Author + information
- Received October 26, 1998
- Revision received February 2, 1999
- Accepted March 15, 1999
- Published online July 1, 1999.
- Greg C. Flaker, MD, FACC∗,* (, )
- J.Wayne Warnica, MD, FACC†,
- Frank M. Sacks, MD‡,
- Lemuel A. Moyé, MD, PhD§,
- Barry R. Davis, MD, PhD§,
- Jean L. Rouleau, MD∥,
- Richard R. Webel, MD∗,
- Marc A. Pfeffer, MD, PhD‡,
- Eugene Braunwald, MD‡,
- for the Cholesterol and Recurrent Events CARE Investigators∗∗
- ↵*Reprint requests and correspondence: Dr. Greg Flaker, University of Missouri Hospital and Clinics, Department of Cardiology, Room MC-501, One Hospital Drive, Columbia, Missouri 65212
This analysis was carried out to determine if revascularized patients derive benefit from the 3-hydroxy-3 methylglutaryl coenzyme A (HMG-CoA) reductase inhibitor pravastatin.
The HMG-CoA reductase inhibitors result in substantial reductions in serum cholesterol and stabilization of atherosclerotic plaques in patients with coronary artery disease.
Pravastatin was found to reduce clinical cardiovascular events in the Cholesterol and Recurrent Events (CARE) trial consisting of 4,159 patients with a documented myocardial infarction and an average cholesterol level (mean 209 mg/dl and all <240 mg/dl). A total of 2,245 patients underwent coronary revascularization before randomization including 1,154 patients with percutaneous transluminal coronary angioplasty (PTCA) alone, 876 patients with coronary artery bypass graft (CABG) alone, and 215 patients with both procedures. Clinical events in revascularized patients were compared between patients on placebo and on pravastatin.
In the 2,245 patients who had undergone revascularization, the primary endpoint of coronary heart disease death or nonfatal myocardial infarction (MI) was reduced by 4.1% with pravastatin (relative risk [RR] reduction 36%, 95% confidence interval [CI] 17 to 51, p = 0.001). Fatal or nonfatal MI was reduced by 3.3% (RR reduction 39%, 95% CI 16 to 55, p = 0.002), postrandomization repeat revascularization was reduced by 2.6% (RR reduction 18%, 95% CI 1 to 33, p = 0.068) and stroke was reduced by 1.5% (RR reduction 39%, 95% CI 3 to 62, p = 0.037) with pravastatin. Pravastatin was beneficial in both the 1,154 PTCA patients and in the 1,091 CABG patients who had undergone revascularization before randomization.
Pravastatin reduced clinical events in revascularized postinfarction patients with average cholesterol levels. This therapy was well tolerated and its use should be considered in most patients following coronary revascularization.
Percutaneous cardiac interventions and bypass surgery play a central role in the management of patients with ischemic heart disease. In the U.S. in 1995, approximately 410,000 percutaneous coronary revascularization procedures and 590,000 coronary bypass procedures were performed (1). Although these procedures frequently relieve symptoms, they prolong life in only a fraction of patients. Progression of atherosclerosis may lead to recurrent angina, myocardial infarction (MI), and death. Concurrent strategies to stabilize or reverse the atherosclerotic process are needed to improve long-term results.
The 3-hydroxy-3methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors cause substantial reductions in serum total and low density lipoprotein cholesterol (LDL-C) and result in arrest of progression or regression of atherosclerotic lesions in up to 77% of hypercholesterolemic patients with coronary artery disease over a two- to four-year follow-up period (2–5). It is important to note that these agents have been shown to improve clinical outcomes in patients with coronary artery disease with average cholesterol values (6)and in patients without established coronary artery disease with elevated (7)or borderline elevated LDL-C (8). Little long-term, controlled information is available on the effects of these agents on cardiovascular events in patients who have had revascularization procedures, particularly in patients with average serum cholesterol values.
