Potential risk factors for cataract.
Mukesh BN, Le A, Dimitrov PN, Ahmed S, Taylor HR, McCarty CA. Development of Cataract and Associated Risk FactorsThe Visual Impairment Project. Arch Ophthalmol. 2006;124(1):79-85. doi:10.1001/archopht.124.1.79
LESLIEHYMANPhDAuthor Affiliations: Marshfield Clinic Research Foundation, Marshfield, Wis (Drs Mukesh and McCarty); Centre for Eye Research Australia, University of Melbourne, Melbourne, Victoria, Australia (Dr Mukesh, Dr Le, Mr Dimitrov, Dr Ahmed, and Dr Taylor).
To evaluate risk factors for the development of cataract in Australian residents.
A total of 3721 participants from 9 randomly selected urban districts within Victoria were recruited and invited to attend comprehensive standardized interviews and ophthalmic examinations at baseline and then 5-year follow-up. Lens opacities were graded clinically and on photographs according to the Wilmer cataract grading system. The development of cortical, nuclear, and posterior subcapsular cataract were assessed separately for associated risk factors. Risk exposure at baseline was used as the predictor for cataract development, which included various sociodemographic, dietary, familial, medical, and ocular characteristics of the participants. Risk factor analyses were performed by univariate and multivariate logistic regression.
Increased age was a risk factor for development of all types of cataract with an increasing risk trend throughout life for nuclear cataract. Female sex, a laborer’s occupation, and myopia were independent risk factors for development of cortical cataract. For development of nuclear cataract, the independent risk factors were having a birthplace outside Australia and New Zealand, current cigarette smoking, and having a history of arthritis. Diabetes mellitus and having taken calcium channel blockers for longer than 5 years were independent risk factors for posterior subcapsular cataract.
The trend of increasing incidence of cataract with increased age is a major public health concern with an aging population in Australia and the world. Among the risk factors identified, cigarette smoking is a factor that is readily modifiable and preventable. The other risk factors identified require further support or clarification of underlying mechanisms to find modifiable features.
Cataract is the leading cause of blindness in the world, and most cataracts are age-related.1 In Australia and other countries, cataract surgery is the most common ophthalmic procedure.2 Corresponding with a world population that is increasing in size and age, the health and economic burden of cataract on societies will escalate, particularly in developing countries, where cataract occurs at an earlier age and cataract surgery is often inaccessible.3 Therefore, even though cataract surgery is an effective cure, research into causative factors and pathomechanisms to delay or prevent the development of cataract is a major challenge for the 21st century.3
Reports from the longitudinal studies of risk factors for cataract development4- 11 are becoming available to clarify risk factors identified in previous cross-sectional and case-control studies. These studies have identified some modifiable and preventable factors that could be targeted by interventions. Some of these factors have generally been accepted, such as smoking for nuclear cataract,1,11- 13 UV-B radiation from sunlight for cortical cataract,1,12- 21 and diabetes mellitus and corticosteroid use for posterior subcapsular (PSC) cataract.1,7,13,20- 23 The protective effects of antioxidant vitamin supplements have been suggested in many studies,24- 33 but the randomized, placebo-controlled trials of the Vitamin E, Cataract and Age-related Macular Degeneration Study34 and the Age-Related Eye Disease Study have recently revealed no significant benefits of antioxidants.35
The purpose of this study was to contribute to the body of evidence that is emerging regarding risk factors associated with the development of age-related cataract. These data could be used to address the increasing global need for public health interventions.
The methodology for the Visual Impairment Project has been previously detailed.36 Baseline assessment was conducted from 1992 to 1994 on 3271 residents, recruited by door-to-door survey from 9 randomly selected adjacent pairs of census collector districts within urban Victoria. Residents needed to be aged 40 years or older and have lived in that address for at least 6 months. The study population was representative of the Victorian and Australian population.37 Assessment included standardized interviews and ophthalmic examinations to collect information on sociodemographic, dietary, familial, medical, and ocular characteristics, including sunlight exposure. Interpreters were used for non–English-speaking participants and home visits were conducted as necessary. Informed consent was obtained from each participant at the beginning of each examination. The study protocol was approved by the Human Research and Ethics Committee of the Royal Victorian Eye and Ear Hospital, Melbourne, Victoria, Australia.
