| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the Departments of Anesthesiology and Public Health Sciences, Wake Forest University School of Medicine, Winston-Salem, North Carolina
Address correspondence and reprint requests to John Butterworth, MD, Professor and Section Head, Cardiothoracic Anesthesiology, Department of Anesthesiology, Wake Forest University School of Medicine, Medical Center Boulevard, Winston-Salem, NC 21757-1009. Address e-mail to jbutter{at}wfubmc.edu
There has long been interest in reducing cardiovascular complications after surgical operations (specifically, myocardial ischemia, myocardial infarction [MI], or cardiac death). In this issue of Anesthesia & Analgesia, two articles are presented on this general topic. The first article systematically reviews various drug options for prevention of adverse cardiac outcomes during and after noncardiac surgery (1). The second article describes a meta-analysis of outcomes with the use of calcium channel blockers (CCBs), concluding, in partial conflict with the results of the first, that CCBs may be underused primarily because of a "North American bias" against this drug class (2).
Stevens et al. (1) systematically reviewed all randomized controlled trials (RCTs) of ß-adrenergic receptor blockers, CCBs, 
2 agonists, or nitrates for prevention of cardiovascular complications in noncardiac surgery. Their objective was to obtain estimates of the relative efficacy of these drug classes for preventing myocardial ischemia, MI, and cardiac death. Twenty-one studies were ultimately included out of 1,093 that were screened. There were analyzable data on 3,646 patients; the largest included trial (of the 2 agonist mivazerol) enrolled 1,897 subjects. Ten trials reported data on prevention of intraoperative myocardial ischemia. ß-adrenergic receptor blockers markedly reduced the risk (the odds ratio [OR] was 0.32 with 95% confidence interval [CI] 0.17–0.58 compared with placebo). For 2 agonists the OR was 0.47 (CI 0.33–0.68). Diltiazem had no significant effect on intraoperative myocardial ischemia in one small study. Five trials reported data on postoperative myocardial ischemia. ß-adrenergic receptor blockers were associated with an OR of 0.46 (CI 0.26–0.81). There were no significant effects of 2 agonists, CCBs, or nitrates.
ß-adrenergic receptor blockers also reduced the likelihood of nonfatal perioperative MI (OR 0.19, CI 0.08–0.48). There were no significant treatment benefits of 2 agonists, CCBs, or nitrates. ß-adrenergic receptor blockers reduced the likelihood of perioperative cardiac death (OR 0.25, CI 0.09–0.73). Alpha2 agonists also reduced the likelihood of cardiac death (OR 0.50, CI 0.28–0.91). There were no trials reporting cardiac deaths with CCBs or nitrates.
These authors appropriately conclude that ß-adrener-gic receptor blockers offer protection against car-diac morbidity and mortality after noncardiac surgery. Moreover, 2 agonists may offer protection against perioperative cardiovascular complications.
The article by Wijeysundera and Beattie (2) presents a contrasting view. These authors identified 11 studies including a total of 1,007 subjects in which the actions of diltiazem, verapamil, or nifedipine at reducing perioperative myocardial ischemia, MI, heart failure, supraventricular tachyarrhythmias, or cardiac death were assessed in those undergoing major noncardiac surgery. The smallest study had 25 patients, the largest had 336. Six of 11 studies were double-blind. Only one study was stated to have both double-blindness and concealed allocation, although we do not understand how double-blindness is possible without concealed allocation. The trials included were not restricted to those conducted for the purpose of evaluating cardiac outcome. Only 2 of the 11 trials were conducted in North America. Approximately half were published in English.
CCBs reduced by half perioperative ischemia (relative risk [RR] 0.49, CI 0.30–0.80) and the incidence of supraventricular tachyarrhythmia (RR 0.52, CI 0.37–0.72). The meta-analysis failed to detect significant effects of CCBs on heart failure (RR 0.60, CI 0.08–4.81), MI (RR 0.25, CI 0.05–1.18), or mortality (RR 0.40, CI 0.14–1.16). Post hoc analyses suggested a significant reduction in combined death and MI (RR 0.35, CI 0.15–0.86) associated with CCBs. These authors concluded that the strongest evidence applied to diltiazem, and that a large RCT was indicated for this drug. They also opined that the CCBs were under used for cardiac protection due to a North American bias. So, how can one resolve the apparent contradictory conclusions of these two studies regarding CCBs?
There are multiple lines of evidence supporting the use of ß-adrenergic receptor blockers for prevention of secondary cardiac outcomes after major noncardiac surgery. First, there is good evidence from large RCTs that these drugs reduce mortality in patients with coronary heart disease in the absence of surgery (3,4). Second, large RCTs confirm that ß-adrenergic receptor blockers reduce mortality in patients with hypertension and chronic heart failure (5,6). Importantly, the side effect profile of ß-adrenergic receptor blockers is well known in patients with known or suspected coronary heart disease. In recent years, convincing evidence has accumulated that perioperative administration of ß-adrenergic receptor blockers will reduce the postoperative incidence of these cardiovascular complications, particularly after major peripheral vascular surgery (7–11). At this point, we believe that investigations should focus on identifying which patients (with or at risk for coronary heart disease) do not need ß-adrenergic receptor blockers when undergoing noncardiac surgery (rather than on whether these drugs are effective) (12).