The Cholesterol and Recurrent Events (CARE) trial offered a unique opportunity to evaluate the effects of the HMG-CoA reductase inhibitor pravastatin in post-MI patients who had undergone coronary revascularization before revascularization and who had a total cholesterol <240 mg/dl (6.2 mmol/liter) and an LDL-C of 115 to 174 mg/dl (3.0 to 4.5 mmol/liter). In this analysis, we compared the five-year follow-up of 2,245 revascularized patients in the CARE study who were randomized to either pravastatin 40 mg/day or placebo.
The design of the CARE trial (9), its baseline findings (10)and final results (6)have been described previously. The CARE was a multicenter, randomized, double blind, placebo-controlled trial designed to assess the efficacy of the HMG-CoA reductase inhibitor pravastatin in reducing fatal coronary heart disease (CHD) and nonfatal MI. Patients who survived an MI (3 to 20 months before randomization), had a plasma total cholesterol <240 mg/dl (6.2 mmol/liter), and LDL-C of 115 to 174 mg/dl (3.0 to 4.5 mmol/liter) were eligible for the trial. From December 1989 through December 1991, a total of 4,159 men and women between 21 and 75 years of age were randomized to the study from 80 centers in North America. Patients were randomized either to placebo or pravastatin 40 mg/day, designed to achieve an average reduction in LDL-C of approximately 30%, and an increase in high density lipoprotein cholesterol (HDL-C) of 5%. The median duration of follow-up was five years. Patients in the CARE study were not enrolled for at least six months after percutaneous transluminal coronary angioplasty (PTCA) and three months after coronary artery bypass graft (CABG) to ensure clinical stability. This analysis evaluates all patients who underwent coronary revascularization before randomization to pravastatin or placebo. All clinical cardiovascular end points were validated by hospital records. Coronary death, nonfatal MI and stroke were evaluated by committees blinded to treatment assignment.
All analyses were performed on an intention-to-treat basis, and p values were two-sided. The comparibility of baseline characteristics in the two treatment groups was ascertained using standard normal tests for continuous variables and chi-square tests for categorical variables. All hypothesis testing and all risk reductions with their confidence intervals were assessed using the Cox proportional hazards model. Kaplan-Meier survival curves for the control and pravastatin groups were calculated (11).
The relationship between treatment assignment and the primary end point in CARE was assessed using a Cox proportional hazards model using time to the primary end point as the dependent variable. Patients who had a primary end point were uncensored and removed from the analysis at the time of the occurrence of a primary end point. Patients without a primary end point were censored at the time of their last follow-up visit.
The influence of on-treatment LDL-C was investigated in time-dependent Cox analyses with a weighted average of the patient’s follow-up LDL-C values. The primary end point of the study was cardiovascular death or nonfatal MI. Multivariate models adjusted for the baseline variables of age, gender, smoking history, diabetes, hypertension and left ventricular ejection fraction.
A total of 2,245 patients had undergone coronary revascularization before randomization in the CARE trial. Included in this group were 1,154 patients who had undergone PTCA alone without a prior CABG. A total of 876 patients had undergone CABG without PTCA before randomization; an additional 215 patients had both PTCA and CABG before randomization, for a total of 1,091 patients with CABG. In 223 patients, PTCA occurred before the index MI; in 220 patients CABG took place before the index MI. In the remaining patients, revascularization was performed between the index MI and randomization.
At the time of randomization, there were differences between the 2,245 revascularized patients and the 1,914 nonrevascularized patients. Revascularized patients were more often male and white, and a higher fraction lived in the U.S. than in Canada. Revascularized patients were less often current smokers. They were less likely to have had a Q wave MI, had a lower peak creatine phosphokinase (CPK), had more often received thrombolysis, had a higher left ventricular ejection fraction and a lower Canadian Cardiovascular Classification compared with nonrevascularized patients. Differences in medications were also noted between revascularized and nonrevascularized patients (Table 1).