From 1997 to 1999, 5-year follow-up assessment was conducted on all available participants. Nonparticipants in the follow-up study were those who moved overseas or interstate, those who refused to participate, or those who had died before the commencement of the 5-year follow-up. Reported deaths were confirmed through the National Death Index at the Australian Institute of Health and Welfare, Canberra, Australian Capital Territory.
Similar standardized ophthalmic examination was conducted at baseline and 5-year follow-up. Presenting and best-corrected visual acuity was measured with a 4-m logMAR chart. Visual field assessment was performed with the Humphrey Field Analyzer using the 24-2 Fastpac statistical program (Humphrey Instruments Inc, San Leandro, Calif). Intraocular pressure was measured with the Tonopen (Oculab, Glendale, Calif) and confirmed by the Goldmann applanation tonometer. Slitlamp biomicroscopic examination was performed after pupil dilation to a minimum of 6 mm with tropicamide.
Lens opacities were graded clinically at the time of the examination and subsequently from photographs according to the Wilmer cataract grading system.38 The cortex was assessed on retroillumination and the nucleus was assessed with a slitlamp. Cortical cataract was defined as 4/16 or greater opacity. Nuclear cataract was defined as Wilmer standard grade 2.0 or greater. Posterior subcapsular cataract was defined as PSC opacity greater than or equal to 1 mm2. The cortical and PSC opacities were photographed with a retroillumination camera (Oxford, Marcher Enterprises, Oxford, England) and 400 film (T-MAX; Kodak Australia, Sydney, Australia). A photographic slitlamp (Topcon America Corporation, Paramus, NJ) with a 0.1-mm slit beam of 9.0 mm in height set to an incident angle of 30° and 200 ASA color slide film (Ektachrome; Kodak Australia, Sydney, Australia) was used to photograph the nuclear opacities. Photographs were graded by 2 research assistants in both baseline and follow-up, and discrepancies were adjudicated by an independent reviewer.
Photographic grades available for cortical, nuclear, and PSC opacities at baseline were 84%, 81%, and 84%, respectively,12 and 58%, 80%, and 58% at follow-up examination. The retroillumination camera malfunctioned at the start of the examination in one of the largest sites, causing us to miss most of the photographs in that test site. It was not possible to re-examine many of these participants. Clinical grades were used when photographic grades were not available, and agreement between photographic grades and clinical grades had been shown to be very high for all 3 types of cataract.12
Participants were queried about the year that any cataract extraction has been performed. The treating ophthalmologist was contacted to provide information on the type of opacity present before cataract extraction. Prior cataract surgery was confirmed on clinical examination and the status of the capsule, if present, was noted. Patients were classified as having incident cataract of lens opacity on the basis of that event occurring in either eye, regardless of the status of the fellow eye at baseline. The specific cataract types are not mutually exclusive, that is, a person may be classified as having any, none, or all 3 of the cataract types.
For each of the separate cataract types, the denominator for the incidence estimates included those participants who did not have that specific cataract type at baseline in either eye. For each specific type of cataract, the numerator for incidence estimates included participants who developed a specific cataract type over the period as well as those who had cataract surgery during the period. To assess measurement error associated with photographic grading of the lens, a sample of photographs from baseline was regraded by one of the photograph graders (P.N.D.) at follow-up.38
Risk factor analyses were performed separately for development of each type of cataract. The potential risk factors considered in the analyses have been summarized in the Figure. The initial value of the factors at baseline was used as the important predictor for cataract development. Cardiovascular-related medications were grouped into β-blockers, α-blockers, angiotensin converting enzyme inhibitors, thiazide diuretics, loop diuretics, calcium channel blockers, and peripheral vasodilators. A food frequency questionnaire was used to determine current consumption of dietary sources of antioxidants and use of vitamin supplements.39 Duration was considered for all dietary supplements, medications, and medical illnesses. Quantitative entities for cigarette smoking (pack-years and duration) and alcohol intake (standard drinks and duration) were considered along with intake status (never, current, past). Lifetime ocular UV-B exposure was determined for each individual from information about place of residence, time spent outdoors, and the use of ocular protective devices, including hats, spectacles, and sunglasses.40 Average annual UV-B exposure normalized with a natural log transformation was used in the analysis.
