 |
The Agency for Healthcare Research and Quality (AHRQ, formerly the Agency for Health Care Policy and Research, AHCPR), through its Evidence-based Practice Centers (EPCs), sponsors the development of evidence reports and technology assessments to assist public- and private-sector organizations in their efforts to improve the quality of health care in the United States. The reports and assessments provide organizations with comprehensive, science-based information on common, costly medical conditions and new health care technologies. The EPCs systematically review the relevant scientific literature on topics assigned to them by AHRQ and conduct additional analyses when appropriate prior to developing their reports and assessments.
To bring the broadest range of experts into the development of evidence reports and health technology assessments, AHRQ encourages the EPCs to form partnerships and enter into collaborations with other medical and research organizations. The EPCs work with these partner organizations to ensure that the evidence reports and technology assessments they produce will become building blocks for health care quality improvement projects throughout the Nation. The reports undergo peer review prior to their release.
AHRQ expects that the EPC evidence reports and technology assessments will inform individual health plans, providers, and purchasers as well as the health care system as a whole by providing important information to help improve health care quality.
We welcome written comments on this evidence report. They may be sent to: Director, Center for Practice and Technology Assessment, Agency for Healthcare Research and Quality, 6010 Executive Blvd., Suite 300, Rockville, MD 20852.
| Douglas B. Kamerow, M.D. | John M. Eisenberg, M.D. |
| Director, Center for Practice and Technology Assessment | Director, Agency for Healthcare Research and Quality |
| The authors of this report are responsible for its content. Statements in the report should not be construed as endorsement by the Agency for Healthcare Research and Quality or the U.S. Department of Health and Human Services of a particular drug, device, test, treatment, or other clinical service. |
This report was developed with support from the Agency for Health Care Policy and Research, now the Agency for Healthcare Research and Quality. Dr. Heidenreich is supported by a career development award from the Veterans Affairs Health Services Research and Development Service. The authors thank Lyn Dupré, Mary Grady, and the peer reviewers for comments on the manuscript, and the members of the University of California, San Francisco (UCSF)-Stanford EPC Governing Council for their suggestions. The authors also thank Francis Chesley, M.D., and the members and support staff of the Stable Angina Guidelines Committee (American College of Cardiology, American Heart Association, American College of Physicians) for guidance throughout the project; Dr. Drummond Rennie and Dr. Lisa Bero for developing a peer review process for evidence reports; and Dr. Patricia Huston for developing a coherent and constructive synthesis of peer review commentaries.
Stable angina is a major health problem that affects over 7 million adults in the United States, with an estimated 350,000 new cases annually. The American College of Cardiology, American Heart Association, and American College of Physicians established a committee to develop guidelines for the diagnosis and treatment of stable angina. This committee of experts in cardiology and internal medicine and the investigators from the University of California, San Francisco (UCSF)-Stanford Evidence-based Practice Center (EPC) prioritized two topics for a thorough systematic review of the literature. The first topic concerned the relative efficacy and tolerability of treatment with beta-blockers, calcium antagonists, and long-acting nitrates for patients who have stable angina. The second topic dealt with the efficacy of alternative therapies in patients who have stable angina.
The authors identified published studies from 1966 through 1997 by searching the MEDLINE and EMBASE databases and by reviewing manually the bibliographies of identified articles.
For the review of traditional therapies, studies that compared two agents from different anti-anginal drug classes (beta-blockers, calcium antagonists, and nitrates) and that were at least 1 week in duration were reviewed. Studies were selected if they reported one of the following outcomes: cardiac death, myocardial infarction, angina frequency, nitroglycerin use, exercise duration, or adverse events leading to withdrawal. For the review of alternative therapies, the authors included all randomized trials of alternative therapies compared with placebo, nitrates, calcium antagonists, or beta-blockers in patients who had stable angina.
For the review of traditional therapies, 91 studies met the inclusion criteria. Each study was abstracted by two independent reviewers. The data were pooled using odds ratios for discrete data and mean differences for continuous data. Studies of calcium antagonists were grouped by duration of action (short- vs. long-acting) and type of drug (nifedipine vs. nonnifedipine).
Rates of cardiac death or myocardial infarction were similar (odds ratio 0.97 [0.67,1.38]), but events were few; the median trial duration was 4 weeks. Beta-blockers provided greater angina relief than calcium antagonists did: mean difference in episodes per week was 0.31 (95 percent confidence interval: 0.00, 0.62). Beta-blockers were discontinued after adverse events less often than calcium antagonists were (odds ratio 0.72 [95 percent confidence interval: 0.60, 0.86]). Trials comparing nifedipine with beta-blockers showed a significantly greater benefit for angina relief for beta-blockers and a decrease in adverse events leading to study withdrawal. Commonly reported side effects were similar or greater in patients taking calcium antagonists compared with patients taking beta-blockers. Randomized trials of alternative therapies for patients who had stable angina were too small and too few to allow conclusions to be drawn.
In trials of patients who had stable angina, beta-blockers provided equivalent or greater angina relief than calcium antagonists and were associated with fewer adverse events. No differences were documented in mortality or myocardial infarction, but trial duration was too short to define clinically important effects on these endpoints. There were too few studies of nitrates and alternative therapies to draw conclusions.
This document is in the public domain and may be used and reprinted without permission.
Heidenreich PA, McDonald KM, Hastie T, et al. An evaluation of beta-blockers, calcium antagonists, nitrates, and alternative therapies for stable angina. (Evidence Report/Technology Assessment No. 10 [Contract 290-97-0013] to the University of California San Francisco-Stanford Evidence-based Practice Center). AHRQ Publication No. 00-E003. Rockville, MD: Agency for Healthcare Research and Quality. November 1999.
Stable angina is a major health problem that affects over 7 million adult men and women in the United States, with an estimated 350,000 new cases annually. Symptomatic therapy is targeted at either reducing oxygen demand by decreasing the work of the heart or increasing oxygen supply by dilating coronary arteries. Acute treatment of an angina episode consists of rest and often sublingual nitroglycerin. Chronic treatment to prevent symptoms has consisted of one or a combination of beta-blockers, calcium antagonists, and long-acting nitrates.
