Cardioselective beta‐blockers for reversible airway disease

Abstract

Background

Beta‐blocker therapy has mortality benefit in patients with hypertension, heart failure and coronary artery disease, as well as during the perioperative period. These drugs have traditionally been considered contraindicated in patients with reversible airway disease.

Objectives

To assess the effect of cardioselective beta‐blockers in patients with asthma or chronic obstructive pulmonary disease (COPD).

Search methods

We performed a search of the Cochrane Airways Group Register of trials up to June 2011. We checked reference lists of trial reports and review articles.

Selection criteria

Randomized, blinded, placebo‐controlled trials of the effects of single‐dose or continued‐treatment cardioselective beta‐blockers in patients with reversible airway disease.

Data collection and analysis

Two independent review authors extracted data from the selected articles, reconciling differences by consensus. We divided beta1‐blockers into those with or without intrinsic sympathomimetic activity (ISA). Interventions were: administration of single‐dose or continued beta1‐blocker, and response to beta2‐agonist given after the study drug.

Main results

Nineteen studies on single‐dose treatment and 10 studies on continued treatment met the inclusion criteria. Single‐dose cardioselective beta‐blocker produced a 7.46% (95% confidence interval (CI) 5.59 to 9.32) reduction in forced expiratory volume in one second (FEV1), but with a 4.63% (95% CI 2.47 to 6.78) increase in FEV1 with beta2‐agonist, compared to placebo. Treatment lasting three to 28 days produced no change in FEV1 (mean difference (MD) ‐0.42% change from baseline; 95% CI ‐3.74 to 2.91), symptoms or inhaler use, whilst maintaining an 8.74% (95% CI 1.96 to 15.52) response to beta2‐agonist. There was no significant change in FEV1 treatment effect for those patients with chronic obstructive pulmonary disease (COPD): single doses (MD ‐5.28%; 95% CI ‐10.03 to ‐0.54); continued treatment (MD 1.07%; 95% CI ‐3.30 to 5.44).

With continued treatment there was no significant difference in FEV1 response for beta1‐blockers without ISA compared to those with ISA: ‐3.22% (95% CI ‐7.79 to 1.36) compared to 2.72% (95% CI ‐2.12 to 7.57). Those without ISA produced a 12.0% increase in FEV1 after beta2‐agonist administration compared to placebo (95% CI 4.12 to 19.87) while beta1‐blockers with ISA produced no change compared to placebo (MD ‐0.60%; 95% CI ‐13.93 to 12.73). These results were obtained in a small number of studies with few patients. The difference was not significant.

Authors' conclusions

Cardioselective beta‐blockers given in mild to moderate reversible airway disease or COPD do not produce adverse respiratory effects. Given their demonstrated benefit in conditions such as heart failure, cardiac arrhythmias and hypertension, these agents should not be withheld from such patients. Long‐term safety still needs to be established.

Plain language summary

Cardioselective beta‐blockers for reversible airway disease

Beta‐blockers reduce mortality in patients with hypertension, heart failure and coronary arterial disease. Traditionally they have not been given to patients with reversible airway disease (asthma or chronic obstructive pulmonary disease with a reversible obstructive component), for fear of adverse respiratory effects. This review of randomized controlled trials, which evaluated cardioselective beta‐blocker use in patients with reversible airway disease, demonstrated no increase in adverse respiratory effects. From the available evidence, it appears to be safe to prescribe these drugs to people with reversible airways disease.

Background

Beta‐adrenergic blocking agents, or beta‐blockers, are indicated in the management of angina pectoris, myocardial infarction, hypertension, congestive heart failure, cardiac arrhythmia and thyrotoxicosis, as well as to reduce complications in the perioperative period (Jones 1980; Koch‐Weser 1984; IPPPSH 1985; Wadworth 1991; Klein 1994; Kendall 1995; Mangano 1996; Doughty 1997; JNC VI 1997; Lechat 1998; Freemantle 1999; Heidenreich 1999). Despite clear evidence of their effectiveness and mortality benefit, clinicians are often hesitant to administer them in the presence of a variety of common conditions for fear of adverse reactions (Kennedy 1995; Viskin 1996; Gottlieb 1998; Chafin 1999). Review articles and practice guidelines have listed asthma as a contraindication to beta‐blocker use, citing cases of acute bronchospasm occurring during non‐cardioselective beta‐blocker use (Tattersfield 1986; Tattersfield 1990; O'Malley 1991; Belli 1995; Craig 1996; JNC VI 1997; Kendall 1997). Cardioselective beta‐blockers, or beta1‐blockers, have over 20 times more affinity for beta‐1 receptors as for beta‐2 receptors, and theoretically should have significantly less risk for bronchoconstriction (Wellstein 1987).

We evaluated only patients with documented reactive airway disease in this review, as these patients have been thought to be particularly susceptible to the respiratory effects of beta‐blockers. Patients with COPD, in general, are at greater risk for ischemic heart disease than asthmatics, so may benefit more from the use of beta‐blockers. This study evaluates a subgroup of patients with a documented chronic obstructive component, but was not designed to make recommendations about patients with COPD. Another recent Cochrane Review evaluated the use of cardioselective beta‐blockers in patients with COPD, given either as a single dose or for continued treatment (Salpeter 2011). Pooled data from 22 trials demonstrated no adverse effect on forced expiratory volume in one second (FEV1) or respiratory symptoms for beta1‐blockers compared to placebo, even for those with severe chronic airway obstruction.

Objectives

1. To evaluate the effect of cardioselective beta1‐blockers on respiratory function in patients with reversible airway disease, as assessed by FEV1, the incidence of symptoms and use of inhaled short‐acting beta2‐agonists.

2. To evaluate the FEV1 response to beta2‐agonists administered after beta1‐blockers in these same participants.

Methods

Criteria for considering studies for this review

Types of studies

We included randomized trials (RCTs) of cardioselective beta‐blockers and included beta‐blockers regardless of whether or not they were considered to possess intrinsic sympathomimetic activity (ISA) (Palmer 1969; Nicolaescu 1972; Skinner 1975a; Singh 1976; Decalmer 1978; Lofdahl 1983b; Falliers 1985; Fitzgerald 1991; McAreavey 1991; Wadworth 1991; Lois 1997; Lois 1999) (Table 1). We included studies of intravenous or oral beta1‐blockers either as a single dose or for continued treatment.

