Plain language summary
Beta-blockers for preventing stroke recurrence
People who have had a stroke or a transient ischaemic attack (TIA) are at risk of having further strokes or heart attacks, or other serious circulatory problems. Beta-blockers are drugs that reduce heart rate and blood pressure, and have other effects that might also reduce the risks of stroke and heart attack. We found two trials involving 2193 participants that tested beta-blockers after stroke in people with a recent stroke or TIA. No clear evidence indicated that beta-blockers reduced the risk of stroke, heart attack, or death from vascular disease. More studies with larger samples are needed.
Les bêta-bloquants pour la prévention de récidives d'accidents vasculaires cérébraux (AVC)
Les bêta-bloquants pour la prévention de récidives d'AVC
Les personnes qui ont subi un accident vasculaire cérébral (AVC) ou un accident ischémique transitoire (AIT) risquent d'avoir d'autres AVC, crises cardiaques ou problèmes circulatoires graves. Les bêta-bloquants sont des médicaments qui abaissent le rythme cardiaque et la tension artérielle, et ont d'autres effets également susceptibles de réduire les risques d'AVC et de crise cardiaque. Nous avons trouvé deux essais impliquant 2193 participants qui avaient testé les bêta-bloquants chez les personnes ayant récemment subi un AVC ou un AIT. Il n'y avait pas de preuve claire que les bêta-bloquants aient réduit le risque d'AVC, de crise cardiaque ou de décès par maladie vasculaire. Des études supplémentaires à plus grands effectifs sont nécessaires.
Notes de traduction
Traduit par: French Cochrane Centre 3rd June, 2013
Traduction financée par: Pour la France : Minist�re de la Sant�. Pour le Canada : Instituts de recherche en sant� du Canada, minist�re de la Sant� du Qu�bec, Fonds de recherche de Qu�bec-Sant� et Institut national d'excellence en sant� et en services sociaux.
Stroke is one of the main causes of death worldwide. It is also a leading cause of disability and dependence, especially in the elderly. Among the survivors, many become functionally dependent, thus imposing a great burden on the family and community. Stroke caused an estimated 5.7 million deaths in 2005, and 87% of these deaths were reported in low-income and middle-income countries. Estimates suggest that by the year 2020, stroke will be the second leading cause of death worldwide (Strong 2007).
Systemic hypertension is an important and well-established risk factor for cardiovascular disease. In a study of the global burden of blood pressure-related disease in 2001, approximately 54% of strokes worldwide were attributable to high blood pressure (Lawes 2008). Evidence suggests that treating hypertension can reduce cardiovascular risk in people with a prior stroke (Gueyffier 1997). Similarly, other studies have found that stroke survivors actually benefited from blood pressure-lowering treatment even if they did not have hypertension at baseline: PROGRESS 2001 found that treatment with perindopril (adding indapamide as required) significantly reduced the risk of subsequent stroke.
Stroke and transient ischaemic attack (TIA) increase the risk of a subsequent stroke. Without preventive treatment, there is a 7% to 12% annual risk of death from vascular causes, non-fatal stroke, or non-fatal myocardial infarction in people suffering a TIA or a non-disabling ischaemic stroke (Hankey 1998). In a prospective observational study, the risk of stroke after TIA or minor stroke was as high as 18.5% at three months (Coull 2004). Treating modifiable risk factors, such as hypertension, diabetes, and tobacco use, may reduce such risk. Also, antiaggregant agents have proven efficacy in preventing stroke recurrence (Shah 2000).
