Prevalence of Age-Related Cataract and Cataract Surgery in a Chinese Adult Population: The Taizhou Eye Study

Abstract

Purpose: To study the prevalence of age-related cataract (ARC), cataract surgery, and visual outcomes in a Chinese adult population in Taizhou, China.

Methods: A population-based, cross-sectional study was conducted using a random cluster sampling method. We evaluated 10,234 eligible subjects 45 years or older (response rate 78.1%) in the Taizhou Eye Study. We conducted a detailed eye examination in all participants, including presenting visual acuity (PVA), best-corrected visual acuity (BCVA), slit-lamp assessment of lens opacities using the Lens Opacities Classification System III (LOCS III), and fundus examination.

Results: The standardized prevalences of cortical, nuclear, and posterior subcapsular cataract (PSC) were 28.6%, 24.3%, and 4.4%, respectively, and combined nuclear and cortical cataract was the most common cataract type (40.0%). According to the US visual impairment (VI) criteria and World Health Organization VI criteria, 40.6% and 21.8% of PSC participants had binocular VI, respectively; these values were higher than the VI rates in cortical and nuclear cataract (all P < 0.001). Of 148 patients (3.5%) who had cataract surgeries, 41.2% had PVA <20/63, and 19.6% had PVA <20/200. The main causes of poor visual outcome after cataract surgery were ocular comorbidities (41.3%), uncorrected refractive error (30.0%), surgical complications (15.0%), and posterior capsular opacification (PCO; 13.7%).

Conclusions: The high prevalence of cataract and high rate of VI from ARC in the adult Chinese population remains a severe public health problem. Cataract surgery remains insufficient in mainland China and poor visual outcomes were frequent. Surgical complications and PCO were important avoidable causes that attributed to poor visual outcomes after cataract surgeries.

Age-related cataract (ARC) remains a leading cause of visual impairment worldwide, especially in China, a developing country that is home to one-fifth of the global population.^1,2 At present, surgery is the only effective treatment for cataract but remains very expensive in developing countries. Cataract causes a heavy socioeconomic burden worldwide.³ According to the United States, the total cost associated with ARC is $5 to $6 billion per year in the United States.⁴ As the aged population increases, the visual impairment caused by ARC and the corresponding socioeconomic burden increase accordingly. 

Age-related cataract estimates are based on population-based epidemiologic studies; however, data on the epidemiologic features of ARC in China have been limited. Most population-based studies on the prevalence of and risk factors for cataract have been conducted in developed countries, such as the United States,^5–10 Australia,¹¹ European countries (e.g., France, United Kingdom, Germany),^12,13 and Singapore.^14,15 Few studies reporting the prevalence of and risk factors for ARC have been conducted in China.^16–18 Moreover, all of these studies on ARC prevalence were conducted before 2006. Due to the rapid economic growth and the changing population of the elderly in China in recent years,¹⁹ these data are of limited use for evaluating the present prevalence of and risk factors for ARC in China. Epidemiologic studies on the current condition of ARC in China are therefore greatly needed. 

To meet this need, we conducted the Taizhou Eye Study from April 2012. This is a large-scale, population-based prospective cohort study focusing on the prevalence, incidence of, and risk factors for ARC and other age-related eye diseases. The present phase of this cross-sectional study evaluated the prevalence of and risk factors for ARC and other age-related eye diseases that affect Chinese people 45 years and older. This report examines the prevalence of ARC and outcomes of cataract surgery in the Taizhou Eye Study and compares the findings with previous studies performed in China and other countries. 

