January 2013
Volume 54, Issue 1
Free
Clinical and Epidemiologic Research  |   January 2013
Metabolic Syndrome and Risk of Age-Related Cataract over Time: An Analysis of Interval-Censored Data Using a Random-Effects Model
Author Affiliations & Notes
  • Haleh Ghaem Maralani
    From the Saw Swee Hock School of Public Health, National University of Singapore and National University Health System, Singapore; the
    Department of Epidemiology, School of Health and Nutrition, Shiraz University of Medical Sciences, Shiraz, Iran; the
  • Bee Choo Tai
    From the Saw Swee Hock School of Public Health, National University of Singapore and National University Health System, Singapore; the
    Yong Loo Lin School of Medicine, National University of Singapore and National University Health System, Singapore; the
  • Tien Y. Wong
    Singapore Eye Research Institute, Singapore National Eye Centre, Singapore;
  • E. Shyong Tai
    Medicine, Yong Loo Lin School of Medicine, National University of Singapore and National University Health System, Singapore; the
  • Jialiang Li
    Department of Statistics and Applied Probability, National University of Singapore, Singapore; and the
  • Jie Jin Wang
    Centre for Eye Research Australia, University of Melbourne, Royal Victorian Eye and Ear Hospital, East Melbourne, Victoria, Australia; the
    Centre for Vision Research, University of Sydney, Sydney, Australia.
  • Paul Mitchell
    Centre for Vision Research, University of Sydney, Sydney, Australia.
  • Corresponding author: Bee Choo Tai, Saw Swee Hock School of Public Health, National University of Singapore and National University Health System, 16 Medical Drive (MD3), Singapore 117597; ephtbc@nus.edu.sg
Investigative Ophthalmology & Visual Science January 2013, Vol.54, 641-646. doi:10.1167/iovs.12-10980
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Haleh Ghaem Maralani, Bee Choo Tai, Tien Y. Wong, E. Shyong Tai, Jialiang Li, Jie Jin Wang, Paul Mitchell; Metabolic Syndrome and Risk of Age-Related Cataract over Time: An Analysis of Interval-Censored Data Using a Random-Effects Model. Invest. Ophthalmol. Vis. Sci. 2013;54(1):641-646. doi: 10.1167/iovs.12-10980.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: To investigate whether the effect of metabolic syndrome (MetS) and its components on the incidence of different cataract subtypes (cortical, nuclear, and posterior subcapsular cataract [PSC]) change with time.

Methods.: A prospective cohort of persons 49 years of age and older were followed over 10 years in the Blue Mountains Eye Study, west of Sydney, Australia. MetS components as defined by the International Diabetes Federation criteria were measured at baseline (1992–1994), after 5 years (1997–1999), and after 10 years (2002–2004). The incidence of different cataract subtypes was obtained from standard photographic grading at these intervals (n = 1997). Using a random-effects complementary log–log regression model with time to cataract development in discrete time interval, we estimated the effect of MetS and its components on the incidence of different cataract subtypes at different time intervals.

Results.: After accounting for changes in MetS components over time and controlling for possible confounders, MetS was found to be associated with an increased 5-year incidence of cortical cataract (hazard ratio [HR] 1.48; 95% confidence interval [CI], 1.05–2.09) and PSC cataract (HR 1.75; 95% CI, 1.01–3.04). Among the five MetS components, high glucose and obesity predicted an increased 5-year incidence of cortical cataract. In addition, low high-density lipoprotein and high glucose were associated with an increased 10-year incidence of cortical and PSC cataracts, respectively.

Conclusions.: Changes in MetS predicted the 5-year incidence of cortical and PSC cataracts. Different MetS components predicted the incidence of cortical and PSC cataracts at varying time intervals.