Beta-blockers have long been used as first-line agents to treat hypertension and have also been used as the reference drug in randomised controlled trials (RCTs), compared with other agents, to treat hypertension. However, since the end of the last decade, systematic reviews, meta-analyses, and RCTs have put in doubt the efficacy of these drugs in preventing outcomes such as death and vascular events in hypertensive patients. In a recent meta-analysis, Messerli and colleagues concluded that, in uncomplicated hypertension, neither diuretics nor beta-blockers are acceptable as first-line treatment (Messerli 2008). Another meta-analysis has shown that, in comparison with other antihypertensive drugs, the effect of beta-blockers is less than optimal, with a raised risk of stroke. The authors concluded that beta-blockers should not remain as the first choice of drug in the treatment of hypertension and should not be used as reference drugs in RCTs of hypertension (Lindholm 2005). The Blood Pressure Lowering Treatment Trialists Collaboration overview found that treatment with any commonly used regimen reduces the risk of cardiovascular events, but with some differences between regimens. Regimens based on beta-blockers showed a trend toward greater risk reduction compared with regimens based on angiotensin-converting enzyme (ACE) inhibitors, and regimens based on calcium antagonists showed a trend toward greater risk reduction compared with those based on beta-blockers (BPLTTC 2003). A Cochrane review evaluating the efficacy of beta-blockers for treating hypertension concluded that available evidence does not support the use of beta-blockers as first-line drugs (Wiysonge 2007). Moreover, RCTs comparing beta-blockers with other drugs in hypertensive patients have shown negative results. In the Losartan Intervention For Endpoint reduction in hypertension study (LIFE), losartan prevented cardiovascular morbidity and death more frequently than atenolol for similar reductions in blood pressure (Dahlöf 2002). Furthermore, in the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT), an amlodipine-based regimen prevented major cardiovascular events more often and induced diabetes less frequently than an atenolol-based regimen (Dahlöf 2005). Nevertheless, diagnostic criteria for hypertension and blood pressure targets have evolved to lower values over the years; the efficacy of beta-blockers was established in populations with higher levels of blood pressure. Hence, a meta-analysis including trials from different decades may underestimate the efficacy of beta-blockers.
Studies showing better outcomes with specific antihypertensive therapy, such as ACE inhibitors or angiotensin receptor blockers (ARBs), compared with beta-blockers or diuretics were unable to address whether there is a mechanism for risk reduction that is independent of blood pressure lowering. In the Valsartan Antihypertensive Long-term Use Evaluation (VALUE) study (Julius 2004), blood pressure lowering was more intense in people who received amlodipine than in those who received valsartan. The cardiovascular endpoint, however, was similar in both groups.
The 'beyond-blood pressure effect' may be related to differences between pulse pressure and central blood pressure. A recent study found that, despite similar results for peripheral blood pressure, atenolol had less impact than eprosartan on central systolic blood pressure , which could explain differences in outcomes, especially stroke (Dhakam 2006).
However, relatively few studies have tested the impact of beta-blockers in people with or without hypertension who have had a TIA or stroke (DUTCH TIA TRIAL 1993). This approach may be effective in preventing stroke recurrence.
Description of the condition
Previous stroke of any severity or TIA due to arterial thrombosis or embolism (symptoms persisting for less than 24 hours).
Description of the intervention
Adrenergic beta-antagonist drugs or beta-blockers: acebutolol, alprenolol, atenolol, betaxolol, bisoprolol, bucindolol, bufuralol, bupranolol, butoxamine, carteolol, carvedilol, celiprolol, epanolol, esmolol, labetalol, metoprolol, nadolol, oxprenolol, pindolol, propranolol, sotalol, and timolol.
How the intervention might work
Beta-blockers might work by reducing arterial pressure and cardiovascular events. They have been used for decades as first-line agents to treat hypertension. Clinical evidence suggests that they reduce hypertension and cardiovascular events (BPLTTC 2003).
Why it is important to do this review
No previous review has focused on beta-blockers for the secondary prevention of stroke. Secondary prevention is very important, given the high rate of stroke recurrence.
To evaluate the efficacy of beta-blockers for preventing stroke recurrence and for reducing death and major vascular events in people with a previous stroke or TIA, and to determine their safety, particularly with regard to the development of diabetes mellitus.
