Beta blockers for peripheral arterial disease
Abstract
Background
Beta (β) blockers are indicated for use in coronary artery disease (CAD). However, optimal therapy for people with CAD accompanied by intermittent claudication has been controversial because of the presumed peripheral haemodynamic consequences of beta blockers, leading to worsening symptoms of intermittent claudication. This is an update of a review first published in 2008.
Objectives
To quantify the potential harmful effects of beta blockers on maximum walking distance, claudication distance, calf blood flow, calf vascular resistance and skin temperature when used in patients with peripheral arterial disease (PAD).
Search methods
For this update, the Cochrane Peripheral Vascular Diseases Group Trials Search Co‐ordinator searched the Specialised Register (last searched March 2013) and the Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, 2013, Issue 2).
Selection criteria
Randomised controlled trials (RCTs) evaluating the role of both selective (β1) and non‐selective (β1 and β2) beta blockers compared with placebo. We excluded trials that compared different types of beta blockers.
Data collection and analysis
Primary outcome measures were claudication distance in metres, time to claudication in minutes and maximum walking distance in metres and minutes (as assessed by treadmill).
Secondary outcome measures included calf blood flow (mL/100 mL/min), calf vascular resistance and skin temperature (ºC).
Main results
We included six RCTs that fulfilled the above criteria, with a total of 119 participants. The beta blockers studied were atenolol, propranolol, pindolol and metoprolol. All trials were of poor quality with the drugs administered over a short time (10 days to two months). None of the primary outcomes were reported by more than one study. Similarly, secondary outcome measures, with the exception of vascular resistance (as reported by three studies), were reported, each by only one study. Pooling of such results was deemed inappropriate. None of the trials showed a statistically significant worsening effect of beta blockers on time to claudication, claudication distance and maximal walking distance as measured on a treadmill, nor on calf blood flow, calf vascular resistance and skin temperature, when compared with placebo. No reports described adverse events associated with the beta blockers studied.
Authors' conclusions
Currently, no evidence suggests that beta blockers adversely affect walking distance, calf blood flow, calf vascular resistance and skin temperature in people with intermittent claudication. However, because of the lack of large published trials, beta blockers should be used with caution, if clinically indicated.
PICOs
Plain language summary
Beta blockers for peripheral arterial disease
Intermittent claudication, the most common symptom of atherosclerotic peripheral arterial disease, results from decreased blood flow to the legs during exercise. Beta blockers, a large group of drugs, have been shown to decrease death among people with high blood pressure and coronary artery disease and are used to treat various disorders. They reduce heart activity but can also inhibit relaxation of smooth muscle in blood vessels, bronchi and the gastrointestinal and genitourinary tracts. The non‐selective beta blockers propranolol, timolol and pindolol are effective at all beta‐adrenergic sites in the body, whereas other beta blockers, such as atenolol and metoprolol, are selective for the heart.
Optimal therapy for people with coronary artery disease or hypertension and intermittent claudication is controversial because of the presumed peripheral blood flow consequences of beta blockers, which lead to worsening of symptoms.
Currently, no evidence from randomised controlled trials suggests that beta blockers adversely affect walking distance in people with intermittent claudication, and beta blockers should be used with caution, if clinically indicated. The review authors identified six randomised controlled trials that involved a total of only 119 people with mild to moderate peripheral arterial disease. The beta blockers studied were propranolol, pindolol, atenolol and metoprolol. None of the trials showed clear worsening effects of beta blockers on time to claudication, claudication distance and maximal walking distance as measured on a treadmill, nor on calf blood flow, calf vascular resistance and skin temperature, when compared with placebo. Trial investigators reported no adverse events or issues regarding taking the beta blockers studied.
Most of the trials were over 20 years old and reported findings between 1980 and 1991. All were small and of poor quality. The drugs were administered over a short time (10 days to two months), and most of the outcome measures were reported in single studies. Additional drugs－calcium channel blockers and combined alpha and beta blockers－were given during some of the trials.
