Previous studies have demonstrated that benefits of intensive statin therapy compared to standard statin therapy begin shortly after an acute event and are continued up to 2 years of follow-up. However, whether efficacy and safety of intensive statin therapy in patients with a recent cardiac event are maintained in longer-term follow-up has not been evaluated. We conducted a post hoc analysis of a subgroup of 999 patients who had a first acute myocardial infarction (MI) <2 months before randomization in a prospective, open-label, blinded end-point evaluation trial of 8,888 patients with a history of MI that compared intensive statin therapy (atorvastatin 80 mg) to standard statin therapy (simvastatin 20 to 40 mg) over approximately 5 years of follow-up. We analyzed the same composite end point used in the Pravastatin or Atorvastatin Evaluation and Infection Therapy (PROVE IT) trial (death, MI, hospitalization for unstable angina, revascularization, and stroke). Rates of the composite end point were 44.7% (n = 226) in the simvastatin group and 37.9% (n = 187) in the atorvastatin group (hazard ratio 0.82, 95% confidence interval 0.67 to 0.99, p = 0.04). Although statistical power was smaller than that of the PROVE IT trial, the relative risk decrease observed at 5 years is consistent with that in the 2-year follow-up in PROVE IT. The 2 treatment regimens were well tolerated. In conclusion, our analysis provides support for the strategy of placing patients with recent MI on intensive statin therapy and maintaining the high dose over the long term, beyond 2 years.
Recent guidelines have advocated early initiation of intensive lipid lowering in patients with acute coronary syndromes based on results from randomized clinical trials, but the short-term benefits of this strategy are still under debate and long-term data are scarce. The Incremental Decrease in End Points through Aggressive Lipid Lowering (IDEAL) trial was a prospective, randomized, open-label, blinded end-point trial that compared the effect on outcome of therapy with atorvastatin 80 mg to that of simvastatin 20 to 40 mg in 8,888 patients with a history of acute myocardial infarction (MI) over a follow-up period of about 5 years. Enrollment into IDEAL was not limited with regard to timing after acute MI and most patients were enrolled several months or years after their MI. For the present analysis we focused on the 999 patients randomized within 2 months after an acute MI, a period when they were still in the unstable phase after an acute coronary syndrome. To facilitate placing the data into the context of recent trials of statins for acute coronary syndromes, the present evaluation focused on analyzing the composite end point that was used as the primary end point of the Pravastatin or Atorvastatin Evaluation and Infection Therapy (PROVE IT) trial.
Methods
The design and principal findings of the trial have been published previously. In brief, IDEAL was a multicenter trial of men and women ≤80 years of age who had a history of confirmed MI. The trial was conducted at 190 centers in Nordic countries and in The Netherlands. The study was approved by institutional review committees, and all subjects gave informed consent. Patients were screened and enrolled from hospitals during hospitalization for an acute MI and outpatient clinics during follow-up visits for hospitalization for recent MI or for long-term management of stable coronary heart disease. The latter setting provided most patients screened and enrolled in the trial. There were 9,689 patients screened, most of whom were already receiving statin therapy. They qualified for randomization unless they were already receiving a dose higher than the equivalent of simvastatin 20 mg/day. Of patients screened, 8,888 were randomized to treatment with atorvastatin 80 mg/day or simvastatin 20 mg/day. There were no run-in or wash-out periods.
Randomization was done through a central interactive voice-response system with allocation balanced by center. Patients randomized in Finland (506 patients in overall trial) received study drugs from clinical centers at no cost. Patients in other countries were given prescriptions for the statin, which they filled at designated pharmacies. To reflect reimbursement rules for statin therapy in the other countries, patients in Norway and Sweden paid part of the medication cost, whereas patients in Denmark and The Netherlands were reimbursed for medication cost. To equalize cost of medication, the sponsor paid the difference in cost between simvastatin and atorvastatin by arrangements made with the governmental agency in each country.
The trial used a study design that was a prospective, randomized, open-label, blinded end-point evaluation. Patients were followed by designated study coordinators and investigators at clinical centers after 12 and 24 weeks and every 6 months thereafter. Blood samples were collected at each visit for measurement of lipids and other components for safety assessment. Patients on simvastatin could be titrated to 40 mg/day if their plasma total cholesterol at the 24-week visit was >190 mg/dl (5.0 mmol/L) on advice from the central laboratory. Except for such cases, lipid levels were not reported to study personnel during the study.
End-point reports from investigators were first screened by an independent center (Inveresk, Raleigh, North Carolina) for blinding of treatment allocation. All end-point events were then adjudicated by an independent end-point classification committee that was thus blinded to treatment allocation. The primary end point was time to first occurrence of a major coronary event, defined as coronary death, confirmed nonfatal acute MI, or cardiac arrest with resuscitation.
For the present post hoc analysis, we only included data from the subgroup of patients who were randomized <2 months after an acute MI. A composite end point that was identical to the primary end point of PROVE IT was evaluated using primary data during 5 years of follow-up. The PROVE IT composite end point consisted of time to occurrence of all-cause death or a cardiovascular event (a composite of hospitalization for nonfatal acute MI, hospitalization for unstable angina, coronary revascularization, and stroke). Time to occurrence of first end point after randomization was examined using Kaplan-Meier hazard rates, and group differences were assessed with log-rank test. Cox proportional hazard models were used to calculate hazard ratios (HRs) and 95% confidence interval (CIs). Statistical analysis was performed using SAS 8.2 (SAS Institute, Cary, North Carolina). All analyses are based on an intention-to-treat principle including all patients. Two-sided p values <0.05 were considered statistically significant; corrections for multiple comparisons were not done.