Criteria for considering studies for this review
Types of studies
We included RCTs of unconfounded comparisons of beta-blocker versus control, or beta-blocker plus other treatment, versus other treatment in people with previous stroke or TIA.
Types of participants
Adults with a prior history of stroke or TIA due to arterial thrombosis or embolism, with and without hypertension. We excluded trials that included participants without a previous stroke or TIA unless we could identify a subgroup of stroke patients for whom separate results were available.
Types of interventions
Beta-blocker versus control, or beta-blocker plus other treatment versus other treatment.
Types of outcome measures
Fatal and non-fatal stroke.
Major vascular events, death from all causes, death from cardiovascular causes, change in blood pressure (mean blood pressure as a continuous outcome), development of diabetes, adverse events, quality of life.
Search methods for identification of studies
See the 'Specialized register' section in the Cochrane Stroke Group module. We searched for trials in all languages.
We searched the Cochrane Stroke Group Trials Register (last searched December 2011), the Cochrane Central Register of Controlled Trials (CENTRAL) and the Cochrane Database of Systematic Reviews (CDSR) (The Cochrane Library 2011, Issue 12), the Database of Abstracts of Reviews of Effects (DARE) (www.crd.york.ac.uk/crdweb/) (December 2011), MEDLINE (1966 to December 2011) (Appendix 1), EMBASE (1980 to December 2011) (Appendix 2), and Latin American and Caribbean Health Sciences Literature (LILACS) (1982 to December 2011) (Appendix 3). We developed the MEDLINE search strategy with the help of the Cochrane Stroke Group Trials Search Co-ordinator and adapted it for the other databases.
We also searched the following ongoing trials and research registers (December 2011):
Searching other resources
In an effort to identify additional published, unpublished, and ongoing trials, we searched reference lists of relevant articles and contacted authors, trialists, and experts in the field.
Data collection and analysis
Selection of studies
Two review authors (LGL and HS) independently scanned the titles and abstracts of records identified by the electronic searches and excluded those articles that clearly were not randomised trials or did not address the effects of beta-blockers in stroke patients. We obtained the full text of the remaining articles, and the same two authors selected those studies that met the predetermined inclusion criteria stated. These two review authors resolved any disagreements by discussion and consultation with a third review author if necessary.
Data extraction and management
Two authors (LGL and BS) independently extracted data on participants, methods, interventions, outcomes, and results. We collected the following information:
general information: published/unpublished, title, authors, reference, contact address, country, language of publication, year of publication, duplicate publications, sponsor, setting;
trial characteristics: design, duration of follow-up, method of randomisation, allocation concealment, blinding (participants and outcome assessors);
interventions: interventions (dose, route, timing), comparison interventions (dose, route, timing), co-medications (dose, route, timing);
participants: total number and numbers in comparison groups, age, baseline characteristics, diagnostic criteria, similarity of groups at baseline, assessment of compliance, losses to follow-up (reasons/description), subgroups; and
outcomes: outcomes specified above, any other outcomes assessed, other events, length of follow-up, quality of reporting of outcomes.
Assessment of risk of bias in included studies
To assess the risk of bias, we independently assessed the quality of the studies included in the review according to the criteria described by Higgins (Higgins 2011).
We assessed the following domains for each trial and rated them at low, unclear, or high risk of bias.
Random sequence generation.
Blinding of participants and personnel.
Blinding of outcome assessment.
Incomplete outcome data.
Selective outcome reporting.
We reported these assessments for each individual study in the 'Risk of bias' tables located in the Characteristics of included studies table.
The review authors resolved disagreements arising at any stage of the process through discussion until consensus was reached.
Measures of treatment effect
As the nature of outcome data was dichotomous, we used the risk ratio (RR) and 95% confidence interval (CI) to measure the intervention effect.
