Author + information
- Published online December 9, 2014.
- Duminda N. Wijeysundera, MD, PhD, Evidence Review Committee Chair & Evidence Review Committee Member,
- Dallas Duncan, MD, MHSc, Evidence Review Committee Member∗,
- Chileshe Nkonde-Price, MD, MS, Evidence Review Committee Member∗,
- Salim S. Virani, MD, PhD, FACC, FAHA, Evidence Review Committee Member∗,
- Jeffrey B. Washam, PharmD, FAHA, Evidence Review Committee Member∗,
- Kirsten E. Fleischmann, MD, MPH, FACC, Perioperative Guideline Vice Chair & Evidence Review Committee Member and
- Lee A. Fleisher, MD, FACC, FAHA, Perioperative Guideline Chair & Evidence Review Committee Member
Objective To review the literature systematically to determine whether initiation of beta blockade within 45 days prior to noncardiac surgery reduces 30-day cardiovascular morbidity and mortality rates.
Methods PubMed (up to April 2013), Embase (up to April 2013), Cochrane Central Register of Controlled Trials (up to March 2013), and conference abstracts (January 2011 to April 2013) were searched for randomized controlled trials (RCTs) and cohort studies comparing perioperative beta blockade with inactive control during noncardiac surgery. Pooled relative risks (RRs) were calculated under the random-effects model. We conducted subgroup analyses to assess how the DECREASE-I (Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography), DECREASE-IV, and POISE-1 (Perioperative Ischemic Evaluation) trials influenced our conclusions.
Results We identified 17 studies, of which 16 were RCTs (12,043 participants) and 1 was a cohort study (348 participants). Aside from the DECREASE trials, all other RCTs initiated beta blockade within 1 day or less prior to surgery. Among RCTs, beta blockade decreased nonfatal myocardial infarction (MI) (RR: 0.69; 95% confidence interval [CI]: 0.58 to 0.82) but increased nonfatal stroke (RR: 1.76; 95% CI:1.07 to 2.91), hypotension (RR: 1.47; 95% CI: 1.34 to 1.60), and bradycardia (RR: 2.61; 95% CI: 2.18 to 3.12). These findings were qualitatively unchanged after the DECREASE and POISE-1 trials were excluded. Effects on mortality rate differed significantly between the DECREASE trials and other trials. Beta blockers were associated with a trend toward reduced all-cause mortality rate in the DECREASE trials (RR: 0.42; 95% CI: 0.15 to 1.22) but with increased all-cause mortality rate in other trials (RR: 1.30; 95% CI: 1.03 to 1.64). Beta blockers reduced cardiovascular mortality rate in the DECREASE trials (RR:0.17; 95% CI: 0.05 to 0.64) but were associated with trends toward increased cardiovascular mortality rate in other trials (RR: 1.25; 95% CI: 0.92 to 1.71). These differences were qualitatively unchanged after the POISE-1 trial was excluded.
Conclusions Perioperative beta blockade started within 1 day or less before noncardiac surgery prevents nonfatal MI but increases risks of stroke, death, hypotension, and bradycardia. Without the controversial DECREASE studies, there are insufficient data on beta blockade started 2 or more days prior to surgery. Multicenter RCTs are needed to address this knowledge gap.