Interview data were entered directly into the data entry system, Paradox (Borland International, Scotts Valley, Calif), and clinical data were entered twice, with all data entry having internal consistency checks. All statistical analyses used SAS statistical software, version 6.10 (SAS Institute Inc, Cary, NC). Univariate association was tested with the Pearson χ2 or Fisher exact test for categorical variables and the Mantel-Haenszel χ2 test for ordinal variables. Age-adjusted and multivariate analyses were performed using logistic regression analysis. Age was considered a continuous variable and age-adjusted relative risks with 95% confidence intervals were presented. P values of less than .05 were considered statistically significant.
There were 3271 participants assessed at baseline, of which 231 participants died before the 5-year follow-up assessment, and 446 participants either migrated out of Victoria or refused to attend follow-up. Therefore, 2594 participants (85% of the surviving participants) were followed up. Of these, 202 (8%) participants who had incomplete lens data for both eyes were excluded from the analyses. Of the remaining 2392 participants, the information on lens grading was available for cortical, nuclear, and PSC cataract for 2113, 2119, and 2193 participants, respectively.
The mean ± SD age of the participants at baseline was 62.5 ± 10.9 years (range, 40-98 years), and 55% were women. The mean ± SD duration between baseline and follow-up assessment was 4.5 ± 0.64 years (range, 4-7 years). There were 555 participants who developed cataract, of whom 100 (18.0%) had mixed cataract, 107 (19.3%) had pure cortical cataract, 268 (48.3%) had pure nuclear cataract, and 80 (14.4%) had pure PSC cataract. Twenty participants with no cataract at baseline had surgery during the follow-up period. Of these, 2 participants had cortical, 6 had nuclear, 4 had PSC, and 8 had mixed cataract. The overall incidence of cortical cataract was 7.7% (95% confidence interval, 5.8-9.5), nuclear cataract was 16.4% (95% confidence interval, 12.1-20.8), and PSC cataract was 7.0% (95% confidence interval, 5.3-8.7).38 Interobserver measurement errors were small and have been outlined in the incidence report.38
Participants with ages older than 50 years were shown to have increased risk of developing all types of cataract, although borderline significance was shown in the 50- to 59-year age group for PSC cataract (data not shown). The age relationship for nuclear cataract was continued throughout life, whereas it seemed to level out after the sixth decade for cortical cataract and the seventh decade for PSC cataract (data not shown). For development of cortical cataract, an increased age-adjusted risk was found for women, people who worked as laborers, people with home duties, and people with myopia, but taking a vitamin C supplement was shown to have a protective effect (Table 1). Significant age-adjusted risk factors for development of nuclear cataract were the following: having a birthplace other than Australia and New Zealand (except the British Isles), current cigarette smoking, having smoked for longer than 30 years, having ever taken systemic corticosteroids, having arthritis, and having had arthritis for more than 5 years (Table 2). For development of PSC cataract, an increased age-adjusted risk was found with having had hypertension longer than 5 years, having diabetes mellitus, having gout, and having taken calcium channel blockers for longer than 5 years (Table 3).
The multivariate model for development of each type of cataract (Table 4) showed that increased age continued to be a significant risk factor for all cataract, with the highest association being with nuclear cataract. Being a woman, having a laborer’s occupation, and having myopia remained significant risk factors for cortical cataract. Having a birthplace other than Australia and New Zealand (except the British Isles), current cigarette smoking, and having arthritis remained significant risk factors for nuclear cataract. For PSC cataract, having diabetes mellitus and having taken calcium channel blockers for longer than 5 years remained significant risk factors. Ever having taken a vitamin C supplement was a protective factor for cortical cataract in the multivariate analysis.
Sound epidemiological studies into risk factors for cataract development are required to plan and implement effective global preventive programs in conjunction with programs that allow cataract surgery to be more accessible in developing countries.3,41 This is an essential step toward reducing the estimate of 40 million people by the year 2020 who will be severely disabled by age-related cataract globally3 and toward curbing the billions of dollars in health costs due to cataract.1 The current study is an extension of the report on the 5-year incidence of cataract from the Visual Impairment Project38 and provides risk factor analyses from prospective data. Such longitudinal data ascertain the temporal relationship between cause and effect. However, according to the Bradford Hill criteria,42 consistency and plausibility are also needed for causation, and we need to assess the consistency of the results with those from other epidemiological and physiological studies.