The case for 2 agonists also having beneficial effects is getting stronger, but further study may be warranted. However, according to the International Conference on Harmonisation and guidance documents from the United States Food and Drug Administration, inert placebos should not serve as the comparison to a new therapy in any randomization trial in this patient population (13,14). In general, the best available therapy should be offered to patients in an ethical clinical trial. Thus, new trials with 2 agonists should be conducted either in comparison with active treatment with a ß-adrenergic receptor blocker or compared to inert placebo in a trial where all patients (in both treatment and control groups) also received a ß-adrenergic receptor blocker.
This proviso regarding placebos also applies to future studies of CCBs. It hardly requires repetition that CCBs were introduced for treatment of angina and hypertension without adequate outcome studies showing mortality/morbidity benefit. Therefore, no one should have been surprised when postmarketing analyses indicated that outcomes were inferior to other, better characterized drugs. Indeed, there is a large literature on the adverse effects of CCBs, including an increased risk of gastrointestinal bleeding, surgical bleeding, cancer, and MI (15–19). The responses of the pharmaceutical industry to this challenge have also been well documented (20).
As a consequence of the CCB controversy, large outcomes trials with CCBs were undertaken in patients with hypertension. The ALLHAT study (21) compared the CCB amlodipine with chlorthalidone (diuretic) and lisinopril (angiotensin converting enzyme inhibitor). Thirty-three thousand, three hundred and fifty-seven subjects were randomized. There was no difference among the groups in the primary outcome (fatal congestive heart disease or nonfatal MI). Amlodipine was associated with a greater risk of heart failure (RR 1.38, CI 1.25–1.52). The CONVINCE Trial (22) of 16,602 subjects at 661 sites, designed to demonstrate the equivalence of cardiovascular outcomes with controlled-onset verapamil as compared with hydrochlorothiazide or atenolol, actually showed that controlled-onset verapamil failed to meet the specified criterion for noninferiority to either hydrochlorothiazide or atenolol. Moreover, controlled-onset verapamil was associated with a greater incidence of hemorrhage (hazard ratio 1.54, CI 1.16–2.04) than the two comparator drugs. The INSIGHT trial (23) of sustained-release nifedipine in hypertension did indeed show comparable outcome between this drug and co-amilozide (combination of hydrochlorothiazide and amiloride) in the 6,321 randomized subjects. There was no between-group difference in the primary composite outcome of cardiovascular death, MI, heart failure, or stroke (RR 1.10 CI 0.91–1.34). However, there were more fatal MIs (OR 3.22, CI 1.18–8.80) and more nonfatal episodes of heart failure (OR 2.20, CI 1.07–4.49) among those receiving nifedipine. The ACTION trial of extended release nifedipine in effort-induced angina is currently underway. In summary, there is no conclusive evidence that treatment with CCBs improves cardiovascular outcomes relative to established drugs in patients with chronic diseases such as hypertension and chronic heart failure (5).
We do not dispute the hypothesis of Wijeysundera and Beattie that CCBs may be effective at improving certain cardiac outcomes after noncardiac surgery. However, an adequate test of this hypothesis requires large, properly designed RCTs. We do not believe that the use of CCBs is determined by a North American bias. Indeed, we note that these authors have recently drawn similar favorable conclusions about the benefit of CCBs in cardiac surgery (24). We believe that most clinicians have wisely elected to postpone their adoption of CCBs for cardiac protection in favor of ß-adrenergic receptor blockers that have been validated in adequately powered RCTs. To those who favor CCBs, rather than ß-adrenergic receptor blockers, we say: Provide the scientific evidence.
References
This article has been cited by other articles:
|
|
 |
|
  P.J. Devereaux, L. Goldman, S. Yusuf, K. Gilbert, K. Leslie, and G. H. Guyatt Surveillance and prevention of major perioperative ischemic cardiac events in patients undergoing noncardiac surgery: a review Can. Med. Assoc. J., September 27, 2005; 173(7): 779 - 788. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P J Devereaux, W S. Beattie, P. T-L Choi, N. H Badner, G. H Guyatt, J. C Villar, C. S Cina, K. Leslie, M. J Jacka, V. M Montori, et al. How strong is the evidence for the use of perioperative {beta} blockers in non-cardiac surgery? Systematic review and meta-analysis of randomised controlled trials BMJ, August 6, 2005; 331(7512): 313 - 321. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Comfere, J. Sprung, M. M. Kumar, M. Draper, D. P. Wilson, B. A. Williams, D. R. Danielson, L. Liedl, and D. O. Warner Angiotensin System Inhibitors in a General Surgical Population Anesth. Analg., March 1, 2005; 100(3): 636 - 644. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
