Impact of Prophylactic β-Blocker Therapy to Prevent Stroke After Noncardiac Surgery
Article Outline
β Blockers are widely used to improve the postoperative cardiac outcome in patients with coronary artery disease scheduled for noncardiac surgery. However, recently serious concerns regarding the safety of perioperative β blockers have emerged. To assess the incidence, risk factors, and β-blocker use associated with postoperative stroke in the Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography (DECREASE) trials, we evaluated all 3,884 patients of the DECREASE trials for postoperative stroke. All cardiac risk factors and medication use were assessed. The incidence of stroke within 30 days after surgery was recorded. The incidence of postoperative stroke in the DECREASE trials was 0.46% (18 of 3,884). For the β-blocker users, the incidence was 0.5%. All the strokes had an ischemic origin. A history of stroke was associated with a greater incidence of postoperative stroke (odds ratio [OR] 3.79, 95% confidence interval [CI] 1.2 to 11.6). Statins and anticoagulants were not associated with postoperative stroke (OR 0.85, 95% CI 0.3 to 2.4; and OR 1.27, 95% CI 0.4 to 4.6, respectively). No association with bisoprolol therapy was found (OR 1.16, 95% CI 0.4 to 3.4). In conclusion, with a low-dose bisoprolol regimen started ≥30 days before surgery, no association was observed between β-blocker use and postoperative stroke.
β Blockers are prescribed in the perioperative period to reduce myocardial oxygen consumption by decreasing sympathetic tone and myocardial contractility. Several randomized clinical trials have shown beneficial effects of initiating β blockers to patients at risk of cardiovascular complications,1, 2 but other trials have failed to show any significant improvement in outcome.3, 4 To resolve these inconsistencies, a large multinational randomized controlled trial was started in 2002: the PeriOperative ischemic Evaluation (POISE) trial.5
The POISE trial randomly assigned 8,351 patients with, or at risk of, atherosclerotic disease who were undergoing noncardiac surgery to receive extended-release metoprolol succinate (n = 4,174) or placebo (n = 4,177). In the POISE trial, high doses of metoprolol, ≤400 mg, were administered in 4 doses starting 2 to 4 hours before surgery and continued for 30 days. The 30-day results showed a significant reduction in cardiac events. However, this came at the cost of a significant increase in the incidence of total mortality and stroke. The high incidence of stroke found in the POISE trial made us question the liberal use of β blockers during the perioperative period.
In the Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography (DECREASE) trials, a low-dose β-blocker regimen was used. No relation was found between stroke and β-blocker use. However, that experience was based on individual DECREASE studies with associated weak statistical power. Therefore, we combined the results of the DECREASE I, II, and IV studies to evaluate the effect of low-dose β-blocker use in the perioperative period on postoperative stroke and to identify the risk factors for postoperative stroke.
Methods
We performed a pooled analysis of all 3,884 patients, who were enrolled in the DECREASE I (1996 to 1999), DECREASE II (2000 to 2005), and DECREASE IV (2004 to 2008) trials.2, 6, 7 We did not include the DECREASE III trial, because it was a randomized controlled trial in which vascular surgery patients were randomized for statins or placebo. The DECREASE I trial was a randomized controlled trial that clearly demonstrated the cardioprotective effect of bisoprolol in high-risk patients undergoing major vascular surgery. The aim of the DECREASE II study was to assess the value of cardiac testing according to the American College of Cardiology/American Heart Association guidelines for intermediate-risk patients receiving β-blocker therapy with tight heart rate control scheduled for major vascular surgery. The DECREASE IV trial was a large prospective, randomized trial that sought to assess the effect of different pharmaceutical strategies (β blockers or statins, or both) to prevent perioperative cardiac complications in intermediate-risk patients. The results and study protocols have been previously published in detail.2, 6, 7
The following potential risk factors for postoperative stroke were recorded from all the trials and included in the present analysis: age >70 years, angina pectoris, previous myocardial infarction on the basis of history or a finding of pathologic Q waves on the electrocardiogram, compensated congestive heart failure or a history of congestive heart failure, drug therapy for diabetes mellitus, renal dysfunction (serum creatinine >160 μmol/L), and previous stroke or transient ischemic attack. Throughout the DECREASE I, II, and IV studies, the definition of the several risk factors was not changed. Also, the type and dosage of perioperative β blockers were noted. The β-blocker dose was converted to a percentage of the maximum recommended therapeutic dose (MRTD) according to the Food and Drug Administration's Center for Drug Evaluation and Research database.8 The MRTD for atenolol is 3.330 mg/kg body weight/day; for bisoprolol, 0.330 mg/kg body weight/day; for metoprolol, 6.670 mg/kg body weight/day; for carvedilol, 0.417 mg/kg body weight/day; for propranolol, 10.700 mg/kg body weight/day; and for labetalol, 40.700 mg/kg body weight/day.
