Long-Term Effects of Cholesterol Lowering and Angiotensin-Converting Enzyme Inhibition on Coronary Atherosclerosis
The Simvastatin/Enalapril Coronary Atherosclerosis Trial (SCAT)
Background—This long-term, multicenter, randomized, double-blind, placebo-controlled, 2×2 factorial, angiographic trial evaluated the effects of cholesterol lowering and angiotensin-converting enzyme inhibition on coronary atherosclerosis in normocholesterolemic patients.
Methods and Results—There were a total of 460 patients: 230 received simvastatin and 230, a simvastatin placebo, and 229 received enalapril and 231, an enalapril placebo (some subjects received both drugs and some received a double placebo). Mean baseline measurements were as follows: cholesterol level, 5.20 mmol/L; triglyceride level, 1.82 mmol/L; HDL, 0.99 mmol/L; and LDL, 3.36 mmol/L. Average follow-up was 47.8 months. Changes in quantitative coronary angiographic measures between simvastatin and placebo, respectively, were as follows: mean diameters, −0.07 versus −0.14 mm (P=0.004); minimum diameters, −0.09 versus −0.16 mm (P=0.0001); and percent diameter stenosis, 1.67% versus 3.83% (P=0.0003). These benefits were not observed in patients on enalapril when compared with placebo. No additional benefits were seen in the group receiving both drugs. Simvastatin patients had less need for percutaneous transluminal coronary angioplasty (8 versus 21 events; P=0.020), and fewer enalapril patients experienced the combined end point of death/myocardial infarction/stroke (16 versus 30; P=0.043) than their respective placebo patients.
Conclusions—This trial extends the observation of the beneficial angiographic effects of lipid-lowering therapy to normocholesterolemic patients. The implications of the neutral angiographic effects of angiotensin-converting enzyme inhibition are uncertain, but they deserve further investigation in light of the positive clinical benefits suggested here and seen elsewhere.
The clearly demonstrated benefits of cholesterol-lowering therapy in reducing coronary atherosclerotic disease (CAD)1 2 3 are thought to be related, at least in part, to the slowed progression or enhanced regression of CAD.4 5 6 7 8 Although individuals with elevated cholesterol levels are at a high risk for CAD, most CAD patients have cholesterol levels that are average or “normal” for the population at large.9 10 It remains unclear whether, and to what extent, cholesterol lowering will affect CAD progression or regression in these patients.11
The biological and pathogenetic complexity of atherosclerotic plaque formation, progression, and rupture suggests that multiple therapies may have cumulative or synergistic effects. For example, various mechanisms12 13 have been proposed to explain the decrease in ischemic events seen with angiotensin-converting enzyme (ACE) inhibition in heart failure patients with or without CAD14 15 or those with preserved left ventricular function.16 Whether the benefits of ACE inhibition are mediated through slowed progression and/or enhanced regression of CAD is not known. Synergism of ACE inhibition and cholesterol lowering therapy is possible, but unproven.
We report the findings of the Simvastatin/Enalapril Coronary Atherosclerosis Trial, a quantitative coronary angiographic study of cholesterol-lowering therapy with simvastatin and ACE inhibition with enalapril, alone or in combination, on CAD progression and regression in normocholesterolemic patients over a period of 3 to 5 years. This was a randomized, double-blind, placebo-controlled, multicenter, 2×2 factorial, clinical trial.
The study design was previously reported in detail.17 Patient eligibility criteria were wider than for previous angiographic trials and included individuals with the following: (1) age ≥21 years and no upper age limit; (2) total serum cholesterol levels between 4.1 and 6.2 mmol/L, HDL cholesterol <2.2 mmol/L, and triglycerides <4 mmol/L and lower than total cholesterol; (3) angiographically detectable coronary atherosclerosis in ≥3 major coronary artery segments; and (4) left ventricular ejection fraction >35%. Patients were not enrolled within 6 months of coronary angioplasty or bypass surgery. Exclusion criteria were the following: clear indications for or contraindications to study drugs, clinical instability, imminent need for intervention, other significant cardiac or systemic diseases, potential noncompliance, and inability to give informed consent.17 Each participating center’s Research Ethics Board approved the protocol.
