopy was performed by an experienced ophthalmologist trained in glaucoma. A Goldmann 3-mirror contact lens was used. In primary angle-closure glaucoma, the anterior chamber angle was occluded or occludable. Using the definition by Foster and associates, the anterior chamber angle was defined as occludable if ≥270 degrees of the posterior trabecular meshwork could not be seen upon gonioscopy. In addition, other features for primary angle-closure glaucoma were iris whirling and glaukomflecken in the anterior subcapsular lens region, in combination with a narrow anterior chamber angle. Those glaucoma subjects who had not undergone gonioscopy (25 subjects out of 158 subjects with glaucoma [16%]) were divided into open-angle glaucoma versus primary angle-closure glaucoma based on the assessment of the anterior chamber angle width as performed by slit-lamp biomicroscopy using van Herick’s method (n = all 25 glaucoma subjects), and additionally based on the assessment of the anterior chamber angle on optical coherence tomographic images, taken at a follow-up examination in 2006 (n = 10 of the 25 subjects). Subjects with a detected reason for elevated intraocular pressure or glaucomatous optic nerve damage, such as iris neovascularization or goniosynechiae attributable to preceding intraocular surgeries, were considered as having secondary glaucoma.

Statistical analysis was performed using SPSS for Windows, version 17.0 (SPSS, Chicago, Illinois, USA). Continuous data were presented as mean ± standard deviation. χ 2 tests were used to compare proportions. Logistic or linear regression models were used to investigate the associations of the presence of glaucoma with the continuous (eg, intraocular pressure) or categorical outcomes (eg, diabetic retinopathy). For continuous variables, beta coefficients (β) are reported; for categorical variables, odds ratios (OR) are presented, and 95% confidence intervals (CI) are described. All P values were 2-sided and were considered statistically significant when the values were less than 0.05.


Optic disc photographs were available for 4315 of the 4439 subjects (97.2%; 8458 eyes), with 1879 subjects (43.5%) living in the rural region and 2436 (56.5%) living in the urban region. For 3926 subjects, optic disc photographs of both eyes could be examined. For 124 participants (2.8%), optic disc photographs, visual field examinations, or records of glaucoma history were not available. The mean age was 55.8 ± 10.3 years (median, 56 years; range, 40–101 years), and the mean refractive error was −0.38 ± 2.23 diopters (median, 0 diopters; range, −20.13 to +7.50 diopters). Out of the 4315 subjects, 7 (0.2%) were aphakic and 78 (1.8%) were pseudophakic.

The mean VCDR was 0.5 ± 0.1 (median, 0.5) for the right eyes and 0.5 ± 0.1 (median, 0.5) for the left eyes ( Table 1 ). Ranging from 0 to 1.0, the VDRC represented optic discs with no cupping to optic discs with total neuroretinal rim loss both in the superior disc region and in the inferior disc region ( Table 1 ). The 97.5th percentile of the vertical cup-to-disc ratio was 0.7 for both eyes, and the 99.5th percentile was 0.8 for both eyes ( Table 1 ). The mean inter-eye asymmetry of the vertical cup-to-disc ratio was 0.1 ± 0.1, with the 97.5th percentile at 0.2 and the 99.5th percentile at 0.3. The mean intraocular pressure was 16.1 ± 3.4 mm Hg, with the 97.5th percentile at 26 mm Hg. A visual field defect defined as sensitivity loss at any location was found for 676 subjects (4.8%). If the visual field defect was compared with the fundus image, the perimetric defect correlated with a glaucomatous appearance of the optic disc and retinal nerve fiber layer for 102 subjects (3.6%). It was then considered to be glaucomatous.


Distribution of the Vertical Cup-to-Disc Diameter Ratio in the Beijing Eye Study (by Eye)

VCDR Frequency Valid Percent (%)
0 37 0.4
0.1 25 0.3
0.2 359 4.2
0.3 1554 18.4
0.4 2042 24.1
0.5 2432 28.8
0.6 1586 18.8
0.7 327 3.9
0.8 74 0.9
0.9 18 0.2
1 4 0.0
Total 8458 100

VCDR = vertical cup-to-disc diameter ratio.

