Rapid Responses to:
|
|
blockade in patients with diabetes undergoing major non-cardiac surgery: randomised placebo controlled, blinded multicentre trial
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
An inconclusive result from wrong patient group
- Andrew C McCulloch (24 June 2006)
-
Perioperative beta blocker therapy is not ready to vanish yet
- Prakash Deedwania, Enrique V. Carbajal (11 July 2006)
-
Who really benefits from beta-blockers perioperatively?
- Maria C. Sitta, Adriana N. Machado, Wilson Jacob Filho, Luiz E. Garcez Leme (26 July 2006)
-
Response to Andrew C McCulloch’s ‘An inconclusive result from wrong patient group’, Prakash Deedwania’s ‘Perioperative beta blocker therapy is not ready to vanish yet’, and Maria C. Sitta et al’s : ‘Who really benefits from beta-blocker perioperatively’
- Anne Benedicte Juul, Jørn Wetterslev, Christian Gluud, Gorm Jensen, Allan Kofoed-Enevoldsen, Torben Callesen (11 August 2006)
-
beta-blockers are effective in preoperative management
- ismael francisco voltolini (27 September 2006)
-
Response to Ismael Francisco Voltolini: Beta-blockers are effective in preoperative management.
- Anne B. Juul, Jørn Wetterslev, Christian Gluud, Gorm Jensen, Torben Callesen, Allan Kofoed-Enevoldsen (10 November 2006)
|
|
|||
|
Andrew C McCulloch, Specialist Registrar in Cardiology Stirling Royal Infirmary , Stirling FK8 2AU
Send response to journal:
|
The study by Juul et al(1) is a valiant attempt to add clarity to an area of controversy. However, their bold statement that metoprolol is of no benefit peri-operatively in diabetic patients is not supported by the evidence they present. The number of cardiac events (table 5 in expanded article) is too small to draw any significant conclusion from. This is because the wrong patient group has been examined. Firstly, the cardiac risk profile of this patient group is too low. The large retrospective study by Lindenauer(2) clearly showed that a patient group with a Revised Coronary Risk Index (RCRI) of 3 or 4 was required to show definite evidence of benefit from peri-operative beta blockade. Those with a score of 2, around 60% of this cohort, showed marginal benefit only. Where the RCRI score was 0 or 1, even where that risk factor was diabetes, there was evidence of potential harm. Studies showing most benefit have studied the highest risk patients - a positive stress echo was required for entry into Poldermans’ trial(3). Secondly, the surgical risk profile of the cohort is not high enough to produce the levels of cardiac stress necessary to produce a sufficient event rate. The authors categorise 39% of operations as low risk. The precise breakdown of operations is unclear but given that only 7% of all operations are vascular, 4% thoracic and 27% intra-abdominal, with not all of these being prolonged procedures resulting in the significant fluid shifts associated with higher risk of cardiac complications, the proportion of operations which fell into a high risk category is probably less than 30%. It should also be noted that metoprolol may not have been commenced sufficiently far in advance of surgery to permit achievement of a steady state, thus diminishing the observed effect on intra-operative and immediate post - operative events. It is not made clear how many patients only received their first dose 2 hours prior to surgery. At present, the available evidence supports the initiation of beta blockers in high risk patients undergoing high risk surgery. Where patients in lower risk groups are already on beta blockers, every effort should be made to continue this in the peri-operative period(4). Low risk patients who have an indication for beta blockade other than surgery related should have that commenced electively a sufficient time in advance of surgery to identify any adverse reactions. References 1 Juul AB, Wetterslev J, Gluud C, Kofoed-Enevoldsen A, et al. Effect of perioperative beta blockade in patients with diabetes undergoing major non - cardiac surgery: randomised placebo controlled, blinded multicentre trial. BMJ 2006; 332: 1482 2 Lindenauer PK, Pekow P, Wang K, et al. Perioperative beta- blocker therapy and mortality after major noncardiac surgery. N Engl J Med 2005; 353: 349-61 3 Poldermans D, Boersma E, Bax JJ, 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: 1789-94 4 Frishman WH. Beta-adrenergic blocker withdrawal. Am J Cardiol 1987; 59: 26F-32F Competing interests: None declared |