Quantitative Coronary Angiography
Quantitative coronary angiography (QCA) was obtained at baseline within 1 week before randomization and at closeout 3 to 5 years later, except for a few patients whose symptoms led to a diagnostic QCA, which duplicated baseline procedural conditions, before scheduled closeout. To duplicate baseline conditions, the enalapril/placebo was stopped ≥1 week before closeout QCA. Sublingual isosorbide dinitrate (5 to 10 mg; 10 minutes before angiography) or nitroglycerin (100 to 300 μg IV just before angiography of each vessel) was given to minimize variations in coronary tone between angiograms. Baseline angiography parameters (distances, angles, and magnification) were reproduced during subsequent and closeout angiograms. Nonionic contrast media were used to reduce variability from contrast-related vasodilation. Multiple angiographic projections of coronary arteries were obtained. Coronary segments were assigned a number based on the published American Heart Association classification.18
QCA analysis used the MEDIS Cardiovascular Measurement System (Medical Imaging Systems by Lijsterbeslaan), which was developed by Reiber et al.19 Validation studies in our QCA Laboratory have yielded interobserver, intraobserver, and intrastudy variation data17 that are consistent with published reports.19 All analyzable coronary segments in the enrollment and closeout studies were included. QCA measures included absolute mean segment lumen diameter, absolute minimum segment lumen diameter, and maximum percent lumen diameter stenosis. The best views and frames were selected from the initial study. Identical views and frames were analyzed in closeout angiograms, and changes in QCA measures were averaged for each patient.
Study End Points
Study end points were QCA measures and prespecified clinical events (death, myocardial infarction, stroke, hospitalization for angina, revascularization, and cancer), although the latter were not powered to detect conclusive differences. Primary angiographic end points were average per-patient changes between baseline and closeout angiograms in mean and minimum absolute diameters and in maximum percent diameter stenosis of all analyzed coronary segments. Patients were categorized as progressors or regressors using the following currently accepted criteria: decreases or increases, respectively, in absolute mean or absolute minimum diameter by ≥0.4 mm (or by an absolute 15% in diameter stenosis) in ≥1 segment while the other segments remained unchanged. Mixed progressors/regressors had ≥ 1 segment with progression plus ≥1 segment with regression. Individuals without changes fulfilling these criteria were classified as “no change.”
Randomization and Follow-up
During a 1-month, single-blind, placebo run-in phase and throughout the trial, patients were instructed to follow the National Cholesterol Education Program–Adult Treatment Panel (NCEP-ATP) step I diet and, when necessary, the step II diet. Randomization followed baseline QCA and fasting lipid levels obtained at the end of the run-in phase. At each follow-up visit (1, 2, 3, 4.5, 6, 9, and 12 months and then every 6 months until closeout), participants underwent a physical examination and blood tests, including fasting lipid levels. Other lipid-lowering medications and ACE inhibitors were prohibited. Patients were encouraged to take acetylsalicylic acid (aspirin) unless contraindicated. Other therapeutic decisions were made by the patients’ own physicians.
With the 1994 publication of the Scandinavian Simvastatin Survival Study (4S),1 we decided it was unethical to keep patients with cholesterol levels persistently >5.5 mmol/L on placebo. With approval of the Data Safety and Advisory Board and the Research Ethics Boards, the protocol was modified to permit the identification of these patients and to reallocate them to active simvastatin, instead of placebo, in a double-blind fashion.
Starting medication doses were simvastatin/placebo 10 mg daily and enalapril/placebo 2.5 mg twice daily. Automatic upward dose titration (done independently for each drug) occurred during the first 3 monthly visits until maximum doses (simvastatin 40 mg daily and enalapril 10 mg twice daily) or, if side effects occurred, maximally tolerated doses were achieved and maintained. These doses could be decreased or discontinued for severe adverse effects.
The sample size of 460 patients, allowing for 15% dropout, had ≥95% power (with a 2-tailed P<0.05) for detecting differences in primary angiographic end points between treatment and control groups. Primary analyses compared the treatment effects of simvastatin with its placebo and those of enalapril with its placebo. Secondary analyses compared the effects in patients receiving simvastatin plus enalapril, simvastatin alone, enalapril alone, and double placebo, as well as in other specified subgroups. Analyses followed the intention-to-treat principle.