Methods

Study Design and Procedures

The Taizhou Eye Study was part of the Taizhou Longitudinal Study,²⁰ which is an ongoing large-scale, population-based cohort study that was initiated by Fudan University in 2007 and is supported by the National Science and Technology Ministry. For the Taizhou Eye Study, we performed a baseline ophthalmologic examination between July 2012 and December 2013 in Taizhou, Jiangsu Province. We followed a random cluster sampling method and constructed a sampling frame using geographically defined clusters that were based on the Public Security Bureau and Community Committee. We selected 20 villages and six communities comprising a sample of 13,106 people 45 years or older. Three to 7 days before the baseline survey, the survey staff distributed recruitment materials to every household in the target communities. All participants were self-identified Han Chinese (Han is the most populous ethnic group in China). The study adhered to the Declaration of Helsinki and was approved by the Human Ethics Committee of the School of Life Science at Fudan University. All participants signed consent before participation. 

Examination Procedure

The examination procedure has been described fully in our previous report.²¹ In brief, participants visited the nearby villages or community offices to undergo a general physical, full ophthalmic examination, and questionnaire on examination days. Home visits were conducted when the participants were physically disabled. 

After patient registration, we conducted general examinations, including measurements of blood pressure, heart rate, body mass index, and body fat percentage. We also collected vein blood for serological and genetic analysis. We measured presenting visual acuity (PVA; wearing present correction if any) using a retro-illuminated logarithm E chart at a distance of 4 m and at 1 m for those failing to read the top line (20/200), as used by Zhao et al.²² during the Nine-Province Eye Survey. Each eye was measured separately. Participants wore glasses during the examination if they had for distance correction. When PVA was ≤20/40 in either eye, the best-corrected visual acuity (BCVA) was measured by subjective refraction without cycloplegia. We measured IOP using Icare rebound tonometry (Icare TAO1i, Helsinki, Finland), and the axial length (AL), central anterior chamber depth, and lens thickness using an A-scan (AL-3000; Tomey, Tokyo, Japan) for all participants. Four experienced technicians from the Taizhou Eye Study Team performed all ocular examinations. All eye technicians had completed standardized ophthalmologic training and were certified to conduct the eye examinations. Examination consistency was greater than 95% between the examiners. 

One ophthalmologist (YT) conducted the examination of the anterior and posterior segments for all participants using a slit-lamp (Topcon SL-8Z; Topcon, Inc., Tokyo, Japan) and +90-diopter (D) lens or direct ophthalmoscopy before and after dilation of the pupil and recorded the lens opacities classification and other diagnoses for the residents. Participants at high risk of angle closure glaucoma were examined only under small pupil condition. Fundus photographs were recorded using a Canon retinal camera system (CX1; Canon, Inc., Tokyo, Japan) for individuals with typical fundus diseases after dilation of the pupil. 

The study clerks conducted face-to-face questionnaire interviews concerning, for example, economic situation, education level, life habits, tobacco and alcohol intake, and drug and disease history, with the aid of computers. For quality control, we have centralized and standardized the sampling, methods, questionnaire design, training, physical examination, laboratory examination, and data management. The general examiners and interviewers had completed a standardized training and were certified to conduct the specific survey. The interviews were tape recorded, and 5% of the tapes were evaluated for interview quality control. All examination and questionnaire data were entered into a computer database on examination days, and the reasonableness of the responses was accessed throughout the study to identify contradictory responses. Phase summaries were prepared to ensure that data were accurate, consistent, and standardized. 

Grading of Cataract and Lens Opacities

We recorded cataracts using the Lens Opacities Classification System III (LOCS III).²³ According to LOCS III, nuclear lens opacities were classified into six grades (NO1 NC1–NO6 NC6), cortical lens opacities were classified into five grades (C1–C5), and posterior subcapsular lens opacities were classified into five grades (P1–P5). 

