Several studies have identified diabetes as a risk factor for age-related (AR) cataracts.1–6 It has been predicted that by the year 2025, >300 million people in the world will have diabetes,7 and a corresponding global increase in AR cataract can be expected. Studies have shown that taking statins (HMG-CoA reductase inhibitors) reduces cardiovascular risks in patients with diabetes, even in those without high LDL-cholesterol.8,9 Consequently, statins are a class of pharmaceuticals commonly prescribed for patients with diabetes. Little work has been done specifically on the wide use of statins in patients with diabetes and their combined impact on cataract development.10 Early clinical trials on patients without diabetes did not find significant lenticular changes with statin use of <5 years.11–14 Subsequently, no adverse side effects on the human lens were listed in current drug compendiums, and manufacturers did not recommend monitoring of the crystalline lens. Furthermore, statin use was generally not controlled for in studies on diabetes and its associations to AR cataract. However, animal studies have clearly shown a correlation between cataract development and chronic statin treatment, although drug dosages have been generally higher than the clinical levels given to humans.15–17 A few human population studies have suggested a protective effect of statin use on cataract risk,18–20 hypothesizing an anti-inflammatory/antioxidant mechanism for the effect.20 Recently, Hippilsley-Cox and Coupland21 looked at data for >2 million patients (ages 30–84) in a prospective cohort study involving 368 general practices in the United Kingdom. The effects of statin type, dose, and duration of use were estimated by Cox proportional hazard models. Refuting earlier findings, statin use was associated with an increased risk of cataract, suggesting that further study is warranted. The type of AR lens opacities [i.e., nuclear sclerosis (NS), cortical cataract (CC), and/or posterior subcapsular cataract (PSC)] involved has not been investigated for specific association with statin use.
Type 2 diabetes accounts for 90% of all cases of diabetes in North America.7 Most studies on the risk factors for AR cataract have included both type 1 and type 2 diabetes. This study compared the prevalence of all clinically apparent AR lens opacities in subgroups with type 2 diabetes and without diabetes from a large clinic population at the University of Waterloo, School of Optometry and Vision Science. The prevalence of statin use was also compared for these subgroups, and any associations determined between statin use, type 2 diabetes, and AR cataract prevalence.
The Waterloo Eye Study (WatES) database was developed from a retrospective file review of 6397 patient visits from January 2007 to January 2008 at the University of Waterloo, School of Optometry and Vision Science. Data were abstracted for several variables for cross-sectional analysis. Abstraction methods, data quality analysis, study strengths and limitations, and population representation of the database have been detailed in an earlier article.22 Furthermore, overall cataract prevalence and the modeling technique using logistic regression have been reported previously for this data set, including a comparison between male and female patients.23 The abstracted ocular health data in the current investigation included the presence of any clinically apparent AR cataract (NS ≥ grade I, LOCS II,2,24 any CC, any PSC, or any history of related lens extraction [LE] and surgical dates of LE). A report of lens opacity in either eye was accepted as a diagnosis. The systemic health information included a diagnosis of type 1 diabetes or type 2 diabetes and any medication being taken. Patient age was electronically calculated in the database from the date of assessment and the patients' birthday, and this information was available for all patient files.
Data Quality and Population Representation
As previously reported, the repeatability of data abstraction was determined through interabstractor agreement rates of 425 double-entered clinic files.22 The Cohen κ was calculated for categorical data and the intraclass correlation coefficient for continuous data. The age and sex distributions of WatES patients with diabetes were compared with provincial diabetes rates available in the Ontario diabetes database.25
Outcome Assessment and Statistical Analysis
Patients with type 1 diabetes (n = 61) were excluded, and the remaining patients (n = 6336) were separated into subgroups of patients having type 2 diabetes and those not having diabetes, and then further sorted into patients having any or no AR lens opacity. The overall prevalence (%) of any AR cataract was determined for patients with type 2 diabetes and without diabetes. Multivariable logistic regression analysis, which controlled for age, was done to determine prevalence probability functions for AR cataract in the two subgroups.
We have previously shown a sex difference in AR cataracts in this population, with being female being associated with an increased risk of CC.23 Smoking is known to be associated with nuclear sclerosis.1–3,26 Recently, the Malay Study found that besides diabetes, the presence of high blood pressure (BP) was associated with increased odds of having cataract.27 Therefore, sex, smoking, and high BP were controlled for, when the odds ratio (OR) for a diagnosis of type 2 diabetes and AR cataract prevalence was calculated through multivariable logistic regression analysis.