1. Beta‐blocker categories.

Nonselective (‐ ISA)	Nonselective (+ ISA)	1: Selective (‐ ISA)	2: Selective (+ ISA)
Propranolol	Oxprenolol	Atenolol	Celiprolol (+ alpha block)
Timolol	Pindolol	Metoprolol	Acebutolol
Nadolol	Dilevalol	Bisoprolol	Xamoterol
Sotalol (antiarrhythmic)	Prenalterol	Esmolol
Ibutmonide (+ alpha block)	Labetalol (+ alpha block)	Practolol
		Pafenolol
		Tolamolol
		Bevantolol (+ alpha agonist)

We included single‐dose trials if they:

1. reported forced expiratory volume at one second (FEV1) at baseline and follow‐up or provided per cent variance;

2. withheld beta2‐agonists for at least eight hours prior to initial FEV1 measurements;

3. did not include or exclude subjects for prior response, positive or negative, to beta‐blockers;

4. were randomized and single or double‐blinded;

5. placebo‐controlled; and

6. included only subjects with documented reversible airway disease, demonstrated by a mean increase of at least 15% in FEV1 in response to beta2‐agonist, response to methacholine challenge, or as asthma defined by the American Thoracic Society (ATS 1993).

We decided, a priori, that the above inclusion criteria (3) through (6) would be applied to trials of continued treatment. We included studies of continued treatment if they did not report FEV1, but instead evaluated the amount of beta2‐agonist use and respiratory symptoms reported compared to placebo. In addition, we included trials even if beta2‐agonists were not withheld during the trial.

We evaluated only cardioselective beta1‐blockers in this study as these are the ones most frequently used in clinical practice. We included studies only if participants had clearly documented reversible airway disease. We chose FEV1 to evaluate respiratory function because it is a standardized and reproducible method of airway function. For single‐dose trials, only studies that withheld beta2‐agonists for at least eight hours prior to measurements were included, to maximize the response seen with beta2‐blockers. We analyzed separately clinical trials that did not meet our inclusion criteria, but gave information on FEV1 responses to beta1‐blockers in patients with reversible airway disease.

Types of participants

We included participants with reversible airway disease, defined as asthma or chronic obstructive pulmonary disease (COPD) with reversible bronchial obstruction. We included definitions of reversible airway disease determined by response to methacholine challenge, an increase in FEV1 of at least 15% to beta2‐agonist administration, or the presence of asthma as defined by the American Thoracic Society (ATS 1993).

Types of interventions

We included two interventions.

1. Intravenous or oral beta1‐blockers versus placebo, given either as a single dose or during continued treatment.

2. Administration of a beta2‐agonist, either intravenously or by inhalation, given after the study medication (beta‐blocker) or after placebo.

Types of outcome measures

Two of us independently extracted data on the following outcomes:

1. change in FEV1 in response to placebo or study drug;

2. FEV1 response to beta2‐agonist administered after placebo or study drug;

3. number of patients with reported symptoms during the trial, such as wheezing, dyspnea or asthma exacerbation; and

4. for trials of continued treatment, the weekly use of inhaled short‐acting beta2‐agonists.

Search methods for identification of studies

Electronic searches

Two investigators jointly developed strategies, with the help of an information service librarian and the Cochrane Airways Group Trial Search Co‐ordinator. We performed a comprehensive search to identify all relevant randomized controlled trials published between 1966 and June 2011 using the Cochrane Airways Group Register (CAGR) (please see Appendix 1). We searched all records in the CAGR concerned with reversible airway disease using the following terms:

(adrenergic* and antagonist*) or (adrenergic* and block*) or (adrenergic* and beta‐receptor*) or (beta‐adrenergic* and block*) or (beta‐blocker* and adrenergic*) or (blockader* or Acebutolol or Alprenolol or Atenolol or Betaxolol or Bisoprolol or Bupranolol or Butoxamine or Carteolol or Celiprolol or Dihydroalprenolol or Iodocyanopindolol or Labetalol or Levobunolol or Metipranolol or Metoprolol or Nadolol or Oxprenolol or Penbutolol or Pindolol or Practolol or Propranolol or Sotalol or Timolol)

There was no restriction on the language of publication.

Searching other resources

We also scanned the reference lists of identified trial reports or review articles, and of abstracts at clinical symposia.

Data collection and analysis

Selection of studies

We performed a search to identify all relevant published clinical trials that addressed the effects of cardioselective beta‐blockers on airway function in patients with reversible airway disease. Two of us independently evaluated studies for inclusion. In choosing articles, investigators were blinded to results but not to journal, author or institution of studies. We calculated the observed percentage agreement between raters for the assessment of inclusion.

Data extraction and management

We classified each beta1‐blocker used into one of two categories. Group 1 included cardioselective beta‐blockers without intrinsic sympathomimetic activity (ISA) and Group 2 included cardioselective beta‐blockers with ISA (Palmer 1969; Nicolaescu 1972; Lofdahl 1983b; McAreavey 1991; Lois 1997; Lois 1999; Fitzgerald 1991) (Table 1).

Two of us extracted data from the selected articles, reconciling differences by consensus. We included only published data from the trials in the analysis. We did not attempt to contact the original authors to verify the data or obtain more information, as all of the trials identified were small and many were published several years ago.

Assessment of risk of bias in included studies

We assessed the methodological quality of each trial, using the following factors.

1. Was the study randomized? If so, was the randomization procedure adequate and was there allocation concealment?

2. Were the patients and people administering the treatment blind to the intervention?

3. Were withdrawals and drop‐outs described and was the analysis by intention‐to‐treat?

Based on these criteria, we broadly subdivided studies into the following three categories: (a) all quality criteria met, (b) one or more of the quality criteria only partially met, or (c) one or more criteria not met.

Measures of treatment effect

We estimated the net treatment effect by measuring the ratio of the lowest FEV1 value seen after study drug to baseline FEV1, for both placebo and active treatment, and recorded the per cent change from baseline. We then compared the treatment response to the placebo response. For studies of the response to beta2‐agonists, we used the baseline of the FEV1 taken after study drug but prior to beta2‐agonist or placebo. We calculated the net treatment effect by calculating the ratio of FEV1 measured after beta2‐agonist to the new baseline for both placebo and active treatment, and then compared the treatment‐agonist response to the placebo‐agonist response.

We measured the results for respiratory symptoms as a risk difference, by subtracting the number of patients with respiratory symptoms after treatment from the number of patients with symptoms after placebo. We then pooled the risk differences using the fixed‐effect model for dichotomous outcomes. For inhaler use during longer duration studies, we averaged the standard deviation (SD) for both placebo and treatment from the available data, and we calculated the weighted mean treatment effects using the fixed‐effect model for continuous outcomes.