- ACC/AHA Clinical Practice Guideline
- adrenergic beta-antagonists
- noncardiac surgery
- perioperative cardiovascular complications
ACC/AHA Task Force Members
Jeffrey L. Anderson, MD, FACC, FAHA, Chair
Jonathan L. Halperin, MD, FACC, FAHA, Chair-Elect
Nancy M. Albert, PhD, RN, FAHA
Biykem Bozkurt, MD, PhD, FACC, FAHA
Ralph G. Brindis, MD, MPH, MACC
Lesley H. Curtis, PhD, FAHA
David DeMets, PhD†
Lee A. Fleisher, MD, FACC, FAHA
Samuel Gidding, MD, FAHA
Judith S. Hochman, MD, FACC, FAHA†
Richard J. Kovacs, MD, FACC, FAHA
E. Magnus Ohman, MD, FACC
Susan J. Pressler, PhD, RN, FAHA
Frank W. Sellke, MD, FACC, FAHA
Win-Kuang Shen, MD, FACC, FAHA
Duminda N. Wijeysundera, MD, PhD
Table of Contents
Eligibility Criteria 2408
Search Strategy 2408
Methods of Review 2409
Statistical Analysis 2409
Nonfatal MI 2410
Nonfatal Stroke 2410
All-Cause Death 2410
Cardiovascular Death 2410
Perioperative Adverse Effects 2410
Post Hoc Analysis 2410
Publication Bias 2410
Influence of the DECREASE Trials 2411
Tables and Figures
Table 1. Summary of Included Studies 2413
Table 2. Perioperative Beta-Blocker Regimens, Duration of Follow-Up, and Comparison Arms in Included Studies 2415
Figure 1. Effect of Perioperative Beta Blockade on In-Hospital or 30-Day Nonfatal MI in RCTs Within Subgroups Defined by DECREASE Trials Versus Other Trials 2417
Figure 2. Effect of Perioperative Beta Blockade on In-Hospital or 30-Day Nonfatal MI in RCTs After Exclusion of the DECREASE Family of Trials, Within Subgroups Defined by the POISE-1 Trial Versus Other Trials 2418
Figure 3. Effect of Perioperative Beta Blockade on In-Hospital or 30-Day Nonfatal Stroke in RCTs Within Subgroups Defined by DECREASE Trials Versus Other Trials 2419
Figure 4. Effect of Perioperative Beta Blockade on In-Hospital or 30-Day Nonfatal Stroke in RCTs After Exclusion of the DECREASE Family of Trials, Within Subgroups Defined by the POISE-1 Trial Versus Other Trials 2420
Figure 5. Effect of Perioperative Beta Blockade on In-Hospital or 30-Day Mortality Rate in RCTs Within Subgroups Defined by DECREASE Trials Versus Other Trials 2421
Figure 6. Effect of Perioperative Beta Blockade on In-Hospital or 30-Day Mortality Rate in RCTs After Exclusion of the DECREASE Family of Trials, Within Subgroups Defined by POISE-1 Trial Versus Other Trials 2422
Author Relationships With Industry and Other Entities (Relevant) 2425
Perioperative cardiac complications are an important concern for the 230 million individuals who undergo surgery worldwide every year (1). After surgery, 2% of these patients suffer major cardiac complications (2), and 8% show evidence of significant myocardial injury (3). Perioperative beta blockade showed early promise as a means of preventing these complications, with enthusiasm driven by promising results in 2 RCTs (4,5).
Consequently, perioperative beta blockade was recommended for a fairly broad spectrum of surgical patients in initial versions of the American College of Cardiology (ACC)/American Heart Association (AHA) clinical practice guidelines (CPGs). For example, among patients with untreated hypertension, known coronary artery disease, or cardiac risk factors, perioperative beta blockade received a Class II recommendation in 1996 (6) and a Class IIa recommendation in 2002 (7). Nonetheless, for several reasons, the strength and scope of these recommendations diminished over successive iterations of these CPGs (8–10). First, subsequent moderate-sized RCTs failed to demonstrate significant benefits from beta blockade (11,12). Second, in the POISE-1 trial of almost 9000 participants, it was found that although perioperative beta blockade prevented perioperative MI, this benefit was accompanied by increased rates of death, stroke, hypotension, and bradycardia (13). Although the POISE-1 trial has been criticized for starting long-acting beta blockers at high doses shortly prior to surgery (14), its results highlighted the potential for important risks from perioperative beta blockade. Third, the validity of work led by Poldermans, including 2 influential perioperative beta-blockade RCTs (5,15), has been scrutinized because of concerns about scientific misconduct (16,17). Consequently, it has been suggested that CPGs re-evaluate and potentially exclude these data from the evidence base used to inform recommendations about perioperative beta blockade (18).
On the basis of the “American College of Cardiology Foundation/AHA clinical practice guideline methodology summit report” (19), the ACC/AHA Task Force on Practice Guidelines (Task Force) recognized the need for an objective review of available RCTs and observational studies by an independent Evidence Review Committee (ERC) to inform any recommendations about perioperative beta blockade in the 2014 ACC/AHA perioperative CPG (20). The ERC undertook this review to address a specific clinical question framed by the writing committee for this CPG (with input from the ERC): What is the evidence that initiating beta blockade within 45 days prior to noncardiac surgery reduces perioperative cardiovascular morbidity and mortality within 30 days after surgery? Our objectives were to summarize evidence relevant to this question and assess the degree to which studies led by Poldermans influenced our overall conclusions.
This systematic review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement (21) and to recommendations of the “American College of Cardiology Foundation/AHA clinical practice guideline methodology summit report” (19).