|||
|
|
|||
|
Prakash Deedwania, Professor of Medicine VACCHCS/ UCSF - Fresno, CA, USA, 93703, Enrique V. Carbajal
Send response to journal:
|
The article by Juul et al (1) showing that perioperative metoprolol succinate CR/ XL did not significantly affect mortality and cardiac morbidity in patients with diabetes undergoing non-cardiac surgery is in contrast with the current guidelines and some previously reported studies demonstrating efficacy of perioperative beta-blockade (2-5). There is a potential explanation for the contrasting results. In the DIPOM study metoprolol was started as a test dose of 50 mg the evening before surgery and only continued until discharge or a maximum of 8 days. However, the primary combined outcome measure was evaluated until 18 months follow up. Based on the Kaplan-Meier curves shown in figure 2 of the article it appears that the curves did indeed show a lower event rate in the metoprolol group during the period the patients were receiving active medication. However, subsequently the event rates increased in the metoprolol group, which is not surprising given that beta-blockade was withdrawn upon discharge or at the end of 8 days and during follow up only 7 patients in the metoprolol group and 5 in the placebo group were taking any beta blockers. The abrupt increase in events could have occurred due to aggravation of cardiovascular events secondary to beta-blocker withdrawal and/ or lack of protection afforded by beta blocker therapy during active therapy. These results are in clear contrast to those reported by Poldermans et al (2) who demonstrated that when beta blocker therapy was given appropriately (started >=7 days before surgery and continued for 30 days post-operatively) it reduced the perioperative incidence of cardiac death and non-fatal MI at 30 days in patients undergoing vascular surgery. If the primary objective of DIPOM study (1) was to examine the event rates during the extended follow up, the active treatment should have been continued accordingly. 1. Juul AB, et al. Effect of perioperative beta blockade in patients with diabetes undergoing major non-cardiac surgery: randomised placebo controlled, blinded multicentre trial. BMJ. 2006;332(7556):1482-1485. 2. Eagle KA, et al. American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1996 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery). ACC/AHA guideline update for perioperative cardiovascular evaluation for noncardiac surgery---executive summary a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1996 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery). Circulation. 2002;105(10):1257-1267. 3. Poldermans D, 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):1789-1794. 4. Raby KE, et al. The effect of heart rate control on myocardial ischemia among high-risk patients after vascular surgery.Anesth Analg. 1999;88(3):477-482. 5. 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;353(4):349-361. Competing interests: None declared |
|||
|
|
|||
|
Maria C. Sitta, Assistant Professor University of São Paulo, Brazil Rua Professor Ovidio Pires de Campos, 333, 3o andar, CEP 05403-01, S.P. - Brazil, Adriana N. Machado, Wilson Jacob Filho, Luiz E. Garcez Leme
Send response to journal:
|
The “Effect of the perioperative beta blockade in patients with diabetes undergoing major non-cardiac surgery: randomized placebo controlled, blinded multicentre trial” (1) deserves consideration regarding the characterization of preoperative cardiac risk. The premise of this trial is that diabetics usually have high cardiac risk. However, all preoperative risk scales available classify diabetics as low-risk patients in the absence of other associated factors. Also, literature suggests that low-risk patients do not benefit from using beta blockers perioperatively. (3) Data available (2) used to characterize diabetics based on the known duration of the disease (11.8±11.7), the treatment with insulin preoperatively (46%) and the complications present, such as neuropathy (26%), retinopathy (19%) and nephropathy (26%). Of these parameters, the only one that is related to higher mortality is the use of insulin preoperatively, a finding that is similar to that found in Lee et al (4). Most patients in this sample have no target organ injury and were not on insulin preoperatively. Would this be the population to benefit from the beta blocker? This question could be answered by classifying patients using one of the risk scales available, such as the RCRI (4), ACP (5) or ACC/AHA algorithms.