Author + information
- Received August 2, 2005
- Revision received November 22, 2005
- Accepted November 28, 2005
- Published online April 18, 2006.
- Mouaz H. Al-Mallah, MD⁎,⁎ (, )
- Imad M. Tleyjeh, MD†,
- Ahmed A. Abdel-Latif, MD‡ and
- W. Douglas Weaver, MD, FACC⁎
- ↵⁎Reprint requests and correspondence:
Dr. Mouaz Al-Mallah, Henry Ford Heart and Vascular Institute, 2799 West Grand Boulevard, K 14, Detroit, Michigan 48202
Objectives This study sought to assess the efficacy of angiotensin-converting enzyme inhibitors (ACEIs) in patients with coronary heart disease and preserved left ventricular (LV) function.
Background The ACEIs have been shown to improve outcomes in patients with heart failure and myocardial infarction (MI). However, there is conflicting evidence concerning the benefits of ACEIs in patients with coronary artery disease (CAD) and preserved LV systolic function.
Methods An extensive search was performed to identify randomized, placebo-controlled trials of ACEI use in patients with CAD and preserved LV systolic function. Of 61 potentially relevant articles screened, 6 trials met the inclusion criteria. They were reviewed to determine cardiovascular mortality, nonfatal MI, all-cause mortality, and revascularization rates. We performed random-effect model meta-analyses and quantified between-studies heterogeneity with I2.
Results There were 16,772 patients randomized to ACEI and 16,728 patients randomized to placebo. Use of ACEIs was associated with a decrease in cardiovascular mortality (relative risk [RR] 0.83, 95% confidence interval [CI] 0.72 to 0.96, p = 0.01), nonfatal MI (RR 0.84, 95% CI 0.75 to 0.94, p = 0.003), all-cause mortality (RR 0.87, 95% CI 0.81 to 0.94, p = 0.0003), and revascularization rates (RR 0.93, 95% CI 0.87 to 1.00, p = 0.04). There was no significant between-studies heterogeneity. Treatment of 100 patients for an average duration of 4.4 years prevents either of the adverse outcomes (one death, or one nonfatal myocardial infarction, or one cardiovascular death or one coronary revascularization procedure).
Conclusions The cumulative evidence provided by this meta-analysis shows a modest favorable effect of ACEIs on the outcome of patients with CAD and preserved LV systolic function.
Angiotensin-converting enzyme inhibitors (ACEIs) have been shown to be beneficial in patients with hypertension (1–3), those with recent myocardial infarction (MI) (4–8), those undergoing percutaneous coronary intervention (PCI) (9), and/or with left ventricular (LV) dysfunction (10–15). In patients with congestive heart failure, analysis from the Studies Of Left Ventricular Dysfunction (SOLVD) and the Survival And Ventricular Enlargement (SAVE) trials suggested that ACEIs may also reduce acute ischemic cardiovascular events by 20% (16–17). The current American College of Cardiology and American Heart Association Guidelines for the management of ST-segment elevation MI recommend that an ACEI should be prescribed at discharge to all patients with ST-segment elevation MI (class I) (18). However, the evidence for routine administration of ACEIs has been conflicting in patients with coronary artery disease (CAD) with preserved LV systolic function. Although two large trials (19–21) showed a decrease in cardiovascular death and nonfatal MI, no such benefit was found in another large trial and in a few smaller trials (22–26). We therefore performed a systematic review of the literature and a meta-analysis of the efficacy of ACEIs in patients with coronary heart disease and preserved LV function. We sought to determine the possible effect of ACEI therapy on all-cause mortality, cardiovascular mortality, nonfatal MI, and revascularization rates in patients with CAD and preserved LV systolic function.
