Klein R, Klein BEK, Tomany SC, Danforth LG, Cruickshanks KJ. Relation of Statin Use to the 5-Year Incidence and Progression of Age-Related Maculopathy. Arch Ophthalmol. 2003;121(8):1151-1155. doi:10.1001/archopht.121.8.1151
Copyright 2003 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2003
To examine the association of hydroxymethyl glutaryl coenzyme A reductase inhibitors (statins) with the 5-year incidence of age-related maculopathy(ARM).
Population-based cohort study. Participants included persons 48 to 91 years old examined March 1, 1993, through June 14, 1995, living in Beaver Dam, Wis (N = 3684), of whom 2780 participated in a follow-up 5 years later.
Standardized procedures were used for physical examinations, blood sample collection, and questionnaire administration. Age-related maculopathy was determined by grading images of the posterior pole using a standard protocol. Standard univariate and multivariate analyses were performed.
Main Outcome Measures
Incidence and progression of ARM was measured over the 5-year interval.
While controlling for age and sex, statin use was not found to be associated with the 5-year incidence of early ARM (odds ratio [OR], 1.12; 95% confidence interval [CI], 0.47-2.67), progression of ARM (OR, 1.22; 95% CI, 0.54-2.76), or incidence of late ARM (OR, 0.41; 95% CI, 0.12-1.45).
These findings do not suggest an association between statin use and incident ARM over a 5-year period. Further investigation of these relationships in larger studies over a longer period is needed.
WHILE photocoagulation and photodynamic therapy have shown some efficacy in reducing severe visual loss in eyes with neovascular age-related macular degeneration, most eyes with advanced age-related macular degeneration do not benefit.1,2 Data from the Age-Related Eye Disease Study showed the limited efficacy of antioxidants and zinc supplements in persons with early age-related maculopathy (ARM) with a 23% rate of progression to advanced ARM in the treated group vs 28% in the control group.3 In addition, 23% of eyes receiving antioxidant and zinc treatment had a 15-letter decrease in the visual acuity score despite such treatment. No other medical interventions have been shown to reduce the incidence and progression of ARM or reduce visual loss.
Recent data from 2 studies showed an inverse association of statins and ARM.4,5 However, the few numbers of persons using statins (27 and 28, respectively) raised concerns about these findings.6 This study examined the association of statins with prevalent and incident ARM in the large population-based Beaver Dam Eye Study.
Methods used to identify and describe the population have appeared previously.7- 11 In brief, a private census of the population of Beaver Dam, Wis, was taken September 15, 1987, to May 4, 1988, to identify all residents in the city or township of Beaver Dam aged 43 to 84 years. Of the 5924 eligible, 4926 participated in the baseline examination between March 1, 1988, and September 14, 1990.7 Of the population, 99% was white. Of the survivors, 3684 (81.1%) participated in the 5-year follow-up examination between March 1, 1993, through June 14, 1995. Of the 3334 surviving participants in the baseline and second examination, 2764 (82.9%) participated in the second follow-up examination between March 1, 1998, and June 9, 2000. Comparisons between participants and nonparticipants at the time of the baseline and 5- and 10-year follow-up examinations have been published elsewhere.7- 9 Nineteen persons (0.5%) were taking statins at the 1988-1990 examination, 143 (3.9%) at the 1993-1995 examination, and 558 (18.9%) at the 1998-2000 examination. Because few persons were taking statins at the 1988-1990 examination, for purposes herein, the 1993-1995 examination was considered "baseline." Because an additional 19 people who did not participate in the first examination participated in the second and third ones, 2783 people were eligible for inclusion in the analysis. Of these, 3 persons were missing data on statin use; therefore, 2780 people contributed to this analysis.
Persons who were alive but did not participate in the 10-year follow-up(n = 425) were older at the 1993-1995 examination than those who did (66.1 vs 63.5 years, P<.001). After adjusting for age, those who were alive during the study period and did not participate were more likely to have fewer years of education completed, a lower annual income, more pack-years smoked, higher systolic blood pressure, and to be retired at the 1993-1995 examination than persons who participated. After adjusting for age and sex, participants with early ARM at baseline were as likely to participate as those in whom ARM was absent (data not shown). While controlling for age, there were no differences in participation at the 10-year follow-up examinations for men and women with a history of statin use and early ARM at the 1993-1995 examination compared with those who had a history of statin use without early ARM (data not shown).
