Rapid Responses to:

PAPERS: P J Devereaux, W Scott Beattie, Peter T-L Choi, Neal H Badner, Gordon H Guyatt, Juan C Villar, Claudio S Cinà, Kate Leslie, Michael J Jacka, Victor M Montori, Mohit Bhandari, Alvaro Avezum, Alexandre B Cavalcanti, Julian W Giles, Thomas Schricker, Homer Yang, Carl-Johan Jakobsen, and Salim Yusuf How strong is the evidence for the use of perioperative blockers in non-cardiac surgery? Systematic review and meta-analysis of randomised controlled trials BMJ 2005; 331: 313-321 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Negative comments are not supported by the reported results

J Robert Sneyd   (6 August 2005)

Leave beta blockade to the anaesthetist and not the epidemiologists

Ewan S Jack   (8 August 2005)

Cardiac risk stratification guides the use of beta blockers in patients undergoing non cardiac surgery

Jawad M Khan, Timothy J Watson Specialist Registrar in Medicine   (9 August 2005)

Not convincing enough to change my practice

Som Kumar   (10 August 2005)

Biased data interpretation should not be allowed

Saul G Myerson   (13 August 2005)

Perioperative blocker therapy in patients undergoing noncardiac surgery

P.J. Devereaux, W. Scott Beattie, Gordon H. Guyatt, Juan C. Villar, Peter T-L Choi, Neal H. Badner, Claudio S. Cin¨¤, Michael J. Jacka, Victor M. Montori, Mohit Bhandari, Alvaro Avezum, Thomas Schricker, Homer Yang, Carl-Johan Jakobsen, Salim Yusuf   (15 August 2005)

Intra-operative hypotension is a major problem of perioperative beta-blockade

Janet T Powell, Matthew Sydes   (17 August 2005)

Why do anaesthetists love beta-blockers?

Mike Grocott, Denny ZH Levett, Monty G Mythen   (22 August 2005)

Peri-operative Beta-blockers: Much room for evidence still exists!

Ashish Aneja, Brian Harte ; Vaishali Singh ; and Amir K Jaffer   (29 September 2005)