Authors' conclusions
Background
Description of the condition
Intermittent claudication, the most common symptom of atherosclerotic peripheral arterial disease (Hiatt 2001), reflects decreased blood flow to the extremities during exercise (Lassila 1986). The incidence of intermittent claudication increases with advancing age, cigarette smoking, impaired glucose tolerance and hypertension (Hughson 1978). Men are twice as likely as women to be affected by intermittent claudication (Kannel 1985).
Patients with peripheral arterial disease (PAD) have increased rates of mortality due to concurrent coronary artery disease and hypertension (Criqui 1985).
Description of the intervention
Beta (β) blockers were thought to decrease all‐cause and cardiovascular mortality and were used as a first‐line medication for primary hypertension. However, recent evidence is counter‐intuitive to this and has demonstrated that beta blockers are less efficacious than placebo, thiazides or angiotensin‐converting enzyme inhibitors in the treatment of primary hypertension (Bangalore 2007; Lavie 2009; Lindholm 2005; Wiysonge 2007; Wright 2009). Further, beta blockers are associated with a higher incidence of all‐cause and cardiovascular mortality compared with other antihypertensive medication. Furthermore, a recent Cochrane review showed that beta blockers fail to improve cardiovascular mortality even when used as a second‐line therapy for treatment of hypertension (Chen 2010). Currently, beta blockers are indicated in the treatment of angina and acute myocardial infarction and heart failure (Javed 2009; NICE 2010). In addition, they are used to treat arrhythmias (variations in the normal rhythm of the heartbeat), migraine headaches, essential tremors, thyrotoxicosis (excessive production of thyroid hormones), glaucoma, anxiety and various other disorders.
Beta blockers make up a large group of drugs, and although all are competitive inhibitors of β receptors, they may have additional pharmacodynamic properties. In addition to increasing the force and rate of myocardial contraction, β1 receptors increase conduction velocity through the atrioventricular (AV) node. Beta1 blockade, therefore, reduces heart rate, blood pressure, myocardial contractility and myocardial oxygen consumption. Beta2 receptor blockade inhibits relaxation of smooth muscle in blood vessels, bronchi and the gastrointestinal and genitourinary tracts. In addition, β2 blockade inhibits the breakdown of glycogen (the main carbohydrate storage compound) to glucose (glycogenolysis) and the formation of sugar from protein and fat in the absence of glucose or carbohydrate (gluconeogenesis). Non‐specific beta blockers such as propranolol, timolol, nadolol, and pindolol demonstrate equal affinity for both β1 and β2 receptors. Commonly used cardioselective (β1) blockers are atenolol and metoprolol.
How the intervention might work
In general, blockade of β receptors results in decreased production of intracellular cyclic adenosine monophosphate (cAMP), with resultant blunting of the multiple metabolic and cardiovascular effects of circulating catecholamines. Beta blockers are indicated for use after myocardial infarction (MI). It has been shown that therapy initiated within 12 hours of MI decreases the rate of in‐hospital cardiovascular mortality by 13% to 15% (MIAMI trial). Beta blockers have also been indicated for use in patients with continuing or recurrent ischaemic pain, tachyarrhythmias (rapid heart rates associated with an irregularity in the normal heart rhythm) and non－ST‐elevation MI (ACCF/AHA 2012).
Optimal therapy for coronary artery disease or hypertension accompanied by intermittent claudication has been controversial because of presumed peripheral haemodynamic consequences of beta blockers that could lead to worsening of symptoms in patients with PAD (George 1974). This has been attributed to decreased cardiac output and unopposed alpha (α)‐adrenergic drive.
Why it is important to do this review
It is well known that acute lowering of blood pressure is contraindicated in critical ischaemia. However, the effect of lowering of blood pressure in people with intermittent claudication is not known. Beta blockers are contraindicated in severe PAD (BNF 2013). The aim of this review is to gather evidence for the contraindication of beta blockers in people with intermittent claudication.