1.  
   Cataract: we defined cataract as any LOCS III grading of ≥2 in either eye. Cortical, nuclear, and posterior subcapsular cataracts (PSC) were defined as LOCS III ≥2 for cortical, nuclear, and PSC opacities in at least one eye, respectively.
2.  
   Any cataract: we defined any cataract as the presence of any cataract (cortical, nuclear, or PSC meeting the above criteria) or a history of previous cataract surgery (pseudophakia or aphakia eyes) in either eye.
3.  
   If one eye had two or three types of cataract, for example, for combined cortical and nuclear cataract, we classified the eye into cortical and nuclear cataract, respectively. One individual could be classified into different groups of cataract types. We excluded participants with congenital or traumatic cataracts. In individuals in whom lens assessment was not possible in one eye (e.g., ocular atrophy or severe corneal opacity), we used the LOCS III score of the contralateral eye.
4.  
   If a participant had undergone unilateral lens extraction, we used the LOCS III grading from the contralateral phakic eye to define the lens opacity types in that individual. If a participant had undergone bilateral lens extraction (bilateral pseudophakic or aphakic eyes), we excluded him or her in the prevalence calculation of specific types of cataract because this was difficult to evaluate.

Statistical Analysis

We used SPSS Statistics 17.0 (IBM SPSS Inc., Chicago IL, USA) for statistical analysis. We calculated the age- and sex-standardized prevalence of each type of cataract (any cataract; cortical, nuclear, and PSC) in our study based on the population distribution data of the China National Census 2012. To achieve this, we calculated the standardized proportion of each age and sex group in our study and then analyzed the age- and sex-standardized prevalence of each type of cataract. We used the US visual impairment (VI) definition (BCVA <20/40) and the World Health Organization (WHO) VI criteria (BCVA <20/63) to calculate the prevalence of VI with cataract and compare the visual impairment prevalences in different groups. We also analyzed the prevalence of cataract surgeries, visual outcomes, and reasons for poor visual outcomes. We used the χ² test to analyze categorical data, and differences were considered statistically significant when P ≤ 0.05. 

Results

Of the 13,106 individuals 45 years or older who were enrolled in the study, 10,234 eligible residents (response rate, 78.1%) completed detailed eye examinations. As previously reported,²¹ the mean age of the participants was 59.5 ± 9.8 (median 59; age range, 45–100 years); this result was similar to the mean age of nonparticipants (60.4 ± 9.5, P = 0.17). There was a higher percentage of females in the study (male-to-female ratio, 4161:6073 vs. 1527:1345, P < 0.001, χ² test). 

In the “any cataract” calculation on a person basis (by person), one participant was excluded because of traumatic cataract in both eyes; therefore, 10,233 individuals were included in the calculation. In the “any cataract” calculation on an eye basis (by eye), 22 right eyes (15 eyes with ocular atrophy or prosthetic eye, 6 eyes with dense central cornea opacities, and 1 eye with traumatic cataract) and 32 left eyes (17 eyes with ocular atrophy or prosthetic eye, 12 eyes with dense central cornea opacities, and 1 eye with traumatic cataract) were excluded; therefore, 20,414 eyes were included in the analysis. For specific cataract calculation, 50 people had cataract surgery or were unavailable for lens grading observation in both eyes; therefore, 10,184 individuals were included in the specific cataract analysis. 

The age- and sex-standardized prevalences of any cataract in adults 45 years or older were 35.1% (95% confidence interval [CI] 34.7–35.5) by eye and 38.1% (95% CI 37.8–38.4) by person. For specific cataract, the age- and sex-standardized prevalences of nuclear, cortical and PSC cataract were 24.3% (95% CI 24.0–24.6), 28.6% (95% CI 28.4–28.9), and 4.4% (95% CI 4.3–4.5), respectively (Table 1). For the 4260 individuals with cataract, the relationship between nuclear, cortical, and PSC cataract was illustrated using a Venn diagram (Fig. 1); combined cortical and nuclear cataract was the most common cataract type (1702/4260, 40.0%). 

Table 1

 

Age- and Sex-Specific Prevalence of ARC in the Taizhou Eye Study

 

Figure 1

 

Venn diagram describing 4260 eligible subjects with nuclear, cortical, and PSC cataracts.