Differences also existed between the baseline characteristics of patients who had previously received PTCA exclusively and those who received CABG without and with PTCA (Table 1). Those patients who received PTCA only were younger, had a lower incidence of hypertension and diabetes but were more frequently current smokers compared with CABG patients. The PTCA-only patients had a lower incidence of congestive heart failure (CHF) and stroke, a higher peak CPK during the index MI but had a higher left ventricular ejection fraction. They had more frequently received thrombolysis, had a slightly higher HDL-C, and had a lower Canadian Cardiovascular Classification compared with CABG patients. Differences in medications received upon entry into the CARE study were also noted between PTCA and CABG patients (Table 1). There were no important differences in characteristics of subgroups based on treatment assignment.
During the follow-up period, the mean total cholesterol, LDL-C, HDL-C, and triglyceride values did not change significantly in the revascularized patients treated with placebo. The average LDL-C in placebo-treated PTCA patients was 136 ± 18 mg/dl, and the average total cholesterol was 211 ± 21 mg/dl. The average LDL-C in placebo-treated CABG patients was 138 ± 19 mg/dl, and the average total cholesterol was 212 ± 22 mg/dl. Pravastatin resulted in marked reductions of LDL-C and total cholesterol in both PTCA and CABG patients. The average LDL-C in PTCA patients treated with pravastatin was 98 ± 18 mg/dl, and the average total cholesterol was 170 ± 21 mg/dl. The average LDL-C in CABG patients treated with pravastatin was 98 ± 20 mg/dl, and the average total cholesterol was 171 ± 26 mg/dl.
Among the 2,245 patients who had undergone coronary revascularization before randomization, those assigned to pravastatin experienced fewer primary end points (CHD death or nonfatal MI), fewer MIs (fatal and nonfatal) and fewer strokes (Table 2). The absolute reduction for CHD death or nonfatal MI, the prespecified primary end point of the CARE trial, was 4.1% (relative risk [RR] reduction 36%, 95% confidence interval [CI] 17 to 51, p = 0.001). For recurrent MI (fatal or nonfatal) the absolute risk reduction was 3.3% (RR reduction 39%, 95% CI 16 to 55, p = 0.002), for stroke it was 1.5% (RR reduction 39%, 95% CI 3 to 62, p = 0.037), and for repeat revascularization it was 2.6% (RR reduction 18%, 95% CI 1 to 33%, p = 0.068).
The beneficial effects of pravastatin were striking among the 1,154 PTCA patients as well as the 1,091 CABG patients who had undergone revascularization before randomization. For PTCA patients, CHD death or nonfatal MI was reduced by 4.5% (RR reduction 39%, 95% CI 10 to 58%, p = 0.011, Fig. 1). For recurrent MI (fatal or nonfatal) the absolute risk reduction was 4.0% (RR reduction 43%, 95% CI 13 to 63%, p = 0.009, Fig. 2). The need for postrandomization revascularization was reduced with pravastatin in PTCA patients by 4.8% (RR reduction 22%, 95% CI 0 to 40%, p = 0.05, Fig. 3). The absolute risk reduction for stroke was 2.6% (RR reduction 72%, 95% CI 31 to 89%, p = 0.006, Fig. 4). Death from all causes occurred in 5.8% of the placebo-treated patients and 6.7% of the pravastatin patients; CHD death occurred in 2.9% of the placebo patients and 3.2% of the pravastatin-treated patients.
In the 1,091 CABG patients, the primary end point (CHD death or nonfatal MI) was reduced by 3.8% (RR reduction 33%, 95% CI 3 to 53%, p = 0.034, Fig. 5). Death from CHD was reduced by 3.2% (RR reduction 44%, 95% CI 7 to 66, p = 0.024). All-cause mortality was reduced by 4.4% (relative risk reduction 38%, 95% CI 9 to 58, p = 0.014). There were trends for reductions in total MI (fatal and nonfatal), need for PTCA, and stroke and a trend toward an increase in repeat CABG in these CABG patients.
The 1,914 patients who had not been revascularized experienced reductions in clinical events with pravastatin, but these reductions were not as striking as those who had been revascularized (Table 2). Nonsignificant reductions in CHD death or nonfatal MI (absolute risk reduction 1.6%, p = 0.37), recurrent MI (absolute risk reduction 1.2%, p = 0.39), and stroke (absolute risk reduction 0.7%, p = 0.38) were noted. The need for revascularization was reduced in the pravastatin group by 7.1% (RR reduction 35%, 95% CI 19 to 48%, p < 0.001). The tests for interaction for the effectiveness of pravastatin between the revascularized and nonrevascularized group were not significant.