Similar procedures have been used at both the 1993-1995 and the 5-year follow-up examinations and are described elsewhere.10,11 Informed consent was obtained from each participant at the beginning of the examination. Medication and vitamin use was assessed using a standardized questionnaire administered by the examiners at each examination.10 Participants were asked to bring to the examination all medications (prescription and over-the-counter) that they were regularly taking. The examiner asked whether there were other medications being taken but not brought. If there were, the subject was asked to call the study examiner with the medication name. In addition, at the baseline examination participants were asked if they had used specific classes of drugs in the past. Information on duration of use was not obtained. Participants were asked to list where they usually obtained their medications. Participants, their physicians, and their pharmacies were called when necessary to verify medication and reason for use.
The name of the drug was entered into a drug database where the record number assigned was associated with the drug use section of the questionnaire through a code table structure. With the drug record, each active ingredient was assigned the appropriate American Hospital Formulary Service code.12 In addition, subclassification information was included(eg, type of lipid-lowering agent: statin [lovastatin, simvastatin, pravastatin, fluvastatin, or atorvastatin], nicotinic acid, fibric acid derivatives [clofibrate and gemfibrozil], and bile acid sequesterants). The examination at baseline and follow-up included measuring weight, height, pulse rate, and blood pressures(using a random-zero sphygmomanometer following the Hypertension Detection and Follow-up Program protocol).13 Casual blood specimens were obtained. Serum total and high-density lipoprotein cholesterol levels were determined by enzymatic procedures.14,15 Stereoscopic 30° color fundus photographs centered on the optic disc (Diabetic Retinopathy Study [DRS] standard field 1), macula (DRS standard field 2), and a nonstereoscopic color fundus photograph temporal to but including the fovea of each eye were taken.
The Wisconsin Age-Related Maculopathy Grading System was used to assess the presence and severity of lesions associated with ARM. Grading procedures, lesion descriptions, and detailed definitions for the presence and severity, as well as the incidence of specific lesions, have been published elsewhere.16- 18 Incidence implies the appearance of a lesion at follow-up when it was absent at baseline in any of the subfields that could be graded at baseline and follow-up examinations. Progression implies the presence of a lesion at baseline with a worsening at follow-up.17,18
Incidence was determined for maximum size and type of each specific drusen class, increased drusen area, increased retinal pigment, retinal pigment epithelial (RPE) depigmentation, pigmentary abnormalities (defined as RPE depigmentation or increased retinal pigment), signs of exudative macular degeneration, and pure geographic atrophy. For example, if none of the subfields had soft indistinct drusen at the 1993-1995 examination, and soft indistinct drusen were present in 1 or more subfields at the 1998-2000 examination, the eye would be considered to have "incident" soft indistinct drusen.
Early ARM was defined by the presence of either soft indistinct drusen or the presence of any type of drusen associated with RPE depigmentation or increased retinal pigment. Late ARM was defined by either exudative macular degeneration or pure geographic atrophy.
For each eye, a 6-level severity scale for ARM was defined as follows19:
Level 10: No drusen of any type or hard drusen or small soft drusen(<125 µm in diameter) only, regardless of area of involvement, and no pigmentary abnormality (increased retinal pigment or RPE depigmentation) present.
Level 20: Hard drusen or small soft drusen (<125 µm in diameter), regardless of area of involvement, with pigmentary abnormalities present or soft drusen (≥125 µm in diameter) with drusen area less than 196 350µm2 (equivalent to a circle with a 500-µm diameter) and no pigmentary abnormalities present.
Level 30: Soft drusen (≥125 µm in diameter) with drusen area less than 196 350 µm2 and pigmentary abnormalities present or soft drusen (≥125 µm in diameter) with drusen area of 196 350µm2 or more with or without increased retinal pigment but no RPE depigmentation present.
Level 40: Soft drusen (≥125 µm in diameter) with drusen area of 196 350 µm2 or more involvement and RPE depigmentation present with or without increased retinal pigment.
Level 50: Pure geographic atrophy in absence of exudative macular degeneration.
Level 60: Exudative macular degeneration with or without geographic atrophy present.
Progression for a participant was defined as an increase in the maculopathy severity in either eye by 2 or more steps from level 10 through level 30 and 1 or more steps from level 40 or level 50 from the 1993-1995 examination to the 1998-2000 examination.