Associations between each cataract type (NS, CC, and PSC) and type 2 diabetes were determined independent of whether they occurred in a mixed-type presentation or as homogeneous opacities. There were a significant number of patients with CC or PSC who also had NS. For this reason, an insufficient number of patients with only CC or PSC were available to consider patients with a single type of opacity in our analysis. Instead, we chose to do a multivariable logistic regression analysis controlling for the aforementioned variables as well as the other lens opacity types when looking at NS, CC, and then PSC. Patients with monocular LE were categorized by the cataract type in the other eye. Patients who had undergone bilateral LE (n = 312) were excluded, as it was not possible to determine from the database which type of AR cataract existed pre-surgically. For the remaining patients (n = 6024), ORs were calculated for any association between a diagnosis of type 2 diabetes and cataract subtype presence.
Finally, the prevalence of reported statin use was determined for patients with type 2 diabetes and those without diabetes. A multivariable analysis was performed to control for age, sex, smoking, high BP, and type 2 diabetes status, to determine any associations (ORs) between statin use and overall AR cataract prevalence, and then for each AR cataract subtype. Prevalence probability functions for AR cataract were determined for each of the following groups: (1) patients with type 2 diabetes not taking statins, (2) patients with type 2 diabetes taking statins, (3) patients without diabetes not taking statins, and (4) patients without diabetes taking statins. The ages at which there was a 50% probability of AR cataracts were determined for patients in each of these groups.
As previously reported, Cohen κ statistic values were high for abstraction repeatability of patient sex and the presence of NS, CC, PSC, and LE, as was the intra-class coefficient value for patient age.23 The Kappa value for the presence of diabetes was high at 0.99 with <1% disagreement rate between abstractors. The Kappa value was somewhat lower for overall medications (κ = 0.84); however, when considered on its own, the interabstractor agreement for statin use was high at κ = 0.95, with a disagreement rate of <1%.
Table 1 details patient characteristics for this study. There were 61(<1%) patients with type 1 diabetes who were not included in the remaining analysis. The mean age for the type 2 diabetes subgroup was approximately 14 years older than that for the patients without diabetes. Patients without diabetes had a slightly higher proportion of female patients compared with male patients. However, there were fewer female patients compared with male patients with type 2 diabetes.
The Canadian Diabetes Association used algorithms applied to health care administrative data from 1995 to 1999 to create the Ontario Diabetes Database (ODD).25Fig. 1 compares the age distribution of WatES patients with diabetes (type 1 and type 2 diabetes) with the 1999 ODD numbers using their age groupings. The overall rate of diabetes in people >19 years of age was 10.5% for the WatES database in 2007 and 6.2% for the ODD in 1999. For both databases, women in the 20- to 34-year age-group had a slightly higher prevalence of diabetes than men, but a lower prevalence in all remaining age-groups. WatES has a lower proportion of patients with diabetes who are >64 years of age, and a slightly higher proportion of patients with diabetes in the 35- to 64-year age range.
The relatively older WatES type 2 diabetes subgroup had 77.0% of its patients with some clinically apparent AR cataract, whereas the proportionately younger no diabetes subgroup only had 32%. Fig. 2 shows the prevalence of AR cataract in the subgroups in yearly age-groups. Prevalence probability functions determined with logistic regression are also shown, and, as expected, AR cataracts occurred significantly earlier in patients with type 2 diabetes compared with those without diabetes. For this population, a diagnosis of type 2 diabetes was associated with increased odds of having AR cataracts (OR = 1.86, 95% CI: 1.34–2.59) across the life span when controlling for age, being female, smoking, and high BP. After multivariable analysis, a diagnosis of type 2 diabetes in this study group was associated with increased odds of having NS (OR = 1.84, 1.32–2.56), CC (OR = 1.38, 1.04–1.82), and PSC (OR = 1.52, 1.04–2.19) compared with patients with no diabetes.
The mean age of patients taking statins was 68.5 years (±11.1). Statin use increased with age such that 0.4% of WatES patients <39 years of age, 9.2% of patients between 39 and 59 years, and 30.5% of patients >60 years were taking statins. In patients >38 years, statin use was reported in 56% of patients with type 2 diabetes, but only 16% of patients without diabetes. Table 2 shows the result of multivariable regression analysis that included statin use. After controlling for age, sex, smoking, high BP, and type 2 diabetes, statin use was significantly associated with AR cataract (OR = 1.57, 1.15–2.13), NS (OR = 1.48, 1.09–2.00), and PSC (OR = 1.48, 1.07–2.04), but not CC (OR = 1.02, 0.80–1.30). PSC was no longer significantly associated with type 2 diabetes when statin use was considered. CC was still associated with being female (OR = 1.59, 1.30–1.94), and in this analysis, with high BP (OR = 1.24, 1.00–1.53). NS was associated with smoking (OR = 1.62, 1.08–2.42).