The choice of a first-line anti-anginal drug has been controversial because all three classes have been shown to be effective in relieving symptoms. There are few long-term trials comparing these drugs in controlled studies, and it is unclear whether any of the drug classes decrease mortality or myocardial infarction in patients with stable angina.
To address these important issues, the University of California, San Francisco-Stanford Evidence-based Practice Center (EPC) conducted a thorough, systematic review and synthesis of the literature on treatment of stable angina and developed an evidence report on the topic. The topic was nominated by the American College of Cardiology, the American Heart Association, and the American College of Physicians. The EPC worked with the nominators' guideline committee to prioritize two topics for this project: (1) the relative efficacy and safety of beta-blockers, calcium antagonists, and long-acting nitrates in patients who have stable angina, and (2) the efficacy of alternative therapies in patients who have stable angina.
The following key questions defined the parameters of the investigation:
Is one class of anti-anginals (beta-blockers, nitrates, or calcium antagonists) superior as monotherapy in terms of occurrence of angina, nitroglycerin use, exercise tolerance, or adverse events leading to study withdrawal in patients with stable angina?
Is one class of drugs superior when outcomes over a longer period of time-such as recurrent myocardial infarction or death-are examined?
Are there any alternative medical treatments including chelation therapy, acupuncture, herbal medications, and garlic that are superior to placebo in any outcome measure for patients with stable angina?
The EPC investigators searched MEDLINE (1966--1997) and EMBASE (1974--1997) and reviewed cited references of retrieved articles to identify published studies. The search criteria were (1) the MESH heading "angina pectoris" or the text word "angina" and (2) publication type "randomized controlled trial" or text word containing a form of the word "random." Using this criteria set, the investigators searched MEDLINE then EMBASE. They also performed searches replacing the second criterion with (3) publication type "controlled" clinical trial, or or (4) text word "double-blind." Finally, the bibliographies of identified trials were reviewed to locate other relevant studies.
The review was limited to randomized controlled trials that directly compared at least two of the three major anti-anginal drug classes: long-acting nitrates, beta-blockers, and calcium antagonists. Studies with the following designs were excluded: one drug versus placebo, dose comparison of one drug, and one drug versus a two-drug combination, as were studies with duration of less than 1 week. Trials that used a study medication during a run-in period prior to randomization were excluded because patients who quickly developed side effects were not included in the randomized trial.
All randomized trials of patients who had stable angina were included. Trials comparing alternative therapies to placebo, nitrates, calcium antagonists, or beta-blockers were abstracted. Because few studies fit these criteria, no limit was imposed on study duration.
To search for alternative therapies for stable angina treatment, the same search criterion noted above was used with the following key words: meditation, prayer, naturopathy, chiropractic, osteopathic, holistic, natural medicine, homeopathy, mind-body, unorthodox, integrative ayurvedic, acupuncture, herbal, relaxation, or chelation.
Study selection was performed initially by title review; candidate abstracts were then reviewed and selected for data retrieval. Two independent reviewers abstracted data for each article on standardized data forms. The reviewers compared their results and settled any differences. In general, one of the two reviewers had cardiology expertise, and the other had expertise in epidemiology or health services research.
Standard methods of meta-analysis were used to combine outcomes data across trials. The following estimates were combined: the mean difference for angina episodes and nitroglycerin use, the standardized mean difference for exercise time, and the odds ratio for cardiac death, myocardial infarction, and all adverse events. The standardized mean difference was used to compare exercise times because exercise protocols varied across studies. The DerSimonian-Laird (random-effects) method was used to estimate summary odds ratios. The investigators examined differences between study subgroups of calcium antagonists using analysis of variance. The prespecified subgroup comparisons of interest were nifedipine versus non-nifedipine and short-acting versus long-acting calcium antagonists. Reported p values are two- tailed with statistical significance at p < 0.05.
Beta-blockers were associated with fewer episodes of angina when compared with calcium antagonists in general and with nifedipine in particular.
Important differences in mortality could not be determined because the trials were short in duration (median 4 weeks).
Fewer adverse events leading to study withdrawal occurred with beta-blockers compared with calcium antagonists.
Commonly reported side effects were similar or greater in the calcium antagonists group compared with the beta-blocker group.
There were no obvious differences in any outcome measures between beta-blockers and non-nifedipine calcium antagonists.
Few studies compared long-acting nitrates with beta-blockers or with calcium antagonists.
Randomized trials of alternative therapies for patients with stable angina were too small and too few to provide conclusive findings.
Few long-term studies have compared treatment with beta-blockers, calcium antagonists, and long-acting nitrates in patients who have stable angina. Additional large randomized trials are needed to determine if clinically important differences in patient survival exist between classes of anti-anginal medication. Further studies are needed to examine the effects of treatments for stable angina in patients with comorbidity such as heart failure or chronic obstructive pulmonary disease. Additional randomized controlled trials of alternative therapies in this patient population are necessary to reach conclusions about efficacy and safety.
Stable angina is a major health problem that affects over 7 million adults in the United States, with an estimated 350,000 new cases annually. The most common cause of angina is atherosclerotic obstruction of the major coronary arteries. Symptoms, which commonly occur for the first time during the fifth or sixth decade of life, result from an inadequate oxygen supply to the myocardium for a given demand. Thus, angina is often precipitated when a person exercises or is subjected to other stress. By definition, the symptoms of stable angina disappear with rest, thereby distinguishing it from unstable angina and myocardial infarction.
Patients with angina are at substantially higher risk of cardiac death or myocardial infarction than are members of the general population. Therapies for patients with stable angina are aimed at reducing symptoms, reducing the risk of death and myocardial infarction, or both. Therapies that act solely to reduce risk without changing symptoms are clearly important components of medical management but are not discussed in this review.