Introduction
Age-related cataract is a leading cause of blindness and poor vision and a major public health concern globally. 1 Metabolic syndrome (MetS) represents a cluster of these metabolic abnormalities involving central obesity, dyslipidemia, hyperglycemia, and high blood pressure (BP). 2 A few studies have investigated the association between cataract and MetS, and whether some individual components are more important risk factors than others for specific cataract subtypes. The Blue Mountains Eye Study (BMES) previously examined baseline MetS with glucose as a mandatory component based on the World Health Organization (WHO) criteria, in relation to 10-year cumulative incidence of the three principal cataract subtypes (i.e., cortical, nuclear, and posterior subcapsular [PSC]) and showed that MetS was associated with an increased risk of all three cataract subtypes. 3 A recent cross-sectional study in Singapore confirmed the relation between baseline MetS, defined by the Third Report of the National Cholesterol Education Program Adult Treatment Panel (ATP-3) and cortical cataract, but not between MetS and either nuclear or PSC cataracts. 4 Other studies have reported that baseline body mass index (BMI), high BP, or diabetes 3,4 is associated with age-related cataracts, whereas elevated serum triglyceride (TG) was a predictor of cataracts among females in another study. 5  
However, there are a number of unanswered questions. First, previous studies used different definitions for MetS, that is, European Group for the Study of Insulin Resistance (EGIR), WHO, and ATP-3, and it has been reported that these definitions were not as successful in predicting diabetes, cardiovascular disease, and other health outcomes. 69 Thus, in this study, we define MetS based on the International Diabetes Federation (IDF) 10 criteria. This is a diagnostic tool for both research purposes and clinical practice, which can be used relatively easily in any country by any physician to identify patients at increased risk of developing health-related outcomes. 2 Moreover, previous studies suggested that IDF provides more reliable criteria for diagnosing MetS in a predictive model for coronary clinical status in type 2 diabetes populations. 11,12 Second, previous studies, including an earlier report of BMES, 3 used only MetS data at baseline in the evaluation of its relationship with cataracts. However, experimental studies in rats and humans have shown that the effect of glucose and lipid abnormalities on cataract formation may change over time 12,13 ; therefore, this underscores the importance of collecting further information on glucose and lipid abnormalities beyond baseline measurements to better detect cataract formation. 
To our best knowledge, this is the first study to have fully utilized the information on MetS and its components that were collected not only at baseline but also at subsequent follow-up visits (i.e., after 5 and 10 years) to examine the risk of different cataract subtypes. Further, since the outcome, different cataract subtypes, was measured in a discrete time interval, we implement the random-effect complementary log–log regression since it may be more appropriate for detecting stronger and more robust relationships 14,15 between individual MetS components, than the logistic regression as used in previous studies. Thus, to evaluate the effect of MetS and its components on the incidence of different age-related cataract subtypes (i.e., cortical, nuclear, and PSC) more precisely, and to determine whether these associations changed with time, we utilized full information that was collected at each follow-up and implemented appropriate statistical models to better describe the relationships. 
Methods
Study Design and Participants
The BMES is a population-based prospective cohort study of vision, common eye diseases, and other health outcomes in a suburban Australian population west of Sydney, Australia. 16 Between 1992 and 1994, noninstitutionalized permanent residents 49 years of age and older were invited to participate, and were requested to return for follow-up examinations after 5 (1997–1999) and 10 years (2002–2004). The recruitment details have been described elsewhere. 3,16,17  
The BMES, approved by the Human Research Ethics Committee of the University of Sydney, was conducted according to the Declaration of Helsinki. Written informed consent was obtained from all participants at each examination. 16 At each visit, trained interviewers completed a comprehensive questionnaire comprising demographic information, eye, and general medical history, including hypertension, diabetes, and preexisting diseases (i.e., angina, acute myocardial infarction [AMI], and stroke), as well as medication used. Height, weight, and seated BP 18 were measured. Fasting pathology tests, including high-density lipoprotein (HDL) cholesterol, TG, 19 and fasting blood sugar (FBS), 20 were also measured within 2 months of each interview. In addition, information on smoking (never, former, and current smoker) and alcohol intake were collected. Moreover, history of eye diseases including cataract, age-related macular degeneration (AMD), myopia, glaucoma, as well as family history of eye disease or blindness were obtained and recorded. 3 Eye iris color and skin sun-tanning characteristics were also estimated on a 4-point scale (always burn, never tan; usually burn, tan with difficulty; burn and tan above average; rarely burn, tan above average). 21  
Cataract Grading
Detailed cataract grading was performed according to definitions described previously. 3 Briefly, the population at risk for cataract comprised participants who had at least one follow-up visit, but whose lens photographs were retrospectively shown not to show signs of cataract at baseline. They also had complete information to define MetS at baseline. Different subtypes of cataract were determined using standard photographic grading at each of the three examinations. The Wisconsin Cataract Grading System was used to perform masked grading of the lens photographs. A 5-point scale was used to assess the presence and severity of nuclear cataract. Nuclear cataract was defined as nuclear opacity worse than standard 3. The extent of cortical or PSC cataract was determined by estimating the lens area involved in segments of a circular grid overlaying the photographs. Cortical opacity involving at least 5% of the total lens area or the presence of any PSC opacity was used to define the presence of the respective cataract subtypes. 3 Thus, distinct types of cataract were categorized and analyzed independently. 
Definition of Metabolic Syndrome
Metabolic syndrome was defined according to IDF criteria 22 as obesity (BMI > 30 kg/m2) plus any two of the following four factors: serum TG level ≥ 1.7 mM or specific treatment for this lipid abnormality; serum HDL cholesterol < 1.03 mM in males and <1.29 mM in females, or specific treatment for this lipid abnormality; systolic BP ≥ 130 mm Hg or diastolic BP ≥ 85 mm Hg, or treatment of previously diagnosed hypertension; or fasting plasma glucose ≥ 5.6 mM, or previously diagnosed type 2 diabetes. 
In this study, the baseline MetS components were measured and recorded when the participant entered the study, and again at the 5- and 10-year examinations after first recruitment. Data are therefore available on how MetS and its components change in each subject throughout the study. 
Statistical Analysis
The χ2 test and independent sample t-test were used to determine the relationship between categorical and continuous covariates included in this study and the 10-year cumulative incidence of cataract, respectively. 
Because an individual's MetS status as well as cataract status were prospectively evaluated at predefined time intervals (i.e., baseline, 5-year, and 10-year), the exact time that cataract—the outcome of interest—developed was therefore not known. Such information was interval censored, and thus the effect of 10-year changes in MetS and its components on the incidence of each cataract subtype was modeled using a random-effects complementary log–log regression model. 14,15 This statistical technique is readily available for survival analysis with discrete time and is one of the most frequently used discrete-time hazard functions. 23 In this study, the outcomes of interest, that is, time to development of different cataract subtypes, were included in the model based on discrete time intervals (i.e., 0–5 years or 5–10 years), in accordance with the follow-up schedule. This approach includes indicator variables for the examination time interval (0–5 and 5–10 years) as covariates. 14,15,24,25 The random-effects model accounts for possible intrasubject correlation in the assessment of MetS and its components, which were repeatedly measured at baseline, 5 years, and 10 years. Of note, when the dichotomous outcome is rare, the complementary log–log regression is more appropriate. However, Nelder (2001) 26 and Hardin and Hilbe (2007) 27 have suggested that when a binary outcome is common, the complementary log–log regression model may also fit the data well. 
Age, sex, smoking, preexisting disease (i.e., angina, AMI, and stroke), family history of eye disease (i.e., cataract, AMD, myopia, glaucoma, and blindness), history of eye disease (i.e., AMD, myopia, and glaucoma), eye iris color, and skin sun-tanning characteristics were considered as possible confounders in the model building. Furthermore, we included interactions between MetS (as well as its components) and the time interval to evaluate whether its relationship with the different cataract subtypes varied according to time interval (i.e., 5- and 10-years). 24,25 We further explored possible interaction between age and sex with MetS and its components. It should be noted that for each cataract subtype, the control group included participants without the same cataract subtype. All statistical evaluations were made assuming a two-sided test based on a 5% level of significance (STATA, version 11; StataCorp, College Station, TX). 
Results
Description of Study Population
A total of 1997 subjects with complete information for the study factors at baseline contributed to the analysis of age-related cataract. Of these, 1820 individuals contributed information for the analysis of cortical cataract, 1357 for nuclear cataract, and 1962 for PSC cataract. Table 1 presents the baseline characteristics for the study subjects according to the cumulative incidence of cataract at 10 years. Over the 10-year follow-up, 857 persons (42.9%) with incident cataract were detected. Of these, 455 (25.0%) were cortical cataract, 436 (32.1%) were nuclear cataract, and 135 (6.9%) were PSC cataract. Of the 455 persons with incident cortical cataract, 252 (53.2%) had more than 5% and less than 10% severity and 213 (46.8%) had 10% or more severity in the worse eye. Of the 135 persons with incident PSC cataract, 104 (77.0%) had less than 5% severity and 31 (23.0%) had 5% or more severity in the worse eye. The mean age of participants was 63.9 years (standard deviation [SD], 8.3) with a female predominance (57.7%). Table 2 shows the changes in MetS and its components over the 10-year study follow-up. Of those included in the analysis, prevalent MetS increased slightly to 16.6% (95% confidence interval [CI], 14.9%–18.3%) after 5 years follow-up from 12.3% (95% CI, 10.9%–13.8%) at baseline, and was 12.8% (95% CI, 11.1%–14.6%) after 10 years (Table 2). 
Table 1. 
 