Unit of analysis issues
We did not find any cluster-randomised or cross-over trials. In future updates of this review, we will exclude cluster-randomised trials. We will include only the first period of cross-over trials, and we will assess trials with multiple observations for the same outcome (e.g. recurring events) if randomised. We will exclude recurring events.
Dealing with missing data
We only analysed the available data. In future updates of this review, we will do best- and worst-case-scenario analyses for incomplete outcome data.
Assessment of heterogeneity
We used I² to assess the likelihood of heterogeneity.
Assessment of reporting biases
Given the limited number of included studies, we did not create funnel plots to assess reporting biases. In future updates of this review, we will interpret any funnel plot asymmetry with caution.
We used the Cochrane Review Manager software (RevMan 5.1) to carry out a quantitative analysis, based on the intention-to-treat principle (RevMan 2011). We performed meta-analyses using a fixed-effect model if no substantial heterogeneity existed and if pooling of results was clinically appropriate.
Subgroup analysis and investigation of heterogeneity
Given the limited number of included studies, we did not undertake subgroup analyses. In future updates of this review, we will explore differences in the results by creating subgroups according to type and dose of beta-blocker (based on data), age of participants (younger than 65 years old versus 65 years of age and older), and ethnicity of participants (white and non-white) for each trial.
Given the limited number of included studies, we did not undertake sensitivity analyses. In future updates of this review, we will assess the robustness of results by performing sensitivity analyses to fixed-effect versus random-effects assumptions and will consider the inclusion or exclusion of studies of poor quality and those that were not placebo controlled or blinded.
Description of studies
Results of the search
We identified a total of 3078 records from the electronic searches. After examination of the titles and abstracts, we excluded 3062 records and obtained full text copies of the remaining 16, which we then subjected to further assessment. We also checked the reference lists of these studies but found no additional studies. Upon verification of all whole articles that appeared to meet our inclusion criteria, only two articles fulfilled all the inclusion criteria of this review (DUTCH TIA TRIAL 1993; Eriksson 1995) (Figure 1).
Both included studies were multicentre, double-blind, randomised, placebo-controlled trials. The mean duration of follow-up was 32 months in one study (DUTCH TIA TRIAL 1993), and 28 months in the other (Eriksson 1995).
One trial included 1473 participants who had had a TIA (symptoms persisting for less than 24 hours) or a non-disabling ischaemic stroke (symptoms persisting for longer than 24 hours) less than three months before randomisation. Patients with a stroke had to be independent in most of their daily activities (score on the modified Rankin Scale of grade 3 or less). Excluded from the study were patients with cerebral ischaemia from identifiable causes other than arterial thrombosis or arterial embolism, and patients with a contraindication against or a strict indication for a beta-blocker. One-third of participants were women, and there was a slight excess of males and of patients older than 65 years of age in the atenolol group (DUTCH TIA TRIAL 1993). The other study included 720 participants recruited within three weeks of an index event of a minor or major stroke or TIA. Patients with subarachnoid haemorrhage, systolic blood pressure < 140 mm Hg and diastolic blood pressure < 80 mm Hg, bradycardia < 50 beats per minute, or manifest heart failure, and those who had atrioventricular block I to III were excluded from the study. Patients in poor general condition or dependent on help for daily activities were not admitted to the trial. Men made up 61% of included participants; mean age of participants was around 70 years (Eriksson 1995).
The two included studies compared atenolol 50 mg once a day with placebo.
The primary outcome event in one study was the combined occurrence of death from vascular causes, non-fatal stroke, or non-fatal myocardial infarction, whichever occurred first (DUTCH TIA TRIAL 1993), and in the other study, death from any cause, non-fatal myocardial infarction, and non-fatal stroke (Eriksson 1995).
We excluded most of the 14 excluded studies because they were RCTs of primary prevention in hypertensive patients. Some studies performed subgroup analyses in participants with prior stroke, but these data were not provided in detail in the results, even after we contacted the authors. For further information, please refer to the Characteristics of excluded studies table.