We included RCTs or cohort studies comparing perioperative beta blockade against inactive control, including placebo, in adults (≥18 years of age) undergoing noncardiac surgery. Otherwise eligible cohort studies were included only if the sample size exceeded 100 participants. Perioperative beta blockade was defined as beta-blocker therapy (except sotalol) started at any point between 45 days prior to surgery and 24 hours after surgery. Treatment also had to be continued until at least hospital discharge or the second day after surgery (whichever occurred first). This minimum duration of postoperative therapy was specified because perioperative MI generally occurs during the first 3 days after surgery (22). Otherwise eligible studies also had to report any of 4 prespecified outcomes: MI, all-cause death, cardiovascular death, or stroke.
Eligible studies were identified using PubMed (up to April 2013), Embase (up to April 2013), and the Cochrane Central Register of Controlled Trials (up to March 2013). The search strategies used within these databases are presented in the Online Data Supplement 1 to 3. The ERC also hand-searched abstracts from conferences of specific scientific societies (ACC, AHA, American Society of Anesthesiologists, European Society of Anesthesiology, European Society of Cardiology, International Anesthesia Research Society, and Society of Cardiovascular Anesthesiologists) occurring between January 2011 and April 2013 and searched bibliographies of previous relevant systematic reviews (18,23–26). No language restrictions were applied. Unpublished trials were not sought, but when necessary, we contacted authors of included studies for additional data.
Methods of Review
Teams of paired reviewers (i.e., D. Duncan and C. Nkonde-Price, S. S. Virani and J. B. Washam) independently performed study eligibility screening, study quality evaluation, and data abstraction. Abstracted data were entered on previously pilot-tested forms developed within the DistillerSR (Evidence Partners Inc., Ottawa, Ontario, Canada) and Indico Clinical Guideline Platform (Indico Solutions Pty. Ltd., Melbourne, Victoria, Australia) web-based software platforms. Disagreements were resolved through consensus and, where necessary, involvement of a third reviewer (D. N. Wijeysundera). For each included study, the ERC abstracted details on participant eligibility criteria, participant number, surgery types, beta-blocker treatment regimen, participant characteristics (i.e., age, sex, coronary artery disease, prior MI, current angina), duration of follow-up, and surveillance protocols for postoperative MI. In addition, the proportion of participants receiving long-term beta-blocker treatment before recruitment was reported for any included RCT. We documented the definition and event rates for the following outcomes occurring during or within 30 days after surgery: nonfatal MI, all-cause death, cardiovascular death, acute stroke, heart failure, significant hypotension, and significant bradycardia. Overall study quality was assessed on the basis of risk of bias, relevance to the study question, and fidelity of implementation (19). With regard to evaluation of risk of bias, we used the Cochrane Collaboration Risk of Bias Tool for RCTs (27) and the Newcastle-Ottawa Scale for cohort studies (28). A RCT was assigned an overall rating of low-to-intermediate risk of bias if the trial was not deemed to be at high risk of bias for any assessed domain of study quality.
Analyses were performed in STATA Version 13 statistical software (StataCorp LP, College Station, Texas). Statistical significance was defined by a 2-tailed p value <0.05, and no adjustment was made for multiple comparisons. Given the major methodological differences between RCTs and cohort studies, the 2 study types were analyzed separately. Initially, we assessed both clinical and statistical heterogeneity across the included studies. Statistical heterogeneity was characterized with the I2 statistic (29), which describes the proportion of total variation explained by between-study variation instead of chance. Higher I2 statistic values imply more heterogeneity between studies than would be expected by chance alone. The random-effects model of DerSimonian and Laird was used to calculate pooled RRs with 95% CIs (30).
We conducted several prespecified subgroup analyses to examine the influence of the DECREASE and POISE-1 trials on the overall results (5,13,15). First, treatment effects within the DECREASE trials were compared against pooled effects in the remaining RCTs. Second, after excluding the DECREASE trials (i.e., DECREASE-I and DECREASE-IV) from the analysis, we compared treatment effects in the POISE-1 trial against pooled effects in the remaining trials. The rationale for this second subgroup analysis was to determine whether there was any signal of a treatment effect independent of the single large RCT (i.e., the POISE-1 trial) in the meta-analysis. Random-effects meta-regression was used to test for statistical significance of any subgroup effects. The ERC visually inspected funnel plots to assess for possible publication bias (31) and also used Egger’s, Harbord’s, and Peters’ tests to formally test for any funnel plot asymmetry (31–33).