Age was defined as the participant's age at the baseline examination. The mean systolic blood pressure was the average of the 2 systolic blood pressure determinations, and the mean diastolic blood pressure was the average of the 2 diastolic blood pressures at baseline. Pulse pressure was defined as the mean systolic blood pressure minus the mean diastolic blood pressure. A person was defined as having a positive history of cardiovascular disease if at baseline he or she responded affirmatively to the questions regarding history of angina, heart attack, or stroke. Cigarette smoking status at the time of the baseline examination was determined as follows. A subject was classified as a nonsmoker if he or she had smoked fewer than 100 cigarettes in his or her lifetime; as an ex-smoker if he or she had smoked more than 100 cigarettes in his or her lifetime but had stopped smoking before the baseline examination; and as a current smoker if he or she had not stopped smoking. A current heavy drinker was defined as a person consuming 4 or more servings of alcoholic beverages daily; a former heavy drinker as having consumed 4 or more servings daily in the past but not in the previous year; and a nonheavy drinker as never having consumed 4 or more servings daily on a regular basis. Participants were classified as current vitamin users if they took at least 1 vitamin per week within 1 month prior to the baseline examination; as past users if they had ever regularly taken vitamins at least once per week, but not within the last month; and as never being users if they never took vitamins regularly at least once a week.
For these analyses, we examined the relationships between statin use at the examination in 1993-1995 and prevalence and the 5-year incidence of each specific maculopathy lesion, and 2 end points of disease severity—early and late ARM and the progression of ARM. We also looked at change in statin use between the second and third examinations and the prevalence and incidence of early and late ARM and its associated lesions at the 1998-2000 examination. A commercially available SAS software program was used to analyze these data.20 Multivariate odds ratios and 95% confidence intervals(CIs) were calculated using logistic regression.21 Age- and sex-adjusted models were constructed by outcome for each of the potential risk factors. A final model was then built by outcome for each risk factor, adjusting for age, sex, vitamin use, pulse pressure, total serum cholesterol level, and history of smoking and heavy drinking. Tests of trend were done treating categorical risk factors as continuous variables in the logistic model and computing the χ2 statistic for the parameter estimate. In these models, age was considered using 3 indicator variables, and smoking status, vitamin use, and heavy drinking status were considered using 2 variables. Total serum cholesterol level was also included as a continuous variable in the models. Change in statin use between the 1993-1995 and the 1998-2000 examinations was modeled with 3 indicator variables. When no participants taking statins developed the outcome of interest, adjusted Cochran-Haenszel estimates of the odds ratio, and the corresponding confidence limits are presented. In these cases, the P value for the odds ratio is computed using the Cochran-Mantel-Haenszel test for association.21
Persons taking statins in 1993-1995 were more likely to be ex-smokers and to have hypertension, a greater body mass index, a higher total serum cholesterol level, a lower high-density lipoprotein cholesterol level, and a history of cardiovascular disease than those not taking statins (Table 1). The 5-year incidence and progression of ARM is given in Table 2.
Controlling for age and sex, there were no statistically significant associations of statin use at the baseline with prevalent ARM (Table 3) or with the 5-year incidence and progression of ARM (Table 4). Multivariable models including total serum cholesterol level, multivitamin use, smoking or heavy drinking status, and pulse pressure at baseline did not change these associations (data not shown). With 2780 participants and an α level of .05, the power to show a 50% decrease in incidence between users and nonusers of statins was 32% for incident early ARM and 34% for progression of ARM.
While correcting for age and sex, persons who began to use statins between the 1993-1995 and 1998-2000 examinations were 32% less likely to have soft indistinct drusen, 36% less likely to have drusen 125 µm or greater in diameter, and 71% less likely to have late ARM present at the 1998-2000 examination compared with persons who never took statins during this period(Table 5). These relations remained while controlling for total serum cholesterol level at the beginning of the period (data not shown).
Because of the discrepancy of the incidence and prevalence data, we examined whether beginning statin use during the period was associated with total serum cholesterol level and characteristics at the start of the period. Those with high total serum cholesterol levels (fourth quartile, ≥267.0 mg/d L [6.90 mmol/L]) and drusen size of 125 µm or more in diameter at the start of the period were less likely to begin statin use than those who had high total serum cholesterol levels and drusen size smaller than 125 µm in diameter (23/106 [21.7%] vs 176/530 [33.2%], P =.02, χ2 test). Similar statistically significant associations were found when soft indistinct drusen were present (20/98 [20.4%]) compared with absent (175/522 [33.5%], P = .01) and when late ARM was present (1/15 [6.7%]) compared with absent (194/606 [32.0%], P = .03, Fisher exact test).