Risk of bias in included studies
The two included studies had high overall methodological quality.
The two included studies used a random permuted blocks design. In one blinded randomisation (DUTCH TIA TRIAL 1993), codes were distributed by telephone, and in the other, no reference was made to allocation concealment (Eriksson 1995).
Both included studies were double-blinded.
Incomplete outcome data
In one study, 36% of participants had discontinued trial medication in the intervention group and 32% in the placebo group after three years of follow-up (DUTCH TIA TRIAL 1993). In the other study, treatment was discontinued in 17% in the intervention group and in 10% in the placebo group (Eriksson 1995). In both studies, the authors performed analysis by intention-to-treat.
There was no concern about selective reporting.
Other potential sources of bias
We believe that the included studies were free of other sources of bias.
Effects of interventions
The two included studies randomised 1104 participants to intervention and 1089 to placebo. There were no statistical differences between the groups for risks of fatal and non-fatal stroke (RR 0.95, 95% CI 0.76 to 1.18) (Analysis 1.1).
For all other outcomes analysed, there were no significant differences between the atenolol group and the placebo group: death from all causes: RR 0.94 (95% CI 0.68 to 1.32) (Analysis 1.2); cardiac death and non-fatal myocardial infarction: RR 0.94 (95% CI 0.63 to 1.41) (Analysis 1.3); and major vascular events: RR 1.01 (95% CI 0.84 to1.21) (Analysis 1.4). Neither of the studies included in this review reported the occurrence of diabetes among their outcomes.
In one study, adverse reactions of any kind were significantly more frequent in participants taking atenolol than in those given placebo: RR 1.50 (95% CI 1.20 to 1.89). Adverse effects were the most common reason given for stopping the trial medication early, and twice as many participants taking atenolol as were given placebo stopped because of an adverse effect (DUTCH TIA TRIAL 1993). In the other included study, 10% of participants in the placebo group and 17% in the atenolol group discontinued treatment because of side effects (Eriksson 1995).
Summary of main results
This review showed no evidence of reduction of recurrent stroke, total mortality, vascular disease, and cardiovascular events in people with previous stroke treated with beta-blockers. Some pathophysiological considerations may in part explain these findings. Atherothrombotic vascular disease manifests, as a rule, as a cerebrovascular event (stroke or TIA), myocardial infarction, or peripheral vascular disease. The predominant risk factors for all these events are quite similar and include hypertension, diabetes mellitus, obesity, dyslipidaemia, and cigarette smoking (Sacco 2006; Smith 2001). This similarity reflects the systemic nature of atherothrombotic vascular disease. However, differences observed between the risk factors specific for vascular disease suggest some degree of specificity in pathophysiological processes. For example, dyslipidaemia is a well-established risk factor for coronary artery disease, but its role in cerebrovascular disease is not well established (Sacco 1997). The theory of divergent pathophysiological mechanisms for stroke and coronary heart disease has been reinforced by data from epidemiological studies and RCTs. Thus, the specificity between different forms of vascular disease may explain the fact that the beneficial effects of beta-blockers demonstrated in ischaemic heart disease may not be valid with respect to cerebrovascular disease.
Overall completeness and applicability of evidence
It should be emphasised that this review was based on studies of prevention, which usually are more complex than studies of therapy, requiring significantly larger samples, because the effects of a preventive intervention tend to be smaller than therapeutic effects. Systematic reviews are particularly important in studies on prevention, with the function to minimize the potential for error type 2 (i.e. lack of effect of insufficient sample size). Therefore, the results of this systematic review did not exclude the possibility that the observed absence of effect of beta-blockers in the secondary prevention of stroke and TIA may be due to lack of further statistical power.