The prevalence of any cataract (by eye) and cortical cataract was greater in females than males (39.0% [95% CI 38.1–39.8] vs. 37.2% [95% CI 36.1–38.2], OR = 1.07 for any cataract by eye, P = 0.009 and 32.5% [95% CI 31.3–33.7] vs. 29.1% [95% CI 27.7–30.5], OR = 1.17 for cortical cataract, P < 0.001, respectively). We also found that the prevalence of any cataract (by eye and by person), nuclear, cortical, and PSC cataracts increased with age (all P < 0.001) (Fig. 2). 

Figure 2

 

Prevalence of ARC in different age groups. The percentage of cataract increased with age in any cataract by eye, in any cataract by person, and in nuclear, cortical, and PSC with significant difference between two age groups. *P < 0.05; **P < 0.01; ***P < 0.001.

Next, we studied VI prevalence in cataract subjects. According to the US VI definition (BCVA <20/40) and WHO definition (BCVA <20/63), the prevalence of VI was significantly greater in PSC than cortical and nuclear cataract (all P < 0.001); 40.6% and 21.8% of subjects with PSC were US bilateral VI and WHO bilateral VI individuals, respectively (based on the better-seeing eyes). Furthermore, when we also included monocular VI, 62.6% and 40.8% of participants with PSC were classified as US VI and WHO VI, respectively. For individuals with “any cataract,” 28.5% and 11.9% were US bilateral VI and WHO bilateral VI, respectively (Fig. 3). 

Figure 3

 

(A) Prevalence of VI (BCVA <20/40 using the US VI definition) in cataract based on the better- and worse-seeing eyes, respectively. (B) Prevalence of VI (BCVA <20/63 using the WHO VI definition) in cataract based on the better- and worse-seeing eyes, respectively. The percentages of bilateral VI were significantly higher in PSC than in cortical and nuclear cataract. Furthermore, when monocular VI was also considered, the percentage of VI in PSC was also higher than cortical and nuclear cataract. *P < 0.05; **P < 0.01; ***P < 0.001.

The overall prevalence of cataract surgery among all 10,233 eligible individuals was 1.45% (148/10233, 95% CI 1.2–1.7). We also calculated the percentage of cataract surgery among the 4260 participants with cataract; 3.5% (95% CI 2.9–4.1) underwent cataract surgeries among the cataract population. The females had a higher tendency to undergo cataract surgeries than males (2.9% vs. 1.9%, OR = 1.5 by eyes and 3.9% vs. 2.8%, OR = 1.4 by person, P = 0.009 and P = 0.046, respectively) (Fig. 4A). Furthermore, we observed a “U”-curve distribution while evaluating the prevalence of cataract surgery according to age. The frequency of cataract surgery decreased from 3.5% (95% CI 1.0–8.8) at 45 to 49 years old to 2.0% (95% CI 1.4–2.7) at 60 to 69 years old (although the difference was not significant); the frequency then increased at 70 to 79 and 80 years and older (both P < 0.001) (Fig. 4B). 

Figure 4

 

Cataract surgery prevalence within the ARC individuals between different sex and age groups. (A) The percentage of cataract surgery was higher in females than in males. (B) There was a decreased trend in cataract surgery in individuals 45 to 69 years old (although this difference was not statistically significant); the percentage obtaining surgery then increased with age and the difference was significant (each age category is compared with the previous one). *P < 0.05; **P < 0.01; ***P < 0.001.

Of the 194 eyes that underwent cataract surgery, 33.0% had a PVA ≥20/40 and 41.2% eyes had a BCVA ≥20/40; 41.2% eyes had a PVA <20/63 and 28.9% eyes had a BCVA <20/63 (Table 2). If PVA <20/63 was defined as poor visual outcome according to the WHO low vision and blindness criteria (VA <20/63), 80 eyes (41.2%) had poor visual outcomes (PVA <20/63). The main causes of poor visual outcome were ocular comorbidities (41.3%), uncorrected refractive error (30.0%), surgical complications (15.0%), and posterior capsular opacification (PCO) (13.7%) (Table 3). 