Symptomatic therapy is targeted at either reducing oxygen demand by decreasing the work of the heart or increasing oxygen supply by dilating coronary arteries (Karliner, 1991). Acute treatment of an episode of angina consists of rest and often includes sublingual nitroglycerin. Chronic treatment to prevent symptoms has consisted of one or a combination of beta-blockers, calcium antagonists, and long-acting nitrates (Gersh, Braunwald, and Rutherford, 1997). The choice of a first-line agent has been controversial because all three drug classes have been shown to be effective in relieving symptoms. Few long-term trials have evaluated the effect of these drugs in patients who have stable angina. Thus, it is unclear whether either of the drug classes improves survival or decreases events in this patient population. Because long-term use of beta-blockers has been shown to reduce mortality following myocardial infarction, researchers have suggested that these drugs also be used as first-line agents in patients with stable angina (Gersh, Braunwald, and Rutherford, 1997). However, calcium antagonists are often used as initial therapy because they are assumed to have equal efficacy in relieving angina and to be associated with fewer side effects than beta-blockers (Chan, Heo, Garibian, et al., 1988; Fox, Jespersen, Ferrari, et al., 1997). Furthermore, long-term nitrate therapy has been recommended as initial therapy by some investigators (Chan, Heo, Garibian, et al., 1988; Hoekenga and Abrams, 1984), but concern about the need for a daily nitrate-free period has led many physicians to use long-term nitrates as second-line agents (Gersh, Braunwald, and Rutherford, 1997). At this time, no single class of therapy has proved superior for treatment of chronic stable angina (Roberts, 1985; Thadani, 1997).
Calcium antagonists are more diverse than beta-blockers in chemical structure and mechanism of action. Short-acting agents such as dihydropyridines, particularly nifedipine, are distinct from other types of calcium antagonists in part because they generally increase rather than decrease heart rate. Recent case-control studies and meta-analyses of randomized trials suggested that short-acting calcium antagonists increase the risk of cardiac events among patients who have hypertension (Psaty, Heckbert, Koepsell, et al., 1995), and nifedipine increases risk of cardiac events following acute ischemic syndromes (Furberg and Psaty, 1996; Furberg, Psaty, and Meyer, 1995). These observations have led to considerable discussion and controversy. Even if the findings are valid in the populations studied, it is uncertain whether they can be extended to the stable angina patient population.
Although all three classes of anti-anginal agents have been proven safe and effective compared with placebo, the efficacy and safety of these agents when compared with one another is not clear. Furthermore, the majority of studies that have compared anti-anginal agents have been small and thus did not have the power to detect clinically important differences between therapies. To determine the relative efficacy and safety of anti-anginal therapies, investigators at the University of California, San Francisco (UCSF)-Stanford Evidence-based Practice Center (EPC) performed a meta-analysis of trials that directly compared beta-blockers, calcium antagonists, and long-acting nitrates in patients who had stable angina.
The American College of Cardiology, American Heart Association, and American College of Physicians established a committee to develop guidelines for the diagnosis and treatment of stable angina. This committee of experts in cardiology and internal medicine and investigators from the UCSF-Stanford EPC prioritized two topics for a thorough and systematic review of the literature. The first topic was designed to address the issue described above to determine the relative efficacy and safety of beta-blockers, calcium antagonists, and long-acting nitrates in patients who have stable angina. The second review examined the efficacy of alternative therapies in patients who have stable angina.
We searched MEDLINE (1966-1997) and EMBASE (1974-1997) and reviewed cited references of retrieved articles to identify published studies. Our search criteria were (1) the MESH heading "angina pectoris" or the text word "angina" and (2) publication type "randomized controlled trial" or text word containing a form of the word "random." Using this criteria set, we searched MEDLINE, then EMBASE. We also performed searches replacing the second criterion with (3) publication type "controlled clinical trial" or (4) text word "double-blind." Finally, we reviewed the bibliographies of identified trials to locate other relevant studies.
To search for alternative therapies for stable angina treatments, we used the same search criteria noted above with the following key words: meditation, prayer, naturopathy, chiropractic, osteopathic, holistic, natural medicine, homeopathy, mind-body, unorthodox, integrative ayurvedic, acupuncture, herbal, relaxation, and chelation.
We restricted our review to controlled trials that directly compared drugs from two or three of the major anti-anginal drug classes: long-acting nitrates, beta-blockers, and calcium antagonists. Trials were included if they enrolled patients who had a history of stable angina. Studies of parallel design were included if they were randomized. We limited the parallel trials to those that were randomized because, unlike cohort and case-control studies, the randomized trial design is not limited by confounding. Blinded crossover studies, in which each patient receives both therapies, were included. Studies of the following designs were excluded: one drug versus placebo, dose comparison of one drug, and one drug versus a two-drug combination. These trial designs do not allow a direct comparison between different drug classes. Studies with duration of less than 1 week were excluded because the short duration would not allow adequate measures of effectiveness and adverse events. We excluded trials that used a study medication during a run-in period prior to randomization because the randomized portion did not include any patients who developed side effects early.
We included all randomized trials of alternative therapies compared with placebo, nitrates, calcium antagonists, or beta-blockers in patients who had stable angina. We abstracted data from these trials. Because few alternative therapy trials fit these criteria, shorter duration (< 1 week) studies were not excluded.
We examined measures of anti-anginal efficacy and tolerability, conducting separate meta-analyses for each outcome. Anti-anginal efficacy measures were number of angina episodes per week (patient recorded), number of nitroglycerin uses per week (patient recorded), total exercise time, and time to 1-mm ST depression on a treadmill or upright bicycle. The tolerability measure was the rate of subject withdrawal from the study as a result of adverse events (cardiac and noncardiac symptoms and death), exclusive of protocol violations or loss to followup. All side effects from the following classes were recorded: worsening angina, headache, other central nervous system (CNS) disturbance, gastrointestinal disturbance, and peripheral edema. We also examined the combined endpoint of cardiac death or nonfatal myocardial infarction for studies in which either was reported.