Table 3 shows the prevalence of cataract in patients (≥30 years), sorted by a diagnosis of type 2 diabetes and statin use, in 10-year intervals. The prevalence of cataract increased at a faster rate in patients with diabetes who used statins. Similar prevalence levels were seen in patients with diabetes who did not use statins and in patients without diabetes who did use statins. The prevalence of cataract increased at the slowest rate in patients without diabetes who did not use statins. Fig. 3 demonstrates the result of logistic regression for each of the four groups. The probability of AR cataract in patients who used statins reached 50% at age 51.7 years for patients with type 2 diabetes and at age 54.9 years for patients without diabetes. In patients who did not use statins, it was later at 55.1 and 57.3 years for patients with type 2 diabetes and without diabetes, respectively.
In WatES patients >38 years of age, reported statin use was almost 3.5 times higher in patients with type 2 diabetes than in those without diabetes. Statin use was significantly associated with AR cataract such that the probability of cataract for patients with type 2 diabetes who did not use statins was similar to patients without diabetes who did use statins. Furthermore, a clinically meaningful difference (5.6 years) was found for the age of 50% probability of AR cataract between WatES patients without diabetes who did not use statins and those with type 2 diabetes who did take statins. Specifically, statin use was associated with NS and PSC types of AR cataracts.
The bio-plausibility of these results lies in the fact that the crystalline lens membrane requires high cholesterol for proper epithelial cell development and lens transparency.28 Increased cataract formation has been seen in both animals and humans with hereditary cholesterol deficiency,28,29 and the risk exists that statins can inhibit cholesterol biosynthesis in the human lens.
Another important finding was that PSC, long associated with diabetes, was no longer significantly associated with type 2 diabetes when statin use was considered. This can be the result of statin use being directly associated with PSC in people with diabetes rather than diabetes itself. But it may also be the result of too few patients with type 2 diabetes who are not using statins to reach statistical significance. Also, only a few studies have found diabetes to be a risk factor for NS.4,6,30 However, most studies looked at much higher grades of NS than WatES, resulting in weak associations between diabetes and NS. It is possible that NS at these levels tends to occur at fairly advanced ages where the prevalence of severe diabetes diminishes because of increased mortality.31
It is likely that some of the associations found in this study, but not in previous investigations, reflect the potential additional years of statin use compared with earlier studies. Statins are intended for long-term use and Neutel et al.,32 in their report on statin use in Canada, found that approximately 75% of users continue to take statins for at least 2 years once they have started. Cenedella28 suggests that the long-term impact of statin use requires study periods of between10 to 20 years.
This study had several strengths. More than 6000 patient visits contributed to the information in the WatES database providing significant statistical power throughout analyses. Age-related cataracts, type 2 diabetes, and statin use were found to be prevalent conditions in this population, especially after 38 years of age, making them appropriate factors for cross-sectional analysis. The study benefited from using the criterion of any clinically apparent AR cataract, as opposed to the more commonly chosen advanced levels of lens opacity. Thus, yearly prevalence levels for overall AR cataract approached 100% by the late 70s. Also, the WatES database included data for all yearly age-groups so that a probability of AR cataract function for the entire human age range could be generated through logistic regression analysis.
Study limitations include the fact that cross-sectional studies cannot determine causation. Because WatES is a clinic population, the results may not be generalizable to the entire population. For example, patients with diabetes are encouraged to get routine eye assessments because of their increased risk of ocular disease, and therefore, we could expect a clinical eye care population to have a somewhat higher percentage of patients with diabetes than the general population.26 Selection bias in clinic-based population studies can occur, as patients with symptoms from maturing cataracts are more likely to seek out vision care than those without.31 Our inclusion of pre-symptomatic levels of cataract should help offset this bias. Limitations in available file information did not allow for analysis of type, dosage, or duration of statin use or patient blood cholesterol levels and their relationships to lens opacity. However, the strong association found between statin use and AR cataract in this study indicates a need for additional study to investigate these more specific aspects of statin use. Given the high cost of LE to the health care system, further work on the impact of statin use on cataract surgery rates is also recommended.