Dealing with missing data

We have changed the methods used to obtain SD in the updated version of the review, to get a better assessment of the net treatment effect in these cross‐over trials. Whenever possible, we calculated the SD for the net treatment effect from individual patient data or P values, and then used these figures to derive the SDs for the analysis. Some trials only provided SDs calculated for treatment response and placebo response separately. For those trials that did not report any information on SDs, we obtained the average SD from trials that provided such data and calculated these separately for placebo, treatment and beta2‐agonist responses. We performed sensitivity analyses to evaluate the effect of including these trials, by using the lowest and highest available SD in place of the pooled SD, and also by excluding these trials from the analysis.

Assessment of heterogeneity

In order to test for inter‐study heterogeneity, we calculated the Chi² for the assumption of homogeneity. In addition, we compared the confidence intervals from the fixed‐effect model to those using the random‐effects model (DerSimonian 1986). We chose the fixed‐effect model to report the results, as minimal heterogeneity was noted in most of the analyses. When heterogeneity was noted, we reported the results from both the random‐effects model and fixed‐effect model. In addition we evaluated the Q‐parameter, as described by Berlin (Berlin 1989) and the values were found to be compatible with homogeneity.

Data synthesis

We then pooled the mean treatment effects to get a weighted average of the study means using a fixed‐effect model for continuous outcomes (Mantel 1959; Yusuf 1985). We obtained confidence intervals (CI) with 95% significance for the pooled study means.

Subgroup analysis and investigation of heterogeneity

We performed a subgroup analysis to compare the treatment effects of cardioselective beta‐blockers with and without ISA. We also performed another analysis to evaluate the response of patients with concomitant chronic airways obstruction, documented by a baseline FEV1 of less than 80% normal predicted value or less than 1.8 liters, or as defined by the American Thoracic Society (ATS 1995). A third subgroup analysis evaluated the treatment response for those participants known to have comorbid cardiovascular conditions such as hypertension.

Sensitivity analysis

Many of the trials were performed 20 or 30 years ago and did not provide adequate information to calculate SDs for the net treatment effect. We performed sensitivity analyses to evaluate the effect of including trials that did not provide any information on SDs.

Results

Description of studies

Results of the search

The database search identified 281 potentially relevant articles. After review of articles and bibliographies, we found 104 trials of beta‐blockers in patients with reversible airway disease. Of these trials 29 met inclusion criteria.

Included studies

Of the 29 included studies, 19 gave information on singe‐dose studies (Johnsson 1975; Skinner 1975b; Benson 1978; Ruffin 1979; Ellis 1981; Lofdahl 1981; van den Bergh 1981; Adam 1982; Lawrence 1982; Greefhorst 1984; Lammers 1984; Lammers 1985a; Chatterjee 1986; Doshan 1986a; Doshan 1986b; Falliers 1986; Lammers 1986a; Chodosh 1988; Lammers 1988; Tantucci 1990) and 10 provided data on treatment of longer duration (Nicolescu 1972; Nicolaescu 1973; Lawrence 1982; Butland 1983; Fenster 1983; Lammers 1985a; Dorow 1986c; van Zyl 1989; Fogari 1990; Bauer 1994). One of the articles (Lawrence 1982) gave data for both. Inter‐rater agreement for study eligibility was 94% (96 of 102 studies). We reached consensus on the remaining six trials.

Excluded studies

We excluded 66 trials for the following reasons: 24 trials evaluated nonselective beta‐blockers only; 27 studies failed to provide baseline and follow‐up FEV1 values; seven studies did not meet definitions for reversible airway disease; three single‐dose studies did not withhold beta2‐agonists for at least eight hours prior to study measurements; one study presented data that were published elsewhere; four studies were not blinded; two were not placebo‐controlled; four included or excluded patients based on a prior response to beta2‐blockers; and one was performed during exercise provocation.

Risk of bias in included studies

To maintain high methodological quality of the studies, we set strict inclusion criteria so that all trials were randomized, blinded and placebo‐controlled. We required all participants to have documented reversible airway disease using a conventional definition, and for single‐dose studies, beta2‐agonists had to be withheld for at least eight hours prior to measurements. We excluded studies that did not meet these methodological criteria, and were evaluated separately for sensitivity analysis.

The dropout rate was low: 2% for single‐dose studies and 1.3% for longer duration trials. All the studies evaluated were small cross‐over trials that received a B quality score on the Jadad scale, due to the fact that the randomization process was not described in detail or the trial was single‐blind instead of double‐blind.

Effects of interventions

Cardioselective blockers without intrinsic sympathomimetic activity (ISA) that were included in the study were atenolol, metoprolol, bisoprolol and practolol. Cardioselective blockers with ISA that were studied were celiprolol, acebutolol and xamoterol (Table 1).

Single‐dose treatment results

Nineteen studies on single‐dose treatment included 240 patients, 79% of whom were men. There was an average of 12.6 patients per study and the total dropout rate was 2.0%. From the available information the age range of participants was 19.5 to 65.1 years and the mean age was 40.1 years. These baseline characteristics were the same for the placebo and treatment groups because all of the trials were cross‐over by design. The baseline forced expiratory volume in one second (FEV1) measured in the treatment group was 2.41 (+/‐ 0.15) L/sec, and in the placebo group was 2.42 (+/‐ 0.2) L/sec.

Compared to placebo, single doses of cardioselective beta‐blockers as a group were associated with a reduction in FEV1 (mean difference (MD) ‐7.46; 95% confidence interval (CI) ‐9.32 to ‐5.59), but with an increase in FEV1 after beta2‐agonist was given (MD 4.63; 95% CI 2.47 to 6.78). There was no increase in respiratory symptoms seen in any of the 19 studies (risk difference (RD) 0.01; 95% CI ‐0.02 to 0.03).

Continued treatment results

Data from 10 studies involving 141 participants (77% of whom were men) were evaluated for response to continued treatment ranging from three days to four weeks. There was an average of 15.4 patients in each study, with a 1.3% dropout rate. From the available information the average age of the participants was 51.3 years. The average baseline FEV1 for the treatment group was 1.81 (+/‐ 0.13) liters and for the placebo group was 1.81 (+/‐ 0.15) liters. Five of the studies (54 participants) did not provide data on FEV1 and were used for information on symptoms and inhaler use.