Comparability of treatment groups with respect to baseline characteristics was assessed using Student’s t tests for continuous variables and χ2 tests for categorical variables. Random effects regression models20 were used to evaluate the effect of treatment on longitudinal changes in lipid and blood pressure levels after adjusting for baseline levels. Angiographic outcomes were compared across treatment groups and subgroups using ANOVA methods. Group differences in the distribution of patients classified as progressors, regressors, mixed progressors/regressors, or no change were determined using χ2 tests. Because of the small number of events, significance levels for comparisons of clinical end points were obtained by Fisher’s exact test. Data on continuous variables were expressed as mean±1SD. Statistical significance was set at P<0.05 (2-tailed).
Screening >16 500 charts and 4000 coronary angiograms led to the enrollment of 460 patients between June 1991 and July 1995 in 4 Canadian centers. One third of patients entering the run-in phase were not randomized. This may have reduced subsequent dropouts. Patients had closeout angiograms on a first-in, first-out basis between July 1996 and July 31, 1998, when the first and last patients had been followed for 5 and 3 years, respectively.
Baseline characteristics are summarized in Tables 1⇓ and 2⇓. The average daily simvastatin dose was 28.5±13.0 mg and that of its placebo, 32.2±11.6 mg; for enalapril, the average daily dose was 7.4±3.3 mg twice a day and that for its placebo, 8.3±2.9 mg twice a day. Average compliance for both drugs and placebos, assessed by pill counts at each visit, was ≈95% throughout the trial.
Average follow-up was 47.8 months. Closeout angiograms were not obtained in 19 patients who died and in 47 others because of refusal or intercurrent disease; 394 patients had paired angiograms for analysis. Baseline characteristics and responses to the treatments were not different between these 394 patients and the overall study population.
Lipid Levels and Blood Pressure
No differences existed in baseline lipid levels between simvastatin and placebo groups. Substantial and highly significant changes in lipid levels were found in patients receiving simvastatin but not placebo. Baseline systolic and diastolic blood pressures did not differ between enalapril and placebo patients. Systolic and diastolic blood pressures decreased significantly in patients on enalapril but not placebo (Table 2⇑).
Angiographic End Points
Effects of Simvastatin
Of the 394 patients with paired angiograms, 194 (2117 segments) were on simvastatin, and 200 patients (2101 segments) were on placebo. No baseline differences existed in mean (2.75 versus 2.72 mm) and minimum (2.03 versus 2.01 mm) absolute diameters or percent diameter stenosis (28.5% versus 27.9%) between simvastatin and placebo patients. During treatment, the average per-patient mean absolute diameter decreased by 0.07±0.20 mm with simvastatin and by 0.14±0.25 mm with placebo (P=0.004). The respective average decreases in minimum absolute diameters were 0.09±0.17 mm and 0.16±0.20 mm (P=0.0001). The average increase in maximum percent diameter stenosis with simvastatin was 1.67±5.01% and with placebo, 3.83±6.58% (P=0.0003) (Figure 1⇓). When the changes were examined by prespecified subgroups (men, women, age<65 years, age ≥65 years, smoking status, diabetics, nondiabetics, hypertensives, normotensives, and degree of baseline lesion severity [≥50%, <50%], significant differences (or strong trends) consistently existed in slowing of disease progression in favor of patients on simvastatin versus placebo. No relationship existed between angiographic changes and baseline lipid levels, but a clear relationship existed between angiographic changes and changes in lipid levels during the study. Simvastatin treatment resulted in fewer progressors, more regressors, and more patients with no changes, compared with placebo, for all 3 QCA end points (Figure 2⇓). Total occlusion of ≥1 coronary artery occurred in 16 simvastatin and 15 placebo patients.
Effects of Enalapril
Paired angiograms in 199 enalapril patients (2146 segments) and 195 placebo patients (2072 segments) showed no differences in QCA end points (Figure 1⇑).
Effects of Combination Therapy
Figure 3⇓ compares QCA end points in the 4 treatment subgroups: both simvastatin and enalapril (n=112), simvastatin plus enalapril placebo (n=118), enalapril plus simvastatin placebo (n=117), and double placebo (n=113). The effects of simvastatin plus enalapril did not differ from those of simvastatin alone, and the effects of enalapril and double placebo were not different.