All randomized, placebo-controlled trials of ACEIs use in patients with CAD and preserved LV function were identified using a two-level search strategy. First, we searched public domain databases, including Medline (1966 to February 14, 2005), the Cochrane Central Register of Controlled Trials (1st Quarter 2005), Database of Abstracts of Reviews of Effects (1st quarter 2005), Cochrane Database of Systematic Reviews (1st quarter 2005), Embase (1980 to 2005 week 6), U.S. Food and Drug Administration website, and Biosis Previews (1969 to 2005 week 8). We used the following key words: coronary artery disease, myocardial infarction, and angiotensin-converting enzyme inhibitors, as well as the following individual medications: captopril, cilazapril, enalapril, enalaprilat, fosinopril, lisinopril, perindopril, ramipril, and quinapril. Second, relevant studies were identified through a manual search of secondary sources, including references of initially identified articles and proceedings from national cardiology scientific sessions at the American Heart Association and American College of Cardiology from 2001 through 2005. The search was performed without any language restrictions. When an abstract from a meeting and a full article referred to the same trial, only the full article was included in the analysis. When there were multiple reports from the same trial, we used the most complete and/or recently reported data.
Inclusion and exclusion criteria
Reports of randomized trials of ACEI use in patients with CAD and preserved LV function were eligible for inclusion in the meta-analysis. Randomized controlled trial was defined according to the National Library of Medicine criteria; CAD was defined as history of prior MI, percutaneous or surgical coronary revascularization, angiographic evidence of atherosclerosis in one or more major coronary arteries, or a positive stress electrocardiogram, echocardiogram, or nuclear stress test. Trials that enrolled patients with recent revascularization were included in the meta-analysis. Preserved LV systolic function was defined as an ejection fraction of ≥40% and/or absence of clinical evidence of congestive heart failure. We excluded trials that did not have a placebo arm. We also excluded trials that had a follow-up duration of <2 years. Trials that did not report mortality, nonfatal MI, or revascularization also were not included. Data for each trial were abstracted by an investigator (M. A.) and confirmed by a second investigator (I. M. T.). All discrepancies were identified and resolved by consensus, or as needed, with a third investigator.
All eligible studies were assessed for the following methodological quality criteria: generation of allocation, allocation concealment, blinding of participants, blinding of caregivers, blinding of outcome assessment, blinding of data analyst, intention to treat analysis, and percentage of patients lost to follow-up (27).
The meta-analyses were performed by computing relative risks (RR) using a random-effects model (28). Quantitative analyses were performed on an intention-to-treat basis and were confined to data derived from the period of follow-up. The RR for all-cause mortality, cardiovascular mortality, nonfatal MI, and revascularization were calculated along with the 95% CIs. The number needed to treat to prevent one event was calculated by the inverse of the pooled absolute risk reduction. Between-study heterogeneity was analyzed by means of I2= [(Q − df)/Q] × 100%, where Q is the chi-square statistic and df is its degrees of freedom. This describes the percentage of the variability in effect estimates that is caused by heterogeneity rather than sampling error (chance) (29). Publication bias was assessed graphically using a funnel plot. All analyses were performed with RevMan Analyses version 4.2.7 (2004, Cochrane Collaboration, Oxford, United Kingdom). To explore heterogeneity, we conducted subgroup analyses, hypothesized a priori, by using a statistical test of interaction (30). Studied subgroups included the study population (diabetes mellitus [DM] <20% vs. ≥20%) and outcome measures (systolic blood pressure decrease <5 mm Hg vs. ≥5 mm Hg and follow-up duration ≤2 years vs. >2 years).
Figure 1shows the results of the literature search. From the 894 reports identified, 6 trials fulfilled the inclusion criteria. We included the Heart Outcomes Prevention Evaluation (HOPE) trial in the meta-analysis even though only 80% of these patients had documented CAD because the rest of the patients were at high risk for having CAD. We also included the Prevention of Atherosclerosis with Ramipril Trial (PART-2) for the same reason. We examined the effects of these two trials by sensitivity analysis. There were 33,500 patients enrolled in the seven trials, 16,772 in the treatment arm and 16,728 in the placebo arm. The mean follow-up duration was 4.4 years.