There is no doubt that aging is a major contributing factor to developing cataract of any type as was clearly shown in this study. The increased risk trend with age is most significant with nuclear cataract. This finding is consistent with those of other studies14,22,43,44 and suggests a natural aging process of the nucleus and possibly a cumulative effect of certain risk exposures throughout life. For cortical and PSC cataract, there is some leveling of the risk around 60 to 70 years of age. This is also reflected to some extent in the other incidence studies43,44 and might suggest a positive change in risk exposures or indicate that the lens is more susceptible to nuclear opacities at older ages.
This study has shown women to be at greater risk for developing cortical cataract: the incidence in the Beaver Dam Eye Study44 has shown that women have a greater risk for developing nuclear cataract, and the increased incidence in women in the Barbados Eye Study43 was shown for both cortical and nuclear cataract. Prevalence studies also support greater association of female sex with nuclear and cortical cataract.12,14,22,45- 47 This finding suggests that there might be differential exposures between the sexes and possibly some hormonal influence. The latter is supported by certain studies that have shown reduced prevalence of cataract with postmenopausal estrogen use14,48- 50 and an association of cataract with multiparity.51 This study, however, did not show any association between cataract and female hormonal use.
When compared with professional work, individuals with a laborer’s occupation or related work were shown to have increased risk of developing cortical cataract; individuals in managerial or administrative work have borderline increased risk. A laborer’s occupation might be associated with a lower socioeconomic status, which has been associated with increased risk of cataract in a number of studies.19- 22 It might also be linked to greater outdoor work and exposure to environmental hazards, in which sunlight exposure is a generally accepted risk factor for cortical cataract.16- 21 However, sunlight exposure as a separate entity was not shown to be a significant risk factor for the development of any type of cataract in this study. In addition, it is difficult to explain the borderline risk significance regarding managerial or administrative work.
Country of birth was shown to provide a significant risk factor for developing nuclear cataract, with participants born outside Australia and New Zealand except for the British Islands having greater risk, most originating from other European countries. It is well documented that there is an excess prevalence of cataract in people of African origin,20,21,52 but no study has yet compared other ethnic backgrounds, probably because the regions where studies have been conducted are not as culturally diverse as Victoria, Australia. The finding in this study warrants further exploration because it could be related to past exposures in the participant’s country of birth or the differences in diet between cultures that have persisted after migration.
This study has corroborated that current smoking is a risk factor for developing nuclear cataract. This finding is consistent with those of a number of prospective studies6,53- 55 and various prevalence or case-control studies.12,14,56 The major pathomechanism behind cigarette smoking is believed to be excessive oxidative stress on the lens, causing protein modification and insolubilization,57 cellular DNA damage,58 and formation of reactive advanced glycation end products.59 From a preventive perspective, evidence has shown that the insult from smoking might be reversible to some extent. This is reflected in this study and other studies, which have shown that increased risk mainly confers to current smokers rather than past smokers6,12- 14,53- 56,60 and studies on smoking cessation have shown modest risk reduction.56,61,62
The risk of developing cataract in people with certain systemic diseases is still unclear, although diabetes mellitus is generally accepted as a risk factor, in particular for cortical and PSC cataract.7,12,13,20- 22 This study has corroborated increased risk for developing PSC cataract in participants with diabetes. The proposed key pathomechanism of diabetes is the formation of advanced glycation end products, leading to lens protein alterations.63 It is unclear whether appropriate diabetic control reduces this pathophysiological event in the lens. However, anti–advanced glycation end product therapeutic strategies are currently being examined and might be a significant preventive treatment for cataract development due to both aging and diabetes.63
The other disease that was shown in this study to confer an increased risk of developing cataract was arthritis. The association was with nuclear cataract, but the prevalence Visual Impairment Project study showed an association with cortical cataract.12 Conversely, in the Italian-American Cataract Study, a positive history of arthritis was associated with reduced risk of cataract.19 These results suggest that random chance is the most likely explanation for the association with arthritis. However, the type of arthritis in this study could not be specified and arthritis with substantial inflammation might be more associated with cataract risk, given that elevated levels of C-reactive protein have been shown to be associated with cataract development from the Physician’s Health Study.64
Out of the list of medications evaluated, only having taken calcium channel blockers for longer than 5 years was associated with development of PSC cataract. Laboratory studies have shown that alterations in calcium homeostasis of the lens, which result in increased cytosolic calcium levels, can lead to cataract formation, particularly cortical cataract.57,65 However, recent laboratory studies with calcium channel blockers have revealed an apparent beneficial effect on cortical cataractogenesis due to their reduction of lenticular calcium levels.66,67 The relationship between calcium regulation and PSC cataract remains to be clarified.