Regardless of the outcome of further study, the benefits of statin use in people with type 2 diabetes are anticipated to continue to outweigh any associated increased risk of AR cataract. However, given the known financial and functional burdens, both in terms of surgical costs and visual impairment when surgery is inaccessible or pending,26,33–36 it is important to identify and minimize factors that accelerate cataract development. Information provided here can serve public health efforts to educate people on the risks of treatments associated with type 2 diabetes. It also supports current efforts to curtail type 2 diabetes prevalence trends. Finally, results from further investigation into these associations may encourage the development of alternative medications for lowering cholesterol that are not associated with an increased risk of cataract development.37
Elizabeth L. Irving
School of Optometry
University of Waterloo
200 University Ave W.
We thank Linda Lillakas of the University of Waterloo, School of Optometry and Vision Science for help with the preparation of the manuscript. This work was supported by NSERC (Canada) and CRC (Canada) to ELI and presented in part at the American Academy of Optometry, Boston MA, October 12–15, 2011.
1. Delcourt C, Cristol JP, Tessier F, Leger CL, Michel F, Papoz L. Risk factors for cortical, nuclear, and posterior subcapsular cataracts: the POLA study. Pathologies Oculaires Liees a l'Age. Am J Epidemiol 2000;151:497–504.
2. Leske MC, Chylack LT Jr., Wu SY. The lens opacities case-control study. Risk factors for cataract. Arch Ophthalmol 1991;109:244–51.
3. Mukesh BN, Le A, Dimitrov PN, Ahmed S, Taylor HR, McCarty CA. Development of cataract and associated risk factors: the Visual Impairment Project. Arch Ophthalmol 2006;124:79–85.
4. 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;15:317–27.
5. Hennis A, Wu SY, Nemesure B, Leske MC. Risk factors for incident cortical and posterior subcapsular lens opacities in the Barbados eye studies. Arch Ophthalmol 2004;122:525–30.
6. Klein BE, Klein R, Lee KE. Diabetes, cardiovascular disease, selected cardiovascular disease risk factors, and the 5-year incidence of age-related cataract and progression of lens opacities: the Beaver Dam Eye Study. Am J Ophthalmol 1998;126:782–90.
7. Stenson S. Healthy sight counseling: diabetes and the eye. Clin Refract Optom 2009;20:248–58.
8. Colhoun HM, Betteridge DJ, Durrington PN, Hitman GA, Neil HA, Livingstone SJ, Thomason MJ, Mackness MI, Charlton-Menys V, Fuller JH. Primary prevention of cardiovascular disease with atorvastatin in type 2 diabetes in the Collaborative Atorvastatin Diabetes Study (CARDS): multicentre randomised placebo-controlled trial. Lancet 2004;364:685–96.
9. Fowler MJ. Microvascular and macrovascular complications of diabetes. Clin Diabetes 2008;26:77–82.
10. Hermans MP, Ahn SA, Rousseau MF. Statin therapy and cataract in type 2 diabetes. Diabetes Metab 2010;37:139–43.
11. Behrens-Baumann W, Thiery J, Fieseler HG, Seidel D. Pravastatin—ocular side effects after a two year follow-up? Lens Eye Toxic Res 1990;7:311–8.
12. Schmitt C, Schmidt J, Hockwin O. Ocular drug-safety study with the HMG-CoA reductase inhibitor pravastatin. Lens Eye Toxic Res 1990;7:631–42.
13. Harris ML, Bron AJ, Brown NA, Keech AC, Wallendszus KR, Armitage JM, MacMahon S, Snibson G, Collins R. Absence of effect of simvastatin on the progression of lens opacities in a randomised placebo controlled study. Oxford cholesterol study group. Br J Ophthalmol 1995;79:996–1002.
14. Lundh BL, Nilsson SE. Lens changes in matched normals and hyperlipidemic patients treated with simvastatin for 2 years. Acta Ophthalmol (Copenh) 1990;68:658–60.
15. Hartman HA, Myers LA, Evans M, Robison RL, Engstrom RG, Tse FL. The safety evaluation of fluvastatin, an HMG-CoA reductase inhibitor, in beagle dogs and rhesus monkeys. Fundam Appl Toxicol 1996;29:48–62.
16. Zakrzewski P, Milewska J, Czerny K. The eye lens evaluation of the atorvastatin-treated white rat. Ann Univ Mariae Curie Sklodowska Med 2002;57:165–71.