Baseline Characteristics of Study Population according to 10-Year Cumulative Incidence of Age-Related Cataract
Table 1. 
 
Baseline Characteristics of Study Population according to 10-Year Cumulative Incidence of Age-Related Cataract
Characteristic Total
n = 1997
No Cataract n = 1140 Cataract
n = 857
P Value
Baseline mean age (SD) 63.9 (8.3) 62.3 (8.3) 65.9 (7.8) <0.0001
Sex (%)
 Male 844 (42.3) 517 (45.3) 327 (38.2) 0.001
 Female 1153 (57.7) 623 (54.7) 530 (61.8)
Family history of blindness (%) 83 (4.2) 39 (3.4) 44 (5.1) 0.058
Family history of any eye disease (%)* 643 (32.2) 366 (32.1) 277 (32.3) 0.918
History of any eye disease at baseline (%)† 400 (20.0) 221 (19.4) 179 (20.9) 0.407
Preexisting disease (%)‡ 308 (15.4) 161 (14.1) 147 (17.2) 0.064
Smoking status (%)
 Nonsmoker 1046 (52.4) 601 (52.7) 445 (51.9) 0.416
 Ex-smoker 698 (34.9) 387 (33.9) 311 (36.3)
 Current smoker 253 (12.7) 152 (13.3) 101 (11.8)
Eye iris color (%)
 Blue 978 (49.0) 565 (49.5) 413 (48.2) 0.897
 Hazel/green 572 (28.6) 323 (28.3) 249 (29.0)
 Tan/brown 251 (12.6) 144 (12.6) 107 (12.5)
 Dark brown 196 (9.8) 108 (9.5) 88 (10.3)
Sun skin-burned (%)
 Always burn, never tan 271 (13.6) 152 (13.3) 119 (13.9) 0.927
 Usually burn, tan with difficulty 496 (24.8) 288 (25.3) 208 (24.2)
 Burn and tan above average 784 (39.3) 448 (39.4) 336 (39.2)
 Rarely burn, tan above average 446 (22.3) 251 (22.0) 194 (22.7)
MetS (%) 246 (12.3) 119 (10.4) 127 (14.8) 0.004
BMI > 30 (%) 346 (17.3) 173 (15.2) 173 (20.2) 0.004
High glucose (%)§ 306 (15.3) 149 (13.1) 157 (18.3) 0.001
Low-HDL (%)‖ 591 (29.6) 318 (27.9) 273 (31.9) 0.055
High TG (%)¶ 815 (40.8) 461 (40.4) 354 (41.3) 0.713
High BP (%)# 1689 (84.6) 941 (82.5) 748 (87.3) 0.004
Table 2. 
 