Titles of identified articles were reviewed by the project director and one other cardiologist. Candidate abstracts were then reviewed and appropriate studies were selected for data retrieval. Two independent reviewers extracted data from each article. The reviewers compared their results and settled any differences by consensus. One of the two reviewers had cardiology training; the other had training in epidemiology or health services research. All outcomes data were abstracted a third time by the project director or another cardiologist.
We examined several subgroups of trials to determine their contribution to the summary differences between beta-blockers and calcium antagonists. The two prespecified comparisons of interest were nifedipine versus non-nifedipine calcium antagonists and short- versus long-acting calcium antagonists. Short-acting preparations were defined as those with usual dosing of three or more times per day according to the Physicians' Desk Reference (1998). We also grouped calcium antagonists as dihydropyridine and non-dihydropyridine. Additional comparisons were made based on funding source (industry versus other), location of trial (Europe versus the United States), and a quality indicator (randomization and double-blinding stated versus both not stated). Because several of the trials examined drugs that are rarely used for stable angina, we repeated our analysis after limiting the trials to those agents most commonly used in the United States. To determine market share, we obtained pharmaceutical data from the National Disease and Therapeutic Index that described the frequency of medications prescribed during the first quarter of 1998 for stable angina (in the absence of hypertension) (IMS Health, 1998). Based on these data, the common (2 percent or more market share) therapies were limited to four calcium antagonists (amlodipine, diltiazem, nifedipine, and verapamil) and five beta-blockers (atenolol, carvedilol, propranolol, metoprolol, and nadolol). We did not evaluate subgroups of long-acting nitrates because of the small number of trials for this class of agents.
We combined estimates of the mean difference for number of angina episodes and number of nitroglycerin tablets used per week; the standardized mean difference for exercise time; and the odds ratio for adverse events, cardiac death, and myocardial infarction (Shadish and Haddock, 1994). The mean difference between the treatment and control group is differencei = meanti - meanci where t is the treatment and c the control group for the ith study. The estimated variance is vi = 2i(1/nti + 1/nci), where n is the within-study sample size for the treatment and control groups and 2 is estimated as the pooled within-group variance.
We used the standardized mean difference to compare exercise times because exercise protocols varied widely across studies. This measure expresses the difference in mean exercise time as a fraction of the standard deviation. The measure is calculated as di = (meanti - meanti)/ pooled SD where SD is the standard deviation. The variance for this value is estimated as vi = [(nti + nci)/ ntinci] + [di2 /2(nti + nci)] . For this comparison, a positive value indicates increased exercise time in the treatment group. We pooled both mean differences and standardized mean differences by weighting the individual differences by the inverse of the variance.
We used the DerSimonian-Laird (random-effects) and Peto (fixed-effects) methods to estimate summary odds ratios for the outcomes of death or myocardial infarction and withdrawal caused by adverse events (DerSimonian and Laird, 1986; Petitti, 1994). Fixed results are presented unless otherwise stated. For studies that reported no events, we substituted 0.1 in place of 0 for the random-effects calculation. Because events were rare, we chose a value below the more commonly used 0.5. We tested homogeneity of study effect size using the Q statistic (Petitti, 1994). A summary rate difference was calculated for trials with one or more adverse events (Petitti, 1994). This measure complements the odds ratio by providing an absolute difference in the adverse event rate. We examined differences between study subgroups of trials using analysis of variance (Hedges, 1994). We did not pool data on individual side effects that did not lead to study withdrawal because these data were inconsistently reported.
For the outcomes of cardiac death, myocardial infarction, and adverse events, the study sample size was defined as the number of patients randomized. For the outcomes of angina episodes, nitroglycerin use, and time spent exercising, the sample size was defined as the number of patients who completed the study. Data are presented as summary odds ratio or summary rate difference (95 percent confidence interval). Reported p values are two-tailed with statistical significance at p < 0.05.
Is one class of anti-anginals (beta-blockers, nitrates, or calcium antagonists) superior as monotherapy in terms of occurrence of angina, nitroglycerin use, exercise tolerance, or adverse events leading to study withdrawal in patients with stable angina?
Is one class of drugs superior when longer term outcomes such as recurrent myocardial infarction or death are examined?
Are any alternative medical treatments including chelation therapy, acupuncture, herbal medications, and garlic superior to placebo in any outcome measure for patients with stable angina?
In addition to the double abstraction method described above, we examined a random sample of 16 studies to determine errors in abstraction or database entry. Few errors were found in abstraction of primary outcomes. However, because of the importance of this area of abstraction to the evidence report, we rechecked all articles for endpoints of mortality, angina episodes, nitroglycerin use, exercise time, and adverse events.
Selected independent experts in the fields of internal medicine, family practice, cardiology, epidemiology, and systematic reviews reviewed the draft manuscript of the evidence report. In addition, the following organizations were invited to nominate individuals to review the manuscript: American Academy of Family Physicians (reviewed), American Academy of Medical Acupuncture and Medical Acupuncture Research Foundation (reviewed), American Academy of Nursing, American Association of Pharmaceutical Scientists (reviewed), American College of Cardiology, American College of Chest Physicians, American College of Clinical Pharmacology, American College of Physicians, American Geriatrics Society (reviewed), American Heart Association, American Nurses Foundation (reviewed), American Society of Health-System Pharmacists (reviewed), Council of Colleges for Acupuncture and Oriental Medicine (reviewed), and the Gerontological Society of America.
The San Francisco Cochrane Center coordinated these reviews and sent them to a criticism editor for synthesis and commentary. We then responded to the editor's commentary and revised the document to its current form.
A total of 3,501 articles were identified with the MEDLINE and EMBASE searches and citation review.
The majority of reviewed articles were excluded because they did not report trials of medical treatments for stable angina. Most anti-anginal trials compared one drug to placebo, or compared different drugs of the same class (e.g., diltiazem with nifedipine). Ultimately, only 3 percent of studies were included in the analysis.