(6) Other factors were described that effectively increase the surgical risk, such as a history of heart failure (10%), chronic atrial fibrillation (8%), arrhythmia (3%), angina pectoris (11%), acute myocardial infarctation (8%) and angioplasty or myocardial revascularization (4%) (2). These numbers convey the idea that less than one fourth of patients have established coronary artery disease and make it difficult to interpret the exclusion criterion employed. Patients with indication for a beta blocker were excluded. The study was designed with 90% power to determine a 10% difference that is considered clinically significant. There was an estimated 30% incidence of events within 12 months, based on Juul and Mangano (7,8). However, there was a 20% incidence of combined events in the placebo group, which reduced the power of the study and confirmed the idea that the sample studied is not representative of patients at high cardiovascular risk. (2) It is worth noting that a multivariate analysis showed that, among the predictive variables of mortality, the age over 65 years variable may have behaved differently in comparison to the group as a whole. Other studies have already pointed out that age is a risk factor for perioperative mortality (9). Is it possible that the diabetic elderly could represent a group that would particularly benefit from the use of a beta blocker? This question still awaits an answer. Maria do Carmo Sitta MD; PhD.
References 1. Juul AB, Wetterslev J, Gluud C, Kofoed-Enevoldsen A, Jensen G, Callesen T, et al. The DIPOM Group. Effect of the perioperative beta blockade in patients with diabetes undergoing major non-cardiac surgery: randomized placebo controlled, blinded multicentre trial. BMJ. 2006; 332(7556):1482. 2. Juul AB, Wetterslev J, Kofoed-Enevoldsen A, Jensen G, Callesen T, Gluud C. The DIPOM Group. The diabetic postoperative mortality and morbidity (DIPOM) trial: rationale and design of a multicenter, randomised, placebo- controlled, clinical trial of metoprolol for patients with diabetes mellitus who are undergoing major noncardiac surgery. Am Heart J 2004;147:677-83. 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 Med 2005;353:349-61. 4. Lee TH, Marcantonio ER, Mangione CM, Thomas EJ, Polanczyk CA, Cook EF, Sugarbaker DJ, Donaldson MC, Poss R, Ho KK, Ludwig LE, Pedan A, Goldman L. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation. 1999;100:1043-9. 5. Guidelines for assessing and managing the perioperative risk from coronary artery disease associated with major noncardiac surgery. American College of Physicians. Ann Intern Med. 1997;127:309-12. 6. Eagle KA, Berger PB, Calkins H, Chaitman BR, Ewy GA, Fleischmann KE, et al. ACC/ AHA guideline update for perioperative cardiovascular evaluation for non- cardiac surgery—executive summary. A report of the American College of Cardiology/ American Heart Association task force on practice guidelines (committee to update the 1996 guidelines on perioperative cardiovascular evaluation for noncardiac surgery). Circulation 2002;105:1257-67. 7. Juul AB, Wetterslev J, Kofoed-Enevoldsen A. Reducing postoperative mortality and cardiovascular morbidity after noncardiac surgery [abstract]. Presented at the Angelo-Danish-Dutch-diabetes-Group meeting, 1998, UK. 8. Mangano DT, Layug EL, Wallace A, et al. Effect of atenolol on mortality and cardiovascular morbidity after noncardiac surgery. Multicenter Study of Perioperative Ischemia Research Group. N Engl J Med 1996;335:1713–1720. 9. Hamel MB, Henderson WG, Khuri SF, Daley J. Surgical Outcomes for Patients Aged 80 and Older: Morbidity and Mortality from Major Noncardiac Surgery. J Am Geriatr Soc 2005; 53:424–9. Competing interests: None declared |
|||
|
|
||||
|
Anne Benedicte Juul, MD, PhD. Copenhagen University Hospital Herlev 2730 Herlev, Jørn Wetterslev, Christian Gluud, Gorm Jensen, Allan Kofoed-Enevoldsen, Torben Callesen
|