In the continued treatment trials the use of cardioselective beta‐blockers as a group was not significantly different from placebo in terms of FEV1 (MD ‐0.42; 95% CI ‐3.74 to 2.91), symptoms (RD 0.01; 95% CI ‐0.02 to 0.04) or inhaler use (MD ‐0.11; 95% CI ‐6.75 to 6.54), and produced an increase in FEV1 compared to placebo after beta2‐agonist was given (MD 8.74; 95% CI 1.96 to 15.52).

Subgroup analysis

For single‐dose trials, beta1‐blockers without ISA were associated with a 6.5% reduction in FEV1 compared to those with ISA (MD ‐9.14%; 95% CI ‐11.31 to ‐6.97) compared to (MD ‐2.66%; 95% CI ‐6.32 to 1.00). However, treatment with beta1‐blockers without ISA was associated with a 9.7% increase in FEV1 in response to beta2‐agonist compared to those with ISA (MD 6.59%; 95% CI 4.18 to 9.01) compared to (MD ‐3.10%; 95% CI ‐7.89 to 1.69). In the continued treatment trials there was no significant change in FEV1 compared to placebo for beta1‐blockers without ISA (MD ‐3.22%; 95% CI ‐7.79 to 1.36) or for those with ISA (MD 2.72%; 95% CI ‐2.12 to 7.57). However, those without ISA produced a 12% increase in FEV1 after beta2‐agonist administration compared to placebo (95% CI 4.12 to 19.87%) while those with ISA produced no increase in FEV1 (MD ‐0.60%; 95% CI ‐13.93 to 12.73). Whilst this difference appears large its is not statistically significant. In addition, it should be noted that these results were obtained in only a small number of studies with a small number of patients.

In order to evaluate the treatment effect in patients with concomitant chronic obstructive pulmonary disease (COPD), we analyzed separately 10 trials that included only patients with documented chronic airways obstruction (Johnsson 1975; Lofdahl 1981; Adam 1982; Butland 1983; Fenster 1983; Lammers 1985a; Chatterjee 1986; Dorow 1986c; Fogari 1990; Bauer 1994). When the analysis was limited to those with COPD there was no significant difference in results for FEV1 treatment effect, in single‐dose trials (MD ‐5.28%; 95% CI ‐10.03 to ‐0.54) or continued treatment (MD 1.07%; 95% CI ‐3.30 to 5.44) and there was no increase in symptoms in any of the trials.

In eight of the trials, all participants had a comorbid condition such as hypertension (Adam 1982; Lawrence 1982; Lammers 1985a; Chatterjee 1986; Dorow 1986c; van Zyl 1989; Fogari 1990; Bauer 1994). When only these trials were included in the analysis there was no significant change in results for FEV1 treatment effect in single‐dose trials (MD ‐6.83%; 95% CI ‐11.46 to ‐2.20) or continued treatment (MD 1.31; 95% CI ‐2.62 to 5.24).

Inter‐study variance

There was no significant inter‐study variance found in the measurements of the FEV1 treatment effect, symptoms and long‐term inhaler use. Heterogeneity was detected between studies in the measurements of FEV1 made after beta2‐agonist use, in both the single‐dose and continued‐treatment studies. This heterogeneity was noted only in the subgroup of beta‐blockers without ISA. When we compared the random‐effects model to the fixed‐effect model for the beta2‐agonist response in the subgroup of beta‐blockers without ISA, there was less than 1% difference for single‐dose studies (5.66%; 95% CI 1.81 to 9.51) compared to 6.59% (95% CI 4.18 to 9.01) and a 1.7% difference for continued treatment (10.32%; 95% CI ‐6.38 to 27.01) compared to 12.0% (95% CI 4.12 to 19.89).

Sensitivity analysis

We performed a sensitivity analysis to evaluate the effect of including studies that did not provide any information on standard deviations (SDs) (Ruffin 1979; Ellis 1981; Chatterjee 1986; Lammers 1988). When we excluded the trials that did not provide SD data from the analysis, there was less than 0.5% difference for all parameters measured. When we performed the analysis by replacing the pooled SD with the lowest and highest available SD, the difference in results between the highest and lowest SD was 2% or less.

We also performed sensitivity analysis to evaluate the effect of including trials that did not meet the inclusion criteria, but that provided information on FEV1 and symptoms for cardioselective beta1‐blocker use in patients with reversible airway disease. Data analysis of 10 excluded studies, with 141 participants, showed no significant difference in any parameters compared to studies that met inclusion criteria (Vilsvik 1976; Boye 1977; Beumer 1978; Mue 1979a; Abraham 1981; Larsson 1982; Schindl 1986; Blaive 1988; Pujet 1992; Cazola 2000).

Discussion

This meta‐analysis pooled the data from 29 randomized, placebo‐controlled trials of the use of cardioselective beta‐blockers in patients with reversible airway disease. The results showed that, compared to placebo, the first dose of active treatment produced a small drop in forced expiratory volume in one second (FEV1), which was not associated with adverse respiratory effects. After continued treatment for a few days to weeks there was no difference in FEV1, symptoms or incidence of inhaler use. Cardioselective beta‐blockers, given in a single dose or for continued treatment, were associated with an increase in response to beta2‐agonists, as compared to placebo. No individual trial demonstrated an increase in respiratory symptoms compared to placebo. In fact, a review of all the 80 trials on cardioselective beta‐blockers identified in the database search found no study that demonstrated an increase in respiratory symptoms for beta1‐blockers compared to placebo or baseline.

We performed subgroup analyses to evaluate the effect of cardioselective beta‐blockers on patients with concomitant chronic obstructive pulmonary disease (COPD) or cardiovascular diseases such as hypertension, as these patients are most often targeted for beta‐blocker treatment. There was no significant difference in the FEV1 treatment effect, number of patients with symptoms, or incidence of inhaler use in these patients.

This meta‐analysis has several limitations (Ioannidis 1999). Most of the participants were relatively young with mild to moderate airway obstruction, and those with recent asthma exacerbation were often excluded from study. In addition, many of the studies were of short duration, so it is not possible to comment on the effect of cardioselective beta‐blockers on the frequency or severity of acute asthma exacerbations after several months of treatment. Furthermore, this analysis was based only on published literature and therefore is subject to publication bias. However, funnel plots of effect size versus standard error for the trials in this analysis showed no evidence of bias. We believe that these pooled results provide valuable information on the safety of cardioselective beta‐blockers in patients with reversible airway disease, with or without concomitant COPD or cardiovascular disease.