Changes in MetS and Its Components over 10-Year Follow-Up
Table 2. 
 
Changes in MetS and Its Components over 10-Year Follow-Up
Factor Baseline
n = 1997
At 5 Years
n = 1926
At 10 Years
n = 1463
MetS (%) 246 (12.3) 319 (16.6) 187 (12.8)
BMI > 30 (%) 346 (17.3) 447 (23.2) 344 (23.5)
High glucose (%)* 306 (15.3) 404 (23.4) 198 (18.5)
Low-HDL (%)† 591 (29.6) 484 (28.0) 418 (32.1)
High TG (%)‡ 815 (40.8) 789 (39.5) 300 (27.3)
High BP (%)§ 1689 (84.6) 1750 (91.3) 1183 (82.6)
The Figure shows the changes in MetS and its components at baseline, 5-year, and 10-year follow-up among individuals with different cataract subtypes. Generally, a lower proportion of individuals with nuclear cataract had MetS or its components as compared with cortical or PSC cataract. MetS increased from baseline to 5-year follow-up and decreased by 10-year follow-up among all three subtypes of cataract. 
Figure. 
 
Changes in metabolic syndrome and its components at baseline, 5-year, and 10-year follow-up among individuals with cortical, nuclear, and PSC cataract.
Figure. 
 
Changes in metabolic syndrome and its components at baseline, 5-year, and 10-year follow-up among individuals with cortical, nuclear, and PSC cataract.
MetS and Its Components and the Incidence of Age-Related Cataract
We found MetS (hazard ratio [HR] 1.48; 95% CI, 1.05–2.09), BMI > 30 (HR 1.59; 95% CI, 1.16–2.17), and elevated glucose (HR 1.60; 95% CI, 1.15–2.23) to be associated with increased 5-year incidence of cortical cataract, whereas low-HDL cholesterol (HR 1.57; 95% CI, 1.10–2.24) was associated with an excess in incidence of cortical cataract at 10-year follow-up (Table 3). 
Table 3. 
 
10-Year Changes in MetS and Its Components on the Incidence of Age-Related Cataract
Table 3. 
 