Please see the Online Data Supplement for more information.
We identified 17 eligible studies: 16 RCTs and 1 cohort study (Online Data Supplement 7). The 16 RCTs contributed data from 12,043 participants (4,5,11–13,15,34–42), and the cohort study contributed relevant data from 348 participants (43). The characteristics of participants, surgical procedures, and perioperative beta-blockade protocols in the included studies are presented in Tables 1 and 2. Except for the DECREASE-I and DECREASE-IV trials (5,15), all RCTs began beta-blocker therapy within 1 day or less prior to surgery.
Of the 16 included RCTs, 8 trials had a low-to-intermediate overall risk of bias (Online Data Supplement 4) (4,11–13,38,40,42,44). Fourteen trials showed intermediate-to-high relevance with regard to their study populations, interventions, and outcomes measures (4,5,11–13,15,35–42), and 10 trials assessed interventions that were implemented with intermediate-to-high fidelity (Online Data Supplement 4) (4,5,11–13,37,38,40–42). When assessed with the Newcastle-Ottawa Scale, the included cohort study did not rate consistently well across all study quality domains (Online Data Supplement 5).
Sixteen RCTs reported effects on nonfatal MI among 11963 participants (Figure 1). Perioperative beta blockade caused an overall moderate reduction in nonfatal MI, based on a RR of 0.68 (95% CI: 0.57 to 0.81; p<0.001) with no measurable statistical heterogeneity (I2=0%). Nonetheless, differences in treatment effects between the DECREASE trials and the remaining RCTs bordered on statistical significance (p=0.08). When the DECREASE trials were excluded (Figure 2), the pooled RR remained essentially unchanged at 0.72 (95% CI: 0.59 to 0.86), with no qualitative differences in effects observed between the POISE-1 trial and the remaining RCTs.
Nonfatal strokes were reported by 10 trials that included 11 611 participants (Figure 3). Beta blockade caused a significant overall increase in the risk of nonfatal stroke (RR: 1.79; 95% CI: 1.09 to 2.95; p=0.02), with no measurable statistical heterogeneity (I2=0%). When DECREASE trials were excluded (Figure 4), the effects in the POISE-1 trial (RR: 1.93; 95% CI: 1.01 to 3.68) were qualitatively similar to those in the remaining trials (RR: 1.72; 95% CI: 0.67 to 4.40).
Sixteen trials reported effects on rates of all-cause death among 11,963 participants (Figure 5). There was a statistically significant subgroup difference (p=0.02) between the DECREASE trials and the remaining RCTs. Among the DECREASE trials, beta blockade was associated with a trend toward a reduced risk of all-cause death (RR: 0.42; 95% CI: 0.15 to 1.22; p=0.11), whereas in the remaining trials, beta blockers significantly increased the risk of all-cause death (RR: 1.30; 95% CI: 1.03 to 1.63; p=0.03), with no measurable statistical heterogeneity (I2=0%). When the DECREASE trials were excluded (Figure 6), effects in the POISE-1 trial (RR: 1.33; 95% CI: 1.03 to 1.73) were qualitatively similar to effects in the remaining trials (RR: 1.17; 95% CI: 0.70 to 1.94).
Cardiovascular deaths were reported by 13 trials encompassing 11,607 participants. There was statistically significant evidence of a subgroup difference (p=0.004) between the DECREASE trials and the remaining RCTs (Online Data Supplement 8). Beta blockers significantly reduced the risk of cardiovascular death in the DECREASE trials (RR: 0.17; 95% CI: 0.05 to 0.64; p=0.008), whereas they showed a trend toward an increased risk of cardiovascular death (RR: 1.25; 95% CI: 0.92 to 1.71; p=0.16) in the remaining trials.
Perioperative Adverse Effects
Eight trials reported effects on heart failure among 11,378 participants (Online Data Supplement 9). Overall, beta blockade had no statistically significant effect on perioperative heart failure (RR: 1.15; 95% CI: 0.91 to 1.45; p=0.23), without measurable statistical heterogeneity (I2=0%). Ten trials reported effects on perioperative hypotension or bradycardia, albeit with highly variable definitions across studies (Online Data Supplement 6) (4,11–13,35–38,41,44). Notably, the DECREASE-I and DECREASE-IV trials did not separately report rates of hypotension or bradycardia (5,15). Nine trials reported effects on hypotension among 10,448 participants (Online Data Supplement 10). Overall, beta blockers significantly increased the risk of perioperative hypotension, with no qualitative differences in effects seen between the POISE-1 trial (RR: 1.55; 95% CI: 1.38 to 1.74) and other studies (pooled RR: 1.37; 95% CI: 1.20 to 1.56). Significant bradycardia was reported by 9 trials encompassing 10,458 participants (Online Data Supplement 11). Risks of bradycardia were significantly increased among patients receiving beta blockers, with no qualitative differences in effects seen between the POISE-1 trial (RR: 2.74; 95% CI: 2.19 to 3.43) and other studies (pooled RR: 2.41; 95% CI: 1.75 to 3.32).