In the past, the standard of care was to consider reversible airway disease to be a contraindication to the use of all beta‐blockers (Tattersfield 1986; Tattersfield 1990; O'Malley 1991; Belli 1995; Craig 1996; JNC VI 1997; Kendall 1997). Due to the proven mortality benefit of beta‐blockers, many of the other relative or absolute contraindications traditionally listed for beta‐blockers have been questioned and disproved, including impaired left ventricular function, peripheral vascular disease, diabetes mellitus, depression and advanced age (Kjekshus 1990; Opie 1990; Wicklmayr 1990; Radack 1991; Beto 1992; Bright 1992; Rosenson 1993; Jonas 1996; Gottlieb 1998; Lechat 1998; Krumholz 1999). Now, with the accumulated evidence of the safety of cardioselective beta‐blockers in reversible airway disease, the most recent expert panel guidelines from the National Asthma Education and Prevention Program state that these agents may be used in patients with asthma (Expert Panel Report 2007).

The original evidence of a potential adverse effect of beta‐blockers in reversible airway disease was based on case reports of acute bronchospasm precipitated by high doses of non‐cardioselective blockers, presumably due to their blockade of beta‐2 receptors on bronchial smooth muscle (McNeill 1964; Zaid 1966; Anderson 1979; Raine 1981). A search of the literature found 16 trials that met the inclusion criteria for this analysis but that evaluated non‐cardioselective beta‐blockers (Johnsson 1975; Skinner 1975a; Benson 1978; Ruffin 1979; Ellis 1981; Sue 1981; Adam 1982; George 1983; Light 1983; Falliers 1985; Lammers 1985b; Doshan 1986b; Giulekas 1986; Chodosh 1988; Fogari 1990; Devereux 1998). Pooled results showed that regular use of nonselective beta‐blockers, compared to placebo, caused a 13.5% reduction (95% confidence interval (CI) ‐23.0 to ‐4.0) in FEV1, and a 22.5% decrement (95% CI ‐32.5 to ‐12.5) in the FEV1 response after beta2‐agonists were given. No significant increases in symptoms or inhaler use were found. However, one cannot rule out the possibility that the decrement in beta2‐agonist response seen with nonselective beta‐blockers may increase the risk of a clinically significant adverse effect occurring during an asthma exacerbation.

Cardioselective beta‐blockers such as atenolol, bisoprolol and metoprolol are at least 20 times more potent at blocking beta‐1 receptors than beta‐2 receptors, so that at therapeutic doses the beta‐2 blocking effect is negligible (Wellstein 1987). The doses of beta1‐blockers evaluated in this analysis varied from therapeutic to supratherapeutic doses, those that are not generally used for initiation of treatment. For example, in single‐dose studies using 50 to 200 mg of atenolol or metoprolol, no clinically apparent effect on respiratory function was demonstrated. Linear regression analysis could not differentiate a treatment effect between low and high doses, as there were so few low‐dose trials.

The results of this meta‐analysis indicate that for cardioselective beta1‐blockers without intrinsic sympathomimetic activity (ISA), the minimal decrement in FEV1 noted on a single dose is attenuated over a few days to weeks. In addition, there is an increase in beta2‐agonist response that is maintained with continued treatment. These results could be explained by an up‐regulation or sensitization of beta‐2 receptors, accompanied by an increased effect of endogenous or exogenous beta2‐agonist stimulation (Hall 1983; Hall 1989; Motomura 1990). This paradoxical effect of upregulation on beta‐2‐agonist receptors may, in fact, be beneficial for patients with asthma (Dickey 2010; Page 2011). There has been recent interest in studying the use of beta‐blockers in the treatment of asthma (Bond 2007; Lipworth 2009).

There is now accumulating evidence that the use of inhaled beta2‐agonists in patients with reactive airway disease is associated with a tolerance to beta2‐agonist stimulation and a worsening of asthma control (Molinoff 1983; Kraan 1985; Borst 1990; Taylor 1992; Cockcroft 1993; Wahedna 1993; Salpeter 2004). Meta‐analyses of randomized, placebo‐controlled trials have shown that the regular use of beta‐agonists in asthma results in an increase in severe asthma exacerbations and asthma‐related deaths, compared with placebo (Salpeter 2006; Salpeter 2010).

There is also evidence that treatment with beta‐blockers that have intrinsic beta2 ISA is associated with down‐regulation of beta2 receptors (Busso 1985; Sbirrazzuoli 1989; Brodde 1990; Jakubetz 1999). This is consistent with data from this analysis that showed that continued treatment with beta1‐blockers with ISA did not produce the increase in beta2‐agonist response that was seen with beta1‐blockers without ISA.

Only a small fraction of patients with heart disease who would benefit from beta‐blockers are presently given this treatment, mainly due to unfounded fears of their adverse effects (Sial 1994; Soumerai 1997; Krumholz 1998; Wang 1998). A study by Gottlieb 1998 on survivors of myocardial infarction included 46,000 patients with asthma and chronic obstructive lung disease, and showed a significant reduction in total mortality for those treated with beta‐blockers compared to those who were not. Other studies that have investigated the use of beta1‐blockers in patients with cardiac disease and concomitant chronic obstructive lung disease or asthma found that these medicines were well tolerated (Quan 1983; Dorow 1986b; Mooss 1994). Trials that evaluated the use of beta1‐blockers in hypertensive patients, many of whom had reversible airway disease, did not demonstrate a worsening of respiratory symptoms or FEV1 in these patients (Formgren 1976; George 1983; Krauss 1984; Falliers 1985). A recent study by Heller 2000 showed that COPD and asthma were the co‐morbidities most commonly associated with beta‐blockers being withheld in elderly patients after a myocardial infarction.

It is generally considered that patients with COPD are at greater risk than those with reversible airway disease for developing ischemic heart disease and other cardiovascular conditions requiring the use of beta‐blockers. There is, however, significant overlap in the presenting features of COPD and reversible airway disease. Another recent meta‐analysis evaluated the effect of cardioselective beta‐blockers in patients with COPD and found no change in FEV1 or respiratory symptoms for single doses or continued use of these agents compared to placebo (Salpeter 2011). Subgroup analyses revealed no difference in results for those with concomitant reversible airway disease, or with severe chronic airways obstruction as demonstrated by a baseline FEV1 of less than 1.4 liters or less than 50% normal predicted values. Three of the trials from that meta‐analysis are also included in this analysis (Adam 1982; Fenster 1983; Fogari 1990). The cumulative evidence from these two meta‐analyses indicates that cardioselective beta‐blockers should not be withheld in patients with reactive airway disease or COPD.

Authors' conclusions

Implications for practice.