10-Year Changes in MetS and Its Components on the Incidence of Age-Related Cataract
Factor Cortical Nuclear PSC
HR (95% CI)* P Value HR (95% CI)* P Value HR (95% CI)* P Value
Incidence at 5 years
 MetS 1.48 (1.05–2.09) 0.025 1.01 (0.75–1.36) 0.941 1.75 (1.01–3.04) 0.045
 BMI > 30 1.59 (1.16–2.17) 0.004 0.84 (0.64–1.10) 0.210 1.30 (0.76–2.20) 0.337
 High glucose† 1.60 (1.15–2.23) 0.006 1.13 (0.87–1.56) 0.372 1.59 (0.93–2.72) 0.091
 Low-HDL‡ 1.14 (0.82–1.58) 0.437 0.94 (0.73–1.22) 0.667 1.43 (0.84–2.45) 0.186
 High TG§ 1.09 (0.82–1.44) 0.570 0.98 (0.78–1.23) 0.874 1.44 (0.90–2.32) 0.129
 High BP‖ 1.12 (0.67–1.89) 0.659 1.08 (0.70–1.68) 0.720 1.10 (0.44–2.74) 0.840
Incidence at 10 years
 MetS 1.40 (0.88–2.20) 0.151 0.93 (0.54–1.61) 0.802 1.39 (0.70–2.76) 0.347
 BMI > 30 1.44 (1.00–2.09) 0.051 1.24 (0.84–1.86) 0.286 1.52 (0.89–2.60) 0.127
 High glucose† 1.24 (0.79–1.97) 0.349 1.30 (0.81–2.08) 0.279 1.90 (1.01–3.61) 0.048
 Low-HDL‡ 1.57 (1.10–2.24) 0.013 1.03 (0.69–1.54) 0.870 1.45 (0.85–2.46) 0.169
 High TG§ 0.76 (0.49–1.16) 0.201 0.97 (0.63–1.51) 0.903 1.50 (0.81–2.76) 0.197
 High BP‖ 0.96 (0.63–1.47) 0.861 1.27 (0.75–2.15) 0.378 1.11 (0.56–2.19) 0.764
However, there was no association between MetS or any of its components with the incidence of nuclear cataract at either 5 or 10 years, even after accounting for information on MetS and its components at baseline and follow-up visits as well as controlling for confounders. 
Conversely, MetS was associated with an increase in 5-year incidence of PSC cataract (HR 1.75; 95% CI, 1.01–3.04), whereas elevated glucose was associated with an increase in 10-year incidence of PSC cataract (HR 1.90; 95% CI, 1.01–3.61) (Table 3). 
Discussion
In this prospective cohort study of an Australian white population participating in the BMES, we found the following: First, after accounting for baseline and further follow-up information on MetS and its components as defined by the IDF criteria, MetS, elevated glucose, and BMI levels >30 contributed to an increase in 5-year incidence of cortical cataract, whereas low-HDL cholesterol was linked to an increase in 10-year incidence of cortical cataract. Second, MetS and elevated glucose were positively associated with the incidence of PSC cataract at 5- and 10-year follow-up, respectively. 
The association between elevated glucose and incidence of cortical and PSC cataract at different time intervals suggests that FBS levels best predicted late incidence of PSC cataract and early incidence of cortical cataract. Mechanisms connecting hyperglycemia with cataract include advanced glycation of lens proteins, 28 hyperosmotic effects of sorbitol on lens fibers via the aldose reductase pathway, 29 with induction of apoptosis in lens epithelial cells leading to the development of cataract. 30  
Additionally, our study demonstrated that BMI levels >30 predicted the 5-year incidence of cortical cataract, suggesting that the contribution of obesity to cortical cataract formation may reduce over time. In the Singapore Malay Eye Study and another study by Lim et al., 31 baseline BMI levels have been shown to contribute to a higher risk of cortical cataract 4 and PSC cataract. 31 The underlying mechanism behind the relationship between obesity and cataract is unclear. 32 To date, there has not been any study that examined how changes in BMI over time would affect cataract formation. However, it has been suggested that obesity was related to cataract by its associated complications such as diabetes, glucose intolerance, insulin resistance, and hyperlipidemia. 32 Moreover, it has been shown that the relationship between glucose and cholesterol with cataract formation changed over time. 12,13 Therefore, this may partly explain why the relationship between obesity and incidence of cataract may also change with time. 
Moreover, our finding has shown an association between low-HDL cholesterol and an excess in 10-year incidence of cortical cataract, suggesting that it takes a longer observation time for low-HDL cholesterol to be confirmed as a predictor of cortical cataract. A previous BMES report, which considered only baseline information using logistic regression, 3 failed to detect such a relationship. This finding thus suggests the importance of full utilization of baseline and follow-up data, to better describe the discrete time to development of cataract using a complementary log–log model. Studies have suggested that inflammation and oxidative stress resulting from low-HDL cholesterol levels might contribute to cataract formation. 3335  