Our report on the DIPOM trial (1) does not state that metoprolol is of no benefit peri-operatively in non-cardiac surgery. As suggested by McCulloch, we humbly state that no evidence supports this intervention. It is correct that a higher event rate or a larger number of participants in the DIPOM trial would have made the trial more conclusive. The number of events in the DIPOM trial during a median 18 months follow up was 192 (21%), which is more than 6 times as many events as in the Mangano et al trial (2) and the Poldermans et al trial (3). Accordingly the DIPOM trial is able to reject an intervention effect of the size suggested by the data of the Poldermans et al trial (3). According to Mangano et al (2) diabetes was the only significant predictor of death among cardiac risk patients undergoing noncardiac surgery (P<0.001) in the multivariate model. Further, betablockers improved two-year survival in diabetic patients with 75% (hazard ratio for death, 0.25; P=0.03) whereas the presence of diabetes in the placebo group was associated with a hazard ratio of 4 (P=0.003). The American Heart Association has stated that diabetic patients belong to the same high-risk category previously reserved for patients with known cardiovascular disease (4), and in the ACC/AHA guidelines 2002 (5) diabetes was stated to be a cardiac risk factor for the development of postoperative cardiovascular events with indication for perioperative beta-blockade. The criteria of Mangano et al provide an alternative approach to choosing patients largely by excluding patients at lowest risk. However, additional trials were and are still required to identify populations of patients in which the perioperative betablockade is cost-effective. We do not agree with McCulloch and Citta et al. that the Lindenauer et al. observational study (6) show that perioperative beta-blockade is of benefit for any group, as it merely indicate that the combination of risk profile and perioperative beta-blockade may be influential on the outcome. However, we agree that there are no contradictions between the results from the Lindenauer et al study (6) and our DIPOM trial results. 496 patients of the DIPOM patients fulfilled the entry criteria regarding cardiovascular risk factors and surgical risk profile of the Mangano et al trial. This is nearly two and a half as many patients as were included in the Mangano et al trial. Also the event rates were comparable in all included patients with the one-year event rate in the DIPOM trial being 16% (1) and the event rate in the Mangano et al trial being 15% (2). In the DIPOM trial a regimen similar to the regimen in the Mangano et al trial was followed - but we agree that both a higher dose and a longer duration of the intervention may have different effects. Only the Poldermans et al trial (3) suggests beneficial effects of perioperative beta-blockade. This trial was unblinded and highly discriminative in the selection of participants based on a stress dobutamine echocardiography. The high risk of bias in this trial and the discriminative selection of patients makes it unsuitable for clinical recommendations. Therefore there is no low bias evidence presently supporting the initiation of perioperative beta-blockade in any patients undergoing non-cardiac surgery (1,7). Contrary to the description by Deedwania the Kaplan-Meier curves, including 95% confidence intervals, in fig 2 in the report of the DIPOM trial (1) does not show any statistically significant differences at any time point during a median of 18 months follow-up. Accordingly no abrupt increase in events took place. However, we agree with Deedwania that a longer duration of the intervention and possible a higher dose might have yielded different results. The DIPOM regimen was similar to the regimen used in the Mangano et al trial (2) except that we only included beta- blocker na¿ve patients and performed intention to treat analysis of the results. As Sitta et al. describe age over 65 years was among the predictive variables in the multivariate analysis - but there was no significant interaction between age and intervention precluding any speculations that the elderly patients with diabetes may benefit from perioperative beta- blockade in the DIPOM trial. References 1. Juul AB, Wetterslev J, Gluud C, Kofoed-Enevoldsen A, G