Review of the Atorvastatin versus Revascularization Treatment Investigators Trial
Atorvastatin vs angioplasty in stable coronary artery disease
Background: Percutaneous transluminal coronary angioplasty (PTCA) was widely used in the 1990s for its potential benefit in improving symptoms in patients with stable coronary artery disease, as discussed in the ACME and RITA-2 trials. However, the RITA-2 trial showed that PTCA worsened hard outcomes, and ACME was underpowered for such comparison. At the same time, trials demonstrating the ability of statin drugs to reduce cardiovascular events and improve survival had led to an increase in use of the lipid-lowering drugs.
The Atorvastatin versus Revascularization Treatment Investigators sought to test the unique hypothesis that atorvastatin is superior to angioplasty in reducing ischemic events in patients with stable coronary artery disease.
*It is important to note that the historical trials of CABG versus medical therapy that led to the widespread adoption of coronary revascularization, in general, were completed prior to the adoption of statin therapy. Thus a weakness of the CABG literature, which we have addressed in detail in prior posts and podcasts, is that the efficacy of CABG is inconsistent at best and medical therapy, at the time, was very limited. This makes this study of atorvastatin versus angioplasty of particular interest.
Patients: Eligible patients had stenosis of 50% or more in one or two coronary arteries and had been recommended for treatment with percutaneous revascularization. The patients were asymptomatic or had Canadian Cardiovascular Society (CCS) class I or II angina. LDL levels had to be least 115 mg/dL and triglyceride levels of 500 mg/dL or less.
Patients were excluded if they had left main coronary artery disease, 3-vessel disease, unstable angina or myocardial infarction within the previous 2 weeks, or if the ejection fraction was <40%.
Baseline characteristics: The trial randomized 341 patients – 177 randomized to the PTCA arm and 164 to the atorvastatin arm.
The average age of patients was 59 years and 84% were men. Approximately 45% had hypertension, 16% had diabetes, 42% had prior myocardial infarction and 22% were current smokers. The average ejection fraction was 61%.
At the time of enrollment, 22% were taking lipid lowering drugs and 21% were taking aspirin or other anticoagulants.
Approximately 57% had single vessel disease, while the rest had 2-vessel disease. More patients in the atorvastatin arm had their target lesion in the left anterior descending artery (43% vs 30%).
Procedures: Patients were randomized to coronary angioplasty versus atorvastatin 80mg PO daily. Randomization was stratified based whether patients have single or 2-vessel disease. Patients assigned to the atorvastatin arm stopped other lipid lowering medications. Patients assigned to the angioplasty arm received the usual care at the time, which could include a lipid lowering drug.
Patients were followed up for 18 months.
Endpoints: The primary outcome was an ischemic event defined as at least one of the following: Death from cardiac causes, resuscitation after cardiac arrest, nonfatal myocardial infarction, cerebrovascular accident, coronary-artery bypass grafting, angioplasty, and worsening angina with objective evidence of ischemia resulting in hospitalization. Secondary endpoints included angina based on the Canadian Cardiovascular Society classification and anti-anginal drugs use.
Ischemic events were adjudicated by a committee blinded to treatment assignment. However, this was not the case for the endpoint of angina.
Analysis was performed based on the intention-to-treat principle. The authors reported that the sample size was planned to achieve 85% power at an alpha level of 0.05. This was based on an assumed event rate for the primary outcome over 18 months of 20% in the atorvastatin arm and 35% in the angioplasty arm, this represents a relative difference of approximately 43%. Stated differently, the trial had 85% power to detect a 43% reduction in events for the atorvastatin group.*
The authors performed 2 interim analyses. Consequently, the significance level for the final analysis of the ischemic events endpoint was reduced from 0.05 to 0.045.
Results: No patients were lost to follow up at the 18 months mark. One patient (0.6%) in the atorvastatin arm did not receive atorvastatin and 11 (6%) patients in the angioplasty arm did not undergo the procedure. The number of treated lesions in the angioplasty arm was 213, with stenting used in 64 of them. At the end of the study, 69% of the patients in the angioplasty arm were using a lipid lowering drug and 93% of patients in the atorvastatin arm were still using it. The reduction in LDL levels at the end of study compared to baseline was 18% in the angioplasty arm and 46% in the atorvastatin arm.
There were less ischemic events with atorvastatin (13.4% vs 20.9%; p= 0.048) - this difference did not reach the level of significance as adjusted for the interim analyses; p=0.045. The reduction in events was driven by less coronary artery bypass surgery with atorvastatin (1.2% vs 5.1%) and less worsening angina with objective evidence of ischemia resulting in hospitalization (6.7% vs 14.1%). Death from cardiac causes was unchanged (0.6% in both treatment groups). Non-fatal myocardial infarctions were not significantly different between both groups (2.4% with atorvastatin vs 2.8% with angioplasty).