Beta‐blocker treatment reduces mortality in patients with cardiovascular disease. The available data suggest that there is a small reduction in forced expiratory volume in one second (FEV1) following a single dose of a cardioselective beta‐blocker, but the response to inhaled beta2‐agonists is preserved, without any adverse respiratory effects. Continued treatment with relatively high doses of cardioselective beta‐blockers without intrinsic sympathomimetic activity (ISA) produces no reduction in FEV1, while increasing the sensitivity to beta2‐agonist stimulation. It would be reasonable to start treatment with a low daily dose of a cardioselective beta‐blocker without ISA and then titrate the dose up as needed.

Implications for research.

From the accumulated evidence we have now, it is apparent that patients with reversible airway disease should not be excluded from future beta1‐blocker trials so that the treatment effect can be studied in this substantial population. It would also be helpful to conduct trials on the regulation of airway responsiveness in patients with reversible airway disease treated with cardioselective beta‐blockers, to understand the effects of these medications on respiratory function. Due to the paradoxical effect of upregulation of beta‐receptors, studies are now emerging to evaluate the potential beneficial effect of beta‐blockers in the treatment of asthma.

What's new

Date	Event	Description
9 June 2011	New search has been performed	New literature search run. No new eligible studies identified. We made minor edits for style.

History

Protocol first published: Issue 2, 2000
 Review first published: Issue 2, 2001

Date	Event	Description
22 July 2008	Amended	Converted to new review format.
3 July 2002	New citation required and conclusions have changed	Substantive amendment.

Appendices

Appendix 1. Sources and search methods for the Cochrane Airways Group Specialised Register (CAGR)

Electronic searches: core databases

Database	Frequency of search
MEDLINE (Ovid)	Weekly
EMBASE (Ovid)	Weekly
CENTRAL (T he Cochrane Library)	Quarterly (4 issues per year)
PsycINFO (Ovid)	Monthly
CINAHL (Ebsco)	Monthly
AMED (Ebsco)	Monthly

Handsearches: core respiratory conference abstracts

Conference	Years searched
American Academy of Allergy, Asthma and Immunology (AAAAI)	2001 onwards
American Thoracic Society (ATS)	2001 onwards
Asia Pacific Society of Respirology (APSR)	2004 onwards
British Thoracic Society Winter Meeting (BTS)	2000 onwards
Chest Meeting	2003 onwards
European Respiratory Society (ERS)	1992, 1994, 2000 onwards
International Primary Care Respiratory Group Congress (IPCRG)	2002 onwards
Thoracic Society of Australia and New Zealand (TSANZ)	1999 onwards

Condition search

1. exp Asthma/

2. asthma$.mp.

3. (antiasthma$ or anti‐asthma$).mp.

4. Respiratory Sounds/

5. wheez$.mp.

6. Bronchial Spasm/

7. bronchospas$.mp.

8. (bronch$ adj3 spasm$).mp.

9. bronchoconstrict$.mp.

10. exp Bronchoconstriction/

11. (bronch$ adj3 constrict$).mp.

12. Bronchial Hyperreactivity/

13. Respiratory Hypersensitivity/

14. ((bronchial$ or respiratory or airway$ or lung$) adj3 (hypersensitiv$ or hyperreactiv$ or allerg$ or insufficiency)).mp.

15. ((dust or mite$) adj3 (allerg$ or hypersensitiv$)).mp.

16. or/1‐15

17. exp Aspergillosis, Allergic Bronchopulmonary/

18. lung diseases, fungal/

19. aspergillosis/

20. 18 and 19

21. (bronchopulmonar$ adj3 aspergillosis).mp.

22. 17 or 20 or 21

23. 16 or 22

24. Lung Diseases, Obstructive/

25. exp Pulmonary Disease, Chronic Obstructive/

26. emphysema$.mp.

27. (chronic$ adj3 bronchiti$).mp.

28. (obstruct$ adj3 (pulmonary or lung$ or airway$ or airflow$ or bronch$ or respirat$)).mp.

29. COPD.mp.

30. COAD.mp.

31. COBD.mp.

32. AECB.mp.

33. or/24‐32

34. exp Bronchiectasis/

35. bronchiect$.mp.

36. bronchoect$.mp.

37. kartagener$.mp.

38. (ciliary adj3 dyskinesia).mp.

39. (bronchial$ adj3 dilat$).mp.

40. or/34‐39

41. exp Sleep Apnea Syndromes/

42. (sleep$ adj3 (apnea$ or apnoea$)).mp.

43. (hypopnea$ or hypopnoea$).mp.

44. OSA.mp.

45. SHS.mp.

46. OSAHS.mp.

47. or/41‐46

48. Lung Diseases, Interstitial/

49. Pulmonary Fibrosis/

50. Sarcoidosis, Pulmonary/

51. (interstitial$ adj3 (lung$ or disease$ or pneumon$)).mp.

52. ((pulmonary$ or lung$ or alveoli$) adj3 (fibros$ or fibrot$)).mp.

53. ((pulmonary$ or lung$) adj3 (sarcoid$ or granulom$)).mp.

54. or/48‐53

55. 23 or 33 or 40 or 47 or 54

Filter to identify RCTs

1. exp "clinical trial [publication type]"/

2. (randomized or randomised).ab,ti.

3. placebo.ab,ti.

4. dt.fs.

5. randomly.ab,ti.

6. trial.ab,ti.

7. groups.ab,ti.

8. or/1‐7

9. Animals/

10. Humans/

11. 9 not (9 and 10)

12. 8 not 11

The MEDLINE strategy and RCT filter are adapted to identify trials in other electronic databases

Data and analyses

Comparison 1. Beta‐blocker vs placebo: Single dose.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 FEV1 treatment effect (Treatment minus placebo, % change from baseline)	19	660	Mean Difference (IV, Fixed, 95% CI)	‐7.46 [‐9.32, ‐5.59]
1.1 Group 1: Cardioselective without ISA vs placebo	19	480	Mean Difference (IV, Fixed, 95% CI)	‐9.14 [‐11.31, ‐6.97]
1.2 Group 2: Cardioselective with ISA vs placebo	6	180	Mean Difference (IV, Fixed, 95% CI)	‐2.66 [‐6.32, 1.00]
2 Patients with symptoms (Treatment minus placebo)	19	845	Risk Difference (M‐H, Fixed, 95% CI)	0.01 [‐0.02, 0.03]
2.1 Group 1: Cardioselective without ISA vs placebo	18	584	Risk Difference (M‐H, Fixed, 95% CI)	0.00 [‐0.03, 0.03]
2.2 Group 2: Cardioselective with ISA vs placebo	7	261	Risk Difference (M‐H, Fixed, 95% CI)	0.02 [‐0.03, 0.06]
3 SUBGROUP: COPD: FEV1 treatment effect (Treatment minus placebo, % change from baseline)	4	74	Mean Difference (IV, Fixed, 95% CI)	‐5.28 [‐10.03, ‐0.54]
3.1 Group 1: Cardioselective without ISA vs placebo	4	74	Mean Difference (IV, Fixed, 95% CI)	‐5.28 [‐10.03, ‐0.54]
4 SUBGROUP: Cardiovascular disease: FEV1 treatment effect (Treatment minus placebo, % change from baseline)	3	72	Mean Difference (IV, Fixed, 95% CI)	‐6.83 [‐11.46, ‐2.20]
4.1 Group 1: Cardioselective without ISA vs placebo	3	72	Mean Difference (IV, Fixed, 95% CI)	‐6.83 [‐11.46, ‐2.20]

