Effect of Statins on Risk of Coronary Disease:  A Meta-analysis of Randomized Controlled Trials FREE

In a very important sense, the cholesterol "controversy" is no more. A spate of recent clinical trials using 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors (statins) to lower low-density lipoprotein cholesterol (LDL-C) have demonstrated beyond reasonable doubt that coronary events, both morbid and mortal, can be prevented.^1- 8 Post hoc analyses of these trials also strongly suggest that stroke rates may be reduced to about the same degree. Observational epidemiology has not identified hypercholesterolemia as a major risk factor for stroke,⁹ although these data did not distinguish hemorrhage from atherothrombotic stroke.

Most subjects in these studies have been middle-aged men. This led to the suggestion that while elevated LDL-C is a risk factor in women as well as in men and in older as well as in middle-aged individuals, the data from these trials cannot be extrapolated to groups other than middle-aged men. It has been further suggested that, in the absence of such sex- and age-specific data, cholesterol screening is optional except in middle-aged men.¹⁰ This has encouraged some insurers, including Medicare, to refuse funding for cholesterol screening in older individuals. While it is difficult to prove, lack of support even for screening may contribute to the rather poor frequency of prescription of cholesterol-lowering regimens even for women with established coronary disease.¹¹

With 1 exception,³ the major statin cholesterol-lowering trials with clinical event end points have included women and those aged 65 years or older. All have reported similar effects in women vs men and in younger vs older subjects. Two trials have issued expanded reports of their findings.^2,5- 6

Meta-analysis of these studies can provide a more accurate and precise estimate of subgroup effects than can be derived from individual trials.¹² It also provides the opportunity to examine, in the aggregate, the effect of statin-induced cholesterol lowering in men and women across a wide age spectrum.

A literature search of the MEDLINE database (1966–December 1998), using the Medical Subject Headings hydroxymethyl-glutaryl-CoA reductase inhibitors, simvastatin, lovastatin, pravastatin, coronary disease, and myocardial infarction as well as the key words statin and coronary heart disease was performed. The search was restricted to studies published in English-language journals, conducted in human subjects, and classified as clinical trials in the MEDLINE database. A manual search was also performed using the authors' reference files and reference lists from original communications and review articles.^13- 15 The contents of 182 abstracts or full-text manuscripts identified during our literature search were reviewed to determine whether they met the criteria for inclusion. Of these abstracts and manuscripts, 29 statin drug treatment trials were identified. Other publications included reviews, letters to the editor, subgroup analysis, and secondary analysis of data from the 29 published trials.

For inclusion, a study had to meet the following criteria: (1) random allocation of study participants to statin or a placebo control group; (2) no intervention difference, other than use of a statin, between the treatment and control groups; (3) intervention duration of at least 4 years; and (4) clinical disease or death as the primary end point. The mean intervention duration in all outcome trials was more than 4 years. Five trials met these criteria and were included in the meta-analysis.^1- 8

Major reasons for exclusion of studies were intervention duration of less than 4 years^16- 36 and trial primary end points that were not clinical events.^16- 39 Additional reasons for exclusion of studies included comparison of different statin drugs³³ and comparison of aggressive or moderate cholesterol lowering with statin drug treatment.³⁹

All data were abstracted in duplicate using a standardized protocol and reporting form. Disagreements were resolved by consensus. We did not contact authors to request additional information. Study characteristics recorded were as follows: (1) first author's name, year of publication, and country of origin; (2) number of participants; (3) mean age and age and sex distributions of participants; (4) presence of preexisting myocardial infarction; (5) baseline mean total, LDL-C, high-density lipoprotein cholesterol (HDL-C), and triglyceride levels; (6) net changes in lipids during intervention; (7) type and dosage of statin drug; and (8) intervention duration.

Major coronary events during treatment were abstracted as the primary outcome. They included coronary death, nonfatal myocardial infarction, silent myocardial infarction, or resuscitated cardiac arrest in the Scandinavian Simvastatin Survival Study (4S)¹; coronary death or nonfatal myocardial infarction in the West of Scotland Coronary Prevention Study (WOSCOPS),³ the Cholesterol and Recurrent Events Trial (CARE),^4- 5 and the Long-term Intervention With Pravastatin in Ischaemic Disease (LIPID) trial⁸; and fatal or nonfatal myocardial infarction, unstable angina, or sudden cardiac death in the Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS).⁷ In addition, data were abstracted on fatal coronary heart disease, cardiovascular disease deaths, noncardiovascular deaths, and all-cause deaths during treatment.