Post Hoc Analysis
In a post hoc analysis, the ERC excluded the DECREASE trials and used pooled RRs from the remaining trials to calculate numbers of avoided or excess nonfatal MIs, all-cause deaths, and nonfatal strokes per 1,000 population. Within a hypothetical population with a baseline 6% risk of nonfatal MI, 2% baseline risk of 30-day all-cause death, and 0.5% baseline risk of nonfatal stroke, perioperative beta blockade leads to 17 fewer nonfatal MIs, 6 excess all-cause deaths, and 4 excess nonfatal strokes in every 1000 treated patients.
Visual inspection of funnel plots showed no clear evidence of publication bias with regard to effects on nonfatal MI, nonfatal stroke, heart failure, hypotension, and bradycardia. These plots did suggest some publication bias with regard to all-cause and cardiovascular death. Specifically, some trials showing an increased mortality rate with beta blockade may not have been published. Nonetheless, formal testing did not reveal any statistically significant evidence of publication bias for any assessed outcomes.
This systematic review found the literature to be consistent with regard to effects of perioperative beta blockade on MI, stroke, hypotension, and bradycardia after noncardiac surgery. Previous trials consistently demonstrated that rates of nonfatal MI were reduced with beta blockade. Although there may be some differences between the DECREASE trials and other trials, these differences relate only to the magnitude of benefit. Whereas the DECREASE trials found larger and, arguably, somewhat implausible effect sizes, with RR reductions ranging from 60% to 95% (5,15), other trials had a more realistic, moderate pooled effect size. Available data also consistently show increased risks of stroke, hypotension, and bradycardia with perioperative beta blockade. These findings are noteworthy because the increased risk of these complications in the POISE-1 trial has often been attributed to the trial’s use of high-dose, long-acting metoprolol (45). The ERC instead found that preceding trials, despite using different dosing regimens, demonstrated a consistent signal of increased stroke (albeit statistically nonsignificant), as well as significant increases in risks for hypotension and bradycardia. Thus, the increased risks of these complications appear to be a more general concern with perioperative beta blockade, as opposed to one associated only with a specific drug-dosing regimen. Notably, there are very few data on stroke, hypotension, and bradycardia from the DECREASE trials, with the only reported events being 7 strokes in the DECREASE-IV trial (15).
The major discrepancy between the DECREASE trials and the other RCTs relates to effects on all-cause and cardiovascular death. The DECREASE trials demonstrated very large reductions in mortality rate, with RR reductions ranging from 58% to 91%. Again, such large benefits attributable to a single intervention are, arguably, somewhat implausible. Conversely, the remaining RCTs found a significant overall increase in mortality rate. Although this pooled estimate was dominated by the POISE-1 trial, which accounted for 80% of the relevant underlying data, it is noteworthy that even when data from the POISE-1 trial were excluded, the pooled effect in the remaining studies was qualitatively similar. Thus, data at the time of publication suggest that the increased mortality rate observed in the POISE-1 trial may not be unique to that specific dosing protocol.
Influence of the DECREASE Trials
The DECREASE trials do influence the overall conclusions of our review, but largely with regard to effects on mortality rates. Specifically, in the absence of the DECREASE trials, other RCTs indicate that beta blockade significantly reduces the risk of postoperative MI but at the cost of increased rates of stroke, hypotension, bradycardia, and death. The major change induced by inclusion of the DECREASE trials in the meta-analysis is a shift of the pooled effect on death to a null effect. Nonetheless, exclusion of these trials has major implications for the generalizability of current RCTs to clinical practice. Aside from the DECREASE trials, all RCTs initiated beta blockade no more than 1 day prior to surgery. Notably, several cohort studies have shown that shorter durations (≤7 days) of preoperative beta-blocker therapy are associated with worse outcomes than are longer durations of preoperative therapy (46–48). Although some authors have emphasized the importance of both longer durations of therapy prior to surgery and preoperative dose-titration to an optimal heart rate, the evidence for substantial preoperative modification of beta-blocker dosing in the DECREASE trials is not compelling. Stated otherwise, the vast majority of patients in these studies presented to surgery receiving the same dose of bisoprolol on which they were started. In addition, once the DECREASE trials were excluded, only 4 included RCTs evaluated oral beta blockers aside from metoprolol (4,40,41,44). Importantly, several cohort studies have found metoprolol to be associated with worse outcomes than those seen with more beta-1 selective agents, such as atenolol or bisoprolol (49–53).