Authors of the most recent study included in this review (Eriksson 1995) commented that more studies with greater statistical power, or meta-analyses, would be needed to confirm whether beta-blockers have a beneficial effect in secondary prevention of stroke, because the study did not reach the desired sample size planned and included fewer than half the participants required. The other study, published in 1993, despite including twice as many participants, was also smaller than planned, and neither study was able to find statistically significant differences between atenolol and placebo in any outcome measured.
Previous studies have shown an increased risk of developing diabetes mellitus in people receiving beta-blockers. However, the studies included in this review did not provide data on the development of diabetes mellitus.
Quality of the evidence
Secondary prevention of stroke is of fundamental importance, and researchers need to explore all potentially beneficial interventions. Antiplatelet agents, particularly aspirin, have shown their effectiveness. It remains for investigators to define other prevention strategies and to answer the question, still uncertain, whether there are differences between different classes of antihypertensive drugs with regard to prevention of clinical outcomes. Reduction of blood pressure, even in normotensive patients, can be an effective measure. In 1997, a meta-analysis concluded that pharmacological interventions for blood pressure reduction reduce the risk of new stroke in people with previous cerebrovascular events (Gueyffier 1997). Similarly, the RCT PROGRESS 2001, which compared perindopril and indapamide (respectively, ACE inhibitor and diuretic) with placebo in participants with prior stroke, showed a reduction of stroke recurrence directly proportionate to arterial pressure reduction in participants who received active treatment. It is interesting to note that in PROGRESS 2001, the difference between intervention and placebo was seen only in those participants who received combination therapy with diuretics and ACE inhibitors, which produced a mean reduction in systolic blood pressure of 12 mm Hg and in diastolic blood pressure of 5 mm Hg. In participants who received only perindopril, investigators described no benefit over placebo, and participants had an average reduction of blood pressure of 5/3 mm Hg.
We included only two studies in this review; despite their high methodological quality, blood pressure reduction in the intervention group was 4/3 mm Hg in one study (Eriksson 1995), and the difference in blood pressure between intervention and placebo groups was 5.8/2.9 mm Hg in the other study (DUTCH TIA TRIAL 1993). It is possible that one explanation for the positive effects not observed in studies with beta-blockers is the relatively minor reduction in blood pressure detected, although this was not the goal of the treatment. Although the inclusion criteria were similar in the two studies, which had sought to select not people with a diagnosis of hypertension but people with prior stroke, in fact one study included people with hypertension or with borderline hypertension, because one of the exclusion criteria was blood pressure less than or equal to 140/80 mm Hg (Eriksson 1995).
Potential biases in the review process
We conducted an extensive literature search without language restrictions, so we believe it is unlikely that we missed any potentially relevant studies. Therefore, we believe that there is no significant bias in the review process.
Agreements and disagreements with other studies or reviews
Recent clinical trials have questioned the efficacy of beta-blockers even in primary prevention. For similar reductions in blood pressure, investigators observed different outcomes in people treated with beta-blockers, diuretics, or newer drugs such as calcium channel blockers, ACE inhibitors, and angiotensin II receptor antagonists (Dahlöf 2002; Dahlöf 2005; Hansson 1999a; Hansson 2000; MacMahon 1990). Discussion of differences between different classes of antihypertensive drugs with regard to reduction of mortality, stroke, and other vascular events has raised suspicion that beta-blockers may produce a smaller benefit in clinical outcomes, despite similar reductions in blood pressure. This question is still controversial and has been restricted to studies on primary prevention (Collins 1990; Lindholm 2005; Messerli 2008; Wiysonge 2007). The latest consensus on the treatment of hypertension still put beta-blockers as first-line drugs, and their easy availability and relatively low cost continue to be important factors in clinical practice (Chalmers 2003).
We thank the Cochrane Stroke Group, especially Hazel Fraser and Brenda Thomas, for their support.