Both proportional (1 − odds ratio [OR]) and absolute risk reduction were used to measure the effect of statin drug treatment on clinical outcomes. The numbers of various outcomes for both the statin and placebo groups were recorded for each study using 2 × 2 tables. The Peto method was used to calculate pooled ORs of outcomes associated with statin therapy.⁴⁰ For each trial, the number of individuals in the treatment group in whom an end point of interest was observed (O) was compared with the number that would, if treatment had no effect, have been expected (E) on the basis of the overall experience in the treatment and control groups combined. If treatment was beneficial, then O − E would tend to be negative. The O − E values from each trial were summed and z statistics were used to test whether the total of O − E values differed from 0. Odds ratios were calculated by using [(O − E)/V], where V was the variance of the O − E total.⁴⁰ To calculate the pooled absolute risk, each study was weighted by its sample size (N_t × N_c)/(N_t + N_c), where N_t and N_c were sample sizes for statin therapy and control groups, respectively. The number needed to treat was calculated by taking the reciprocal of the absolute risk reduction. This represents the number of patients who would have to be treated to prevent 1 outcome event.

A series of prestated subgroup analyses was performed to examine the effect of statin drug treatment on major coronary events. First, the risk reductions of major coronary events and deaths were compared between 2 primary prevention trials^3,7 and 3 secondary prevention trials.^1,4,8 Furthermore, the risk reductions of major coronary events were calculated by sex and age groups. Four of the 5 trials included participants who were women or aged at least 65 years.^1,4,7- 8 Three of these trials planned subgroup analyses by sex and age group (≥65 years) in their original proposal.^1,4,8 One trial⁷ only reported subgroup analysis by median age, which was 57 years for men and 62 years for women.

A sensitivity analysis was conducted to explore the impact of excluding small nonoutcome trials with short intervention duration (<4 years) on risk estimates. Twelve trials that were excluded from the primary analysis and reported at least 1 major coronary event were included in this analysis.^{16- 20,23,25,27- 30,37}

Study design and participant characteristics for the 5 randomized, placebo-controlled, double-blind trials included in our meta-analysis are presented in Table 1. A total of 30,817 participants were included in these trials. Mean follow-up time was 5.4 years and mean age was 59 years. Only men younger than 65 years were included in WOSCOPS, while women and participants who were aged 65 years or older were included in the remaining 4 trials. Three trials (4S, CARE, and LIPID) were conducted in patients with a history of coronary heart disease (secondary prevention trials) and 2 trials (WOSCOPS and AFCAPS/TexCAPS) were conducted in a healthy population (primary prevention trials).

In 4S, patients with serum total cholesterol levels of 5.5 to 8.0 mmol/L (213-309 mg/dL) and triglyceride levels of 2.5 mmol/L (221 mg/dL) or less while following a lipid-lowering diet were eligible for participation.¹ In WOSCOPS, men with serum LDL-C levels of at least 4.0 mmol/L (155 mg/dL) at screening visits (at least 1 LDL-C measurement of ≥4.5 mmol/L [174 mg/dL] and 1 of ≤6.0 mmol/L [232 mg/dL]) while following a lipid-lowering diet were eligible for participation.³ In CARE, patients with total serum cholesterol levels of less than 6.2 mmol/L (240 mg/dL), LDL-C levels between 3.0 and 4.5 mmol/L (116-174 mg/dL), and triglyceride levels of less than 4.0 mmol/L (354 mg/dL) were eligible.⁴ Total cholesterol levels between 4.65 and 6.82 mmol/L (180-264 mg/dL), LDL-C levels between 3.36 and 4.91 mmol/L (130-190 mg/dL), and HDL-C levels of 1.16 mmol/L (45 mg/dL) or less for men and 1.22 mmol/L (47 mg/dL) or less for women were used as entry criteria in AFCAPS/TexCAPS.¹⁰ A broad range of serum total cholesterol levels (4.0-7.0 mmol/L [155-271 mg/dL]) and triglyceride levels of less than 5.0 mmol/L (443 mg/dL) were used as entry criteria for LIPID.⁸ Mean baseline levels of serum lipids in the 5 trials reflected these inclusion criteria.

The net changes in lipids (percentage change in treatment group − percentage change in placebo group) among those attending follow-up for lipid measurements are shown in Table 1. The mean reduction (weighted by sample size) in total cholesterol, LDL-C, and triglyceride levels was −20%, −28%, and −13%, respectively, and HDL-C was increased by an average of 5% among the 5 trials. The estimates excluded participants who were lost to follow-up before the end of the trial, although those participants were still included in end-point analysis. Assuming that there were few differences in changes in serum lipids between treatment and control groups among dropouts, the differences in lipid changes between treatment and control groups among all participants were likely somewhat smaller than these estimates.