The only ocular factor that was shown to have an increased risk of developing cataract in this study was myopia of greater than 1.0 diopter. The association was shown only with cortical cataract, whereas the prevalence Visual Impairment Project study showed associations with all types of cataract.12 The incidence Barbados Eye Study, which so far has only performed analyses for nuclear cataract, also revealed an associated risk between myopia and nuclear cataract,4 but the incidence Beaver Dam Eye Study revealed no association with any types of cataract.9 A prospective study is particularly important in assessing the risk of myopia for nuclear cataract development because a nuclear cataract itself can contribute to increased lens power and hence increased myopia. In general, the association between myopia and nuclear opacity is still inconclusive and further clarification is required from other prospective studies.
An area of great interest in recent years is whether the use of antioxidant supplementation has any beneficial effect in preventing cataract formation, based on the generally agreed fact that cataract can develop due to a lack of lenticular antioxidant micronutrients. Many studies have reported an associated protective effect for various vitamin supplementations.24- 33 This study has revealed that supplementation with vitamin C, a major antioxidant required daily, is associated with a significantly reduced risk of developing cortical cataract. The ultimate test regarding the effectiveness of antioxidant vitamin supplements will come from randomized clinical trials. Recently, the Age-Related Eye Disease Study has revealed no significant benefits in cataract development with vitamin C, vitamin E, and β-carotene supplementation.35 However, the Age-Related Eye Disease Study population consisted of relatively well-nourished participants aged 55 years or older. Therefore, any extra beneficial effect from supplementation is less likely to be detected and the beneficial effect to be gained might have been administered too late in life. The current results of the Vitamin E, Cataract and Age-related Maculopathy Degeneration Study34 have also revealed no benefits with vitamin E supplementation in preventing cataract.
Many studies often overlook the null results, which could be just as important when used to plan other studies and public health interventions. For this study, the important null associations with cataract were sunlight exposure and corticosteroid intake. The null result for sunlight UV-B radiation could be due to the method used to calculate annual ocular sunlight exposure, and any method based on a number of self-reported variables could have inherent inaccuracies. This will occur for both the case and control groups, which create an undifferential bias that causes a tendency toward the null. Corticosteroid intake on the other hand did show a significant age-adjusted risk for nuclear cataract but was not significant in the multivariate model. Undifferential bias could also exist in a participant’s recall of drug history. Other factors that have had some anecdotal evidence for risk from other studies but were found to have null associations with cataract development in this study include alcohol intake, aspirin use, family history, hypertension, and low educational status.
The majority of the findings in this study agree with results from previous cross-sectional studies and recent reports from other prospective studies. The strengths of this study lie in its high participation rate and standardized measurement protocols. Increased age, current cigarette smoking, diabetes mellitus, and female sex are risk factors that are generally consistent between different types of studies. Among these factors, and from a public health perspective, cigarette smoking remains highest on the list as a socially preventable and modifiable risk factor. The other factors will require further physiological studies, and any modification or prevention is likely to be pharmacological. Findings related to occupation, country of birth, arthritis, myopia, and calcium channel blockers are still inconclusive and require further clarification from future prospective studies.
Correspondence: Bickol N. Mukesh, PhD, Marshfield Clinic Research Foundation (ML2), 1000 N Oak Ave, Marshfield, WI 54449 (email@example.com).
Submitted for Publication: April 12, 2004; accepted July 28, 2005.
Financial Disclosure: None.
Funding/Support: This study was supported by grants from the National Health and Medical Research Council, Canberra, Australian Capital Territory, Australia; the Victorian Health Promotion Foundation, Carlton South, Victoria, Australia; the Dorothy Edols Estate, Melbourne, Victoria; the Ansell Ophthalmology Foundation, Melbourne; the Jack Brockhoff Foundation, Melbourne; the Eye Ear Nose and Throat Research Institute, Melbourne; the Felton Bequest, Melbourne; the Hugh D.T. Williamson Foundation, Melbourne; and the Appel Family Bequest, Melbourne. Dr McCarty was the recipient of the Wagstaff Research Fellowship in Ophthalmology from the Royal Victorian Eye and Ear Hospital, Melbourne.