Table 2

 

Visual Outcome of Cataract Surgery (Eye Specific)

 

Table 3

 

Causes of Poor Visual Outcome After Cataract Surgeries (PVA <20/63)

 

Discussion

Our study provides a detailed report of ARC prevalence in the adult Chinese population in Taizhou. We used the age- and sex-standardized prevalence from the 2012 China census. Approximately 515.6 million adult people were estimated to have ARC, and 147 million people were estimated to have bilateral VI with cataract according to the US VI criteria. Combined cortical and nuclear cataract was the most common cataract type in China. The prevalence of cataract surgery was very low among cataract patients (3.5%), and the poor visual outcome was relatively high (41.2% of eyes had a postoperative PVA <20/63). Surgical complications and PCO were important avoidable causes (28.7% in total) of poor visual outcome after cataract surgery. To our knowledge, the Taizhou Eye Study is the largest, single-population–based, cross-sectional eye study to date in China. 

Several studies have reported the prevalence of cataract in China^16,18,24 and other countries^{10,14,25–29} (Table 4). We did not make a detailed, direct comparison between these studies because the lens opacities grading system, cataract definition method, examination technique, and population and age distribution varied greatly among the studies. However, based on these studies, we found that the Chinese population might have a higher prevalence of cortical cataract than the Caucasian population. In our study, the prevalence of any cortical cataract (28.6%) was slightly higher than any nuclear cataract (24.3%), and the PSC was the least common type (4.4%). Our results are consistent with the Doumen Guangzhou Study,²⁴ the Singapore Tanjong Pagar Survey,¹⁴ the Los Angeles Latino Eye Study,²⁸ the Barbados Eye Study,¹⁰ and the Salisbury Eye-Evaluation Project (nonwhite population),²⁷ in which the prevalence of cortical cataract was higher than that of nuclear cataract. Racial differences and other factors (including sunlight exposure, hypertension, and diabetes) may explain this difference.²⁸ However, in the Beijing Eye Study and the Shipa Taiwan Eye Study, nuclear cataract was found more commonly than the cortical cataract.^16,18 The discrepancy might be due to differences between the cataract grading systems used and the sampled populations. Another reason for the discrepancy might be the higher incidence rate of cortical cataract than nuclear cataract. The 5-year incidence of cortical cataract (11.8%) in the Beijing Eye Study was higher than that of the nuclear cataract (4.3%),³⁰ indicating a large increase in the prevalence of cortical cataract in China in recent years. 

Table 4

 

Comparison of Population-Based Prevalence Studies of Cataract

 

In our study, women exhibited a higher tendency to have “any cataract” and cortical cataract than men. A higher prevalence of cataract in females has also been found in previous studies.^{10,18,26,29,30} In the Blue Mountain Eye study, females exhibited a significantly higher prevalence of cortical cataract.²⁹ In the Beijing Eye Study, the Los Angeles Latino Eye Study, and the Shipa Taiwan Eye Study, females exhibited a higher prevalence of “any cataract” and three types of cataract than males.^16,18,28 However, in other studies (e.g., the Beaver Dam Eye Study), no difference between sexes was observed.²⁶ In our study, no sex difference was found for “any cataract” by person. This might be the reason that many females had severe cataract and more eyes affected in comparison with males, while no per-person difference was observed. 

In our study, only 1.45% of the entire population had undergone previous cataract surgeries; that is, approximately 14,500 people had cataract surgery per million of population, which was similar to the percentage of cataract surgeries (1.46%) that was found in Jiangsu Province during the Nine-Province Eye Survey (2006–2007)³¹ and higher than that (1.3%) found in the Beijing Eye Study (2001)¹⁶; however, the percentage is lower than that found in the Singapore Tanjong Pagar Survey (5.1%),¹⁸ the Los Angeles Latino Eye Study (3.9%),²⁸ the Beaver Dam Eye Study (7%),²⁶ and in Australia (6%).²⁹ The fundamental medical access and economic situation are the two predominant reasons that explain the differences in the percentage of cataract surgeries between China and developed countries, especially in rural areas. Many rural people in China continue to regard poor vision as a natural process of aging and do not consult doctors about cataract treatment. 