Finally, as a specific entity, our finding demonstrated a positive association between MetS and 5-year incidence of cortical and PSC cataracts. In particular, MetS as a whole rather than its individual components, best predicted PSC cataract at 5 years. Interestingly, when we considered only the effect of MetS components at baseline (data not shown) without accounting for its changing status at follow-up, we did not detect any effects of BMI or low-HDL cholesterol on cortical cataract, even after adjusting for age, sex, eye diseases, and preexisting disease status. Therefore, the effects of MetS and its components on the incidence of cortical cataract at different time intervals would not have been detected if we had only utilized the baseline data. 
However, based on a previous report of BMES, 3 MetS as defined by the WHO criteria, was a predictor of nuclear cataract, whereas our study did not confirm this association. We did an analysis of the data based on MetS defined by WHO criteria, which was used by a previous report of BMES. Using a complementary log–log regression model, the results of MetS defined by the IDF and WHO criteria were the same for nuclear cataract (result not shown). Therefore, this difference may be due to differences in technique and sample size. Moreover, obesity has been shown to have a different pattern in its relation with nuclear cataract, with a lower BMI level being an independent risk factor for nuclear cataract. 32 This suggests the possibility of distinct etiologic pathways for different cataract subtypes. 32 Since obesity was a mandatory component in defining MetS in our study, this may partially explain why there was no effect of MetS on nuclear cataract. 
The strengths of this study include (1) its representative elderly Australian population, (2) high-quality data collection, (3) population-based study of long-term incidence of cataract, (3) full use of MetS data at each visit, (4) applying a random-effects complementary log–log regression model that can detect stronger and more robust relationships 14,15 between individual MetS components and cataract subtypes where occurrences are measured in discrete time interval, and (5) cataract diagnosis based on standardized lens photographic grading, which has been shown to have high reproducibility. 17,36 However, some drawbacks should also be noted. Statistical significance was not achieved for the relationship between MetS components and PSC cataract, mainly because of the lack of statistical power due to the limited number of PSC cases. This warrants further investigation in studies with more PSC cases. 
In conclusion, changes in MetS predicted the 5-year incidence of cortical and PSC cataract. Different MetS components predicted the incidence of cortical and PSC cataracts at varying time intervals. This underscores the importance of fully accounting for the data on MetS components collected at each follow-up visit for physicians to better predict the risk of different cataract subtypes in older persons at varying time intervals. 
Acknowledgments
The authors thank Elena Rochtchina, Ava Grace Tan, and Erdahl Teber (Centre for Vision Research Westmead Millennium, Institute University of Sydney, Australia) for helping us access the database of the Blue Mountains Eye Study. 
References
Attebo K Mitchell P Smith W. Visual acuity and the causes of visual loss in Australia. The Blue Mountains Eye Study. Ophthalmology . 1996; 103: 357–364. [CrossRef] [PubMed]
Alberti KGMM Zimmet P Shaw J. Metabolic syndrome--a new world-wide definition. A Consensus Statement from the International Diabetes Federation. Diabet Med . 2006; 23: 469–480. [CrossRef] [PubMed]
Tan JS Wang JJ Mitchell P. Influence of diabetes and cardiovascular disease on the long-term incidence of cataract: the Blue Mountains eye study. Ophthalmic Epidemiol . 2008; 15: 317–327. [CrossRef] [PubMed]
Sabanayagam C Wang JJ Mitchell P Metabolic syndrome components and age-related cataract: the Singapore Malay eye study. Invest Ophthalmol Vis Sci . 2011; 52: 2397–2404. [CrossRef] [PubMed]
Bojarskiene F Cerniauskiene LR Paunksnis A Luksiene DI. Association of metabolic syndrome components with cataract [in Lithuanian]. Medicina (Kaunas) . 2006; 42: 115–122. [PubMed]
World Health Organization (WHO). Definition, Diagnosis and Classification of Diabetes Mellitus and Its Complications. Report of a WHO consultation. Geneva: WHO; 1999.
Balkau B Charles MA. Comment on the provisional report from the WHO consultation. European Group for the Study of Insulin Resistance (EGIR). Diabet Med . 1999; 16: 442–443. [CrossRef] [PubMed]
Executive Summary of the Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA . 2001; 285: 2486–2497. [CrossRef] [PubMed]
Stern MP Williams K González-Villalpando C Hunt KJ Haffner SM. Does the metabolic syndrome improve identification of individuals at risk of type 2 diabetes and/or cardiovascular disease? Diabetes Care . 2004; 27: 2676–2681. [CrossRef] [PubMed]
Ahmed A Khan TE Yasmeen T Awan S Islam N. Metabolic syndrome in type 2 diabetes: comparison of WHO, modified ATPIII & IDF criteria. J Pak Med Assoc . 2012; 62: 569–574. [PubMed]