Negative comments are not supported by the reported results 6 August 2005
J Robert Sneyd, Professor of Anaesthesia Peninsula medical School, Plymouth PL6 8DH Send response to journal: Re: Negative comments are not supported by the reported results	Peri-operative betablockade is a 'hot' topic in anaesthesia/peri- operative medicine and this meta-analysis is therefore timely. It is however dissapointing that the authors (and possible the BMJ editorial team) go on to make non evidenced comments which are at odds with common sense and clinical experience. Let us start with the editorial team. In the section 'This week in the BMJ' I read 'Perioperative beta blockers may do more harm then good' and the authors themselves seem to regard all peri -operative adverse events as equivalent in importance. Both are wrong. The listed authors all come from an epidemiology department and none are listed as practising anaesthesia - perhaps this is the problem. Bradycardia and hypotension are common during and after anaesthesia and are usually easily managed. In contrast, non-fatal myocardial infarction, non-fatal cardiac arrest are uncommon and hard to manage. The anaesthesia community waits anxiously for large prospective studies such as POISE. In the meantime, a simple and economical intervention which appears to reduce serious cardiovascular outcomes is hard to ignore. Competing interests: None declared
Leave beta blockade to the anaesthetist and not the epidemiologists 8 August 2005
Ewan S Jack, SpR Intensive care/anaesthesia Victoria Infirmary, GLASGOW G42 9TY Send response to journal: Re: Leave beta blockade to the anaesthetist and not the epidemiologists	Editor - perioperative beta blockade in at risk patients is a movement that is gaining widespread support in the field of anasethesia. It is, therefore, somewhat disappointing to read the conflicting material in this weeks BMJ. The authors of the meta-analysis (Devereaux et al) are not perioperative physicians/anaesthetists and fail to realise the frequency and ease with which bradycardia and hypotension occur during anaesthesia. Cardiac complications are much more serious and harder to remedy, the editorial comments in 'this week in the BMJ' will cause more harm than good. Doubling the risk of innocent side effects is immaterial if we have a treatment that can reduce cardiovascular morbidity. The printed word in this issue then confuses us even more with the summarisation of the excellent paper in the New England Journal of Medicine in the short cuts section. This study, admittedly retrospective, stratifies patients into cardiac risk groupings in much the way that practicing anaesthetists do on a day to day basis. This paper results in the advice to continue prescribing beta blockers. The context of beta blockade and who prescribes it should be kept in mind when making editorial comments about such papers. We all await the POISE study with interest. Competing interests: None declared
Cardiac risk stratification guides the use of beta blockers in patients undergoing non cardiac surgery 9 August 2005
Jawad M Khan, Specialist Registrar Cardiology University Department of Medicine, City Hospital, Birmingham, B18 7QH, Timothy J Watson Specialist Registrar in Medicine Send response to journal: Re: Cardiac risk stratification guides the use of beta blockers in patients undergoing non cardiac surgery	The authors conclude ‘the evidence that perioperative beta blockers reduce major cardiovascular events is encouraging but too unreliable to allow definitive conclusions to be drawn’. The meta-analysis included all studies that evaluated beta blocker treatment in patients having non- cardiac surgery. The resultant heterogeneity caused by this approach explains the dilution of the benefits of beta blockade. A number of randomised controlled studies(1,2,3) have shown clear benefit from the use of beta blockers in patients with or at risk of coronary artery disease (two of the following: age >65yrs, hypertension, smoker, diabetes mellitus, cholesterol > 6.2 mmol/L.)(1) in terms of overall survival and reduction in cardiovascular morbidity. Benefits of up to 90% reduction in death from cardiac causes and myocardial infarction have been shown with the use of beta blockers in high risk patients undergoing vascular surgery.(3) In a meta-analysis of six randomized trials involving 694 surgical patients, beta-blockers were associated with a 75 percent reduction in the risk of perioperative death from cardiac causes.(4) The main mechanism of action is multifactorial. Increased levels of serum catecholamines correlate with the peak incidence of tachycardia, myocardial ischaemia and myocardial ischaemic complications postoperatively.(5) This is driven by the sympathetic nervous system and results in elevation of myocardial oxygen consumption and hence myocardial ischaemia in patients with coronary artery disease.(7) Beta blockade statistically decreases duration and frequency of silent ischaemia during surgery.