1.1. Analysis.

Comparison 1 Beta‐blocker vs placebo: Single dose, Outcome 1 FEV1 treatment effect (Treatment minus placebo, % change from baseline).

1.2. Analysis.

Comparison 1 Beta‐blocker vs placebo: Single dose, Outcome 2 Patients with symptoms (Treatment minus placebo).

1.3. Analysis.

Comparison 1 Beta‐blocker vs placebo: Single dose, Outcome 3 SUBGROUP: COPD: FEV1 treatment effect (Treatment minus placebo, % change from baseline).

1.4. Analysis.

Comparison 1 Beta‐blocker vs placebo: Single dose, Outcome 4 SUBGROUP: Cardiovascular disease: FEV1 treatment effect (Treatment minus placebo, % change from baseline).

Comparison 2. Beta‐blocker + agonist vs placebo + agonist: Single dose.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 FEV1 treatment effect after Beta‐agonist (Treatment minus placebo, % change)	15	444	Mean Difference (IV, Fixed, 95% CI)	4.63 [2.47, 6.78]
1.1 Group 1: Cardioselective without ISA vs placebo	15	332	Mean Difference (IV, Fixed, 95% CI)	6.59 [4.18, 9.01]
1.2 Group 2: Cardioselective with ISA vs placebo	5	112	Mean Difference (IV, Fixed, 95% CI)	‐3.10 [‐7.89, 1.69]

2.1. Analysis.

Comparison 2 Beta‐blocker + agonist vs placebo + agonist: Single dose, Outcome 1 FEV1 treatment effect after Beta‐agonist (Treatment minus placebo, % change).

Comparison 3. Beta‐blocker vs placebo: Longer duration.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 FEV1 treatment effect (Treatment minus placebo, % change from baseline)	5	174	Mean Difference (IV, Fixed, 95% CI)	‐0.42 [‐3.74, 2.91]
1.1 Group 1: Cardioselective without ISA vs placebo	4	86	Mean Difference (IV, Fixed, 95% CI)	‐3.22 [‐7.79, 1.36]
1.2 Group 2: Cardioselective with ISA vs placebo	2	88	Mean Difference (IV, Fixed, 95% CI)	2.72 [‐2.12, 7.57]
2 Inhaler use per patient‐week (Treatment minus placebo)	3	166	Mean Difference (IV, Fixed, 95% CI)	‐0.11 [‐6.75, 6.54]
2.1 Group 1: Cardioselective without ISA vs placebo	2	48	Mean Difference (IV, Fixed, 95% CI)	0.00 [‐16.97, 16.97]
2.2 Group 2: Cardioselective with ISA vs placebo	3	118	Mean Difference (IV, Fixed, 95% CI)	‐0.13 [‐7.35, 7.10]
3 Patients with symptoms (Treatment minus placebo)	8	504	Risk Difference (M‐H, Fixed, 95% CI)	0.01 [‐0.02, 0.04]
3.1 Group 1: Cardioselective without ISA vs placebo	8	410	Risk Difference (M‐H, Fixed, 95% CI)	0.01 [‐0.02, 0.04]
3.2 Group 2: Cardioselective with ISA vs placebo	2	94	Risk Difference (M‐H, Fixed, 95% CI)	0.0 [‐0.06, 0.06]
4 SUBGROUP: COPD: FEV1 treatment effect (Treatment minus placebo, % change from baseline)	2	108	Mean Difference (IV, Fixed, 95% CI)	1.07 [‐3.30, 5.44]
4.1 Group 1: Cardioselective without ISA vs placebo	1	20	Mean Difference (IV, Fixed, 95% CI)	‐6.2 [‐16.37, 3.97]
4.2 Group 2: Cardioselective with ISA vs placebo	2	88	Mean Difference (IV, Fixed, 95% CI)	2.72 [‐2.12, 7.57]
5 SUBGROUP: Cardiovascular disease: FEV1 treatment effect (Treatment minus placebo, % change from baseline)	3	134	Mean Difference (IV, Fixed, 95% CI)	1.31 [‐2.62, 5.24]
5.1 Group 1: Cardioselective without ISA vs placebo	2	46	Mean Difference (IV, Fixed, 95% CI)	‐1.40 [‐8.10, 5.31]
5.2 Group 2: Cardioselective with ISA vs placebo	2	88	Mean Difference (IV, Fixed, 95% CI)	2.72 [‐2.12, 7.57]

3.1. Analysis.

Comparison 3 Beta‐blocker vs placebo: Longer duration, Outcome 1 FEV1 treatment effect (Treatment minus placebo, % change from baseline).

3.2. Analysis.

Comparison 3 Beta‐blocker vs placebo: Longer duration, Outcome 2 Inhaler use per patient‐week (Treatment minus placebo).

3.3. Analysis.

Comparison 3 Beta‐blocker vs placebo: Longer duration, Outcome 3 Patients with symptoms (Treatment minus placebo).

3.4. Analysis.

Comparison 3 Beta‐blocker vs placebo: Longer duration, Outcome 4 SUBGROUP: COPD: FEV1 treatment effect (Treatment minus placebo, % change from baseline).

3.5. Analysis.

Comparison 3 Beta‐blocker vs placebo: Longer duration, Outcome 5 SUBGROUP: Cardiovascular disease: FEV1 treatment effect (Treatment minus placebo, % change from baseline).