Jensen, T Callesen et al. Effect of perioperative beta blockade in patients with diabetes undergoing major non - cardiac surgery: randomised placebo controlled, blinded multicentre trial. BMJ 2006; 332: 1482 2. Mangano DT, Layug EL, Wallace A, Tateo I . Effect of atenolol on mortality and cardiovascular morbidity after non-cardiac surgery. Multicenter Study of Perioperative Ischemia Research Group. N Engl J Med 1996; 335(23):1713-1720 3. Poldermans D, Boersma E, Bax JJ, Thomson Jr, 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: 1789-94 4. Redberg RF, GreenlandP, Fuster V, Pyorala K, Blair SN, Folsom AR et al. Prevention Conference VI: Diabetes and Cardiovascular Disease: Writing Group III: risk assessment in persons with diabetes. Circulation 2002; 105:e 144-52. 5. Eagle KA, Berger PB, Calkins H, Chaitman BR, Ewy GA, Fleischmann KE, et al. ACC/AHA guideline update for perioperative cardiovascular evaluation for non-cardiac surgery---executive summary a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1996 Guidelines on Perioperative Cardiovascular Evaluation for Non-cardiac Surgery). Circulation 2002; 105:1257-1267. 6. Lindenauer PK, Pekow P, Wang K, Mamidi DK, Gutierrez B, Benjamin EM et al. Perioperative beta- blocker therapy and mortality after major noncardiac surgery. N Engl J Med 2005; 353: 349-61 7. Wetterslev J, Juul AB. Benefits and harms of perioperative beta- blockade. Best Practice and Research, Clinical Anaesthesiology 2006; 20(2): 285-302. Competing interests: None declared |
|||
|
|
|||
|
ismael francisco voltolini, phisitian pucrs(hospital são lucas) av: ipiranga 6681--porto alegre-rs-brasil-90619900
Send response to journal:
|
The article, effect of perioperative beta-blockade in patients with diabetes undergoing major non-cardiac surgery: randomised placebo controlled, blinded multicentre trial, is of high relevance, considering that many patients with diabetes I and II undergo elective and urgent surgery every day. It's necessary therefore to establish the best evidence to offer optimal cardiovascular protection in such cases. As described in the introduction of the article, beta-blockers in patients with diabetes and additional risk factors for coronary artery disease seemed to be beneficial with a two year hazard ratio of death of 0,25 (p=0,03) and because of this the use of beta -blockers is recommended by the American College of Cardiology and the American Heart Association. Epidemiologically the article is very well done. The design (randomised placebo controlled and blinded) is very adequate to test the operational hypothesis. The authors avoided the occurrence of bias, beginning by the establishment of the beta and alpha errors, calculating the size of the sample and setting a two sided alpha = 0,05. Confusion bias was avoided with adequate randomisation, good stratification and establishing rigorous exclusion and inclusion criteria. This can be sought in the first table's study. Measurement bias was submitted to control when the measurement tools were defined and with the blinding of the researchers. This last bias is not completely eliminated. Even being blinded the measurers could know who is using metoprolol controlling the heart ratio. This is the only potential bias in the study. The most fragile point of the article is found in table 6. There you can see that the mean heart ratio of the patients taking metoprolol was 78 in the first day of the study. In the second day the mean heart ratio was 72. A difference of only 6 beats per minute, this means a reduction of only 7,6% in the mean heart ratio, and worst, this difference diminished to 4 beats per minute on the fourth and to 2 beats per minute on the eighth day. Is a patient given beta-blockers with a heart ratio of 72bpm or 75bpm? Do inflammatory cytokines and neurohumoral mediators completely eliminated, caused by pain and surgical trauma? Considering the short time of administration of the study's drug (only 4,9 days) and the insufficient beta-blockade, I disagree with the authors when they say that the evidence is insufficient to recommend the use perioperative beta-blockers. We may follow the Association’s guidelines until newer evidence is established. Competing interests: None declared |
|||
|
|
|||
|