An analysis for adherence was also conducted. Overall adherence for each subject was defined as total study medication exposure as percent total follow-up time for all-cause death. Total study medication exposure was defined as the interval between dates of first and last dose, regardless of the number of doses taken during that interval. Categories of adherence were defined as adherence ≥80% (adherers) and adherence <80% (nonadherers).
Statistical analyses were performed by the sponsor’s statistician (MS) and independently verified by the steering committee’s independent academic statistician (IH) who had full access to the data. The authors had full access to the data and take responsibility for the integrity of the data. All authors have read and agree to the report as written.
Results
From March 1999 to March 2001, 9,689 patients with a confirmed history of MI were screened for the trial ( Figure 1 ). Of these patients, 1,073 (11%) had a screening visit <2 months after hospitalization for an acute MI. Of these patients 74 were excluded, mostly because they were already receiving a statin dose higher than simvastatin 20 mg/day or an equipotent dose of another statin. Thus, 999 were randomized, 506 to simvastatin and 493 to atorvastatin. Mean time from index MI to randomization was 46 days. All randomized patients were included in the intention-to-treat analyses for efficacy and safety. Only 5 subjects (0.5%) withdrew their consent or were lost to follow-up, but vital status was known for 3 of these subjects (0.3%) at the end of the study.
Baseline characteristics were similar in the 2 treatment groups ( Table 1 ), except that slightly more patients in the atorvastatin group were receiving calcium antagonists than in the simvastatin group.
| Characteristics | Simvastatin (n = 506) | Atorvastatin (n = 493) |
|---|---|---|
| Age (years), mean ± SD | 59.8 ± 9.6 | 60.6 ± 9.7 |
| Men | 395 (78.1%) | 376 (76.3%) |
| Current smoker | 88 (17.4%) | 81 (16.4%) |
| Hypertension | 159 (31.4%) | 153 (31.0%) |
| Diabetes mellitus | 58 (11.5%) | 55 (11.2%) |
| Previous myocardial infarction | 506 (100%) | 493 (100%) |
| >1 previous myocardial infarction | 85 (16.8%) | 77 (15.6%) |
| Previous coronary artery bypass grafting | 25 (4.9%) | 25 (5.1%) |
| Previous percutaneous coronary intervention | 96 (19.0%) | 97 (19.7%) |
| Cerebrovascular disease | 27 (5.3%) | 32 (6.5%) |
| Peripheral vascular disease | 17 (3.4%) | 19 (3.9%) |
| Congestive heart failure | 24 (4.7%) | 31 (6.3%) |
| Prerandomization statin therapy | ||
| Simvastatin | 206 (40.7%) | 219 (44.4%) |
| Atorvastatin | 65 (12.9%) | 55 (11.2%) |
| Pravastatin | 40 (7.9%) | 31 (6.3%) |
| Other statins | 8 (1.6%) | 4 (0.8%) |
| Concomitant therapy | ||
| Aspirin | 424 (83.8%) | 416 (84.4%) |
| Warfarin or dicoumarol | 26 (5.1%) | 32 (6.5%) |
| β blockers | 446 (88.1%) | 437 (88.6%) |
| Calcium antagonists | 53 (10.5%) | 84 (17.0%) |
| Angiotensin-converting enzyme inhibitors | 177 (35.0%) | 165 (33.5%) |
| Angiotensin II blockers | 22 (4.4%) | 27 (5.5%) |
| Blood pressure (mm Hg), mean ± SD | ||
| Systolic | 134.4 ± 20.2 | 134.9 ± 21.5 |
| Diastolic | 80.2 ± 10.9 | 79.0 ± 10.5 |
| Body mass index (kg/m 2 ), mean ± SD | 27.1 ± 3.6 | 27.2 ± 3.7 |
| Patients with lipid values | 504 | 489 |
| Lipid values (mg/dl), mean ± SE | ||
| Total cholesterol | 201 ± 2.0 | 203 ± 2.0 |
| Low-density lipoprotein cholesterol | 128 ± 1.8 | 129 ± 1.8 |
| High-density lipoprotein cholesterol | 44 ± 0.5 | 44 ± 0.5 |
| Triglycerides | 152 ± 3.3 | 160 ± 3.7 |
At 24 weeks, 117 patients (23.1%) randomized to simvastatin had their dose titrated to 40 mg/day, whereas 30 patients (6.1%) in the atorvastatin group had their dose decreased to 40 mg/day at the discretion of the investigator who might have related patients’ adverse reports to the high dose. After 1 year, 2% of patients on simvastatin had switched to a different statin, most often atorvastatin; the corresponding proportion in the atorvastatin group was 5%, most of whom had switched to simvastatin. At the end of the trial, 27% of patients being prescribed simvastatin were taking 40 instead of 20 mg/day, whereas 12% of those being prescribed atorvastatin were taking 40 instead of 80 mg/day.
At 12 weeks, mean levels of low-density lipoprotein (LDL) cholesterol were 104 mg/dl (2.7 mmol/L) in the simvastatin group and 79 mg/dl (2.0 mmol/L) in the atorvastatin group ( Figure 2 ). At 1 year, mean LDL cholesterol levels were 102 mg/dl (2.6 mmol/L) in the simvastatin group and 81 mg/dl (2.1 mmol/L) in the atorvastatin group, representing an absolute difference of 21 (p <0.0001). Over the course of the trial, mean LDL cholesterol levels were 105 mg/dl (2.7 mmol/L) in the simvastatin group and 83 mg/dl (2.1 mmol/L) in the atorvastatin group. Only small changes occurred in high-density lipoprotein cholesterol levels, with minor differences between treatment groups. Triglycerides were lowered by 7% in the simvastatin group to a mean of 145 mg/dl (1.64 mmol/L) during the trial. In the atorvastatin group, triglyceride levels were lowered by 21% to a mean of 118 mg/dl (1.33 mmol/L).