Objectives
To quantify the potential harmful effects of beta blockers on maximum walking distance, claudication distance, calf blood flow, calf vascular resistance and skin temperature when used in patients with PAD.
Methods
Criteria for considering studies for this review
Types of studies
Randomised controlled trials (RCTs) comparing beta blockers and placebo in participants with PAD.
Types of participants
Patients with moderate to severe PAD, with or without additional co‐morbidity (an additional disease or condition, for example, diabetes mellitus), were included. Peripheral arterial disease was defined as a typical history of intermittent claudication and reduced ankle brachial pressure index (ABPI)－less than 0.9.
Types of interventions
We included studies that used selective beta blockers (β1) and non‐selective beta blockers (β1 and β2). We excluded studies using some beta blockers with additional alpha (α) blocking properties (for example, labetalol).
Types of outcome measures
Primary outcomes
Primary outcome measures considered for this review were initial claudication distance in metres, time to claudication in minutes and maximal walking distance in metres and minutes (as assessed by treadmill).
Secondary outcomes
Secondary outcome measures included calf blood flow (mL/100 mL/min), calf vascular resistance and skin temperature (ºC). Mortality data and complications associated with the use of beta blockers, such as development of critical ischaemia, cardiovascular morbidity and mortality; all‐cause mortality; and drug withdrawal, were also recorded for participants with stable PAD.
Search methods for identification of studies
No restrictions on language of publication were applied.
Electronic searches
For this update, the Cochrane Peripheral Vascular Diseases Group Trials Search Co‐ordinator (TSC) searched the Specialised Register (last searched March 2013) and the Cochrane Central Register of Controlled Trials (CENTRAL) (2013, Issue 2; part of The Cochrane Library, www.thecochranelibrary.com). See Appendix 1 for details of the search strategy used to search CENTRAL. The Specialised Register is maintained by the TSC and is constructed from weekly electronic searches of MEDLINE, EMBASE, CINAHL and AMED and handsearching of relevant journals. The full list of the databases, journals and conference proceedings that have been searched and the search strategies used are described in the Specialised Register section of the Cochrane Peripheral Vascular Diseases Group module in The Cochrane Library (www.thecochranelibrary.com).
Searching other resources
We searched bibliographies of all identified research articles and review articles.
Data collection and analysis
Selection of studies
All three review authors independently reviewed abstracts of potential trials involving beta blockers in PAD. We obtained full papers for those fulfilling the relevant criteria or when clarification was required. We excluded trials that compared different types of beta blockers.
Studies that met the inclusion criteria were independently selected by SCVP and DAM, and this process was overseen by ADS. No disagreements among SCVP and DAM regarding inclusion were reported; however, both could not decide on inclusion of one study, and this decision was made by ADS.
Data extraction and management
SCVP and DAM independently extracted data from the included studies using the proforma designed by the Cochrane PVD Group.
Some of the results, for example, calf blood flow (Hiatt 1985; Lepantalo 1984a; Lepantalo 1985), claudication and maximal walking distance (Roberts 1987), were presented only as graphs, hence we excluded them from the analyses.
Vascular resistance is described in arbitrary units. The review authors calculated vascular resistance by dividing mean arterial blood pressure by calf blood flow.
Assessment of risk of bias in included studies
All three review authors independently assessed the methodological quality of selected trials using the Cochrane 'Risk of bias' tool provided in the Cochrane Handbook for Systematic Reviews of Interventions (version 5.1.0) (Higgins 2011). The results for each included trial are summarised in the 'Risk of bias' tables.
Measures of treatment effect
We planned to use odds ratio (OR) with 95% confidence interval (CI) for dichotomous variables and mean difference (MD) with 95% CI for continuous variables.