Thus, the strength of evidence that a longer duration of preoperative therapy with selective oral beta blockers safely reduces the risk of perioperative MI is entirely dependent on the DECREASE trials. Such reliance on controversial studies points to the need for new, adequately powered RCTs of clinically sensible, perioperative beta-blockade regimens. The ERC proposes that such trials evaluate beta-blockade regimens started at least several days prior to surgery, preferably with more beta-1 selective agents. In light of the consistent signals of increased harm associated with beta blockade initiated very close to surgery, the onus lies with the perioperative medical community to demonstrate that such alternative dosing regimens are safe and efficacious with regard to prevention of perioperative MI.
The present systematic review also has several important limitations. First, exclusion of the DECREASE-I, DECREASE-IV, and POISE-1 trials leaves few data from which to make firm conclusions about the efficacy and safety of perioperative beta blockade. For example, none of the pooled effects on MI, death, and stroke were statistically significant within this smaller subgroup of studies. Consequently, comparison of this subgroup with the POISE-1 trial focused simply on qualitative comparisons of pooled effects. Second, as with most systematic reviews, our review is limited by the possibility of unpublished data and heterogeneity of outcome definitions used in the original studies. Third, the included trials did not systematically report treatment effects in clinically sensible subgroups, such as strata defined by Revised Cardiac Risk Index scores (54). Several prior observational studies have suggested that the treatment effects of perioperative beta blockade vary across these strata, with benefits confined to high-risk individuals who have at least 2 to 3 clinical risk factors (50,55). Further exploration of such subgroup differences within the context of RCTs will entail an individual patient data meta-analysis. Fourth, we did not adjust for multiple comparisons when conducting subgroup analyses. The results of these subgroup analyses should therefore be viewed as hypothesis generating as opposed to definitive.
Fifth, in spite of an extensive literature search for observational studies, only 1 cohort study was included in this systematic review (43). Despite still informing an overall understanding of the risks and benefits of beta blockade in noncardiac surgery, several potentially relevant observational studies did not meet our inclusion criteria (50,55,56). For example, a 2005 multicenter cohort study did not capture preadmission beta-blocker use (55). Thus, the investigators could not differentiate between ongoing, long-term beta-blocker use and new perioperative beta-blocker use for cardiac risk reduction. Similarly, a 2010 single-center study also included patients undergoing cardiac surgery and grouped all individuals receiving beta blockers prior to surgery into a single category, regardless of the duration of preoperative therapy (56). Most recently, a 2013 multicenter cohort study defined perioperative beta blockade on the basis of on any relevant prescription on either the day of surgery or the subsequent day (50). Nonetheless, the study did not analyze the dose or duration of inpatient beta-blocker prescriptions; hence, it did not meet our criterion for minimum duration of postoperative beta-blocker therapy. Additionally, the data sources in all 3 studies could not differentiate between beta blockade for preventing cardiac events and beta blockade for treating postoperative complications (e.g., myocardial ischemia). The importance of distinguishing between prophylactic and therapeutic interventions is underscored by the observation that, in some RCTs of perioperative beta blockade, 7% to 10% of participants in the control arm still received open-label beta blockers, possibly to treat new postoperative complications (4,12).
In summary, this systematic review found that perioperative beta blockade started within 1 day or less before noncardiac surgery helps prevent nonfatal MI but at the cost of increased risks of stroke, death, hypotension, and bradycardia. The DECREASE-I and DECREASE-IV trials differed from other trials with regard to design in that they are the only RCTs that assessed beta blockade started 2 or more days before surgery. Their results differed significantly from other RCTs in that perioperative mortality rate was decreased, as opposed to increased, with beta-blocker therapy. In the absence of these controversial studies, there are insufficient robust data on the efficacy and safety of perioperative beta-blocker regimens that use agents aside from metoprolol or initiate treatment 2 to 45 days prior to surgery.