Using the criteria of the ISGEO, glaucoma was detected in 158 subjects (prevalence: 3.7% ± 0.3% [95% CI 3.1%–4.2%]) ( Table 2 ). The number of subjects in each diagnostic category is listed in Table 2 . Out of the 158 glaucoma subjects, 111 subjects (2.6%, 95% CI 2.1%–3.0%) fulfilled the criteria of OAG and 44 subjects (1.0%, 95% CI 0.7%–1.3%) fulfilled the criteria of PACG. In 3 subjects (0.07%, 95% CI, 0%–0.1%) with secondary glaucoma, glaucoma was considered to be attributable to uveitis (n = 1 subject), attributable to a preceding pars plana vitrectomy that was performed because of proliferative diabetic retinopathy (n = 1), or lens-related (n = 1). The glaucoma prevalence stratified by age and gender is shown in Table 3 .


Definition of Glaucoma by the International Society of Geographical and Epidemiological Ophthalmology and the Number of Subjects of the Beijing Eye Study in Each Category

Category Definition Number of Subjects
Category 1 Definition 1A: VCDR ≥97.5th percentile + VFD 49
Definition 1B: VCDR asymmetry ≥97.5th percentile + VFD 8
Definition 1C: Rim width ≤0.1VCDR + VFD 22
Category 2 Definition 2A: VCDR ≥99.5th percentile 72
Definition 2B: VCDR asymmetry ≥99.5th percentile 73
Category 3 Definition 3: Visual acuity <3/60 + IOP >99.5th percentile/filtering surgery 5
Glaucoma total 158 a

IOP = intraocular pressure; VCDR = vertical cup-to-disc ratio; VFD = visual field defect.

a The total sum of subjects adds up to more than the real total number of glaucoma subjects, since there was an overlap between the categories or definitions within each category.


Number of Glaucoma Subjects in the Beijing Eye Study Stratified by Type of Glaucoma, Gender, and Age

Age Item Men and Women Men Women
n % (95% CI) n % (95% CI) n % (95% CI)
40-49 Glaucoma 32 2.2 (1.5–3.0) 11 1.9 (0.8–3.0) 21 2.5 (2.0–3.0)
OAG 26 1.8 (1.1–2.5) 9 1.5 (0.5–2.5) 17 2.0 (1.5–2.5)
PACG 6 0.4 (0.1–0.8) 2 0.3 (0–0.8) 4 0.5 (0.2–0.7)
Secondary 0 0 (−) 0 0 (−) 0 0 (−)
All 1436 592 844
50–59 Glaucoma 30 2.5 (1.6–3.4) 13 2.8 (1.3–4.4) 17 2.3 (1.8–2.9)
OAG 19 1.6 (0.9–2.3) 11 2.4 (1.0–3.8) 8 1.1 (0.7–1.5)
PACG 10 0.8 (0.3–1.4) 1 0.2 (0–0.6) 9 1.2 (0.8–1.6)
Secondary 1 0.1 (0–0.3) 1 0.2 (0–0.6) 0 0 (−)
All 1181 454 727
60–69 Glaucoma 56 4.4 (3.3–5.5) 31 5.1 (3.4–8.5) 25 3.7 (3.0–4.5)
OAG 42 3.3 (2.3–4.3) 27 4.5 (2.8–7.3) 15 2.2 (1.7–2.8)
PACG 13 1.0 (0.5–1.6) 4 0.7 (0–0.7) 9 1.3 (0.9–1.8)
Secondary 1 0.1 (0–0.2) 0 0 (−) 1 0.2 (0–0.3)
All 1277 605 672
70+ Glaucoma 40 9.5 (6.7–12.3) 24 10.1 (6.3–13.9) 16 8.7 (6.7–10.8)
OAG 24 5.7 (3.5–7.9) 16 6.7 (3.5–9.9) 8 4.4 (2.9–5.9)
PACG 15 3.6 (1.8–5.3) 7 2.9 (0.8–5.1) 8 4.4 (2.9–5.9)
Secondary 1 0.2 (0–0.7) 1 0.4 (0–1.2) 0 0 (−)
All 421 238 183
All Glaucoma 158 3.7 (3.1–4.2) 79 4.2 (3.3–5.1) 79 3.3 (2.6–4.0)
OAG 111 2.6 (2.1–3.0) 63 3.3 (2.5–4.1) 48 2.0 (1.4–2.5)
PACG 44 1.0 (0.7–1.3) 14 0.7 (0.4–1.1) 30 1.2 (0.8–1.7)
Secondary 3 0.07 (0–0.1) 2 0.1 (0–0.3) 1 0.04 (0–0.1)
All 4315 1889 2426

CI = confidence interval; OAG = open-angle glaucoma; PACG = primary angle-closure glaucoma.