The end point for the present study was stroke, either a transient ischemic attack or a cerebrovascular accident within 30 days after surgery. Stroke was confirmed by computed tomography according to the criteria stated in the guidelines from the Stroke Council, American Heart Association.9 As per protocol, all patients suspected to have had an in-hospital stroke were evaluated by computed tomography.
Because the results of the POISE trial were our incentive to perform the present analysis, we also compared the rate of strokes recently reported for the POISE trial.5 We also compared both studies concerning the possible risk factors for postoperative stroke.
The continuous data are described as the mean ± SD or the median and range. Dichotomous data are described as numbers and percentages. Logistic regression analysis was used to identify the clinical characteristics and medical therapy (e.g., statins, β blockers, antiplatelet agents, oral anticoagulants) associated with postoperative stroke. Interaction terms were included if statistically significant. Odds ratios (ORs) are reported with the corresponding 95% confidence intervals (CIs). For all tests, p <0.05 (2-sided) was considered significant. All statistical analyses were performed using Statistical Package for Social Sciences statistical software (SPSS, Chicago, Illinois).
Results
The baseline clinical characteristics of the patients included in the DECREASE I, II, and IV trials are listed in Table 1. The key characteristics were as follows: mean age 67 years, 71% men, 16% with diabetes mellitus, 28% with angina pectoris, 26% with previous myocardial infarction, and 11% with a history of cerebrovascular disease. Overall, 37% of the patients were receiving statin therapy; the patients in the DECREASE IV study were randomized for statin therapy. Most patients (73%) underwent vascular surgery. The incidence of postoperative stroke in the DECREASE trials was 0.46% (18 of 3,884). In the DECREASE I, II, and IV trials, 3, 8, and 7 patients experienced stroke, respectively. The baseline characteristics of the patients with stroke are listed in Table 2. Perioperative β-blocker use was present in 12 (67%) of 18 patients who experienced a postoperative stroke. The average dose of bisoprolol used among the stroke patients was 15% of the MRTD. Among the β-blocker users, the incidence was 0.5% (12 of 2,366). All strokes had an ischemic origin. The median day on which the strokes had occurred was the second postoperative day (interquartile range 1 to 4). A history of stroke was associated with postoperative stroke (OR 3.79, 95% CI 1.2 to 11.6). Statins and anticoagulants were not associated with postoperative stroke (OR 0.85, 95% CI 0.3 to 2.4; and OR 1.27, 95% CI 0.4 to 4.6, respectively). No association with bisoprolol therapy was found (OR 1.16, 95% CI 0.4 to 3.4; Table 3).