Presidents and Staff
American College of Cardiology
Patrick T. O’Gara, MD, FACC, President
Shalom Jacobovitz, Chief Executive Officer
William J. Oetgen, MD, MBA, FACC, Executive Vice President, Science, Education, and Quality
Amelia Scholtz, PhD, Publications Manager, Science, and Clinical Policy
American College of Cardiology/American Heart Association
Lisa Bradfield, CAE, Director, Science and Clinical Policy
Emily Cottrell, MA, Quality Assurance Specialist, Science and Clinical Policy
American Heart Association
Elliott Antman, MD, FAHA, President
Nancy Brown, Chief Executive Officer
Rose Marie Robertson, MD, FAHA, Chief Science Officer
Gayle R. Whitman, PhD, RN, FAHA, FAAN, Senior Vice President, Office of Science Operations
Anne Leonard, MPH, RN, FAHA, Science and Medicine Advisor, Office of Science Operations
Jody Hundley, Production Manager, Scientific Publications, Office of Science Operations
We thank Mr. Jonothan Earnshaw, Dr. Martin London, Dr. Homer Yang, and Dr. Michael Zaugg for responding to our enquiries about their publications. We are grateful to Dr. Hance Clarke, Dr. Zhigang Duan, and Dr. Rita Katznelson for their invaluable help in interpreting non–English language papers identified by the systematic review.
Appendix 1 Author Relationships With Industry and Other Entities (Relevant)∗—Perioperative Beta Blockade in Noncardiac Surgery: A Systematic Review for the 2014 ACC/AHA Guideline on Perioperative Cardiovascular Evaluation and Management of Patients Undergoing Noncardiac Surgery (July 2014)
|Committee Member||Employment||Consultant||Speakers Bureau||Ownership/Partnership/Principal||Personal Research||Institutional, Organizational, or Other Financial Benefit||Expert Witness|
|Duminda N. Wijeysundera (ERC Chair)||Li Ka Shing Knowledge Institute of St. Michael’s Hospital—Scientist; Toronto General Hospital—Staff, Department of Anesthesia and Pain Management; University of Toronto—Assistant Professor, Department of Anesthesia and Institute of Health Policy Management and Evaluation; Institute for Clinical Evaluative Sciences—Adjunct Scientist||None||None||None||None||None||None|
|Dallas Duncan||University of Toronto—Anesthesiology Residency, Clinical Investigator Program||None||None||None||None||None||None|
|Lee A. Fleisher (Perioperative Guideline Chair)||University of Pennsylvania Health System Department of Anesthesiology and Critical Care—Chair||None||None||None||None||None||None|
|Kirsten E. Fleischmann, (Perioperative Guideline Vice Chair)||UCSF School of Medicine, Division of Cardiology—Professor of Clinical Medicine||None||None||None||None||None||None|
|Chileshe Nkonde-Price||Yale University School of Medicine—Cardiovascular Disease Medicine Fellow; University of Pennsylvania School of Medicine—Robert Wood Johnson Clinical Scholars Program Fellow||None||None||None||None||None||None|
|Salim S. Virani||Michael E. DeBakey VA Medical Center—Staff Cardiologist; VA Health Services Research and Development Center for Innovations in Quality, Effectiveness and Safety—Investigator; Baylor College of Medicine—Assistant Professor, Section of Cardiovascular Research; Associate Director for Research, Cardiology Fellowship Training Program||None||None||None||None||None||None|
|Jeffrey B. Washam||Duke University Medical Center, Duke Heart Center—Clinical Pharmacist, Cardiac Intensive Care Unit||None||None||None||None||None||None|
This table represents the relationships of ERC members with industry and other entities that were determined to be relevant to this initiative. These relationships were reviewed and updated in conjunction with all conference calls of the ERC during the evidence review process. The table does not necessarily reflect relationships with industry at the time of publication. A person is deemed to have a significant interest in a business if the interest represents ownership of ≥5% of the voting stock or share of the business entity, or ownership of ≥$10,000 of the fair market value of the business entity; or if funds received by the person from the business entity exceed 5% of the person’s gross income for the previous year. Relationships that exist with no financial benefit are also included for the purpose of transparency. Relationships in this table are modest unless otherwise noted.
According to the ACC/AHA, a person has a relevant relationship IF: a) the relationship or interest relates to the same or similar subject matter, intellectual property or asset, topic, or issue addressed in the document; or b) the company/entity (with whom the relationship exists) makes a drug, drug class, or device addressed in the document, or makes a competing drug or device addressed in the document; or c) the person or a member of the person’s household, has a reasonable potential for financial, professional, or other personal gain or loss as a result of the issues/content addressed in the document.