(8) A large retrospective cohort study of perioperative beta-blocker therapy showed a reduced risk of in-hospital death among high-risk, but not low-risk, patients undergoing major noncardiac surgery. The relationship between perioperative beta-blocker treatment and the risk of death varied directly with cardiac risk; among the 580,665 patients with an Cardiac Risk Index (RCRI) score of 0 or 1, treatment was associated with no benefit and possible harm, whereas among the patients with an RCRI score of 2, 3, or 4 or more, the adjusted odds ratios for death in the hospital were 0.88 (CI 0.80-0.98), 0.71 (CI 0.63-0.80), and 0.58 (CI 0.50 to 0.67), respectively. (9) Therefore, in summary, patients with coronary artery disease or at high risk of having undiagnosed or sub-clinical coronary artery disease should be treated with perioperative beta blockade. However, there is no clear data to suggest patients at ‘low risk’ should be routinely treated. This demonstrates the importance of cardiac risk statification prior to elective non cardiac surgery. However, there is lack of data on the timing of prescriptions for beta-blockers relative to surgery and this requires further research. References: 1. Mangano D, Layug E, Wallace A, et al. The effect of Atenolol on mortality and cardiovascular morbidity after noncardiac surgery. The New England Journal of Medicine 1996;335:1713-1720. 2. Wallace A, Layug, B, Tateo M, et al. Prophylactic Atenolol reduces postoperative Myocardial Ischaemia. Anesthesiology 1998;88:7-17. 3. Poldermans D, Boersma E, Bax J, et al. The effect of Bisoprolol on perioperative mortality and myocardial infarction in high-risk patients undergoing vascular surgery. The New England Journal of Medicine 1999;341(24):1789-1794. 4. Stevens RD, Burri H, Tramer MR. Pharmacologic myocardial protection in patients undergoing noncardiac surgery: a quantitative systematic review. Anesth Analg 2003;97:623-633. 5. Badner N, Knill R, Brown J. Myocardial infarction after noncardiac surgery. New England Journal of Medicine 1996;335:1713-1720. 6. Mangano D, Holleneberg M, Fegert G. Perioperative myocardial ischemia in-patients undergoing surgery-1. Incidence and severity during 4 day postoperative period. Journal American College of Cardiology 1991;17:843-850. 7. Sutton T, Rock P. Beta-Blockers should be used in all patients undergoing vascular surgery. http://www.jcardioanaesthesia.com/abs13_4/v13n4p490.html 8 . Pasternack P, Imparato A, Baumann G, et al. The Hemodynamics of Beta-blockade in patients undergoing abdominal aortic aneurysm repair. Circulation 1987;76(3):1-7. 9. Lindenauer PK, Pekow P, Wan K, et al. Perioperative Beta-Blocker Therapy and Mortality after Major Noncardiac Surgery. 2005;353(4):349-361. Competing interests: None declared
Not convincing enough to change my practice 10 August 2005
Som Kumar, SpR in Cardiology Wythenshawe Hospital, Southmoor Road, M23 9LT Send response to journal: Re: Not convincing enough to change my practice	I read your article with interest. As a senior SpR in Cardiology in a tertiary cardiac centre I see a large number of referrals for patients with angina due for urgent non cardiac surgery. A particularly difficult subgroup is patients with mild predictable angina (no ACS) at moderate workload who need imminent non cardiac surgery. I always advocate aggressive beta blockade in such patients who do not have extensive myocardial ischemia as the combined risk of coronary angiography,revascularisation and subsequent (often vascular) surgery is lower than the 3.4% incidence of major perioperative complications(POLDERMANS, NEJM 1999)with aggresive bisoprolol therapy. As pointed out in the article, caution is needed if such patients need thoracic epidural as the risks of hypotension and bradycardia is high. However the latter are easily detected early by close hemodynamic monitoring, easily corrected and has therefore never deterred my colleague anaesthesists from using perioperative Beta blockers. Perhaps more important issues which this article did not address are to considerthe inadequacy of beta blockade due to late initiation, reluctance to use intravenous short acting agents when oral intake is delayed due to surgical complications and the abrupt discontinuation in the post operative phase leading to rebound ischemia and hypertension. Competing interests: None declared
Biased data interpretation should not be allowed 13 August 2005