). Another 959 studies were excluded because they compared a single drug with placebo. An additional 255 studies were intraclass comparisons (e.g., different beta-blockers compared) and thus were excluded. In 78 studies a new drug was added to background therapy without a drug-drug comparison. We excluded 77 studies because they evaluated anti-anginal drugs other than nitrates, beta-blockers, or calcium antagonists. Another 54 studies were excluded because the study duration was less than 1 week. Seventeen articles were duplicate reports or substudies of an included study, and nine evaluated medication withdrawal as the intervention. We excluded 25 studies that may have fulfilled the inclusion criteria but were written in a language other than English. Six trials were excluded because a study drug was required during the run-in period prior to randomization. Five articles did not report any of the primary outcomes (angina, nitroglycerin use, exercise time, adverse events, cardiac death, or myocardial infarction). The remaining 90 articles were abstracted (Evidence Tables 1-21). The MEDLINE search identified 86 articles (96 percent); EMBASE identified one that MEDLINE did not; and the citation search identified three not found in our MEDLINE or EMBASE searches.
Most trials (76%) in our analysis compared beta-blockers with calcium antagonists. Only 20% of trials evaluated a long-acting nitrate. Thus, the power to detect differences between long-acting nitrates and other classes is limited.
). The mean age of enrolled patients was 57 years; 84 percent were male; and 35 percent had a history of myocardial infarction. The included studies had a median length of 4 weeks (interquartile range 3-6 weeks), and only two trials (Dargie, Ford, and Fox, 1996; Rehnqvist, Hjemdahl, Billing, et al., 1996) lasted longer than 6 months.
BB = beta-blocker; CA = calcium antagonists
This figure shows summary odds ratios for benefit with beta-blockers (BB) versus calcium antagonists (CA) for cardiac death or myocardial infarction. The two long-term studies are displayed separately along with a pooled estimate from the 60 short-term studies.
| Outcome | Trials | Beta-Blockers | Calcium Antagonists | Measure | Effect Size | 95% CI | p | ||
|---|---|---|---|---|---|---|---|---|---|
| N | Events | N | Events | ||||||
| Cardiac death or MI | 61 | 3,090 | 57 | 3,054 | 59 | Odds ratio (< 1 favors BB) | 0.97 | 0.67, 1.38 | 0.9 |
| Angina episodes (per week) | 32 | 1,127 | 1,107 | Mean difference (< 0 favors BB) | -0.31 | -0.62, 0.00 | 0.05 | ||
| Nitroglycerin use (per week) | 25 | 736 | 726 | Mean difference (< 0 favors BB) | -0.14 | -0.41, 0.32 | 0.3 | ||
| Time to 1-mm ST depression | 20 | 537 | 557 | Standardized* mean difference (> 0 favors BB) | 0.06 | -0.06, 0.18 | 0.3 | ||
| Total exercise time | 32 | 840 | 831 | Standardized* mean difference (> 0 favors BB) | -0.10 | -0.20, 0.00 | 0.05 | ||
| Adverse events leading to withdrawal | 51 | 2,811 | 239 | 2,778 | 317 | Odds ratio (< 1 favors BB) | 0.72 | 0.60, 0.86 | < 0.001 |
BB = beta-blocker; CA = calcium antagonists
This figure shows summary odds ratios for benefit with beta-blockers versus calcium antagonists for cardiac death or myocardial infarction (a) and adverse events (b). Values less than 1.0 favor beta-blockers. A greater rate of adverse events leading to withdrawal was observed with calcium antagonists compared with beta-blockers for the nifedipine, long-acting, and short-acting calcium-antagonist subgroups. BB = beta-blocker, CA = calcium antagonist, Nifed = nifedipine, long = long-acting calcium antagonist, short = short-acting calcium antagonist.
). The pooled odds ratio of these two trials for risk with beta-blockers was 0.98 (0.66,1.47). For the 59 short-term trials, there was no obvious difference in cardiac death or myocardial infarction, odds ratio 0.90 (0.40, 2.00). When both long- and short-term trials were combined (Table 1), the odds ratio for risk with beta-blockers was 0.97 (0.67,1.38). A random-effects model produced similar results: 0.98 (0.67, 1.43). No significant differences were found when the trials were grouped by type of calcium antagonist (nifedipine vs. non-nifedipine) or calcium antagonist duration of action (Figures 4a and 4b).
This figure shows the difference in angina episodes per week between beta-blockers (BB) and calcium antagonists (CA).
The figure shows summary odds ratios for benefit with beta-blockers versus calcium antagonists for angina episodes per week (a) and exercise time to 1-mm ST depression (b). Values less than 0 indicate fewer angina episodes and shorter exercise time with beta-blockers. Because exercise protocols varied, a standardized mean difference (mean difference / standard deviation) in exercise time to ischemia (1-mm ST depression) is displayed. Both nifedipine and short-acting calcium antagonists were associated with a greater frequency of angina compared with beta-blockers. The abbreviations are the same as in Figure 4.
). When compared with calcium antagonists, beta-blockers were associated with 0.31 (0.00, 0.62) fewer episodes of angina per week. When the comparison was restricted to trials comparing beta-blockers with nifedipine, there were 0.63 (0.21, 1.0) fewer angina events per week with beta-blockers (Figure 6a). We did not observe a significant effect on the duration of action of calcium antagonists when the short-acting and long-acting calcium antagonists were compared separately with beta-blockers (Figures 6a and 6b).
This figure shows the difference in nitroglycerin use per week between beta-blockers (BB) and calcium antagonists (CA).
). There was a nonsignificant reduction of 0.14 (-0.19, 0.46) fewer nitroglycerin tablets per week with beta-blockers in those studies that compared beta-blockers with nifedipine. We observed no differences when short-acting and long-acting calcium antagonist preparations were examined separately.
This figure shows the difference in exercise time to ischemia between beta-blockers (BB) and calcium antagonists (CA). Ischemia is defined as 1-mm of ST depression on the electrocardiogram. A standardized mean difference is displayed which equals the difference in means divided by the pooled standard deviation.
). Similarly, no difference in exercise time to ischemia was observed when only trials comparing beta-blockers with nifedipine were evaluated (Figure 6b). No differences were noted when short-acting and long-acting preparations of calcium antagonists were evaluated separately.