More patients in the angioplasty arm experienced reduction in angina (53.7% vs 40.9%; p= 0.009). The use of anti-anginal drugs was similar in both groups at the beginning as well as at the end of the study; approximately 90% at both time points.
Conclusion: In patients with single or 2-vessel stable coronary artery disease, atorvastatin 80 mg daily did not significantly reduce ischemic events compared to an initial strategy of angioplasty.
It may be tempting for readers to interpret these findings as: a strategy of statin therapy is as effective or non-inferior to angioplasty for stable coronary artery disease. Such a framing would be an incorrect interpretation of the trial—in our opinion.
The trial was specifically designed to assess whether atorvastatin was superior to angioplasty, and, strictly speaking, it did not meet the prespecified criteria to claim superiority.
Looking beyond the strict (p-value driven) interpretation of the results, the trial design was very limited. It was a small trial (N=341) that tested for an unrealistically large difference in events (15% absolute risk reduction, 43% relative risk reduction), over a relatively short period, using a composite endpoint consisting of multiple hard and soft outcomes. Furthermore, it’s prespecified analysis overestimated event rates nearly two-fold.
We believe, therefore, that this should be considered a negative trial that yields little actionable information for patient care.
It makes our series because it has historical significance in terms of trying to understand the widespread adoption of angioplasty.
Even at the time it was published, there were serious issues with AVERT. Here's our letter to the editor , published in 1999. I had actually forgotten about this letter entirely, but in retrospect, it seems pretty prescient (considering that this was published 8 years before COURAGE and 20 years before ISCHEMIA)...
Aggressive Lipid-Lowering Therapy Compared with Angioplasty in Stable Coronary Artery Disease
Published December 9, 1999
N Engl J Med 1999;341:1853-1855
DOI: 10.1056/NEJM199912093412415
To the Editor:
We would like to call attention to two critical limitations of the Atorvastatin versus Revascularization Treatment (AVERT) study (July 8 issue).1 First, we take exception to the authors' conclusion that “aggressive lowering of LDL [low-density lipoprotein] cholesterol levels . . . is at least as effective as angioplasty . . . in reducing the incidence of ischemic events in low-risk patients.” This statement implies that angioplasty is an effective means of reducing ischemic events in this target population. In fact, the results of previous controlled clinical trials suggest exactly the opposite.2,3 In the Angioplasty Compared to Medicine (ACME) trial, the six-month incidence of death, myocardial infarction, or repeated revascularization was 21 percent in the group assigned to percutaneous transluminal coronary angioplasty (PTCA) as compared with 14 percent in the medical-therapy group (relative risk, 1.5; 95 percent confidence interval, 0.9 to 2.1). Similarly, in the second Randomized Intervention Treatment of Angina (RITA-2) trial, the three-year cumulative incidence of death or nonfatal myocardial infarction was 6.3 percent in the PTCA group as compared with 3.3 percent in the medical-therapy group (relative risk, 1.9; 95 percent confidence interval, 1.1 to 3.4). The selection of a PTCA-based strategy as the control therapy in patients with minimal (if any) myocardial ischemia would thus seem to be a straw man that inherently biased the study toward a positive result for medical therapy, with or without aggressive lipid lowering.
Furthermore, the authors claim that restenosis was responsible for only a “small percentage of the events” between 6 and 18 months in the angioplasty group. Close inspection of the Kaplan–Meier curves (Figure 2 of the article), however, indicates that the main divergence in the curves occurred between six and eight months, an interval that is entirely compatible with the time course of clinical restenosis — particularly after coronary stenting. In fact, in the Stent Anticoagulation Restenosis Study,4 50 percent of the ischemic events due to restenosis occurred between 6 and 12 months after stenting. Thus, considering the known time course of coronary restenosis and in the absence of published data from an independent, core angiographic laboratory, attribution of the excess of ischemic events in the PTCA group to anything but restenosis seems highly implausible.
In summary, catheter-based percutaneous coronary intervention is a safe and effective means of controlling symptoms of ischemia in patients with one- or two-vessel disease, and aggressive lipid lowering is an effective means of stabilizing and reducing the progression of plaque. The fact that PTCA had little benefit in the AVERT study relates mainly to the selection of a patient population that had very little to gain from revascularization rather than to any specific benefit of lipid-lowering treatment. These therapies should thus be seen as additive and complementary5 rather than as mutually exclusive approaches, as the authors suggest.
David J. Cohen, M.D.
Joseph P. Carrozza, M.D.
Donald S. Baim, M.D.
Beth Israel Deaconess Medical Center, Boston, MA 02215
5 years ago had heart attack at an airport. Cardiac arrest during transport. CPR and defib. ROSC on arrival in ER. Quickly to Cath Lab. 2 stents. On Atorvastatin 80mg since. As a layman, I have an interest in following topics related to my specific health and I appreciate your review of this study.