Overall, 2042 major coronary events and 748 coronary deaths occurred in participants assigned to placebo and 1490 events and 543 deaths occurred in those allocated to active treatment (Table 2). When the results from the 5 trials were pooled, a significant reduction in the odds of major coronary events and coronary deaths (P<.001 for both) was observed among the participants allocated to active treatment. The reduction in coronary events was 31% (95% confidence interval [CI], 26%-36%) and the reduction in fatal coronary disease was 29% (95% CI, 20%-36%). Compared with control groups, active treatment was associated with an absolute risk reduction in coronary disease of 36 events and 13 deaths per 1000 patients. The number needed to treat was 28 to prevent a major coronary event and 75 to prevent a death from coronary disease.

Overall, 1297 deaths (868 from cardiovascular disease and 429 from noncardiovascular disease) occurred in control participants and 1046 (646 from cardiovascular disease and 400 from noncardiovascular disease) occurred in those assigned to active treatment (Table 2). Compared with controls, those receiving active treatment had a 21% reduction in the odds of total mortality (95% CI, 14%-28%; P<.001) and a 27% reduction in the odds of cardiovascular mortality (95% CI, 19%-34%; P<.001). Active treatment was also associated with an absolute risk reduction of 16 deaths from all causes and 14 deaths from cardiovascular disease per 1000 patients. The number needed to treat was 61 to prevent a death from all causes and 69 to prevent a death from cardiovascular disease. Noncardiovascular mortality was similar in active treatment and control groups (OR, 0.93; 95% CI, 0.81-1.07; P = .29).

Two primary prevention trials with 13,200 participants and 3 secondary prevention trials with 17,617 participants were included in this meta-analysis. Active treatment was associated with a 34% risk reduction (95% CI, 23%-43%; P<.001) in major coronary events in the 2 primary prevention trials and a 30% risk reduction (95% CI, 24%-35%; P<.001) in the 3 secondary prevention trials. Compared with controls, active treatment was associated with a lower risk of coronary disease mortality (OR, 0.73; 95% CI, 0.51-1.05; P = .09), cardiovascular mortality (OR, 0.68; 95% CI, 0.50-0.93; P = .01), and all-cause mortality (OR, 0.87; 95% CI, 0.71-1.06; P = .18) in the 2 primary prevention trials. Likewise, active treatment was associated with a lower risk of coronary disease mortality (OR, 0.71; 95% CI, 0.63-0.80; P<.001), cardiovascular mortality (OR, 0.73; 95% CI, 0.66-0.82; P<.001), and all-cause mortality (OR, 0.77; 95% CI, 0.70-0.85; P<.001) in the 3 secondary prevention trials. Active treatment was not significantly associated with change in noncardiovascular mortality in the 2 primary prevention trials (OR, 1.04; 95% CI, 0.80-1.35; P = .75) or in the 3 secondary prevention trials (OR, 0.89; 95% CI, 0.75-1.04; P = .15).

Odds ratios of major coronary events associated with statin treatment from individual trials stratified by sex are given in Figure 1, top. In all trials, the odds of coronary events were reduced for those assigned to active treatment compared with controls, and the risk reduction was statistically significant in 2 of 4 trials in women and all 5 trials in men. The overall proportional risk reduction was similar for women (29%; 95% CI, 13%-42%; P<.001) and men (31%; 95% CI, 26%-35%; P<.001) (Table 3). The absolute risk reduction was also similar in women (33 per 1000; 95% CI, 13-52 per 1000) and men (37 per 1000; 95% CI, 29-44 per 1000).

In all trials, the odds of coronary events were reduced in active treatment compared with controls stratified by age group (Figure 1, bottom). The risk reduction was statistically significant in all 4 trials among persons aged at least 65 years and in 4 of 5 trials among persons younger than 65 years. The overall proportional risk reduction was similar for persons aged at least 65 years (32%; 95% CI, 23%-39%; P<.001) and persons younger than 65 years (31%; 95% CI, 24%-36%; P<.001) (Table 3). The absolute risk reduction, however, was slightly higher in persons aged at least 65 years (44 per 1000; 95% CI, 30-58 per 1000) compared with persons younger than 65 years (32 per 1000; 95% CI, 24-40 per 1000).