Clinical End Points
As summarized in Table 3⇓, no differences existed between simvastatin and placebo or enalapril and placebo in all-cause mortality, in cardiovascular events (myocardial infarction, stroke, or hospitalization for angina), or incidence of cancer. Fewer revascularization procedures (6% versus 12%, P=0.021) and angioplasties (3% versus 9%, P=0.020) were required in simvastatin patients compared with placebo. Compared with placebo, enalapril was associated with a decrease in the combined end point of death/myocardial infarction/stroke (7% versus 13%; P=0.043) (Table 3⇓). Clinical events did not differ significantly among patients treated with simvastatin and enalapril, simvastatin alone, enalapril alone, and double placebo (Table 4⇓).
Both agents were well tolerated. Serial biochemical monitoring showed no differences in the frequency of elevated creatine kinase and liver enzyme abnormalities between patients on simvastatin and those on placebo and no differences in the frequency of elevated serum potassium and creatinine levels between patients on enalapril and those on placebo.
Important findings from this trial are that long-term lipid-lowering therapy with simvastatin resulted in significant slowing of CAD progression in normocholesterolemic patients that was independent of baseline lipid levels but related to changes in lipid levels during treatment. Long-term ACE inhibition with enalapril had a neutral effect on CAD progression as determined by QCA. Adding enalapril to simvastatin did not result in an incremental angiographic effect.
Simvastatin was associated with less revascularization and enalapril with fewer events (combined end point of death/myocardial infarction/stroke). Patients receiving both drugs had no fewer clinical events than those on either drug alone or on placebo, although events were so few that conclusions are subject to chance. The effects of individual drugs on clinical events may be due to chance, but they are consistent with other large trials of lipid-lowering therapy and ACE inhibition.1 2 3 14 15 16
The impact of simvastatin on CAD progression in normocholesterolemic patients was similar to that in earlier studies which enrolled patients with higher cholesterol levels.4 5 6 7 8 The magnitude of changes in lumen diameters may be too small to be clinically relevant.21 However, this may be because the changes in lumen diameters reported in this and other studies are, for the purpose of group comparisons, “statistical measures” derived from averaging effects observed in a large number of segments instead of “actual measures” in one individual. On a per-patient basis, lipid lowering in this group of normocholesterolemic CAD patients resulted in more patients having regression or no change in their CAD and fewer having progression. The magnitude of these changes on the arterial wall is consistent with the proposed mechanism of plaque shrinkage due to reduction in and stabilization of the lipid-rich plaque.21
The smaller sample size and shorter follow-up of normocholesterolemic patients in the Harvard Atherosclerosis Reversibility Project (HARP) trial may explain the reported neutral effect on angiographic end points, unlike our conclusively positive results. This observation was supported by the findings of the Multicentre Anti-Atheroma Study (MAAS), which examined follow-up data at 2 and 4 years. Benefits became clear and significant at 4 years.7
It had been anticipated a priori that ACE inhibition would have a beneficial effect on CAD and that adding ACE inhibition to cholesterol lowering would have a synergistic effect. Our angiographic results failed to substantiate these expectations. Reasons for these outcomes are unclear. One important difference between this trial and others that have shown the angiographic benefits of ACE inhibition is the different procedural conditions. For example, the Trial on Reversing Endothelial Dysfunction (TREND) study, which examined coronary vasomotor dysfunction using acetylcholine-provoked constriction of target segments, avoided vasoactive drugs for ≥12 hours before the procedure.22 Because we were interested in changes in anatomic atherosclerosis, we gave nitrates before and during the procedure to minimize variability in coronary lumen diameter due to vasomotor tone. This pretreatment likely abolished any differences in lumen diameters mediated by differing vasomotor tone between treatment groups. Preliminary data from a trial of ACE inhibition suggested benefits when ultrasound measurement of the carotid artery intima-media thickness was used. Changes in ultrasonic measures of wall thickness, detected with intravascular ultrasound, would precede QCA lumen diameter changes. In the absence of intravascular ultrasound data, it cannot be concluded that enalapril had no effect.