17. Gerson RJ, MacDonald JS, Alberts AW, Chen J, Yudkovitz JB, Greenspan MD, Rubin LF, Bokelman DL. On the etiology of subcapsular lenticular opacities produced in dogs receiving HMG-CoA reductase inhibitors. Exp Eye Res 1990;50:65–78.
18. Chodick G, Heymann AD, Flash S, Kokia E, Shalev V. Persistence with statins and incident cataract: a population-based historical cohort study. Ann Epidemiol 2010;20:136–42.
19. Tan JS, Mitchell P, Rochtchina E, Wang JJ. Statin use and the long-term risk of incident cataract: the Blue Mountains Eye Study. Am J Ophthalmol 2007;143:687–9.
20. Klein BE, Klein R, Lee KE, Grady LM. Statin use and incident nuclear cataract. JAMA 2006;295:2752–8.
21. Hippisley-Cox J, Coupland C. Unintended effects of statins in men and women in England and Wales: population based cohort study using the QResearch database. BMJ 2010;340:c2197.
22. Machan CM, Hrynchak PK, Irving EL. Waterloo Eye Study: data abstraction and population representation. Optom Vis Sci 2011;88:613–20.
23. Machan CM, Hrynchak PK, Irving EL. Modeling the prevalence
of age-related cataract: Waterloo Eye Study. Optom Vis Sci 2012; 89:130–6.
24. Chylack LT Jr., Leske MC, McCarthy D, Khu P, Kashiwagi T, Sperduto R. Lens opacities classification system II (LOCS II). Arch Ophthalmol 1989;107:991–7.
25. Institute for Clinical Evaluative Sciences (ICES). Diabetes in Ontario: An ICES Practice Atlas: June 2003. Available at: http://www.ices.on.ca/webpage.cfm?site_id=1&org_id=31&morg_id=0&gsec_id=0&item_id=1312
. Accessed June 11, 2012.
26. Hodge WG, Whitcher JP, Satariano W. Risk factors for age-related cataracts. Epidemiol Rev 1995;17:336–46.
27. Sabanayagam C, Wang JJ, Mitchell P, Tan AG, Tai ES, Aung T, Saw SM, Wong TY. Metabolic syndrome components and age-related cataract: the Singapore Malay Eye Study. Invest Ophthalmol Vis Sci 2011;52:2397–404.
28. Cenedella RJ. Cholesterol and cataracts. Surv Ophthalmol 1996;40:320–37.
29. Mori M, Li G, Abe I, Nakayama J, Guo Z, Sawashita J, Ugawa T, Nishizono S, Serikawa T, Higuchi K, Shumiya S. Lanosterol synthase mutations cause cholesterol deficiency-associated cataracts in the Shumiya cataract rat. J Clin Invest 2006;116:395–404.
30. McCarty CA, Nanjan MB, Taylor HR. Attributable risk estimates for cataract to prioritize medical and public health action. Invest Ophthalmol Vis Sci 2000;41:3720–5.
31. Ederer F, Hiller R, Taylor HR. Senile lens changes and diabetes in two population studies. Am J Ophthalmol 1981;91:381–95.
32. Neutel CI, Morrison H, Campbell NR, de Groh M. Statin use in Canadians: trends, determinants and persistence. Can J Public Health 2007;98:412–6.
33. Resnikoff S, Pascolini D, Etya'ale D, Kocur I, Pararajasegaram R, Pokharel GP, Mariotti SP. Global data on visual impairment in the year 2002. Bull World Health Organ 2004;82:844–51.
34. Javitt JC, Wang F, West SK. Blindness due to cataract: epidemiology and prevention. Annu Rev Public Health 1996;17:159–77.
35. American Academy of Ophthalmology. Cataract in the Adult Eye, Preferred Practice Patterns: 2008. Available at: http://one.aao.org/CE/PracticeGuidelines/PPP.aspx
. Accessed October 8, 2008.
36. American Optometric Association. Optometric Clinical Practice Guideline: Care of the Adult Patient with Cataract. Available at http://www.aoa.org/documents/CPG-8.pdf
. Accessed June 11, 2012.
37. Culver AL, Ockene IS, Balasubramanian R, Olendzki BC, Sepavich DM, Wactawski-Wende J, Manson JE, Qiao Y, Liu S, Merriam PA, Rahilly-Tierny C, Thomas F, Berger JS, Ockene JK, Curb JD, Ma Y. Statin use and risk of diabetes mellitus in postmenopausal women in the Women's Health Initiative. Arch Intern Med 2012;172:144–52.