Table 1. Baseline characteristics of patients in Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography (DECREASE) and PeriOperative ischemic Evaluation (POISE) trials
| Variable | DECREASE I Trial (n = 1,351) | DECREASE II Trial (n = 1,467) | DECREASE IV Trial (n = 1,066) | DECREASE Trials Overall (n = 3,884) | POISE Trial (n = 8,351) |
|---|---|---|---|---|---|
| Age (years) | 67.5 | 67.0 | 66.8 | 67.1 | 69.0 |
| Men | 1,050 (77%) | 1,075(73%) | 639(60%) | 2,764(71%) | 5,293(63%) |
| Ischemic heart disease | 1,596(41%) | 3,589(43%) | |||
| Myocardial infarction | 562(42%) | 396(27%) | 54(5%) | 1,012(26%) | |
| Angina pectoris | 229(17%) | 808(55%) | 55(5%) | 1,092(28%) | |
| Heart failure | 96(7%) | 168(11%) | 8(1%) | 272(7%) | 220(3%) |
| Diabetes mellitus | 187(13%) | 320(21%) | 115(11%) | 622(16%) | 2,427(29%) |
| Cerebrovascular disease | 104(8%) | 280(19%) | 46(4%) | 428(11%) | 1,263(15%) |
| Renal dysfunction | 59(4%) | 124(8%) | 11(1%) | 194(5%) | 401(5%) |
| Medication use | |||||
| Statins | 285(21%) | 617(42%) | 533(50%) | 1,435(37%) | 2,677(32%) |
| Angiotensin-converting enzyme inhibitors | 394(29%) | 478(32%) | 96(9%) | 968(25%) | 3,717(44%) |
| Aspirin | 406(30%) | 664(45%) | 102(9%) | 1,174(30%) | 3,011(36%) |
Table 2. Baseline characteristics of patients with postoperative stroke
| Variable | n = 18 |
|---|---|
| Age (years) | 71.6 |
| Men | 13(72%) |
| Myocardial infraction | 5(27%) |
| Angina pectoris | 1(5%) |
| Heart failure | 0(0%) |
| Diabetes mellitus | 3(16%) |
| Cerebrovascular disease | 6(33%) |
| Renal dysfunction | 1(5%) |
| Medication use | |
| β Blocker | 12(67%) |
| Statins | 11(32%) |
| Angiotensin-converting enzyme inhibitors | 4(22%) |
| Aspirin | 8(44%) |
| Surgery type | |
| Colorectal | 1(5%) |
| Esophagogastric | 3(16%) |
| Ear/nose/throat | 1(5%) |
| Urologic | 1(5%) |
| Orthopedic | 1(5%) |
| Vascular | 11(61%) |
Table 3. Adjusted odds ratios for perioperative stroke in relation to risk factors and medication using multivariate analysis
| Variable | Odds Ratio | 95% Confidence Interval |
|---|---|---|
| Medical risk factors | ||
| Age (per year increase) | 1.04 | 0.99–1.10 |
| Male gender | 1.41 | 0.45–4.42 |
| Ischemic heart disease | 0.4 | 0.13–1.21 |
| Cerebrovascular disease | 3.79 | 1.24–11.6 |
| Diabetes mellitus | 0.88 | 0.31–3.93 |
| Medication use | ||
| β Blockers | 1.16 | 0.40–3.35 |
| Statins | 0.85 | 0.30–2.40 |
| Antiplatelet therapy | 1.75 | 0.60–5.15 |
| Oral anticoagulants | 1.27 | 0.35–4.64 |
Discussion
The results of the present study have shown that low-dose bisoprolol started ≥30 days before surgery is not associated with an increased risk of postoperative stroke. Compared to the increased risk of postoperative stroke found in the POISE study (OR 2.2, 95% CI 1.3 to 3.8), we did not find a significant increased risk of postoperative stroke (OR 1.16, 95% CI 0.4 to 3.4). The key question is how this difference could be explained. To explain the difference in the ORs found, several aspects should be considered, including the indication for perioperative β-blocker therapy, timing of the initiation of therapy, type of β blockers used, and dosage and treatment targets.
The 2007 update on perioperative β-blocker therapy in the American College of Cardiology and American Heart Association guidelines for perioperative cardiovascular assessment for noncardiac surgery has recommended β blockers for patients already receiving therapy or those who are undergoing vascular surgery and have ischemia on preoperative testing (class I). β-blocker therapy has also been recommended for patients undergoing vascular surgery or intermediate- or high-risk nonvascular surgery, who have a high risk of coronary disease or those with established disease (class II).10 The results of some randomized trials, including the POISE study, have not supported the recommendations in the guidelines and have shown no beneficial effect of perioperative β blockade.3, 4, 5, 11
In a review, Lindenauer et al12 attempted to identify the risk categories that might benefit or be harmed by β blockers. They found that patients with the greatest risk, stratified according to the revised cardiac risk index by Lee et al,13 showed the greatest benefit from β blockade. However, those in the low- to moderate-risk categories showed a trend toward harm from β blockade due to bradycardia and hypotension. Therefore, patient evaluation and selection are the first and most import elements before β-blocker therapy should be initiated. The DECREASE I trial included patients within the highest risk category: inducible ischemia found on the dobutamine stress echocardiogram. The POISE study, as well as the DECREASE II and IV trials, included patients with low to moderate risk. Although the inclusion criteria of both studies were roughly the same, the results were different. Therefore, other aspects of β-blocker initiation should be evaluated.