Anne B. Juul, md, ph.d KAS Herlev Hospital, Copenhagen University Hospital, DK-2730 Herlev, Jørn Wetterslev, Christian Gluud, Gorm Jensen, Torben Callesen, Allan Kofoed-Enevoldsen
Send response to journal:
|
Thank you for your interest in our article on the DIPOM trial:Effect of perioperative beta-blockade in patients with diabetes undergoing major, non-cardiac surgery: randomised placebo controlled, blinded, multicentre trial(1). Special efforts were carried out maintaining blinding through monitoring. A study nurse measured and registered heart rate and blood pressure before and after administration of study drug and conducted the reporting of these variables in the case record form. All investigators, medical staff, and the patients were sought blinded. And blinding was also maintained during follow-up, adjudication of outcome measures, data management, and data analysis. However, we agree that complete elimination of bias is not possible. We suspect that such bias may be more troublesome in beta-blocker trials finding positive effects of the drug. During treatment, the average heart rate was significantly lower in the beta-blocker group compared with the placebo group (75 beats per minute versus 82 beats per minute in the placebo group, P<0.001) (1). Mangano et al (2) found the same average heart rate of 75 beats per minute in the beta-blocker group. We have therefore reason to believe that the beta-blockade received by the patients in the DIPOM trial was similar to the beta-blockade received by the patients in the Mangano et al. trial (2). A possibly beneficial effect of perioperative beta-blockade in cardiac high-risk patients cannot be excluded. However, all randomised clinical trials (the DIPOM (1), the MAVS (3), and the POBBLE (4) trials), which have been published since Mangano et al (2) and Poldermans et al (5) published their results, have not found any significant beneficial effects of perioperative beta-blockade in cardiac risk patients undergoing non- cardiac surgery. In addition, meta-analyses (6,7) have been unable to demonstrate statistically significant difference in fatal events or in non -fatal myocardial infarction when trials with inadequate methodology were excluded. Thus, there is no valid evidence to recommend perioperative beta -blockade on the sole indication of diabetes mellitus. Large randomised trials using larger dose and/or longer duration of beta-blockade than used in the DIPOM trial are needed to answer all questions regarding potential optimal dosage and duration and to identify potential populations in which beta-blockade is beneficial and safe. Such a trial, conducted by the POISE investigators (8), is currently under way. References 1. Juul AB, Wetterslev J, Gluud C, Kofoed-Enevoldsen A, et al. Effect of perioperative beta blockade in patients with diabetes undergoing major non - cardiac surgery: randomised placebo controlled, blinded multicentre trial. BMJ 2006; 332: 1482 1. 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: 1713-20 2. Yang H, Raymer K, Butler R, et al., Metoprolol after vascular surgery (MAVS). Can J Anesth 51 (2004) (supplement 1), p. A7 3. POBBLE Trial investigators. Perioperative beta-blockade (POBBLE) for patients undergoing infrarenal vascular surgery: results of a randomized double-blind controlled trial. J Vasc Surg 2005;41: 602-9 4. Poldermans D, Boersma E, Bax JJ. 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: 1789-94 5. Devereaux PJ, Scott Beattie W, Choi PT-L, Badner NH, Guyatt G, Villar JC, et al. How strong is the evidence for the use of perioperative ß blockers in non-cardiac surgery? Systematic review and meta-analysis of randomised controlled trials. BMJ 2005;331: 313-21 6. Wetterslev J, Juul AB. Benefits and harms of perioperative beta- blockade. Best Pract Res Clin Anaesthesiol. 2006 ;20(2):285-302. 7. Wetterslev J, Juul AB, Gluud C. How to meta-analyze trials of perioperative beta blockade.Surgery. 2006 Apr;139(4):583 8. POISE Trial Investigators; Devereaux PJ, Yang H, Guyatt GH, Leslie K, Villar JC, Monteri VM, Choi P, Giles JW, Yusuf S. Rationale, design, and organization of the PeriOperative ISchemic Evaluation (POISE) trial: a randomized controlled trial of metoprolol versus placebo in patients undergoing noncardiac surgery. Am Heart J. 2006 Aug;152(2):223-30. Competing interests: None declared |
|||