After including 12 small nonoutcome trials, estimates for risk reduction were virtually unchanged. Among 17 trials, 19,597 participants were allocated to placebo and 24,601 to active treatment. Of those, 2159 and 1618 developed major coronary events, 776 and 589 died from coronary heart disease, 905 and 697 died from cardiovascular disease, 447 and 410 died from noncardiovascular disease, and 1352 and 1107 died from all causes for the control and active treatment groups, respectively. The corresponding proportional risk reduction was 31% (95% CI, 26%-35%; P<.001) for major coronary events, 28% (95% CI, 19%-35%; P<.001) for fatal coronary disease, 27% (95% CI, 19%-34%; P<.001) for cardiovascular disease mortality, and 22% (95% CI, 15%-28%; P<.001) for all-cause mortality. The mortality from noncardiovascular disease was not significantly different between placebo and treatment groups (OR, 0.91; 95% CI, 0.80-1.04; P = .18).

Overall, 1021 cancer cases occurred in participants who were allocated to placebo and 1009 in participants who were allocated to active treatment (OR, 0.99; 95% CI, 0.90-1.08; P = .76). Forty study participants had asymptomatic episodes of elevated creatine kinase concentrations (>10 times the upper reference limit) in the control group compared with 50 in the active treatment group (OR, 1.25; 95% CI, 0.83-1.89; P = .29). Two hundred fifty-eight participants had increased aspartate or alanine aminotransferase levels (>3 times the upper reference limit) in the control group compared with 290 in the active treatment group (OR, 1.13; 95% CI, 0.95-1.33; P = .17).

A clear and consistent effect of statin-induced LDL-C lowering in significantly reducing the risk of coronary events, independent of sex or age, has been demonstrated in the individual studies and the meta-analysis presented here. A roughly 30% decline in coronary events is seen in all sex and age groups studied. There are insufficient data from published results (except in CARE and 4S)^2,5- 6 to draw any conclusion about sex- or age-specific mortality. Ongoing trials will provide additional information on the effects of LDL-C lowering by statins by sex and age.^41- 43

It is clear that there is no effect of these interventions on noncardiovascular mortality. Neither those studies whose participants had initially high cholesterol levels (4S, LIPID, and WOSCOPS) nor those whose initial cholesterol levels were lower (CARE and AFCAPS/TexCAPS) reported any increase in noncardiovascular mortality. Because these studies were limited for approximately 5 years, they cannot offer information other than the possible effects of longer exposure to statins or cholesterol lowering. Some of the guidelines and policies that resulted from an unwillingness to extrapolate earlier reported data to women and the elderly should be revisited. Policies designed to limit screening in women and elderly persons¹⁰ make no sense and, in fact, are potentially harmful because they diminish in the eyes of both the public and the practicing physician the importance of cholesterol interventions in these groups.

The recent results of the Heart and Estrogen/Progestin Replacement Study (HERS)⁴⁴ failed to demonstrate benefit on cardiovascular events with a regimen of conjugated equine estrogen and medroxyprogesterone (in dosages of 0.625 and 2.5 mg/d, respectively). At this point, therefore, hormone replacement therapy should not be looked on as an alternative to vigorous interventions to lower LDL-C levels in women, who make up two thirds of the population aged 65 years or older.⁴⁵

Finally, currently available data do not allow us to draw any conclusions about the effects on total mortality in these sex and age groups beyond those that are already published. It would, however, be wrong to conclude that an effect on morbidity has no implications for the potential effects on mortality. The prevention of a morbid event also prevents that individual from graduating into a much higher risk category for subsequent mortality.

In fact, the benefits of LDL-C lowering on morbidity, particularly in older age groups, have been underappreciated. Prevention of morbid events results in lower prevalence of congestive heart failure, angina, significant arrhythmia, and debilitating strokes. Such interventions are likely to have a beneficial effect on both the quality of life for the individual patients and the cost of caring for older subjects imposed on families and society. By placing undue attention on mortality alone as a measure of the success or failure of an intervention, clinicians fail to account for the importance of avoiding such disabilities. It is irrational to deny the demonstrable benefits of LDL-C lowering by sex and age while we await more definitive information about mortality, particularly given the lack of evidence that statin-induced LDL-C lowering in any way increases noncardiovascular morbidity or mortality.¹⁵

In summary, the benefits of LDL-C lowering induced by statins appear to be universal, not defined by sex or age. It is important now to work to extend these benefits to all who are at risk for atherosclerotic cardiovascular disease.