Meta‐analyses were not performed, hence data from individual trials are discussed and presented.
Unit of analysis issues
Each participant was considered an individual unit of analysis for this review, as meta‐analyses were not performed. However, in future reviews, if new trials have been published, for cross‐over trials, when available, paired data will be considered to avoid unit of analysis error. If it becomes difficult to identify data clearly, individual participant data will be considered for meta‐analysis, but results will be presented with caution.
Dealing with missing data
Some key data were missing from some of the trials. As they were old trials, study co‐ordinators could not be contacted to obtain more information. With regards to some trials, details of standard deviation were calculated from the provided standard error and mean, as discussed by Higgins 2011.
Assessment of heterogeneity
Variability in participants, interventions and outcomes was considered in the assessment of clinical heterogeneity. As a meta‐analysis was not performed, statistical heterogeneity was not relevant.
If new trials are included in future updates of this review, when possible, we will perform statistical analysis according to the statistical guidelines for review authors put forth by the Cochrane PVD Group. We will assess the degree of heterogeneity amongst trials by using the I2 statistic according to the formula I2 = 100% × (Q ‐ degrees of freedom)/Q, where Q is the Chi2 statistic (Higgins 2011). We will present a summary statistic for each outcome using a fixed‐effect model for homogenous data and a random‐effects model for heterogenous data.
Assessment of reporting biases
The review authors intended to use funnel plots for publication bias; however, as no meta‐analysis was performed, this could not be done. All studies were carefully assessed to look for selective reporting bias. Whilst no uniformity was observed in the results presented, outcome reporting was quite varied among trials.
Data synthesis
Because data from the included studies were insufficient, meta‐analysis was not performed. If new trials are included in future updates of this review, we will present a summary statistic for each outcome using a fixed‐effect model for homogenous data and a random‐effects model for heterogenous data.
Subgroup analysis and investigation of heterogeneity
Subgroup analysis was not feasible because data were insufficient; however, such an analysis will be carried out if sufficient data are provided by future trials.
Sensitivity analysis
We planned to undertake sensitivity analyses to examine the stability of the results in relation to a number of factors, including study quality, the source of the data (published and unpublished) and participant type. Such analyses were not undertaken because sufficient data were lacking.
Results
Description of studies
Results of the search
See Figure 1.
Study flow diagram.
Included studies
No additional studies are included in this update.
We have included six RCTs that fulfilled the inclusion criteria for this review (Clement 1980; Hiatt 1985; Lepantalo 1984a; Lepantalo 1985; Roberts 1987; Solomon 1991). Summary details of included studies are given in the Characteristics of included studies table.
Various beta blockers were used and compared with placebo in the above‐mentioned trials. Propranolol, atenolol and metoprolol were the most commonly evaluated beta blockers.
In a randomised, double‐blind, placebo‐controlled trial, Clement 1980 evaluated the effects of propranolol (80 mg twice daily) and metoprolol (100 mg twice daily) on claudication and maximal walking times in participants with chronic intermittent claudication. In this cross‐over trial, 10 participants with chronic intermittent claudication were given a wash‐out period with placebo for four weeks and were subsequently randomly assigned to receive metoprolol, propranolol or placebo, each for two months.
Hiatt 1985 compared the effects of metoprolol (50 mg) and propranolol (40 mg) with those of placebo on calf blood flow and vascular resistance. In this randomised, double‐blind, placebo‐controlled trial, 19 participants with chronic stable intermittent claudication received placebo in the run‐in phase for three weeks. Participants were later randomly assigned to receive either metoprolol or propranolol for two weeks. Subsequently, they were administered placebo for two weeks before the cross‐over phase. As in the earlier studies, calf blood flow was measured using venous occlusion plethysmography.