Bosevski M Pemovska G Bosevska G Georgievska-Ismail L. Clinical role of estimation metabolic syndrome's components in type 2 diabetic population with symptomatic coronary artery disease—a comparison of two criteria. Med Arh . 2010; 64: 144–146. [PubMed]
Kyselová Z Garcia SJ Gajdosíková A Gajdosík A Stefek M. Temporal relationship between lens protein oxidation and cataract development in streptozotocin-induced diabetic rats. Physiol Res . 2005; 54: 49–56. [PubMed]
Jacob RF Mason RP. Lipid peroxidation induces cholesterol domain formation in model membranes. J Biol Chem . 2005; 280: 39380–39387. [CrossRef] [PubMed]
Rabe-Hesketh S Skrondal A. Multilevel and Longitudinal Modeling Using Stata. 2nd ed. College Station, TX: Stata Press; 2008.
Neuhaus JM. Statistical methods for longitudinal and clustered designs with binary responses. Stat Methods Med Res . 1992; 1: 249–273. [CrossRef] [PubMed]
Mitchell P Smith W Wang JJ Cumming RG Leeder SR Burnett L. Diabetes in an older Australian population. Diabetes Res Clin Pract . 1998; 41: 177–184. [CrossRef] [PubMed]
Kanthan GL Wang JJ Rochtchina E Ten-year incidence of age-related cataract and cataract surgery in an older Australian population. The Blue Mountains Eye Study. Ophthalmology . 2008; 115: 808–814. [CrossRef] [PubMed]
Whitworth JA; World Health Organization, International Society of Hypertenion Writing Group. 2003 World Health Organization (WHO)/International Society of Hypertension (ISH) statement on management of hypertension. J Hypertens . 2003; 21: 1983–1992. [CrossRef] [PubMed]
Cugati S Wang JJ Rochtchina E Mitchell P. Ten-year incidence of diabetes in older Australians: the Blue Mountains Eye Study. Med J Aust . 2007; 186: 131–135. [PubMed]
World Health Organization (WHO). Definition, Diagnosis and Classification of Diabetes Mellitus and Its Complications. Part 1: Diagnosis and Classification of Diabetes Mellitus. Geneva: WHO; 1999.
Mitchell P Smith W Wang JJ. Iris color, skin sun sensitivity, and age-related maculopathy. The Blue Mountains Eye Study. Ophthalmology . 1998; 105: 1359–1363. [CrossRef] [PubMed]
Zimmet P Magliano D Matsuzawa Y Alberti G Shaw J. The metabolic syndrome: a global public health problem and a new definition. J Atheroscler Thromb . 2005; 12: 295–300. [CrossRef] [PubMed]
Singer JD Willett JB. Applied Longitudinal Data Analysis: Modeling Change and Event Occurrence . Oxford, UK: Oxford University Press; 2003: 407–463.
Allison PD. Discrete-time methods for the analysis of event histories. In: Leinhardt S ed. Sociological Methods and Research . San Francisco, CA: Jossey-Bass; 1982: 61–98.
Lecaire T Palta M Zhang H Allen C Klein R D'Alessio D. Lower-than-expected prevalence and severity of retinopathy in an incident cohort followed during the first 4–14 years of type 1 diabetes: the Wisconsin Diabetes Registry Study. Am J Epidemiol . 2006; 164: 143–150. [CrossRef] [PubMed]
Nelder JA. Letter to the editor. Stat Med . 2001; 20: 2205. [CrossRef] [PubMed]
Hardin JW Hilbe JM. Generalized Linear Models and Extensions. 2nd ed. College Station, TX: Stata Press; 2007.
Stitt AW. Advanced glycation: an important pathological event in diabetic and age related ocular disease. Br J Ophthalmol . 2001; 85: 746–753. [CrossRef] [PubMed]
Kador PF Kinoshita JH. Diabetic and galactosaemic cataracts. CIBA Found Symp . 1984; 106: 110–131. [PubMed]
Pollreisz A Schmidt-Erfurth U. Diabetic cataract-pathogenesis, epidemiology and treatment. J Ophthalmol . 2010; 2010: 608751. [PubMed]
Lim LS Tai ES Aung T Relation of age-related cataract with obesity and obesity genes in an Asian population. Am J Epidemiol . 2009; 169: 1267–1274. [CrossRef] [PubMed]
Cheung N Wong TY. Obesity and eye diseases. Surv Ophthalmol . 2007; 52: 180–195. [CrossRef] [PubMed]
Klimov AN Gurevich VS Nikiforova AA Antioxidative activity of high density lipoproteins in vivo. Atherosclerosis . 1993; 100: 13–18. [CrossRef] [PubMed]
Varma SD Chand D Sharma YR Kuck JF Jr Richards RD. Oxidative stress on lens and cataract formation: role of light and oxygen. Curr Eye Res . 1984; 3: 35–57. [CrossRef] [PubMed]
von Eckardstein A Hersberger M Rohrer L. Current understanding of the metabolism and biological actions of HDL. Curr Opin Clin Nutr Metab Care . 2005; 8: 147–152. [CrossRef] [PubMed]
Panchapakesan J Cumming RG Mitchell P. Reproducibility of the Wisconsin cataract grading system in the Blue Mountains Eye Study. Ophthalmic Epidemiol . 1997; 4: 119–126. [CrossRef] [PubMed]
Footnotes
 The Blue Mountains Eye Study was supported by Australian National Health and Medical Research Council Grants 974159, 991407, and 211069.
Footnotes
 Disclosure: H. Ghaem Maralani, None; B.C. Tai, None; T.Y. Wong, None; E.S. Tai, None; J. Li, None; J.J. Wang, None; P. Mitchell, None
Figure. 
 