Table 1summarizes the baseline characteristics of the patients. All patients were followed up for more than 2 years. Patients enrolled in the HOPE trial were older, and more often were women and diabetic. On the other hand, a history of prior MI and prior revascularization were more common among patients enrolled in the European Trial on Reduction of Cardiac Events With Perindopril in Stable Coronary Arteries (EUROPA) and Angiotensin-Converting Enzyme Inhibition in Stable Coronary Artery Disease (PEACE) trial. The use of evidence-based therapies (antiplatelet agents, beta-blockers, and lipid-lowering therapies) was highest among the recently published PEACE trial patients.
The quality of the trials was assessed. Generation of allocation and allocation concealment was adequate in most of the trials. All trials were double blinded; however, the blinding of outcome assessment and blinding of data analyst was not clear in all of the trials. All trials used an intention-to-treat analysis.
The ACEI therapy resulted in a mean decrease of 3.9 mm Hg in the systolic blood pressure and a mean decrease of 1.8 mm Hg in the diastolic blood pressure. Data on blood pressure response to therapy were not available in the Quinapril Ischemic Event Trial (QUIET). Patients enrolled in the HOPE and EUROPA trials had a larger decrease in systolic (4.1 mm Hg vs. 3.5 mm Hg) and diastolic (2.0 mm Hg vs. 1.4 mm Hg) blood pressure. Only the PEACE and HOPE trials reported a decrease in new-onset diabetes over the follow-up duration. Pooled data from both trials confirmed the above observation (RR 0.76, 95% CI 0.60 to 0.95, p = 0.02).
Figure 2shows that ACEI therapy was associated with a decrease in cardiovascular mortality (RR 0.83, 95% CI 0.72 to 0.96, p = 0.01), nonfatal MI (RR 0.84, 95% CI 0.75 to 0.94, p = 0.003), all-cause mortality (RR 0.87, 95% CI 0.81 to 0.94, p = 0.0003), and coronary revascularization rates (RR 0.93, 95% CI 0.87 to 1.00, p = 0.04). The combined data for coronary revascularization were not available for the QUIET and PART-2 trials and were not included in the analysis. The number needed to treat with ACEIs to prevent either of the adverse outcomes (one cardiovascular death or any death, or nonfatal MI, or revascularization) is 100.
The test for heterogeneity showed no difference in effect among the studies as evident by I2estimates for different outcomes. The funnel plot shows symmetrical distribution of RR estimates with no evidence of publication bias.
Sensitivity and subgroup analyses
There was also a strong trend toward decreased all-cause mortality (RR 0.89, 95% CI 0.77 to 1.02, p = 0.09), even when combining data from the PEACE, QUIET, Effect of Antihypertensive Agents on Cardiovascular Events in Patients With Coronary Disease and Normal Blood Pressure Trial (CAMELOT), and PART-2, which individually did not show any significant beneficial effect of ACEIs. On the other hand, we found no significant difference in nonfatal MI (RR 0.96, 95% CI 0.82 to 1.12, p = 0.60), cardiovascular mortality (RR 0.87, 95% CI 0.58 to 1.30, p = 0.5), or revascularization (RR 0.98, 95% CI 0.9 to 1.07, p = 0.61).
We examined the effect of excluding the HOPE trial from the analysis because of its large weight in the pooled estimate. The ACEI therapy was associated with a lower cardiovascular death (RR 0.88, 95% CI 0.76 to 1.02), all-cause mortality (RR 0.89, 95% CI 0.81 to 0.98), revascularization (RR 0.97, 95% CI 0.90 to 1.03), and nonfatal MI (RR 0.86, 95% CI 0.75 to 0.99).
Because absence of clinical evidence of heart failure is not an adequate surrogate for preserved LV function, we performed another sensitivity analysis by excluding the EUROPA and PART-2 study. The combined estimates were: CV death, RR 0.85, 95% CI 0.69 to 1.05; all-cause mortality, RR 0.87, 95% CI 0.80 to 0.95; revascularization, RR 0.92, 95% CI 0.84 to 1.02; and nonfatal MI, RR 0.86, 95% CI 0.72 to 1.03.
These point estimates obtained by different sensitivity analyses testing multiple assumptions remained consistent with the modest effect of ACEI similar to the primary results. However, they had wider confidence intervals because of the decrease in sample size when fewer trials were included.