Lepantalo 1984a studied the effects of metoprolol and methyldopa on calf blood flow and vascular resistance in participants with peripheral vascular disease. In this randomised, double‐blind, placebo‐controlled trial, 14 participants were given placebo in the run‐in phase for three weeks and were subsequently randomly assigned to receive metoprolol (100 to 200 mg), methyldopa (500 to 1000 mg) or placebo, each for three weeks. Calf blood flow was measured using venous occlusion plethysmography.
In another randomised, double‐blind, placebo‐controlled trial, Lepantalo 1985 evaluated the effects of propranolol (80 mg), pindolol (5 mg), labetalol (200 mg), labetalol (400 mg) and placebo on hyperaemic calf blood flow, skin temperature and vascular resistance in seven participants with hypertension and chronic intermittent claudication. All drugs were administered for 10 days.
Roberts 1987 administered atenolol (100 mg), labetalol (200 mg), pindolol (5 mg), captopril (25 mg) and placebo to a total of 20 participants and studied their effects on mean blood pressure, calf blood flow, mean pain‐free walking distance and mean maximal walking distance. All participants received placebo for a month as an initial wash‐out phase. Later, they were randomly assigned to receive atenolol, labetalol, pindolol, captopril or placebo, each for one month. Calf blood flow was measured using venous occlusion plethysmography, and walking distances were assessed using a treadmill.
Solomon 1991 evaluated the effects of atenolol (50 mg), nifedipine (20 mg) and a combination of atenolol and nifedipine, as well as placebo, on skin temperature and walking distance in a total of 49 participants, who were randomly assigned after a run‐in phase with placebo. In this cross‐over trial, drugs were administered, each for four weeks. No wash‐out period preceded cross‐over of participants.
None of the primary outcomes were reported by more than one study. Similarly, secondary outcome measures, with the exception of vascular resistance (as reported by three studies), were reported, each by only one study. Pooling of such results was deemed inappropriate, hence results of individual series are presented.
Excluded studies
An additional four studies (Diehm 2011; Ehrly 1987; Espinola‐Klein 2011; van de Ven 1994) were excluded in this update, making a total of 22 excluded studies, with the reasons stated in the table Characteristics of excluded studies.
Risk of bias in included studies
Moderate to high risk of bias was seen in the included trials, as most of the methodological details, especially details on method of randomisation, could not be obtained from trial authors. Further details are provided in Figure 2 and Figure 3.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Allocation
Details on method of randomisation was mentioned only in Roberts 1987 (randomisation codes) and Solomon 1991 (computer randomisation).
Allocation concealment for most trials was unclear because of lack of reporting. Roberts 1987 reported that the study tablets were dispensed monthly by hospital staff, but no further details were provided.
Blinding
Most trials were reported to be double‐blinded; however, in Roberts 1987, an independent observer who was aware of the randomisation codes assessed symptoms of participants at each visit.
Incomplete outcome data
Whilst Clement 1980, Hiatt 1985 and Solomon 1991 used intention‐to‐treat analysis, it was unclear whether such analysis was undertaken in the remaining trials (Lepantalo 1984a; Lepantalo 1985; Roberts 1987). These were therefore judged to be at unclear risk of bias.
Selective reporting
All trials (Clement 1980; Hiatt 1985; Lepantalo 1984a; Lepantalo 1985; Roberts 1987; Solomon 1991) reported their predetermined outcomes and therefore were judged to be at low risk of bias. Outcomes were quite varied, and the method of reporting differed between trials such that a meta‐analysis could not be performed.
Other potential sources of bias
Details regarding cross‐over were unclear in the Clement 1980 trial, hence the potential for other bias exists. In the Lepantalo 1985 trial, exclusion criteria were unclear, as were losses to follow‐up. No suggestions can be made regarding other potential sources of bias in the remaining trials (Hiatt 1985; Lepantalo 1984a; Roberts 1987; Solomon 1991).