Appendix 1. MEDLINE search strategy
1. cerebrovascular disorders/ or exp basal ganglia cerebrovascular disease/ or exp brain ischemia/ or exp carotid artery diseases/ or exp intracranial arterial diseases/ or exp "intracranial embolism and thrombosis"/ or exp intracranial hemorrhages/ or stroke/ or exp brain infarction/ or vasospasm, intracranial/ or vertebral artery dissection/
2. (stroke or cerebrovasc$ or brain vasc$ or cerebral vasc$ or cva$ or apoplex$ or isch?emi$ attack$ or tia$1).tw.
3. ((brain$ or cerebr$ or cerebell$ or vertebrobasilar or intracran$ or intracerebral or infratentorial or supratentorial or MCA) adj10 (isch?emi$ or infarct$ or thrombo$ or emboli$)).tw.
4. ((brain$ or cerebr$ or cerebell$ or intracerebral or intracran$ or subarachnoid) adj10 (haemorrhage$ or hemorrhage$ or haematoma$ or hematoma$ or bleed$)).tw.
5. 1 or 2 or 3 or 4
6. exp Adrenergic beta-Antagonists/
7. (beta$ adj5 block$).tw.
8. (acebutalol or alprenolol or atenolol or betaxolol or bisoprolol or bupranolol or bucindolol or butoxamine or bufuralol or carteolol or carvedilol or celiprolol or esmolol or epanolol or labetalol or labetolol or metoprolol or nadolol or oxprenolol or pindolol or propranolol or sotalol or timolol).tw.
9. 6 or 7 or 8
10. 5 and 9
11. limit 10 to humans
12. Randomized Controlled Trials as Topic/
13. random allocation/
14. Controlled Clinical Trials as Topic/
15. control groups/
16. clinical trials as topic/ or clinical trials, phase i as topic/ or clinical trials, phase ii as topic/ or clinical trials, phase iii as topic/ or clinical trials, phase iv as topic/
17. double-blind method/
18. single-blind method/
20. placebo effect/
21. cross-over studies/
22. Therapies, Investigational/
23. Drug Evaluation/
24. Research Design/
25. randomized controlled trial.pt.
26. controlled clinical trial.pt.
27. (clinical trial).pt.
29. (controlled adj5 (trial$ or stud$)).tw.
30. (clinical$ adj5 trial$).tw.
31. ((control or treatment or experiment$ or intervention) adj5 (group$ or subject$ or patient$)).tw.
32. (quasi-random$ or quasi random$ or pseudo-random$ or pseudo random$).tw.
33. ((singl$ or doubl$ or tripl$ or trebl$) adj5 (blind$ or mask$)).tw.
34. (coin adj5 (flip or flipped or toss$)).tw.
35. latin square.tw.
37. (cross-over or cross over or crossover).tw.
39. (assign$ or alternate or allocat$).tw.
42. 11 and 41
Appendix 2. EMBASE search strategy
#1. cerebrovascular disease/ or basal ganglion hemorrhage/ or exp brain hematoma/ or exp brain hemorrhage/ or exp brain infarction/ or exp brain ischemia/ or exp carotid artery disease/ or cerebral artery disease/ or cerebrovascular accident/ or exp intracranial aneurysm/ or exp occlusive cerebrovascular disease/ or stroke/ or stroke unit/ or stroke patient/
#2 (stroke or cerebrovasc$ or brain vasc$ or cerebral vasc$ or cva$ or apoplex$ or isch?emi$ attack$ or tia$1).tw.
#3 ((brain$ or cerebr$ or cerebell$ or vertebrobasilar or intracran$ or intracerebral or infratentorial or supratentorial or MCA) adj10 (isch?emi$ or infarct$ or thrombo$ or emboli$)).tw.
#4 ((brain$ or cerebr$ or cerebell$ or intracerebral or intracran$ or subarachnoid) adj10 (haemorrhage$ or hemorrhage$ or haematoma$ or hematoma$ or bleed$)).tw.