ACC indicates American College of Cardiology; AHA, American Heart Association; ERC, Evidence Review Committee; UCSF, University of California, San Francisco; and VA, Veterans Affairs.
↵∗ These members of the Evidence Review Committee are listed alphabetically and all participated equally in the process.
↵† Former Task Force member; current member during the writing effort.
This document was approved by the American College of Cardiology Board of Trustees and the American Heart Association Science Advisory and Coordinating Committee in July 2014.
The American College of Cardiology requests that this document be cited as follows: Wijeysundera DN, Duncan D, Nkonde-Price C, Virani SS, Washam JB, Fleischmann KE, Fleisher LA. Perioperative beta blockade in noncardiac surgery: a systematic review for the 2014 ACC/AHA guideline on perioperative cardiovascular evaluation and management of patients undergoing noncardiac surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2014;64:2406–25.
This article has been copublished in Circulation.
Copies: This document is available on the World Wide Web sites of the American College of Cardiology (www.cardiosource.org) and the American Heart Association (my.americanheart.org). For copies of this document, please contact the Elsevier Inc. Reprint Department via fax (212) 633-3820 or e-mail .
Permissions: Multiple copies, modification, alteration, enhancement, and/or distribution of this document are not permitted without the express permission of the American College of Cardiology. Requests may be completed online via the Elsevier site (http://www.elsevier.com/authors/obtainingpermission-to-re-useelsevier-material).
- American College of Cardiology Foundation and the American Heart Association, Inc.
- Devereaux P.J.,
- Goldman L.,
- Cook D.J.,
- et al.
- Poldermans D.,
- Boersma E.,
- Bax J.J.,
- et al.
- Eagle K.A.,
- Brundage B.H.,
- Chaitman B.R.,
- et al.
- Eagle K.A.,
- Berger P.B.,
- Calkins H.,
- et al.
- Fleisher L.A.,
- Beckman J.A.,
- Brown K.A.,
- et al.
- Fleisher L.A.,
- Beckman J.A.,
- Brown K.A.,
- et al.
- Fleisher L.A.,
- Beckman J.A.,
- Brown K.A.,
- et al.
- Juul A.B.,
- Wetterslev J.,
- Gluud C.,
- et al.
- Poldermans D.,
- Devereaux P.J.
- Dunkelgrun M.,
- Boersma E.,
- Schouten O.,
- et al.
- ↵Erasmus MC Follow-up Investigation Committee. Investigation Into Possible Violation of Scientific Integrity: Report Summary. November 16, 2011.
- Erasmus MC Follow-up Investigation Committee. Report on the 2012 Follow-Up Investigation of Possible Breaches of Academic Integrity. September 30, 2012.
- Bouri S.,
- Shun-Shin M.J.,
- Cole G.D.,
- et al.
- Jacobs A.K.,
- Kushner F.G.,
- Ettinger S.M.,
- et al.
- Fleisher L.A.,
- Fleischmann K.E.,
- Auerbach A.D.,
- et al.
- Devereaux P.J.,
- Beattie W.S.,
- Choi P.T.-L.,
- et al.
- Higgins J.P.,
- Altman D.G.,
- Gotzsche P.C.,
- et al.
- ↵Wells GA, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Ottawa, ON—Ottawa Hospital Research Institute, 2013. Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed May 29, 2014.
- Higgins J.P.T.,
- Thompson S.G.,
- Deeks J.J.,
- et al.
- Egger M.,
- Davey Smith G.,
- Schneider M.,
- et al.
- Zaugg M.,
- Bestmann L.,
- Wacker J.,
- et al.
- Flu W.-J.,
- van Kuijk J.-P.,
- Chonchol M.,
- et al.
- Redelmeier D.,
- Scales D.,
- Kopp A.
- Lee T.H.,
- Marcantonio E.R.,
- Mangione C.M.,
- et al.
- ACC/AHA Task Force Members
- Table of Contents
- Presidents and Staff
- Appendix 1 Author Relationships With Industry and Other Entities (Relevant)∗—Perioperative Beta Blockade in Noncardiac Surgery: A Systematic Review for the 2014 ACC/AHA Guideline on Perioperative Cardiovascular Evaluation and Management of Patients Undergoing Noncardiac Surgery (July 2014)