We also observed a “U”-shaped curve describing the prevalence of cataract surgery with age. A higher prevalence of cataract surgery was found in those aged 45 to 49 than those aged 50 to 69. The urgent need for good vision and more convenient access to medical care are the main reasons for a higher percentage of cataract surgery in adults aged 45 to 49. Cataracts develop and severely affect the visual acuity of people aged 70 years or older; therefore, the percentage of cataract surgery increased. In our study, the prevalence of cataract surgery was more common in females than in males. In the Beaver Dam Study, women were more severely affected by cataract than men in the group 75 years or older.²⁶ This might have caused the result that females were more aware of visual decrease and that more females obtained cataract surgery than males. Overall, the government and hospitals should conduct more effective programs to increase the popularity and availability of cataract surgery. 

Cataract surgery outcome also warrants attention. In our study, 21.6% of eyes had 20/200 ≤ PVA < 20/63, and nearly 20% of eyes had PVA <20/200 after cataract surgery. This result is consistent with those reported in the Hainan Province Survey, in which 40.6% of eyes exhibited PVA <20/63 after cataract surgery.³² The visual outcomes were not satisfying. In our study, ocular comorbidities (41.3%), including AMD, high myopic atrophy, glaucoma, and corneal opacity, were the main reasons for poor visual outcome after cataract surgery. This deserves more attention in clinics, as these ocular comorbidities might affect postoperative visual outcomes. Detailed and adequate preoperative examinations and ocular evaluations are urgently needed in clinics before cataract surgery; this is very important not only for the prospect of appropriate visual outcome, but also for better preoperative communication. In our study, we also found that surgical complications and PCO, an avoidable or treatable complication after cataract surgery, accounted for 28.7% of poor visual outcomes; this finding implies that hospitals should also conduct regular, long-term visits postoperatively to ensure the quality of cataract surgery. 

The strengths of our study are that it included a large and representative population, enjoyed a good response, used a standardized examination process, used the LOCS III system to grade lens opacities, employed experienced ophthalmologists to measure lens opacities and fundus changes, and was based on US and WHO visual impairment criteria and comparisons with other studies. However, some limitations occurred. First, the response rate (78.1%) was not very high. Furthermore, we conducted only fundus examination and took fundus photographs for individuals with fundus diseases, rather than using more accurate fundus examination methods, such as optical coherence tomography scanning. Moreover, in cases with severe lens opacity, we might have underestimated some posterior segment diseases, including diabetic retinopathy, glaucoma, and AMD. 

In conclusion, in the Taizhou Eye Study, the age- and sex-standardized prevalence of ARC was very high. Combined cortical and nuclear cataract was the most common cataract in our study. The prevalence of each type of cataract increased with age, and females had a higher prevalence of any cataract by eye and cortical cataract than males. In contrast, cataract surgery percentage was lower than that in developed countries, and poor visual outcomes are frequent. Government and hospitals should work more effectively to promote the popularity and quality of cataract surgery. 

Acknowledgments

The authors thank Walter J. Stark, David S. Friedman, Sheila West, and Jing Tian from Wilmer Eye Institute, Johns Hopkins Hospital for their helpful advice and technical support. We also thank the members of the Taizhou Eye Study team for their contribution to this study. 

Supported by grants from the Natural Science Foundation of China (NSFC81270989), Key Projects in the National Science & Technology Pillar Program (2011BAI09B00), the New One Hundred People's Plan of Shanghai Health Bureau (XBR2011056), and the Visual Impairment and Reconstruction Key Laboratory of Shanghai (12DZ2260500). 

Disclosure: Y. Tang, None; X. Wang, None; J. Wang, None; W. Huang, None; Y. Gao, None; Y. Luo, None; J. Yang, None; Y. Lu, None