Saul G Myerson, Clinical Lecturer and SpR in Cardiology Oxford University Dept. of Cardiovascular Medicine, John Radcliffe Hospital, Headley Way, Oxford OX3 Send response to journal: Re: Biased data interpretation should not be allowed	I am amazed that this paper has been allowed to be published in its current form, when the manipulation of the data and the conclusions are so incredibly biased. This paper is not an unbiased reporting of a meta- analysis - it is nothing short of a propanganda exercise to justify the purpose of the POISE trial. That all the authors are participating in the POISE trial should raise serious questions about their impartiality. I am surprised that the BMJ and its reviewers have allowed this through. I have no view on the issue to defend - I do not know whether beta- blockers are useful in peri-operative situations, and further doubt has also been cast in last week’s New England Journal of Medicine (Lindenauer et al). In addition, I accept all the limitations of meta-analyses, especially when there is heterogeneity in the studies, and the important need for a randomised controlled trial, such as the POISE trial. However, the presentation and interpretation of the data in this paper are hopelessly biased and there is no need for this. They describe a relative risk of death, MI or cardiac arrest of 0.44 as ‘slight’. Had this been a trial which the authors supported, it would probably have read in bold letters: “beta-blockers halve the rate of serious cardiac events”. A risk reduction of 56% is not what I would consider “slight” and when the standard confidence limits of 95% meant that this fell into ‘statistical significance’, the limits were upped to 99%. I wonder if the same would be applied to the POISE trial results? I agree with Prof Sneyd above that from a clinical perspective, bradycardia and hypotension are far less serious than death, cardiac arrest or MI and if there truly was a halving of these, this would be worth the increase in bradycardia and hypotension. We desperately need a randomised controlled trial, and the POISE trial and its participants should be commended for conducting one, so there is no need for shoddy reporting such as this. Competing interests: None declared
Perioperative blocker therapy in patients undergoing noncardiac surgery 15 August 2005
P.J. Devereaux, cardiologist Clinical Epidemiology & Biostatistics, McMaster University Health Sciences Centre, Room 2C8, 1200 Ma, W. Scott Beattie, Gordon H. Guyatt, Juan C. Villar, Peter T-L Choi, Neal H. Badner, Claudio S. Cin¨¤, Michael J. Jacka, Victor M. Montori, Mohit Bhandari, Alvaro Avezum, Thomas Schricker, Homer Yang, Carl-Johan Jakobsen, Salim Yusuf Send response to journal: Re: Perioperative blocker therapy in patients undergoing noncardiac surgery	Dr. Jack charges, "the authors of the meta-analysis are not perioperative physicians/anaesthetists," suggesting that this compromises our ability to undertake and interpret a perioperative beta-blocker meta- analysis. We agree that it is important that a systematic review team include both clinical experts and methodologic experts. As it turns out, all the authors are practicing physicians and 9 are practicing anesthesiologists. Dr. Sneyd states that, "the authors themselves seem to regard all peri-operative adverse events as equivalent in importance." This is erroneous. Our meta-analysis points out that the evidence that beta- blockers prevent major cardiovascular events (i.e., cardiovascular death, nonfatal myocardial infarction, and nonfatal stroke) is weak, whereas the evidence that perioperative beta-blockers cause adverse events (i.e., hypotension and bradycardia requiring treatment) is much stronger. If subsequent large trials show beta-blockers do not prevent major cardiovascular events then patients receiving beta-blockers are subject to risks without benefit. Dr. Jack states, "doubling the risk of innocent side effects is immaterial." We disagree: perioperative bradycardia and hypotension requiring treatment, in the setting of hemodynamically significant coronary artery disease, may lead to myocardial infarction.[1] Drs. Khan and Watson are concerned that our, "meta-analysis included all studies that evaluated beta blocker treatment in patients having non- cardiac surgery," and that "the resultant heterogeneity... explains the dilution of the benefits of beta blockade." They are right that the impact of beta-blockers on major cardiovascular events differed substantially between studies (I2 = 42%). As, we pointed out, however, the only variable that explained this heterogeneity is the quality of the trials. The 3 methodologically weak trials (these trials were stopped early after an interim analysis suggested a much larger than predicted treatment effect or there was failure to blind patients, health care providers, or data collectors) by Poldermans, Zaug, and Urban suggested a sixfold larger relative risk reduction than did the high quality trials that failed to demonstrate a statistically significant result.[2-4] Drs. Khan and Watson go on to selectively reference two of the weaker studies, those by Mangano and Poldermans. The trial by Mangano and colleagues randomised 200 patients undergoing noncardiac surgery to receive atenolol or placebo for a maximum of 7 days post surgery.[5] This trial reported a statistically significant 55% relative risk reduction in death (analysis based upon 30 deaths) at 2 year follow-up with atenolol therapy. The authors, however, excluded the deaths that occurred while patients were taking the study drug. When one adheres to the intention-to -treat principle and includes these deaths, the results are no longer statistically significant.