The timing that initiation plays is a second key element. By starting β-blocker therapy within hours before surgery, any adverse effects from the β blockers might be overlooked. The response to β-blocker therapy in these patients cannot be adequately assessed within this short period, and the danger of an overdose to these patients seems obvious. In studies that have started β-blocker therapy hours before surgery, the incidence of postoperative stroke was greater than in studies that started β-blocker therapy ≥1 week before surgery.2, 3, 4, 5, 7, 11, 14 Evidence has also suggested that the acute effects of β blockade (i.e., decreased heart rate and systolic pressure and reduced myocardial ischemia) might not fully explain the benefit of perioperative therapy, because anti-inflammatory and plaque-stabilizing properties take days to develop.15 In patients with heart failure, it has been shown safe and effective to start β-blocker therapy at a relatively low dose and subsequently slowly increase the dose to control the blood pressure and heart rate.16
Third, the type of β blocker used might have led to different study outcomes. Although both the POISE and the DECREASE trials clearly demonstrated cardioprotective effects from β blockers, some evidence suggested that β blockers that are highly β1 selective show better results than those that are moderately β1 selective. This might explain why bisoprolol (highly β1 selective) has been associated with better results, and metoprolol and atenolol (moderately β1 selective) have been associated with mixed results in clinical trials. In a study by Redelmeier et al,17 of 37,151 patients receiving perioperative β blockers, longer acting agents (e.g., atenolol) demonstrated greater cardioprotection than short-acting agents (e.g., metoprolol).
Fourth, genetic and cultural differences might play an important role. Lemaitre et al18 investigated the interaction of variations in β-adrenegeric receptor genes with β-blocker use on the risk of myocardial infarction and ischemic stroke. Several genetic variations in the β1-adrenergic receptor gene interacted with β-blocker use in the risk of both myocardial infarction and ischemic stroke. β-Blocker use was associated with a greater risk of combined myocardial infarction and ischemic stroke in carriers of the rs#2429511 (OR 1.24, 95% CI 1.03 to 1.50) compared to carriers of the common allele (OR 0.70, 95% CI 0.51 to 0.94). Two other major single nucleotide polymorphisms in the β-adrenergic receptors genes (Ser49Gly and Arg389Gly) have been identified in several studies and might affect drug responses.19, 20 The role of these polymorphisms in the perioperative setting is still unclear. Similarly, interethnic differences in the response to β-blocker therapy have been reported. African American patients have been shown to respond poorly and Chinese patients to respond well to β blockers.21 These intercultural differences might easily lead to overdosing or underdosing and therefore do not favor a fixed dose regimen. Because >40% of the patients included in the POISE trial were of Asian and South-American origin, a subanalysis of these patient groups would be justified.
Finally the balance between the β-blockers does to achieve the target heart rates in relation to the drug side effects should be optimized. The bisoprolol dose used in the DECREASE studies was 10% to 20% of the MRTD. In contrast, the POISE trial used 50% to 100% of the MRTD of metoprolol. Both POISE and DECREASE withheld the administration of β blockers if the heart rate decreased to <50 beats/min or if the systolic blood pressure decreased to <100 mm Hg. The importance of heart rate control has been shown in several studies. The trials that achieved the most effective control of the heart rate were associated with a reduced incidence of postoperative myocardial infarction, suggesting that effective control of the heart rate is a key item for achieving cardioprotection.22, 23, 24 Both the DECREASE and the POISE studies showed that, although they had different treatment protocols, β blockers were effective in reducing perioperative cardiac complications. The POISE trial prescribed a high dose immediately before surgery, and the DECREASE trials favored a dose titration approach for a prolonged period. However, the cardioprotective effect of a high-dose regimen resulted in an increased incidence of side effects, such as stroke.
In a recent review by Bangalore et al,25 112 randomized controlled trials of perioperative β-blocker use were analyzed. They concluded that for the overall cohort treatment of 1,000 patients, β-blocker use resulted in 16 fewer nonfatal myocardial infarctions in survivors but at the expense of 3 disabling strokes, 45 patients with clinically significant perioperative bradycardia, 59 with hypotension, and potentially increased mortality. The overall OR for perioperative nonfatal stroke was 2.16 (95% CI 1.27 to 3.68). This was mainly driven by the results from the POISE study, especially because the other randomized controlled trials failed to show a significant correlation between perioperative β-blocker use and postoperative stroke. However, the results from the DECREASE studies were not included in their analysis.