Table 2summarizes subgroup analyses. There was no significant interaction between different outcomes and any of the studied subgroups: proportions of DM <20% vs. ≥20%, systolic blood pressure decrease <5 mm Hg vs. ≥5 mm Hg, and follow-up duration ≤2 years vs. >2 years.
Based on the findings of this meta-analysis from more than 33,000 patients, our results confirm that ACEIs, when added to conventional therapy (aspirin, beta-blockers, and statins), are beneficial in patients with CAD and preserved LV function. Death, nonfatal MI, and the need for coronary revascularization were all modestly reduced after ACEI treatment. This benefit is, however, modest. Treatment of 100 patients for an average duration of 4.4 years prevents either of the poor outcomes (one death, or one nonfatal MI, or one cardiovascular death or one coronary revascularization procedure).
The ACEIs have multiple potential cardioprotective effects. In addition to controlling blood pressure and reducing LV hypertrophy, they have multiple potential anti-atherosclerotic effects. They improve endothelium-dependent vasodilatation, and by increasing the level of bradykinin, increase the expression and activity of nitric oxide synthase and reduce the production of smooth muscle proliferating agents (31–33). They are also associated with a decrease in vascular nicotinamide adenine dinucleotide activity and reactive oxygen species. Use of ACEIs reduces activation of important signaling pathways (including pathways involving nuclear factor-B and activator protein 1); reductions in vascular inflammation, endothelial dysfunction, progression of atherosclerosis, and activation of metalloproteinases 2 and 9; an improvement in plaque stability; a decrease in the tendency toward thrombosis; and an improvement in fibrinolysis (34,35).
Each of the above changes could decrease cardiovascular adverse events. However, not all of the individual trials showed a statistically significant benefit of treatment. Multiple reasons could explain this. Except for the CAMELOT trial, all trials showed a reduction in all-cause mortality in patients receiving ACEIs as evidenced by the individual RR from each study. However, this reduction did not reach statistical significance in some because of a small sample size and lack of statistical power. A meta-analysis can overcome this limitation and detect smaller changes in outcome.
Differences in the risk profile and baseline therapies of patients enrolled among different trials might have impacted the final results by reducing event ratio. The baseline characteristics of the patients and adjunctive co-therapies in the trials were consistently different. Patients in the HOPE trial had the highest risk of cardiac events (older, and a higher prevalence of female gender and diabetes). On the other hand, patients in the PEACE trial were more often on adjunctive therapies for CAD and had better risk factor modification. Because relative risk estimates for all of these groups were not available from the eligible trials, we were not able to explore heterogeneity because of the different study populations. We attempted to explore heterogeneity by using study-level data such as percent of the study population with diabetes mellitus. This subgroup analysis may be underpowered to detect a difference.
The follow-up duration, which was not the same among different trials, can also explain between-study heterogeneity. The QUIET and CAMELOT trials, for example, followed up patients for only two years, compared with the HOPE, EUROPA, and PEACE trials, in which follow-up was four years or more. The beneficial effects of therapy started after one year in EUROPA and two years in the HOPE trials. We did not find a statistically significant interaction between follow-up duration and effect size.
The fact that different ACEIs have different pharmacokinetic and pharmacodynamic characteristics is not a plausible explanation because the majority of the patients received a tissue-specific ACEI. However, the dose of the ACEI was different among included trials. In the QUIET trial, 20 mg quinapril did not affect the overall progression of coronary atherosclerosis. This dose is less than “the standard” 40-mg dose most often used.
Another possible explanation of between-study heterogeneity might be related to differences in blood pressure decrease among different trials. Numerous experimental studies and clinical trials support the emerging realization that ACEIs restore endothelial function or prevent endothelial dysfunction more than what is expected from blood pressure reduction alone (32). The HOPE and PEACE trials were associated with the same degree of blood pressure decrease, yet the impact of ACEI treatment on the primary end point of both trials was different. Thus, the difference in the magnitude of blood pressure decrease does not seem to be the only reason for inconsistency between these two trials. In addition, subgroup analysis from HOPE and EUROPA indicates that the benefits of ACEIs were universal in normotensive as well as hypertensive patients (19,21).