[6] Poldermans and colleagues randomised 112 patients who had a positive dobutamine stress echocardiography study and were undergoing elective vascular surgery.2 These investigators discontinued their unblinded trial after an interim analysis based upon 20 outcomes within 30 days of surgery had demonstrated an implausible 100% relative risk reduction in myocardial infarction and an equally implausible 80% relative risk reduction in cardiac death with bisoprolol therapy. Empiric data warns us to be skeptical of unexpected large treatment effects in studies terminated early.[7 8] Further, these results are inconsistent with the much larger and methodologically stronger vascular surgery trial by Yang and colleagues that failed to demonstrate any effect on major perioperative cardiovascular events.[9] Drs. Khan and Watson also point to a recent retrospective observational database study that was published in the New England Journal of Medicine to support their position that beta-blockers are effective in patients undergoing noncardiac surgery.[10] History, however, warns us to interpret observational drug studies with caution (e.g., hormone replacement therapy). Dr. Myerson charges that our meta-analysis is a, "manipulation of the data and the conclusions are so incredibly biased," and he states that our meta-analysis is nothing more than, "shoddy reporting." Dr. Myerson's strong language does not strengthen the weakness of the evidence supporting the effect of beta-blockers on peri-operative cardiac event. He states that we, "describe a relative risk of death, MI or cardiac death of 0.44 as 'slight.'" Had Dr. Myerson read our paper more carefully, he would have seen that the very opposite is true. Our paper states, "the evidence on perioperative beta-blockers from our meta-analyses suggests a large treatment effect (56% relative risk reduction in major perioperative cardiovascular events). This treatment effect, however, is inconsistent with the results of the beta-blocker trials in myocardial infarction and congestive heart failure that have randomised more than 50 000 patients and have shown moderate treatment effects (i.e., relative risk reductions of 15-35%).[11-15] If perioperative beta-blockers prevent major perioperative cardiovascular events they probably do so through suppressing adrenergic activity. Therefore, large treatment effects are unlikely, because a substantial number of perioperative cardiovascular pathogenic mechanisms that beta-blockers do not affect remain (increased platelet reactivity, plasminogen activator inhibitor I, factor VIII- related antigen levels, and inflammation; decreased antithrombin III concentrations)."[16-19] Dr. Myerson states, "when the standard confidence limits of 95% meant that this fell into 'statistical significance', the limits were upped to 99%." In our paper we state, "when (as with small trials with few or a moderate number of events) statistical significance depends on a difference of only a handful of events, the 99% confidence interval may better convey our confidence in the estimate of the treatment effect." Dr. Myerson also states, "I wonder if the same would be applied to the POISE trial results." The POISE trial will observe a much larger number of events and our rationale suggests that a 95% confidence interval is appropriate in this setting. Physicians are inundated with guidelines and opinions recommending beta-blocker therapy for patients undergoing noncardiac surgery. Our meta -analysis should encourage physicians to set aside these opinions and evaluate the evidence. Three of the authors share our anticipation for the results of POISE. This suggests that, putting aside the strong language they have used, there is little disagreement in our positions. Beta-blockers are a promising intervention for reducing perioperative cardiovascular events, but appreciable doubt regarding their impact remains. Given the well- established adverse effects of beta-blockers in this context, it becomes important to definitively answer the question. Fortunately, the answer is imminent. We invite and encourage all physicians involved in the care of patients undergoing noncardiac surgery to join the POISE trial, or other ongoing trials, to resolve a controversy that, as the authors have demonstrated, remains of great interest. P.J. Devereaux, W. Scott Beattie, Gordon H. Guyatt, Juan C. Villar, Peter T-L Choi, Neal H. Badner, Claudio S. Cina, Michael J. Jacka, Victor M. Montori, Mohit Bhandari, Alvaro Avezum, Julian Giles, Thomas Schricker, Homer Yang, Carl-Johan Jakobsen, Salim Yusuf REFERENCES: 1. Devereaux P, Goldman L, Cook D, Gilbert K, Leslie K, Guyatt G. Perioperative cardiac events in patients undergoing noncardiac surgery - the magnitude of the problem, the pathophysiology, and methods to estimate and communicate risk: A review. CMAJ (in press) 2005. 