An on-treatment analysis was performed to determine the most desirable LDL-C values in revascularized patients. Patients who received pravastatin were divided into quintiles based on the average LDL-C from all follow-up values. The primary end point for PTCA patients in the lowest quintile (LDL-C <84 mg/dl) was not significantly different when compared to PTCA patients in the highest quintile (LDL-C >112 mg/dl). Similarly, the primary end point for CABG patients in the lowest quintile (LDL-C <82 mg/dl) was not statistically significant when compared with CABG patients in the highest quintile (>114 mg/dl). The RR for the primary end point was not significantly different for PTCA and CABG patients in all quintiles, even when adjusted for differences in baseline variables; however, the side confidence intervals prevented a definite statistical conclusion.
In summary, pravastatin reduced coronary heart death and nonfatal MI in patients who had undergone revascularization before randomization into CARE. Pravastatin reduced MI, prevented the need for repeat revascularization, and reduced stroke in PTCA-treated patients. In nonrevascularized patients, pravastatin reduced the need for subsequent PTCA or CABG.
The HMG-CoA reductase inhibitors have earned a pivotal place in the management of patients with hypercholesterolemia. Reductions in cardiovascular mortality and MI have been demonstrated in hypercholesterolemic patients with and without established coronary artery disease with some of these agents (6,12). The current analysis demonstrates the importance of this form of therapy in post-MI patients with average cholesterol values and whose coronary artery disease was deemed severe enough to warrant revascularization before enrollment in the trial. Pravastatin was well tolerated in this (6)and other trials (7).
Recent studies have shown that lipid-lowering agents are infrequently used in patients undergoing revascularization following myocardial infarction. In the VANQWISH trial (13)only 17% of patients with a non-Q-wave MI had an elevated cholesterol level. Only 13% of patients in this category received a lipid-lowering agent (13). Given the favorable results of HMG-CoA reductase inhibitor therapy in revascularized patients, this therapy should be considered for almost all patients after PTCA and CABG.
Lipid therapy and PTCA
Previous studies have demonstrated the ineffectiveness of several HMG-CoA reductase inhibitors in preventing restenosis after PTCA (14,15). Following PTCA, injury to the vessel wall is deep and the vessel is expanded (16). A multifactorial inflammatory response occurs, with smooth muscle cells and neointima formation as the predominant cellular response (17,18). Angiographic trials have demonstrated restenosis in 30% to 60% of patients within six months (19,20). Because patients in the CARE study were enrolled at least six months after PTCA, the early problems with restenosis would not be addressed with this analysis. Instead, the reduction in events associated with pravastatin in PTCA patients observed in CARE was not an early but rather a late effect and likely due to alteration in the athersclerotic process. The current analysis is the first demonstration of long-term clinical benefit of cholesterol lowering in patients who have undergone PTCA.
Lipid therapy and CABG
In CABG patients, LDL-C reduction has been shown to reduce atherosclerosis in bypass grafts and native vessels. In a substudy of the Cholesterol Lowering Athersclerosis Study (CLAS), 162 men with an average baseline LDL-C of 170 mg/dl received either colestipol and niacin or placebo. In the group receiving active therapy, the LDL-C value was reduced by 43% and new lesions were reduced in native vessels and bypass grafts (21). In the Post Coronary Artery Bypass Trial (22)patients with elevated LDL-C values (mean 155 mg/dl) were randomized to receive moderately aggressive lipid therapy, designed to reduce LDL-C to a goal of 130 to 140 mg/dl, or aggressive lipid therapy to reach a goal of 60 to 85 mg/dl. Patients in the moderately aggressive arm were treated with lovastatin and had LDL-C values of 132 to 136 mg/dl, similar to the baseline LDL-C value of 138 mg/dl in the current study. Patients in the aggressive arm received more intensive therapy and attained LDL-C values of 93 to 97 mg/dl. These patients were found to have reduced graft atherosclerosis compared with the patients who were treated with moderately aggressive therapy.