Changes in metabolic syndrome and its components at baseline, 5-year, and 10-year follow-up among individuals with cortical, nuclear, and PSC cataract.
Figure. 
 
Changes in metabolic syndrome and its components at baseline, 5-year, and 10-year follow-up among individuals with cortical, nuclear, and PSC cataract.
Table 1. 
 
Baseline Characteristics of Study Population according to 10-Year Cumulative Incidence of Age-Related Cataract
Table 1. 
 
Baseline Characteristics of Study Population according to 10-Year Cumulative Incidence of Age-Related Cataract
Characteristic Total
n = 1997
No Cataract n = 1140 Cataract
n = 857
P Value
Baseline mean age (SD) 63.9 (8.3) 62.3 (8.3) 65.9 (7.8) <0.0001
Sex (%)
 Male 844 (42.3) 517 (45.3) 327 (38.2) 0.001
 Female 1153 (57.7) 623 (54.7) 530 (61.8)
Family history of blindness (%) 83 (4.2) 39 (3.4) 44 (5.1) 0.058
Family history of any eye disease (%)* 643 (32.2) 366 (32.1) 277 (32.3) 0.918
History of any eye disease at baseline (%)† 400 (20.0) 221 (19.4) 179 (20.9) 0.407
Preexisting disease (%)‡ 308 (15.4) 161 (14.1) 147 (17.2) 0.064
Smoking status (%)
 Nonsmoker 1046 (52.4) 601 (52.7) 445 (51.9) 0.416
 Ex-smoker 698 (34.9) 387 (33.9) 311 (36.3)
 Current smoker 253 (12.7) 152 (13.3) 101 (11.8)
Eye iris color (%)
 Blue 978 (49.0) 565 (49.5) 413 (48.2) 0.897
 Hazel/green 572 (28.6) 323 (28.3) 249 (29.0)
 Tan/brown 251 (12.6) 144 (12.6) 107 (12.5)
 Dark brown 196 (9.8) 108 (9.5) 88 (10.3)
Sun skin-burned (%)
 Always burn, never tan 271 (13.6) 152 (13.3) 119 (13.9) 0.927
 Usually burn, tan with difficulty 496 (24.8) 288 (25.3) 208 (24.2)
 Burn and tan above average 784 (39.3) 448 (39.4) 336 (39.2)
 Rarely burn, tan above average 446 (22.3) 251 (22.0) 194 (22.7)
MetS (%) 246 (12.3) 119 (10.4) 127 (14.8) 0.004
BMI > 30 (%) 346 (17.3) 173 (15.2) 173 (20.2) 0.004
High glucose (%)§ 306 (15.3) 149 (13.1) 157 (18.3) 0.001
Low-HDL (%)‖ 591 (29.6) 318 (27.9) 273 (31.9) 0.055
High TG (%)¶ 815 (40.8) 461 (40.4) 354 (41.3) 0.713
High BP (%)# 1689 (84.6) 941 (82.5) 748 (87.3) 0.004
Table 2. 
 
Changes in MetS and Its Components over 10-Year Follow-Up
Table 2. 
 
Changes in MetS and Its Components over 10-Year Follow-Up
Factor Baseline
n = 1997
At 5 Years
n = 1926
At 10 Years
n = 1463
MetS (%) 246 (12.3) 319 (16.6) 187 (12.8)
BMI > 30 (%) 346 (17.3) 447 (23.2) 344 (23.5)
High glucose (%)* 306 (15.3) 404 (23.4) 198 (18.5)
Low-HDL (%)† 591 (29.6) 484 (28.0) 418 (32.1)
High TG (%)‡ 815 (40.8) 789 (39.5) 300 (27.3)
High BP (%)§ 1689 (84.6) 1750 (91.3) 1183 (82.6)
Table 3. 
 
10-Year Changes in MetS and Its Components on the Incidence of Age-Related Cataract
Table 3. 
 
10-Year Changes in MetS and Its Components on the Incidence of Age-Related Cataract
Factor Cortical Nuclear PSC
HR (95% CI)* P Value HR (95% CI)* P Value HR (95% CI)* P Value
Incidence at 5 years
 MetS 1.48 (1.05–2.09) 0.025 1.01 (0.75–1.36) 0.941 1.75 (1.01–3.04) 0.045
 BMI > 30 1.59 (1.16–2.17) 0.004 0.84 (0.64–1.10) 0.210 1.30 (0.76–2.20) 0.337
 High glucose† 1.60 (1.15–2.23) 0.006 1.13 (0.87–1.56) 0.372 1.59 (0.93–2.72) 0.091
 Low-HDL‡ 1.14 (0.82–1.58) 0.437 0.94 (0.73–1.22) 0.667 1.43 (0.84–2.45) 0.186
 High TG§ 1.09 (0.82–1.44) 0.570 0.98 (0.78–1.23) 0.874 1.44 (0.90–2.32) 0.129
 High BP‖ 1.12 (0.67–1.89) 0.659 1.08 (0.70–1.68) 0.720 1.10 (0.44–2.74) 0.840
Incidence at 10 years
 MetS 1.40 (0.88–2.20) 0.151 0.93 (0.54–1.61) 0.802 1.39 (0.70–2.76) 0.347
 BMI > 30 1.44 (1.00–2.09) 0.051 1.24 (0.84–1.86) 0.286 1.52 (0.89–2.60) 0.127
 High glucose† 1.24 (0.79–1.97) 0.349 1.30 (0.81–2.08) 0.279 1.90 (1.01–3.61) 0.048
 Low-HDL‡ 1.57 (1.10–2.24) 0.013 1.03 (0.69–1.54) 0.870 1.45 (0.85–2.46) 0.169
 High TG§ 0.76 (0.49–1.16) 0.201 0.97 (0.63–1.51) 0.903 1.50 (0.81–2.76) 0.197
 High BP‖ 0.96 (0.63–1.47) 0.861 1.27 (0.75–2.15) 0.378 1.11 (0.56–2.19) 0.764
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×