2. Poldermans D, Boersma E, Bax JJ, Thomson IR, van de Ven LL, Blankensteijn JD, et al. The effect of bisoprolol on perioperative mortality and myocardial infarction in high-risk patients undergoing vascular surgery. Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography Study Group. N Engl J Med 1999;341(24):1789-94. 3. Zaugg M, Tagliente T, Lucchinetti E, Jacobs E, Krol M, Bodian C, et al. Beneficial effects from beta-adrenergic blockade in elderly patients undergoing noncardiac surgery. Anesthesiology 1999;91(6):1674-86. 4. Urban MK, Markowitz SM, Gordon MA, Urquhart BL, Kligfield P. Postoperative prophylactic administration of beta-adrenergic blockers in patients at risk for myocardial ischemia. Anesth Analg 2000;90(6):1257-61. 5. Mangano DT, Layug EL, Wallace A, Tateo I. Effect of atenolol on mortality and cardiovascular morbidity after noncardiac surgery. Multicenter Study of Perioperative Ischemia Research Group. N Engl J Med 1996;335(23):1713-20. 6. Swedberg K, Wedel H. Effect of atenolol on mortality and cardiovascular morbidity after noncardiac surgery. Evidence-Based Cardiovascular Medicine 1998;March; 2:33. 7. Wheatley K, Clayton D. Be skeptical about unexpected large apparent treatment effects: the case of an MRC AML12 randomization. Control Clin Trials 2003;24(1):66-70. 8. Montori V, Devereaux PJ, Adhikari N, Burns K, Eggert C, Darling L, et al. Randomized trials stopped early for benefit: a systematic survey of the literature. JAMA (In Press) 2005. 9. Yang H, Raymer K, Butler R, Parlow J, Roberts R. Metoprolol after vascular surgery (MaVS). Can J Anesth 2004;51:A7. 10. Lindenauer P, Pekow P, Wan K, Mamidi D, Gutierrez B, Benjamin E. Perioperative Beta-Blocker Therapy and Mortality after Major Noncardiac Surgery. N Engl J Med 2005;353:349-361. 11. Yusuf S, Peto R, Lewis J, Collins R, Sleight P. Beta blockade during and after myocardial infarction: an overview of the randomized trials. Prog Cardiovasc Dis 1985;27(5):335-71. 12. Randomised trial of intravenous atenolol among 16 027 cases of suspected acute myocardial infarction: ISIS-1. First International Study of Infarct Survival Collaborative Group. Lancet 1986;2(8498):57-66. 13. Lechat P, Packer M, Chalon S, Cucherat M, Arab T, Boissel JP. Clinical effects of beta-adrenergic blockade in chronic heart failure: a meta- analysis of double-blind, placebo-controlled, randomized trials. Circulation 1998;98(12):1184-91. 14. The Cardiac Insufficiency Bisoprolol Study II (CIBIS-II): a randomised trial. Lancet 1999;353(9146):9-13. 15. Effect of metoprolol CR/XL in chronic heart failure: Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure (MERIT-HF). Lancet 1999;353(9169):2001-7. 16. McDaniel MD, Pearce WH, Yao JS, Rossi EC, Fahey VA, Green D, et al. Sequential changes in coagulation and platelet function following femorotibial bypass. J Vasc Surg 1984;1(2):261-8. 17. Rosenfeld BA, Beattie C, Christopherson R, Norris EJ, Frank SM, Breslow MJ, et al. The effects of different anesthetic regimens on fibrinolysis and the development of postoperative arterial thrombosis. Perioperative Ischemia Randomized Anesthesia Trial Study Group. Anesthesiology 1993;79(3):435-43. 18. Schillinger M, Domanovits H, Bayegan K, Holzenbein T, Grabenwoger M, Thoenissen J, et al. C-reactive protein and mortality in patients with acute aortic disease. Intensive Care Med 2002;28(6):740-5. 19. Flinn WR, McDaniel MD, Yao JS, Fahey VA, Green D. Antithrombin III deficiency as a reflection of dynamic protein metabolism in patients undergoing vascular reconstruction. J Vasc Surg 1984;1(6):888-95. Competing interests: Drs. P.J. Devereaux, Scott Beattie, Gordon Guyatt, Juan Villar, Peter Choi, Neal Badner, Claudio Cina, Michael Jacka, Victor Montori, Alvaro Avezum, Thomas Schricker, Homer Yang, and Salim Yusuf are all members of the POISE trial. Dr. Salim Yusuf has received honoraria and research grants from Astra Zeneca, who manufacture metoprolol CR.
Intra-operative hypotension is a major problem of perioperative beta-blockade 17 August 2005
Janet T Powell, Professor Imperial College at Charing Cross, W6 8RP, Matthew Sydes Send response to journal: Re: Intra-operative hypotension is a major problem of perioperative beta-blockade	Intra-operative hypotension is a major problem of perioperative b -blockade We are pleased to see that the systematic review of Devereaux and colleagues focused on hypotension as a problem following peri-operative b -blockade [1]. During the double-blind placebo-controlled POBBLE trial of peri-operative b -blockade with metoprolol in vascular surgical patients, which was published too recently to be included within the Devereaux review, intra-operative and post-operative hypotension was of major concern to the anaethetists. These concerns even led some of them to refuse entry of their patients to the trial [2]. The extent of intra-operative hypotension monitored in the POBBLE trial is shown in Table 1. In addition, a reduction in pulse rate to <50 bpm