In the Beaver Dam Eye Study, statin use was not statistically significantly associated with the prevalence, incidence, or progression of ARM. These findings are inconsistent with data from an earlier cross-sectional cohort study of 66- to 75-year-olds.4 In that study, 76 (22%) of the 352 persons who did not take statins and 1 (4%) of the 27 who were taking statins had signs of ARM (odds ratio, 0.14; 95% CI 0.02, 0.83). McCarty et al found a statistically nonsignificant (P = .11) 4-fold increase in progression of large drusen (≥125 µm in diameter but <250 µm in diameter) in those not taking blood cholesterol–lowering medications (72 [13%] of 552 subjects) compared with those taking them (1[3.6%] of 28 subjects) and advocated "a randomized clinical trial to assess the effect"5(p340) of these drugs(especially the statins) based on their findings. However, no such relationship was found cross sectionally in their study.22 The POLA (Pathologies Oculaires Liées B l'Age), Rotterdam, Amsterdam, and Beaver Dam Eye Studies did not find a relationship of lipid-lowering agents with ARM.23- 25
Hypercholesterolemia, oxidative damage, and endothelial dysfunction have been hypothesized as pathogenetic mechanisms in the incidence and progression of ARM.26- 28 The cholesterol-lowering, antioxidant, and antiendothelial dysfunction effect of statins have been hypothesized to explain why statins may reduce the incidence and progression of ARM.4 It is, however, possible that those studies' results may be due not to a protective effect of statins but to chance alone given the large number of drugs studied and the small numbers of persons taking statins or cholesterol-lowering drugs.6 In addition, in some studies, a high total serum cholesterol level has been associated with reduced prevalence of advanced ARM.29,30 It is also possible that the relationship found by Hall et al4 and McCarty et al5 may be caused by unmeasured confounding or to a biological relationship of a protective effect of high cholesterol levels incidentally treated with statin. Our study notes the cholesterol levels were still higher in those using statins compared with nonusers.
In the Beaver Dam Eye Study, there were no statistically significant associations of statin use and ARM prevalence at the 1993-1995 examination or 5-year incidence. However, in persons who began using statins between the 1993-1995 and 1998-2000 examinations, statins seemed to have a protective effect with statistically significantly reduced prevalence of soft distinct and large drusen and incident late ARM. These findings among new users of statins were consistent with the cross-sectional findings of Hall et al4 and may be partly explained in our study by our finding that persons with a high total serum cholesterol level and ARM at the 1993-1995 examination were less likely to start taking statins than those with a high total serum cholesterol level without ARM, leading to the finding of a "protective" effect when the data are examined cross sectionally. Further follow-up of the cohort will provide information on the association of change in statin use with incident ARM.
While the Beaver Dam Eye Study is among the largest population-based studies to date to examine the association of statins with ARM, care must be taken in interpreting our study results. Although we report no association of statins to incident ARM, we have limited power to assure that such a relation does not exist. For example, with 2780 participants of whom 118 took statins and a 5-year cumulative incidence of 5.6% for early ARM, the power to detect a 50% reduction of incidence at the α level of .05 in statin users was 30% for early ARM. With the increased frequency of statin use in the population, adequate power to examine this association may be more feasible in the future. We also have no data regarding the dosage of statin use, although it is not known whether this or duration of use would influence these associations.
We found no consistent or significant association of statin use to the incidence or progression of ARM. Further analyses of data from other large population-based studies of longer periods are warranted before advocating a clinical trial of this class of drugs for ARM.
Corresponding author: Ronald Klein, MD, MPH, Department of Ophthalmology and Visual Sciences, University of Wisconsin, Madison, 610 N Walnut St, 417 WARF, Madison, WI 53726 (e-mail: Kleinr@epi.ophth.wisc.edu).
Submitted for publication December 10, 2002; final revision received March 7, 2003; accepted March 28, 2003.
This study was supported by grant EY06594 from the National Institutes of Health, Bethesda, Md (Drs R. Klein and B. E. K. Klein); and in part by Research to Prevent Blindness Inc, New York, NY (Dr R. Klein, Senior Scientific Investigator Award).