Effects of interventions
Claudication time
One trial evaluated claudication time (Clement 1980). In this study, propranolol and metoprolol were compared with placebo. Mean claudication times were 5.20 minutes (standard deviation (SD) 2.25), 4.90 minutes (SD 2.04) and 4.53 minutes (SD 1.43) with propranolol, metoprolol and placebo, respectively.
Maximum walking time
One trial evaluated maximum walking time (Clement 1980). Propranolol and metoprolol were compared with placebo. Mean maximal walking times were 8.18 minutes (SD 2.54), 8.18 minutes (SD 2.47) and 7.75 minutes (SD 2.20) with propranolol, metoprolol and placebo, respectively.
Claudication distance
Only one trial evaluated claudication distance (Solomon 1991) and compared atenolol with placebo. Mean claudication distance was 62.6 metres for those who received atenolol versus 66.5 metres for those in the placebo group. The author reported a mean change reduction of 6% (95% CI 1% to ‐13%), which was considered to be clinically and statistically insignificant.
Maximum walking distance
Solomon 1991 evaluated maximum walking distance while comparing atenolol with placebo. Maximum walking distances with atenolol and placebo were 110.8 metres and 113.8 metres, respectively. The author reported a mean change reduction of 2% (95% CI 4% to ‐8%), which was considered to be clinically and statistically insignificant.
Calf blood flow after exercise
Roberts 1987 compared the effects of atenolol and pindolol with those of placebo on calf blood flow after exercise. Mean calf blood flow after exercise was 25.5 (SD 15.6), 20.1 (SD 10.2) and 29.9 (SD 17.8) mL/dL/min with atenolol, pindolol and placebo, respectively.
Calf vascular resistance
The effects of propranolol and pindolol versus those of placebo on vascular resistance after reactive hyperaemia were evaluated in one trial (Lepantalo 1985). Vascular resistance was 2.5 units (SD 1.2), 2.1 units (SD 1.3) and 2.1 units (SD 1.3) with propranolol, pindolol and placebo, respectively.
The effect of metoprolol on calf vascular resistance was evaluated by Lepantalo 1984a. Results showed vascular resistance to be 1.7 units (SD 0.9) with metoprolol and 1.5 units (SD 1.0) with placebo.
Hiatt 1985 also compared the effect of metoprolol on vascular resistance after exercise. Vascular resistance was 11.5 units (SD 4.7) and 11.2 units (SD 4.4) for metoprolol and placebo, respectively.
Skin temperature
The effects of propranolol and pindolol on skin temperature were evaluated in one trial (Lepantalo 1985). Mean skin temperature was 24.9°C (SD 3.8), 25.5°C (SD 4.0) and 27°C (SD 4.1) with propranolol, pindolol and placebo, respectively.
Adverse events
For all of the trials that we reviewed, no reports described adverse events related to the use of beta blockers, including critical ischaemia, cardiovascular morbidity and mortality. These trials described no issues regarding participant compliance with medication.
Discussion
Summary of main results
This systematic review identified six randomised, double‐blind, placebo‐controlled, cross‐over trials with a total of 119 participants. Most were small trials that recruited between 10 and 20 participants, and the largest study enrolled 49 participants (Solomon 1991).
None of the trials showed statistically significant worsening effects for beta blockers on the primary or secondary outcomes in participants with claudication. All trials were of poor quality, with the drugs administered over a short time (10 days to two months).
Investigators in the trial by Clement (Clement 1980) when using propranolol did not demonstrate a statistically significant decrease in claudication time or maximum walking time. This was a small trial that included only 10 participants. Details and data regarding the cross‐over phase were unclear.
Lepantalo (Lepantalo 1984a) studied the effects of metoprolol on calf blood flow and vascular resistance. Results showed no deleterious effects in the metoprolol group. In this cross‐over trial, participants were also randomly assigned to receive methyldopa, further confounding study results.