was noted in 57% of patients randomised to metoprolol versus 14% of those randomised to placebo (p<0.001). The incidence of major cardiovascular events (myocardial infarction, stroke or death) in the POBBLE trial was 15% (7/48) in the placebo group versus 13% (7/55) in the metoprolol group, so that even with the high rate of events in this group of "arteriopaths", there was no clear benefit of metoprolol on major cardiovascular events within 30 days of surgery. Our unpublished findings support the concerns of Devereaux et al [1] about the effect of peri-operative b -blockade on intraoperative hypotension. Table 1: Reported fall in systolic blood pressure of >25% during POBBLE Intraoperative reduction of >25% in SBP Placebo Metoprolol Test statistic N % N % No 10 23% 3 6% Yes 34 77% 49 94% Ppn test: z=-2.42, p=0.016 Missing 0 - 1 - Total 44 100% 53 100% Number of falls Median (IQR) 2 (1-5) 5 (3-9) MW-test: z=-3.52, p=0.0004 Mean (sd) 4.1 (5.8) 7.3 (8.5) Time fallen (mins) Median (IQR) 20 (5-38) 30 (15-60) MW-test: z=-3.52, p=0.002 Mean (sd) 29.0 (30.8) 44.6 (49.2) References 1 Devereaux JP, Beattie WS, Choi P et al How strong is the evidence for use of perioperative b -blockers in non-cardiac surgery? Systematic review and meta-analsysis of randomised controlled trials. BMJ 2005;331:313-6 2 POBBLE trial investigators, Peri-operative b -blockade (POBBLE) for patients undergoing infrarenal vascular surgery: results of a randomized double-blind controlled trial. J Vasc Surg 2005;41:602-9 Yours sincerely, Janet Powell, Imperial College, London (j.powell@imperial.ac.uk) Matthew Sydes, MRC Clinical Trials Unit, London (matthew.sydes@ctu.mrc.ac.uk) for POBBLE Trial Investigators Competing interests: None declared
Why do anaesthetists love beta-blockers? 22 August 2005
Mike Grocott, Senior Lecturer in Intensive Care Medicine University College London, WC1E 6BT, Denny ZH Levett, Monty G Mythen Send response to journal: Re: Why do anaesthetists love beta-blockers?	The Systematic Review and Meta-analysis by Devereaux and colleagues concludes "The evidence that perioperative beta-blockers reduce major cardiovascular events is encouraging but too unreliable to allow definitive conclusions to be drawn." The resulting correspondence is laced with misquotation from the primary paper, outrage that non-anaesthetists should dare to question anaesthetic dogma (misguided given that nine of 18 authors are practicing anaesthetists) and strong opinion supported by weak evidence. The two commonly cited primary sources in this field are both deeply flawed. The paper by Mangano reports a statistically significant reduction in mortality for patients given Atenolol when compared with placebo. However, as noted by Devereaux, when early deaths which occurred whilst the patients were receiving the study drug (Atenolol or placebo) are included the result is not statistically significant. The paper by Poldermans reports a compelling finding. But generalisability of this result is highly questionable. The study group comprises 112 patients selected from the 173 who had positive Dobutamine stress echo out of 846 patients identified at screening to have one or more cardiac risk factors. How can we separate the intervention from the diagnostic test and other exclusions? The meta-analysis by Devereaux found a statistically significant reduction in major perioperative cardiovascular events (cardiovascular death, non-fatal myocardial infarction, or non-fatal cardiac arrest) with 95% confidence limits but not with 99% confidence limits. When lower quality studies were removed the result was no longer significant at a 95% level. When the Poldermans study is removed the overall result is not significant and none of the other studies has a statistically significant result. This fact on its own should give us cause for concern given the limitations of the Poldermans study. The recent large (>100,000 patients) retrospective cohort study by Lindenauer suggest a reduction in mortality in those whose baseline risk of cardiac complications is high but for those at low risk there was no benefit (and possibly harm) and for those without risk factors there was an increase in mortality. The evidence supporting the perioperative prescription of beta- blockers to reduce mortality following surgery remains enticing but inconclusive. In addition there is reason to believe that this intervention may cause harm particularly in patients with a low base-line risk of cardiac complications. The reviewer's conclusion is well balanced and we should all await the results of POISE and other large randomized controlled studies. The best that can be said at the moment is that this intervention may reduce mortality in patients at high risk of cardiac complications but should be prescribed with caution in those at lower risk given the significant possibility that it may cause harm. 1. Devereaux JP, Beattie WS, Choi P et al How strong is the evidence for use of perioperative-blockers in non-cardiac surgery? Systematic review and meta-analysis of randomised controlled trials. BMJ 2005;331:313 -6 2. Mangano D, Layug E, Wallace A, et al. The effect of Atenolol on mortality and cardiovascular morbidity after noncardiac surgery. The New England Journal of Medicine 1996;335:1713-1720. 