In univariate analysis, the glaucoma frequency increased significantly with age ( P < .001). The prevalence of glaucoma for the age groups of 40–49 years, 50–59 years, 60–69 years, and 70+ years was 2.2%, 2.5%, 4.4%, and 9.5%, respectively ( Table 3 ). The prevalence of glaucoma was significantly associated with intraocular pressure ( P < .001), myopic refractive error ( P = .03), and the degree of nuclear cataract ( P < .001). The prevalence of glaucoma was not significantly associated with gender ( P = .11), area of habitation ( P = .27), known diagnosis of arterial hypertension ( P = .22), known diagnosis of arterial hypotension ( P = .82), self-reported diagnosis of diabetes mellitus ( P = .36), known diagnosis of cerebral hemorrhage ( P = .56), known diagnosis of hyperthyroidism ( P = .94), self-reported diagnosis of coronary heart disease ( P = .75), and known diagnosis of hyperlipidemia ( P = .53).

In a binary regression analysis with the presence of glaucoma as dependent variable and age, intraocular pressure, refractive error, and nuclear cataract as independent variables, the presence of glaucoma was still significantly associated with age ( P = .001; OR: 1.05 [95% CI 1.02–1.07]), intraocular pressure ( P < .001; OR: 1.10 [95% CI 1.06–1.14]), and myopic refractive error ( P = .002; OR: 0.91 [95% CI: 0.85–0.97]), while the association between glaucoma and nuclear cataract was no longer statistically significant ( P = .78).

In a further step, the whole glaucoma group was differentiated into subjects with open-angle glaucoma (n = 111) and subjects with primary angle-closure glaucoma (n = 44). The prevalence of OAG was significantly higher in people with higher age ( P < .0001), myopic refractive error ( P < .001), male gender ( P = .006), and nuclear cataract ( P = .002). The association between OAG and intraocular pressure was marginally significant ( P = .05). The area of habitation was not significantly associated ( P = .07) with the prevalence of OAG. In a multivariate analysis, OAG remained to be significantly associated with age ( P < .001), male gender ( P = .022), and myopic refractive error ( P < .001) and marginally significantly with intraocular pressure ( P = .05). The prevalence of PACG was related with higher age ( P < .001), higher hyperopic refractive error ( P = .002), higher intraocular pressure ( P < .001), and higher nuclear cataract ( P < .001). It was not associated with gender ( P = .11) or area of habitation ( P = .57). In a multivariate analysis, PACG remained to be significantly associated with intraocular pressure ( P < .001) and marginally significantly with hyperopic refractive error ( P = .05).

In the whole study population (n = 4315 subjects), cataract surgery had been performed in 30 subjects (0.7%). Within this group of subjects who had undergone cataract surgery, 5 (17%) were classified to have glaucoma, with 1 (3%) having open-angle glaucoma and 4 (13%) having primary angle-closure glaucoma. This ratio of 1 to 4 was considerably lower than the ratio in the whole study population (111/44 or 2.5:1).

According to the World Health Organization criteria, low vision and blindness were defined as visual acuity in the better-seeing eye of <20/60 to 20/400 and <20/400, respectively. In the 111 subjects with open-angle glaucoma, 2 (1.8%) had low vision attributable to glaucoma and none of the subjects with open-angle glaucoma was blind ( Table 4 ). Unilateral glaucoma-related blindness occurred in 6 OAG subjects (5.4%): 3 subjects (3/39, or 7.7%) from the rural region and 3 subjects (3/72, or 3.2%) from the urban region. Out of 44 subjects with primary angle-closure glaucoma, 3 (7%) had low vision and 3 (7%) were blind because of glaucoma. Unilateral glaucoma-related blindness occurred in 8 subjects (18%) of the PACG group: 5 subjects (24%) from the rural region and 3 (13%) from the urban region. The prevalence of bilateral blindness and unilateral blindness was significantly ( P = .02 and P = .03) higher in the PACG group than in the OAG group. Correspondingly, all 3 subjects with glaucoma-related blindness belonged to the PACG group. The frequency of glaucoma-associated blindness and low vision was higher in the rural area than in the urban region (6/62 vs 2/96, P = .04).

Jan 17, 2017 | Posted by in OPHTHALMOLOGY | Comments Off on Prevalence of Glaucoma in North China: The Beijing Eye Study

Full access? Get Clinical Tree

Get Clinical Tree app for offline access