In another study, Lepantalo (Lepantalo 1985) randomly assigned a small number of participants to receive placebo, propranolol, pindolol and different strengths of labetalol (200 mg, 400 mg). The results, as described previously, showed no statistically significant worsening effects on the studied outcome measures. This small group of participants lacked proper randomisation and the cross‐over phase was unclear, raising additional questions about the validity of study conclusions.
The overall effects of atenolol on claudication and maximal walking distances could not be estimated because sufficient data were lacking. In the trial by Solomon (Solomon 1991), nifedipine (a calcium channel blocker) was also used in one of the drug groups, but no wash‐out period was provided before the cross‐over phase. This would have affected the outcome measures, which may not truly represent the effects of each individual drug.
A few limitations of this review are worth noting. Most of the trials were over 20 years old, and none of the reports supplied power calculations or a valid method of randomisation. Most important, no consistency of the inclusion criteria was noted between trials, and most of the included participants suffered from mild or moderate peripheral vascular disease. The duration of the course of beta blockers and of the cross‐over phases varied between trials, and no wash‐out periods were provided in some trials. In other trials, calcium channel blockers and combined alpha and beta blockers were used. Furthermore, primary outcome measures were available in only two trials, and complete outcome data were lacking in one of them. Finally, data were insufficient to pool individual results, hence individual data are presented. In conclusion, despite the inclusion of six RCTs, evidence supporting or refuting the use of beta blockers in peripheral arterial disease remains elusive.
Overall completeness and applicability of evidence
This review addressed whether evidence is available to suggest that beta blockers are harmful for patients with intermittent claudication. Not all outcome measures were available from all studies, and in some cases, trials reported incomplete data. These trials were very old, were not of good quality and consisted of only a minimal number of participants. The evidence, therefore, is not contemporary; however, no major adverse events resulting from the use of beta blockers were reported, which suggests that beta blockers may be used if required in patients with intermittent claudication. A cautious approach should be taken in patients with critical ischaemia.
All included trials were over 20 years old. Today, patients are increasingly aware of their health conditions, the importance of addressing their lifestyles and their co‐morbidities are well managed. It is therefore possible that outcomes may vary considerably if the trials included in this review were to be repeated now. Therefore, the results of these trials should be interpreted with caution.
Quality of the evidence
The trials included in this review were not of high quality, and most failed to mention the randomisation technique or allocation concealment process used or losses to follow‐up. Further, data were not clearly presented, and some of the outcomes of this review were not available in the included studies.
Potential biases in the review process
Two review authors (SCVP, DAM) independently assessed the articles for inclusion and exclusion criteria, thereby minimising the risk of any potential selection bias of articles. All data were extracted by using the proforma developed by the Cochrane PVD group. Data analysis was performed by two review authors (SCVP, DAM), and the senior review author (ADS) independently checked all results before final assessment and drafting of the manuscript.
Some results, for example, calf blood flow (Hiatt 1985; Lepantalo 1984a; Lepantalo 1985), claudication distance and maximal walking distance (Roberts 1987), were presented only as graphs, hence we excluded them from the analyses, as it was not possible to derive data suitable for inclusion.
Agreements and disagreements with other studies or reviews
Miyajima et al (Miyajima 2004) conducted a review of RCTs that compared the effects of beta blockers versus placebo on intermittent claudication. Their review was slightly different in that they collated different beta blockers (although of the same type) versus placebo and found significant worsening of claudication distance with beta blockers. We do not agree with this method, as different drugs have different pharmacological properties. For example, pindolol and acebutolol have some intrinsic sympathomimetic activity as well (BNF 2013). Miyajima 2004 identified this and performed a subanalysis of beta blockers with and without intrinsic sympathomimetic activities and found no significant difference in walking distance. We think this grouping is also not appropriate because in the group without sympathomimetic activity, the authors included labetalol, which has an additional action of lowering peripheral vascular resistance that is not provided by the rest (BNF 2013). Although we disagree with the methods used by these review authors, their overall results suggested no significant effects of beta blockers on claudication time.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