In our pravastatin-treated patients, an on-treatment LDL-C of 98 mg/dl is similar to the aggressively treated arm in the post-CABG trial. In contrast to the post-CABG results, CARE demonstrated a reduction in “hard” coronary end points including CHD death and total mortality in CABG patients. The results in CARE were achieved with monotherapy using normal doses of pravastatin, 40 mg. The relationship between on-treatment LDL-C and event rates in CARE has been described elsewhere (23).
The reduction in clinical events with pravastatin was especially striking in revascularized CARE patients. The PTCA patients and CABG patients treated with pravastatin had fewer primary end points (deaths due to coronary heart disease or nonfatal MIs). The PTCA patients in the pravastatin group had significantly fewer MIs, less repeat revascularizations, and fewer strokes. Nonrevascularized patients assigned to pravastatin had significantly fewer revascularization procedures but did not have significantly fewer primary end points or fewer MIs or strokes. The reason for this observation is unclear and may be due to chance because the test for interactions between pravastatin treatment and prior revascularization was not significant. It is also possible that pravastatin is especially effective in more extensive forms of atherosclerosis that are severe enough to warrant some form of revascularization.
In summary, pravastatin has been shown to reduce important clinical events in patients who had previously undergone coronary revascularization. The favorable effects became obvious over an extended period of time. This analysis of revascularized patients was part of the larger CARE cohort in which pravastatin was shown to reduce clinical events in post-MI patients with average cholesterol values. This is a limitation of the study, but until further information becomes available, it appears that HMG-CoA reductase inhibitors should be considered in most patients who have undergone coronary revascularization.
↵∗∗ A complete list of Care Investigators appears in reference .
☆ The Cholesterol and Recurrent Events trial was funded by an investigator-initiated grant from the Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton, New Jersey.
- coronary artery bypass graft
- Cholesterol and Recurrent Events trial
- coronary heart disease
- congestive heart failure
- creatine phosphokinase
- high density lipoprotein cholesterol
- 3-hydroxy-3 methylglutaryl coenzyme A
- low density lipoprotein cholesterol
- myocardial infarction
- percutaneous transluminal coronary angioplasty
- Received October 26, 1998.
- Revision received February 2, 1999.
- Accepted March 15, 1999.
- American College of Cardiology
- ↵American Heart Association; Heart and Stroke Statistical Update, 1998.
- MAAS Investigators
- CCAIT study group,
- Waters D.,
- Higginson L.,
- Gladstone P.,
- et al.
- REGRESS Study Group,
- Jukema J.W.,
- Bruschke A.V.G.,
- vanBoven A.J.,
- et al.
- Kaplan E.L.,
- Meier P.
- Scandinavian Simvastatin Survival Study Group. Randomized trial of cholesterol lowering in 4,444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383–89.
- Veterans Affairs Non-Q-Wave Infarction Strategies in Hospital (VANQWISH) Trial Investigators,
- Boden W.E.,
- O’Rourke R.A.,
- Crawford M.H.,
- et al.
- PREDICT Trial investigators,
- Bertrand M.E.,
- McFadden E.P.,
- Fruchart J.,
- et al.
- Steele P.M.,
- Chesebro J.H.,
- Stanson A.W.,
- et al.
- Ip J.H.,
- Fuster V.,
- Badimon L.,
- Badimon J.,
- Taubman M.B.,
- Chesebro J.H.
- Austin G.E.,
- Ratliff N.B.,
- Hollman J.,
- Tabeil S.,
- Phillips D.
- ↵Holmes DR, Vlietstra R, Smith H, et al. Restenosis after coronary angioplasty (PTCA): a report from the PTCA registry of the National Heart, Lung, and Blood Institute. Am J Cardiol 1984;53:77–81C.
- Serruys P.W.,
- Luijten H.E.,
- Beatt K.J.,
- et al.
- Sacks F.M.,
- Moye L.A.,
- Davis B.R.,
- et al.