We believe that the protocol used in the DECREASE studies (low-dose, long-acting agents titrated to effect ≥30 days in advance) is associated with an overall benefit compared to the risk and that high-dose therapy started the morning of surgery is associated with an increased risk rather than benefit.
References
- . 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
- . The effect of bisoprolol on perioperative mortality and myocardial infarction in high-risk patients undergoing vascular surgery. N Engl J Med. 1999;341:1789–1794
- . Effect of perioperative beta blockade in patients with diabetes undergoing major non-cardiac surgery: randomised placebo controlled, blinded multicentre trial. BMJ. 2006;332:1482
- . The effects of perioperative beta-blockade: results of the Metoprolol after Vascular Surgery (MaVS) study, a randomized controlled trial. Am Heart J. 2006;152:983–990
- . Effects of extended-release metoprolol succinate in patients undergoing non-cardiac surgery (POISE trial): a randomised controlled trial. Lancet. 2008;371:1839–1847
- . Bisoprolol and fluvastatin for the reduction of perioperative cardiac mortality and myocardial infarction in intermediate-risk patients undergoing noncardiovascular surgery: a randomized controlled trial (DECREASE-IV). Ann Surg. 2009;249:921–926
- . Should major vascular surgery be delayed because of preoperative cardiac testing in intermediate-risk patients receiving beta-blocker therapy with tight heart rate control?. J Am Coll Cardiol. 2006;48:964–969
- . Maximum recommended therapeutic dose database. http://www.fda.gov/cder/Offices/OPS_IO/MRTD.htmAccessed November 6, 2009
- . Practice guidelines for the use of imaging in transient ischemic attacks and acute stroke: a report of the Stroke Council, American Heart Association. Stroke. 1997;28:1480–1497
- . ACC/AHA 2007 guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery) developed in collaboration with the American Society of Echocardiography, American Society of Nuclear Cardiology, Heart Rhythm Society, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society for Vascular Medicine and Biology, and Society for Vascular Surgery. J Am Coll Cardiol. 2007;50:1707–1732
- . Perioperative beta-blockade (POBBLE) for patients undergoing infrarenal vascular surgery: results of a randomized double-blind controlled trial. J Vasc Surg. 2005;41:602–609
- . Perioperative beta-blocker therapy and mortality after major noncardiac surgery. N Engl J Med. 2005;353:349–361
- . Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation. 1999;100:1043–1049
- . Adrenergic receptor genotype but not perioperative bisoprolol therapy may determine cardiovascular outcome in at-risk patients undergoing surgery with spinal block: the Swiss Beta blocker in Spinal Anesthesia (BBSA) study: a double-blinded, placebo-controlled, multicenter trial with 1-year follow-up. Anesthesiology. 2007;107:33–44
- . Early use of beta-blockers is associated with attenuation of serum C-reactive protein elevation and favorable short-term prognosis after acute myocardial infarction. Cardiology. 2003;99:47–53
- . ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2008: the Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2008 of the European Society of Cardiology: developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur Heart J. 2008;29:2388–2442
- . Beta blockers for elective surgery in elderly patients: population based, retrospective cohort study. BMJ. 2005;331:932
- . Beta1- and beta2-adrenergic receptor gene variation, beta-blocker use and risk of myocardial infarction and stroke. Am J Hypertens. 2008;21:290–296
- . Beta1- and beta2-adrenoceptor polymorphisms and cardiovascular diseases. Fundam Clin Pharmacol. 2008;22:107–125
- . Beta 1-adrenergic receptor polymorphisms confer differential function and predisposition to heart failure. Nat Med. 2003;9:1300–1305
- . Altered sensitivity to and clearance of propranolol in men of Chinese descent as compared with American whites. N Engl J Med. 1989;320:565–570
- . Does tight heart rate control improve beta-blocker efficacy? (An updated analysis of the noncardiac surgical randomized trials). Anesth Analg. 2008;106:1039–1048
- . High-dose beta-blockers and tight heart rate control reduce myocardial ischemia and troponin T release in vascular surgery patients. Circulation. 2006;114:I344–I349
- . Beta-blockade in noncardiac surgery: outcome at all levels of cardiac risk. Arch Surg. 2008;143:940–944
- . Perioperative beta blockers in patients having non-cardiac surgery: a meta-analysis. Lancet. 2008;372:1962–1976
PII: S0002-9149(09)02211-5
doi:10.1016/j.amjcard.2009.08.646
© 2010 Elsevier Inc. All rights reserved.