#5 1 or 2 or 3 or 4
#6 exp beta adrenergic receptor blocking agent/
#7 (beta$ adj5 block$).tw.
#8 (acebutalol or alprenolol or atenolol or betaxolol or bisoprolol or bupranolol or bucindolol or butoxamine or bufuralol or carteolol or carvedilol or celiprolol or esmolol or epanolol or labetalol or labetolol or metoprolol or nadolol or oxprenolol or pindolol or propranolol or sotalol or timolol).tw.
#9 6 or 7 or 8
#10 5 and 9
#11 limit 10 to human
#12 Controlled study/
#14 Double blind procedure/
#15 Single blind procedure/
#16 Clinical trial/
#17 (clinical adj5 trial$).ti,ab,hw.
#18 ((doubl$ or singl$ or tripl$ or trebl$) adj5 (blind$ or mask$)).ti,ab,hw.
#23 latin square.ti,ab,hw.
#26 Crossover Procedure/
#27 Drug comparison/
#28 Comparative study/
#29 (comparative adj5 trial$).ti,ab,hw.
#30 (control$ or prospectiv$ or volunteer$).ti,ab,hw.
#31 exp "Evaluation and Follow Up"/
#32 Prospective study/
#34 11 and 33
Appendix 3. LILACS search strategy
#1 Acidente Cerebral Vascular
#2 Derrame Cerebral
#5 Acidente Cerebrovascular
#6 Apoplexia Cerebral
#7 Apoplexia Cerebrovascular
#8 Icto Cerebral
#9 Acidente Vascular Encefálico
#11 Acidente Vascular do Cérebro
#12 Acidente Cerebral Vascular
#13 Acidentes Cerebrais Vasculares
#14 Acidentes Cerebrovasculares
#15 Acidentes Vasculares Cerebrais
#16 #1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15
#17 Antagonistas de Receptores Adrenérgicos beta 1
#18 Antagonistas de Receptores beta-1 Adrenérgicos
#19 Antagonistas de Receptores Adrenérgicos beta-1
#20 Antagonistas dos Receptores beta-1 Adrenérgicos
#21 (acebutalol or alprenolol or atenolol or betaxolol or bisoprolol or bupranolol or bucindolol or butoxamine or bufuralol or carteolol or carvedilol or celiprolol or esmolol or epanolol or labetalol or labetolol or metoprolol or nadolol or oxprenolol or pindolol or propranolol or sotalol or timolol)
#22 #17 OR #18 OR #19 OR #20 OR #21
#23 ((Pt RANDOMIZED CONTROLLED TRIAL OR Pt CONTROLLED CLINICAL TRIAL OR Mh RANDOMIZED CONTROLLED TRIALS OR Mh RANDOM ALLOCATION OR Mh DOUBLE-BLIND METHOD OR Mh SINGLE-BLIND METHOD OR Pt MULTICENTER STUDY) OR ((tw ensaio or tw ensayo or tw trial) and (tw azar or tw acaso or tw placebo or tw control$ or tw aleat$ or tw random$ or (tw duplo and tw cego) or (tw doble and tw ciego) or (tw double and tw blind)) and tw clinic$)) AND NOT ((CT ANIMALS OR MH ANIMALS OR CT RABBITS OR CT MICE OR MH RATS OR MH PRIMATES OR MH DOGS OR MH RABBITS OR MH SWINE) AND NOT (CT HUMAN AND CT ANIMALS)) [Palavras]
#24 #16 AND #22 AND #23
Contributions of authors
Luiz Gustavo Lima (LGL) and Humberto Saconato (HS) were responsible for the conception of this review. LGL, HS, Álvaro Nagib Atallah (ANA), Edina MK da Silva (EMKS), and Bernardo Soares (BS) were responsible for developing the search strategy, screening search results, obtaining papers, screening retrieved papers against the inclusion criteria, appraising the quality of papers, and extracting data. LGL and EMKS were responsible for writing the review.