This figure shows the individual trial odds ratios for risk of adverse events leading to withdrawal for beta-blockers versus calcium antagonists. Individual trial data shown for studies with > 1 withdrawal in each study arm. Summary estimate based on all trials.
). The absolute difference in the adverse event rate was 2.0 (0.5, 3.4) fewer events with beta-blockers compared with calcium antagonists for every 100 patients treated.
and 4b). In contrast, there was no significant difference in the rate of adverse events between non-nifedipine calcium antagonists and beta-blockers. Adverse events were not significantly different between long-acting and short-acting calcium antagonists when these agents were compared separately with beta-blockers (Figures 4a and 4b).
When we examined individual calcium antagonists, both verapamil and nifedipine had more side effects than the beta-blockers with which they were compared. There were 76 verapamil side effects in 508 patients treated with verapamil (six trials), compared with 47 side effects in 512 patients treated with beta-blockers. There were 56 nifedipine side effects in 380 patients treated with nifedipine (eight trials) compared with 44 beta-blocker side effects in 397 patients treated with beta-blockers. Too few data were available for diltiazem (four trials, 124 patients) to draw conclusions. Peripheral edema was more common with nifedipine (16 vs. 0 with beta-blockers) and diltiazem (4 vs. 0 with beta-blockers). Headache was also more common with nifedipine (26 vs. 11 with beta-blockers), while gastrointestinal disturbance was more frequent with verapamil (43 vs. 15 with beta-blockers) and diltiazem (4 vs. 0 with beta-blockers).
When we limited the trials to those that studied the four most commonly used calcium antagonists and the five most commonly used beta-blockers, we noted significantly fewer angina episodes per week, odds ratio 0.40 (0.05, 0.75), and fewer adverse events, 0.73 (0.59, 0.92) for beta-blockers compared with calcium antagonists. No significant differences in any of the outcome measures were found when the trials were grouped by funding source or study location. Similar results were found when trials of crossover and parallel design were evaluated separately.
We separated trials into high (randomization and double blinding stated) and low (both not stated) quality. Separate analyses of high- and low-quality studies demonstrated similar results.
We compared beta-blockers with dihydropyridine calcium antagonists (nifedipine n = 26, nicardipine n = 5, amlodipine n = 1, felodipine n = 1) separately and found fewer adverse events, odds ratio 0.63 (0.49, 0.80), and angina episodes per week, 0.61 (0.23, 0.99) with beta-blocker therapy. However, nifedipine accounted for 79 percent of all dihydropyridines, and an effect of dihydropyridine therapy independent of nifedipine could not be demonstrated.
| Outcome | Trials | Nitrates | Calcium Antagonists | Measure | Effect Size | 95% CI | p | ||
|---|---|---|---|---|---|---|---|---|---|
| N | Events | N | Events | ||||||
| Cardiac death or MI | 10 | 377 | 0 | 376 | 0 | Odds ratio (< 1 favors Nitrates) | 1.00 | 0.30 3.5 | 1.0 |
| Angina episodes (per week) | 5 | 143 | 143 | Mean difference (< 0 favors Nitrates) | 0.52 | -0.12 1.2 | 0.10 | ||
| Nitroglycerin use (per week) | 5 | 143 | 143 | Mean difference (< 0 favors Nitrates) | 0.19 | -0.23 0.60 | 0.4 | ||
| Time to 1-mm ST depression | 3 | 81 | 81 | Standardized* mean difference (> 0 favors Nitrates) | -0.19 | -0.50 0.11 | 0.2 | ||
| Total exercise time | 5 | 126 | 126 | Standardized* mean difference(> 0 favors Nitrates) | -0.11 | -0.36 1.4 | 0.4 | ||
| Adverse events leading to withdrawal | 10 | 340 | 12 | 339 | 9 | Odds ratio (< 1 favors Nitrates) | 1.48 | 0.26 8.5 | 0.9 |
This figure shows the difference in angina episodes per week between long-acting nitrates (NI) and calcium antagonists (CA).
This figure shows the difference in nitroglycerin use per week between long-acting nitrates (NI) and calcium antagonists (CA).
This figure shows the difference in exercise time to ischemia between long-acting nitrates (NI) and calcium antagonists (CA). Ischemia is defined as 1 mm of ST depression on the electrocardiogram. A standardized mean difference is displayed which equals the difference in means divided by the pooled standard deviation.
-12). Patient characteristics were similar in trials previously described comparing beta-blockers and calcium antagonists. There were no significant differences in outcomes between these classes of medications; however, a trend ( p = 0.1) was noted for greater number of angina episodes per week in patients who were taking long-acting nitrates.
| Outcome | Trials | Beta-Blockers | Nitrates | Measure | Effect Size | 95% CI | p | ||
|---|---|---|---|---|---|---|---|---|---|
| N | Events | N | Events | ||||||
| Cardiac death or MI | 6 | 129 | 1 | 129 | 1 | Odds ratio (< 1 favors BB) | 1.00 | 0.21, 4.7 | 1.0 |
| Angina episodes (per week) | 5 | 86 | 84 | Mean difference (< 0 favors BB) | -0.83 | -4.7, 3.1 | 0.7 | ||
| Nitroglycerin use (per week) | 3 | 53 | 53 | Mean difference (< 0 favors BB) | -1.9 | -0.40, 0.20 | 0.08 | ||
| Time to 1-mm ST depression | 1 | 27 | 27 | Standardized* mean difference (> 0 favors BB) | 0.12 | -0.42, 0.65 | 0.7 | ||
| Total exercise time | 1 | 21 | 21 | Standardized* mean difference (> 0 favors BB) | 0.29 | -0.90, 0.32 | 0.3 | ||
| Adverse events leading to withdrawal | 5 | 124 | 6 | 124 | 13 | Odds ratio (< 1 favors BB) | 0.56 | 0.20, 1.6 | 0.3 |
This figure shows the difference in angina episodes per week between long-acting nitrates (NI) and beta-blockers (BB).
This figure shows the difference in nitroglycerin use per week between long-acting nitrates (NI) and beta-blockers (BB).