3. Wallace A, Layug, B, Tateo M, et al. Prophylactic Atenolol reduces postoperative Myocardial Ischaemia. Anesthesiology 1998;88:7-17. 4. Poldermans D, Boersma E, Bax J, et al. The effect of Bisoprolol on perioperative mortality and myocardial infarction in high-risk patients undergoing vascular surgery. The New England Journal of Medicine 1999;341(24):1789-1794. 5. Lindenauer PK, Pekow P, Wan K, et al. Perioperative Beta-Blocker Therapy and Mortality after Major Noncardiac Surgery. The New England Journal of Medicine 2005;353(4):349-361. Competing interests: None declared
Peri-operative Beta-blockers: Much room for evidence still exists! 29 September 2005
Ashish Aneja, Associate Staff Cleveland Clinic Foundation, Cleveland, OH, Brian Harte ; Vaishali Singh ; and Amir K Jaffer Send response to journal: Re: Peri-operative Beta-blockers: Much room for evidence still exists!	We at the Cleveland Clinic Foundation IMPACT (Internal Medicine Perioperative Assessment and Treatment) Center perform perioperative risk assessment in approximately 12,000 surgical patients annually. It is unlikely that we will change our current beta-blocker prescribing practice patterns based upon the exhaustive yet flawed meta-analysis by Devereaux et al. A closer look at the studies chosen by the authors highlights the heterogeneity and their lack of applicability to our practice (1). We believe that beta-blockers appear most likely to be beneficial in patients with a high (>2) RCRI (Relative Cardiac Risk Index) as defined by Lee et al (2). A modification of the RCRI was used by Lindenauer et al in a recent NEJM article to predict the usefulness and safety of beta-blocker therapy which supports their efficacy in patients with higher RCRI scores (3). It appears that Devereaux et al chose studies that included patients who were not at a significantly elevated perioperative cardiac risk as defined by the ACC/AHA guidelines, or who were not undergoing high-risk surgery, and therefore would have been unlikely to benefit substantially from beta-blocker therapy. Furthermore, a significant number of studies chosen for the meta-analysis lacked any follow through beyond the recovery room and oftentimes used 1-2 doses of parenteral beta-blocker therapy while the patient was in the operating room or in the recovery room. This may appear acceptable to the anesthesia community because of the sense of control over the situation, but from internists’ or cardiologists’ perspective, it appears that the much of the protection from adverse cardiac events is derived from higher-risk patients initiated on beta- blocker therapy in the pre-operative period and continued for a reasonable duration following the surgery. The bold conclusions that appear to have been extracted in this study, including the uncertain lack of benefit with these agents cannot conceivably be drawn from the trials that were selected because of little or no follow up in a significant percentage of studies chosen. The large number of patients included in this analysis may have helped beef up the statistical aspects but have failed to adequately address the core issue of cardiovascular benefit derived from beta-blocker therapy. With regards to adverse events reporting, Devereux et al regard bradycardia and hypotension as major problems with this class of medications. The authors have failed to elucidate the downstream effects of these adverse events as they may translate into excessive mortality, a higher incidence of acute coronary syndromes, need for temporary pacing and prolongation of hospital stay. Short of documentation of these hard clinical end-points, these adverse events may have little clinical significance and appear to relatively easily to manage. Despite the criticisms, this analysis has succeeded in creating the necessary controversy that in turn would facilitate the performance and success of the forthcoming POISE study. 1) P J Devereaux, W Scott Beattie, Peter T-L Choi, Neal H Badner, Gordon H Guyatt, Juan C Villar, Claudio S Cinà, Kate Leslie, Michael J Jacka, Victor M Montori, Mohit Bhandari, Alvaro Avezum, Alexandre B Cavalcanti, Julian W Giles, Thomas Schricker, Homer Yang, Carl-Johan Jakobsen, and Salim Yusuf. How strong is the evidence for the use of perioperative blockers in non-cardiac surgery? Systematic review and meta -analysis of randomised controlled trials. BMJ 2005. bmj.38503.623646.8Fv1. 2) Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation 1999;100:1043-1049 3) Lindenauer PK, Pekow P, Wang K, Mamidi DK, Gutierrez B, Benjamin EM. Perioperative beta-blocker therapy and mortality after major noncardiac surgery. N Engl J Med. 2005 Jul 28;353(4):349-61 Competing interests: None declared