If, as has been suggested, aspirin has a large negative interaction with enalapril, concomitant aspirin use might confound the results and negate the effects of ACE inhibition in this study. Secondary analysis of data from heart failure trials suggests that such an effect is possible, although its magnitude is probably small. This effect has not been consistently reported.23 24 The Heart Outcomes Prevention Evaluation (HOPE) trial reported clear clinical benefits with ACE inhibition in the same type of patients, most of whom were on aspirin,16 as those enrolled in the Simvastatin/Enalapril Coronary Atherosclerosis Trial.
Potential mechanisms of the benefit of ACE inhibition include normalization of endothelial dysfunction and plaque formation and stabilization.12 13 These effects, which are not easily detected by QCA analysis, may have been operative in large trials demonstrating clinical benefits.14 15 16
The angiographic and clinical results from this long-term lipid-lowering trial confirm the beneficial effects of therapy and extend the observation of positive angiographic effects to normocholesterolemic CAD patients. In other words, these results support the concept that nearly all CAD patients may benefit from treatment with simvastatin.
The implications of the neutral angiographic effects of ACE inhibition are uncertain, but they deserve further investigation in light of the positive clinical benefits suggested here and seen elsewhere. Although no definitive conclusions can be drawn from the clinical effects of combination therapy, this important issue may be further resolved with the pending results of other large randomized trials.
The late Dr N.J. Davies contributed substantially to the planning and implementation of this study. The committed participation of the patients in this study and the cooperation of their physicians are gratefully acknowledged. We gratefully acknowledge the financial and in-kind support of the Medical Research Council of Canada, Merck Frosst Canada & Co, the Alberta Heritage Foundation for Medical Research, University of Alberta Hospitals, and Safeway Canada.
The principal investigator, Dr Teo, has received unrestricted grants from Merck Frosst Canada & Co, as part of the Medical Research Council of Canada University-Industry Program, in order to carry out this study. Dr. Montague was a Co-Principal Investigator and contributor to the study while he was head of Cardiology at the University of Alberta; he is now an employee of Merck Frosst Canada & Co.
A complete list of all SCAT Investigators and contributors has been published previously.17
- Received March 30, 2000.
- Revision received May 11, 2000.
- Accepted May 11, 2000.
- Copyright © 2000 by American Heart Association
Waters D, Higginson L, Gladstone P, et al, for the CCAIT Study Group. Effects of monotherapy with an HMG-CoA reductase inhibitor on the progression of coronary atherosclerosis as assessed by serial quantitative arteriography: the Canadian Coronary Atherosclerosis Intervention Trial. Circulation. 1994;89:959–968.
Lonn EM, Yusuf S, Jha P, et al. Emerging role of angiotensin-converting enzyme inhibitors in cardiac and vascular protection. Circulation. 1994;90:2056–2069.
Dzau VJ. Mechanism of protective effects of ACE inhibition on coronary artery disease. Eur Heart J. 1998;19(suppl J):J2–J6.
Teo KK, Burton JR, Buller C, et al, on behalf of the SCAT Investigators. Rationale and design features of a clinical trial examining the effects of cholesterol lowering and angiotensin-converting enzyme inhibition on coronary atherosclerosis: Simvastatin/Enalapril Coronary Atherosclerosis Trial (SCAT). Can J Cardiol. 1997;13:591–599.
Reiber JHC, Serruys PW, Kooijman CJ, et al. Assessment of short-, medium- and long-term variations in arterial dimensions from computer-assisted quantitation of coronary cineangiograms. Circulation. 1985;71:280–288.
Laird NM, Ware JH. Random effects models for longitudinal data. Biometrics. 1984;38:963–974.
Brown BG, Zhao XQ, Sacco DE, et al. Lipid lowering and plaque regression: new insights into prevention of plaque disruption and clinical events in coronary disease. Circulation. 1993;87:1781–1791.
Mancini GBJ, Henry GC, Macaya C, et al. Angiotensin-converting enzyme inhibition with quinapril improves endothelial vasomotor dysfunction in patients with coronary artery disease: the TREND (Trial on Reversing Endothelial Dysfunction) study. Circulation. 1996;94:258–265.
Pitt B, Yusuf S, for the SOLVD Investigators. Studies of left ventricular dysfunction (SOLVD) sugbroup results. J Am Coll Cardiol. 1992;19:215A.