,14). Patient characteristics were similar in trials that compared beta-blockers with calcium antagonists. There were no significant differences between beta-blockers and long-acting nitrates in any of the outcomes measured. A trend ( p = 0.08) toward increased nitroglycerin use per week was noted for patients who were taking long-acting nitrates.To determine the effect of excluding articles that were written in languages other than English, we reviewed the abstracts (in English) that were available for 12 studies (50 percent) whose full text was not written in English. None of the studies reported a significant difference in death or myocardial infarction rates, exercise time, angina episodes, or nitroglycerin use for comparisons between beta-blockers and calcium antagonists (n = 6), beta-blockers and nitrates (n = 2), or calcium antagonists and nitrates (n = 4). Two of the studies that compared calcium antagonists and nitrates noted fewer side effects with calcium antagonists.
We identified 18 trials of alternative therapies used for patients who had stable angina. We excluded two studies that were not randomized and four studies of herbal medications that used another herbal therapy as a control. The remaining 12 articles included four of acupuncture (three English, one Chinese language) (Ballegaard, Meyer, and Trojaborg, 1991; Ballegaard, Pedersen, Pietersen, et al., 1990; Richter, Herlitz, and Hjalmarson, 1991; Zhou and Liu, 1993) and eight of herbal therapy (one English, seven Chinese language) (Cai, Xu, and Shen, 1996; Chen, Zhou, and Gao, 1996; Fang, Jiang, and Luo, 1987; Liao, Chen, Wu, et al., 1989; Wang, Shi, and Tu, 1996; Wu, 1990; Zhang, Liang, and Ma, 1995; Zhu, 1994). There were no randomized trials of garlic or chelation therapy in patients with stable angina. Our review was limited to those studies with at least an abstract written in English (all four acupuncture, seven herbal articles).
Three trials randomized patients to acupuncture or sham acupuncture, in which needles are not placed at traditional points (Ballegaard, Meyer, and Trojaborg, 1991; Ballegaard, Pedersen, Pietersen, et al., 1990; Zhou and Liu, 1993). No significant differences were noted in angina episodes or nitroglycerin use; one study, however, noted increased work capacity in patients randomized to acupuncture. The fourth study (Richter, Herlitz, and Hjalmarson, 1991) noted reduced angina and increased time to angina with exercise for patients treated with acupuncture compared with placebo.
Of the seven Chinese herbal articles, six provided abstracts in English (Cai, Xu, and Shen, 1996; Chen, Zhou, and Gao, 1996; Wang, Shi, and Tu, 1996; Wu, 1990; Zhang, Liang, and Ma, 1995; Zhu, 1994). The controls used in these trials were dipyridamole and aspirin (n = 2), nifedipine (n = 1), isosorbide dinitrate (n = 1), and unstated (n = 1). Four trials (80 percent) reported significantly lower angina (Cai, Xu, and Shen, 1996; Wang, Shi, and Tu, 1996; Wu, 1990; Zhu, 1994), and one reported decreased ST depression during exercise in patients randomized to herbal therapy (Zhu, 1994). In the one herbal trial published in English, 32 patients were randomized to either an intravenous combination of radix ginseng, radix astragali and radix angelicae sinesis, or 10 percent dextrose. Both therapies were given as daily infusions for 2 weeks (Liao, Chen, Wu, et al., 1989). Significant improvements in angina and exercise endurance and ST depression were noted in the herbal group.
This quantitative overview shows that calcium antagonists were associated with a greater number of angina episodes and adverse events when compared with beta-blockers in trials of patients who have stable angina. Beta-blockers and nonnifedipine calcium antagonists had similar effects on adverse events, symptoms, and exercise tolerance. Trial duration was too short to determine differences in long-term rates of survival or myocardial infarction between calcium antagonists and beta-blockers.
Our finding that beta-blockers were tolerated as well as calcium antagonists was not expected. Past reviews have noted frequent side effects with beta-blockers (Hoekenga and Abrams, 1984 ), whereas calcium antagonists have been recommended and widely used because of their low side-effect profile (Chan, Heo, Garibian, et al., 1988 ; Fox, Jespersen, Ferrari, et al., 1997 ). Our findings must be tempered by noting that patients who were thought unlikely to tolerate beta-blockers (patients with congestive heart failure, heart block, or significant pulmonary disease) were generally excluded from the randomized studies. Thus, our findings cannot be extrapolated to patients who have these conditions. However, our findings suggest that patients who do not have these contraindications should have equal or greater success in finding angina relief and freedom from adverse events when they use a beta-blocker instead of a calcium antagonist.
Recently, the safety of calcium antagonists for patients who have coronary artery disease has been questioned. Prior meta-analyses and case-control studies have demonstrated an increased risk of myocardial infarction in hypertensive patients who take short-acting calcium antagonists (Psaty, Heckbert, Koepsell, et al., 1995 ). In patients who have acute coronary syndromes, mortality increased with increasingly large doses of short-acting nifedipine (Furberg and Psaty, 1996 ; Furberg, Psaty, and Meyer, 1995 ). In a recent case-control study, short-acting calcium antagonists were associated with a four-fold increase in the risk of cardiovascular deaths and of hospital admission when compared with the risk from beta-blockers (Alderman, Cohen, Roque, et al., 1997 ). Our study was unable to determine whether important differences in mortality or myocardial infarction rates exist in the stable angina population between anti-anginals because followup was less than 6 months in all but two studies. This important question will have to be addressed by larger, long-term comparisons of these agents.
Our study suggests that nifedipine may not be as well tolerated as other calcium antagonists. This finding is consistent with past investigations of long-acting nifedipine. In six small randomized studies of at least 1-week duration that compared long-acting nifedipine with other calcium antagonists, long-acting nifedipine was associated with more symptoms and side effects in two studies (Licciardello, Privitera, Calvi, et al., 1997 ; Pandolfo, Pajes, Zardi, et al., 1997) and was equivalent to other calcium antagonists in four studies (Donaldson, Dawkins, and Waller, 1993 ; Schulte, 1995 ; Siu, Jacoby, Phillips, et al., 1993 ; Zanolla, Franceschini, Rossi, et al., 1997 ).