Finally, there was a difference in the rate of statin use between the different trials. The ACEIs and statins inhibit superoxide production in the cell (36,37) and may have similar anti-inflammatory effects. Thus, a high statin use rate may mask the anti-atherosclerotic benefit of an ACEI. In the EUROPA trial subgroup analysis, the effect of perindopril was independent of the use of statins. On the other hand, there was a trend toward benefit from the ACEI inhibitor in patients who were not on statins in the PEACE trial.
Our systematic review has several strengths. It helps to answer the general question of the beneficial effects of ACEIs in patients with CAD and preserved LV function with different treatment regimens. The findings in a randomized clinical trial may be valid only for a population of patients with the same characteristics as those investigated in the trial. If many trials exist in different groups of patients with different co-treatments, with similar results in the various trials, then it can be concluded that the intervention under study has general applicability. In addition, by integrating the actual evidence, our systematic review allowed a more objective appraisal of the literature by resolving uncertainties when the original randomized trial data did not agree.
On the other hand, our analysis has possible limitations. Disease misclassification is a potential bias. Several studies have used a broad definition of preserved LV function and CAD (19,21). In addition, we did not have access to the individual patient data, and we used the data in the published manuscripts. Publication bias is another possible limitation. Our process of literature identification was comprehensive and should have captured the majority of published studies. There was no evidence of publication bias on visual inspection of the funnel plot. We also searched for unpublished studies, and thus we think we have identified all relevant literature on this topic.
The results of this meta-analysis confirm the beneficial effects of ACEIs. This is in line with other meta-analyses that showed that ACEIs are useful in preventing new-onset diabetes (38) and atrial fibrillation (39). However, these meta-analyses included trials that also examined angiotensin receptor blockers. The current cumulative evidence supports the use of an ACEI and should be considered in patients with pre-diabetic conditions, hypertension, impaired fasting glucose levels, congestive heart failure, or coronary heart disease.
In conclusion, our findings show that treatment with ACEIs, when added to conventional therapy, has a modest effect in reducing all-cause morality, cardiovascular mortality, nonfatal MI, and subsequent revascularization in patients with CAD who have preserved LV function.
- Abbreviations and Acronyms
- angiotensin-converting enzyme inhibitor
- coronary artery disease
- left ventricular
- myocardial infarction
- percutaneous coronary intervention
- Received August 2, 2005.
- Revision received November 22, 2005.
- Accepted November 28, 2005.
- American College of Cardiology Foundation
- Ambrosioni E.,
- Borghi C.,
- Magnani B.,
- Survival of Myocardial Infarction Long-Term Evaluation (SMILE) Study Investigators
- Pfeffer M.A.,
- Greaves S.C.,
- Arnold M.O.,
- et al.,
- Healing and Early Afterload Reducing Therapy (HEART) Trial Investigators
- Rutherford J.D.,
- Pfeffer M.A.,
- Moyeé L.A.,
- et al.
- Antman E.M.,
- Anbe D.T.,
- Armstrong P.W.,
- et al.
- Dagenais G.,
- Yusuf S.,
- Bourassa M.,
- et al.
- Fox K.M.,
- European Trial on Reduction of Cardiac Events with Perindopril in Stable Coronary Artery Disease Investigators
- MacMahon S.,
- Sharpe N.,
- Gamble G.,
- et al.
- Jüni P.,
- Altman D.,
- Egger M.
- Higgins J.P.,
- Thompson S.G.,
- Deeks J.J.,
- Altman D.G.
- Altman D.G.,
- Bland J.M.
- Mancini G.B.,
- Henry G.C.,
- Macaya C.,
- et al.
- Wagner A.H.,
- Kohler T.,
- Ruckschloss U.,
- et al.
- Griendling K.K.,
- Minieri C.A.,
- Ollerenshaw J.D.,
- et al.
- Abuissa H.,
- Jones P.,
- Marso S.,
- et al.
- Healey J.,
- Baranchuk A.,
- Crystal E.