European data are consistent with the U.S. findings that calcium blockers are preferred over beta-blockers for first-line therapy for patients who have stable angina (Fox, Jespersen, Ferrari, et al., 1997 ). These findings contrast with our results that beta-blockers were equivalent to or better than long-acting nitrates and calcium antagonists in tolerability and in angina reduction. Because patients frequently tolerate the first drug with which they are treated, it is likely that the reported drug use data reflect the physicians' first choice of agents. An alternative explanation for the disparity is that effectiveness and tolerability differ substantially between the trial and practice setting, such that patients are frequently switched from beta-blockers to other therapy because of problems not measured in the trials.
The choice of a first-line agent for use in patients who have stable angina has been addressed by recent clinical guidelines. The European Society of Cardiology recommends beta-blockers for patients who have clear-cut effort-related angina or a prior myocardial infarction (Management of stable angina pectoris, 1997 ). The North of England Stable Angina Guideline Development Group recommends beta-blockers for all patients who have no contraindications (North of England evidence-based guidelines, 1996 ), citing evidence that beta-blockers lower mortality after myocardial infarction, reduce mortality for patients who do have a subsequent infarction while on medication, and reduce adverse vascular events in patients who have hypertension. Data from our study support these recommendations. We add evidence that beta-blockers are tolerated as well as or better than calcium antagonists and long-acting nitrates, and they provide angina relief equivalent to these agents.
Despite the effectiveness of current medical treatment, there are patients whose symptoms are refractory to beta-blockers, calcium antagonists, nitrates, and revascularization, or who are unable to tolerate these conventional treatments. These patients may seek nonconventional (alternative) treatments for their angina. Although patients frequently note relief with these therapies, often it is not clear whether the benefit is caused by a direct effect of the therapy or a placebo effect. Randomized placebo-controlled trials are needed to separate direct effect from placebo. Our review identified only two therapies (acupuncture and herbal therapy) that have been examined with this degree of rigor. In two small trials, traditional acupuncture was found to be superior to sham acupuncture in terms of improved exercise time but was not shown to affect symptoms. Our review of herbal therapy is limited because the few trials using placebo, calcium antagonist, or nitrate controls were published in Chinese, and the English abstracts were unclear. Randomized trials of alternative treatments for angina are just beginning, and only a limited number of patients have been evaluated. Thus, at this time we have no clear indication of the benefit or harm from these therapies for patients who have stable angina.
Like those of any meta-analysis, our findings may have been affected by publication bias (Easterbrook, Berlin, Gopalan, et al., 1991 ). It is possible that studies that failed to demonstrate significant differences were not submitted or were not accepted for publication. However, few (25 percent) of the studies included in the meta-analysis demonstrated significant differences between drugs, suggesting that any such bias may be small. Rarely did the studies include patients who had congestive heart failure, chronic obstructive lung disease, or nonexertional chest pain; thus, our findings are not applicable to these types of patients. Only a few small studies compared long-acting nitrates with calcium antagonists or beta-blockers. Therefore, our analysis was unlikely to detect any clinically important differences between these agents.
The majority of patients enrolled in the trials were men; thus, it is unclear if women respond similarly to treatment. Outcomes for each trial were not listed separately by age, sex, or ethnicity subgroup so that the effect of these variables on response to treatment could not be determined. Our search strategy using MEDLINE and EMBASE may not have identified all randomized trials of alternative therapies. We did not include studies reported in foreign-language publications in our meta-analysis. A recent study has found that studies published in English are more likely to have p values < 0.05 (Egger, Zellweger-Zahner, Schneider, et al., 1997 ). However, a review of the English abstracts that were available (50 percent of foreign language studies) did not demonstrate a difference in significant findings.
Our study did not examine measures of quality of life. Although adverse events severe enough to lead to withdrawal from the study occurred less frequently in patients treated with beta-blockers than in those treated with calcium antagonists, less severe side effects were not consistently reported, so we could not adequately measure differences in their rates. Commonly reported side effects were similar or greater in patients treated with calcium antagonists compared with patients who were given beta-blockers. Longer term studies are required if we are to determine differences in safety between classes of anti-anginals. Since the studies in our analyses were predominantly of short duration, cardiac events were rare and the power to detect differences in cardiac events was low.
Our quantitative review suggests that beta-blockers provide equivalent or greater reduction in angina episodes per week and lead to similar or reduced rates of adverse events, when compared with either calcium antagonists or long-acting nitrates. The differences in rate of adverse events and frequency of angina between beta-blockers and calcium antagonists are accounted for in large part by nifedipine, which was associated with a higher rate of adverse events and angina symptoms. There were no differences in angina episodes per week or adverse event rates between nonnifedipine calcium antagonists and beta-blockers. Further studies are required to determine whether the different therapies are associated with a significant difference in mortality in patients who have stable angina. There is insufficient evidence to comment on the effectiveness of alternative therapies.
Few long-term studies have compared treatment with beta-blockers, calcium antagonists, and long-acting nitrates in patients who have stable angina. Additional large randomized trials are needed to determine if clinically important differences in patient survival exist between classes of anti-anginal medication. Further studies are needed to examine the effects of treatments for stable angina in patients with comorbidity such as heart failure or chronic obstructive pulmonary disease. Additional randomized controlled trials of alternative therapies in this patient population are necessary to reach conclusions about efficacy and safety.
Evidence Tables Abbreviation Key
BB: beta-blockers
CA: calcium antagonists
CABG: coronary artery bypass graft
CAD: coronary artery disease
CNS: central nervous system
COPD: chronic obstructive pulmonary disease
Diff: difference
GI: gastrointestinal
MI: myocardial infarction
N: number
NI: nitrate
SD: standard deviation
Free Full text in PMC] [PubMed]Free Full text in PMC] [PubMed]Free Full text in PMC] [PubMed]Free Full text in PMC] [PubMed]Free Full text in PMC] [PubMed]