Comparison 4. Beta‐blocker + agonist vs placebo + agonist: Longer duration.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 FEV1 treatment effect after Beta‐agonist (Treatment minus placebo, % change)	3	80	Mean Difference (IV, Fixed, 95% CI)	8.74 [1.96, 15.52]
1.1 Group 1: Cardioselective without ISA vs placebo	3	60	Mean Difference (IV, Fixed, 95% CI)	12.00 [4.12, 19.87]
1.2 Group 2: Cardioselective with ISA vs placebo	1	20	Mean Difference (IV, Fixed, 95% CI)	‐0.60 [‐13.93, 12.73]

4.1. Analysis.

Comparison 4 Beta‐blocker + agonist vs placebo + agonist: Longer duration, Outcome 1 FEV1 treatment effect after Beta‐agonist (Treatment minus placebo, % change).

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Adam 1982.

Methods	Single‐dose Double‐blind Cross‐over Dropout: 0%
Participants	n = 10 Mean age = 65.1 Inclusion: hypertension and reversible chronic airway obstruction, with > 15% improvement FEV1 and FEV1/FVC < 70% Exclusions: none given
Interventions	#1: Intervention: Metoprolol 100 mg Atenolol 100 mg Placebo #2: Inhaled salbutamol after Tx or placebo
Outcomes	FEV1 Symptoms
Notes	Nonselectives studied: labetalol, propranolol

Bauer 1994.

Methods	Longer duration: 1 week treatment and placebo, cross‐over Double‐blind Dropout: 0%
Participants	n = 18 Mean age = 48.6 Inclusion: hypertension and stable asthma with > 15% improvement, FEV1 > 50% predicted Exclusion: angina, heart block, HR < 50, impaired renal or hepatic function, diabetes mellitus, pregnancy, lactation or adverse reaction to beta‐blocker
Interventions	Metoprolol 14/190 mg Atenolol 100 mg Placebo
Outcomes	FEV1 Symptoms
Notes

Benson 1978.

Methods	Single‐dose Cross‐over Single‐blind Dropout prior to treatment: 14% Dropout with treatment: 0%
Participants	n = 12 Mean age = 32.2 Inclusion: reversible airway obstruction, with spontaneous variability > 15%. All stable at time of study Exclusion: none given
Interventions	#1: Acebutolol 300 mg Atenolol 100 mg Placebo #2: Inhaled isoprenaline after Tx or placebo
Outcomes	FEV1 Symptoms
Notes	Nonselectives studied: propranolol, pindolol

Butland 1983.

Methods	Longer duration: 4 weeks treatment and placebo Cross‐over Double‐blind Dropout: 0%
Participants	n = 12 Mean age = 61 Inclusion: reversible COPD: FEV1 < 1 L with > 20% reversal. Could continue taking inhalers, steroids. Exclusion: other lung disease or serious disease of other organs
Interventions	Atenolol 100 mg/d Metoprolol 100 mg/d Placebo
Outcomes	Symptoms Exercise tolerance
Notes	FEV1 reported only as % of predicted normal value

Chatterjee 1986.

Methods	Single‐dose Cross‐over Double‐blind Dropout: 0%
Participants	n = 12 Mean age = 60 Inclusion: hypertension and bronchial asthma, with FEV1/vital capacity < 30% and > 15% improvement with agonist Exclusion: pregnancy, overt heart failure, renal failure, heart block. Current antihypertensive and inhaler treatment prohibited.
Interventions	#1: Atenolol 100 mg Bisoprolol 10 mg Bisoprolol 20 mg #2: Inhaled salbutamol after Tx and placebo
Outcomes	FEV1 Symptoms
Notes

Chodosh 1988.

Methods	Single‐dose Cross‐over Double‐blind Dropout: 11% prior to study beginning
Participants	n = 16 Mean age = 39 Inclusion: normotensive patients with stable bronchial asthma, > 15% increase FEV1 and FEV1 60% to 90% predicted Exclusion: taking cromolyn, their steroid or inhaler use changed, recent asthma attack requiring Tx , upper respiratory infection within 2 weeks or recent status asthmaticus
Interventions	#1: Metoprolol 200 mg Placebo #2: Inhaled isoproterenol after Tx and placebo
Outcomes	FEV1 Symptoms
Notes	Nonselective studied: dilevalol

Dorow 1986c.

Methods	Longer duration: 12 weeks treatment, 4 weeks placebo Cross‐over Double‐blind Dropout: 0%
Participants	n = 34 Inclusion: hypertension + reversible airway obstruction, > 15% increase FEV1 Exclusion: severe hypertension
Interventions	Celiprolol 100 to 600 mg/d Placebo
Outcomes	FEV1 Symptoms Inhaler use/wk
Notes

Doshan 1986a.

Methods	Single‐dose Cross‐over Double‐blind Dropout: 6% prior to study
Participants	n = 15 Age: 19 to 55 Inclusion: mild asthma, otherwise good health, with >15% increase FEV1, and FEV1 > 50% predicted Exclusion: those that require asthma medications other than theophylline or inhalers
Interventions	#1: Celiprolol 400 mg Celiprolol 600 mg Atenolol 100 mg Placebo #2: Inhaled albuterol after Tx and placebo
Outcomes	FEV1 Symptoms
Notes

Doshan 1986b.

Methods	Single‐dose Cross‐over Double‐blind Dropout: 0%
Participants	n = 34 Age: 18 to 57 Inclusion: normotensive + asthma, with > 15% increase and FEV1 > 50% predicted Exclusion: those requiring cromolyn or steroids
Interventions	Celiprolol 200 mg Celiprolol 400 mg Atenolol 100 mg Placebo
Outcomes	FEV1 Symptoms
Notes	Nonselective studied: propranolol

Ellis 1981.

Methods	Single‐dose Cross‐over Double‐blind Dropout: 0%
Participants	n = 14 Inclusion: reversible airways obstruction, average increase FEV1 23% Exclusion: heart failure, heart block, irreversible airways obstruction
Interventions	#1: Atenolol 50 mg Atenolol 100 mg Atenolol 200 mg Placebo #2: Inhaled isoprenaline after Tx or placebo
Outcomes	FEV1 Symptoms
Notes	Nonselective studied: propranolol

Falliers 1986.

Methods	Single‐dose Cross‐over Double‐blind Dropout: 0%
Participants	n =18 Age: 21 to 60 yrs Inclusion: asthma, with < 80 predicted and >15% improvement FEV1 Exclusion: hypertension, hematological dz, CVA within 3 months, cardiac dysrhythmia, heart block, recent asthma attack or upper respiratory infection in past 2 weeks, or status asthmaticus, or if taking cromolyn
Interventions	#1: Intervention: Metoprolol 100 mg Metoprolol 200 mg Placebo #2: Inhaled isoproterenol after Tx or placebo
Outcomes	FEV1 Symptoms
Notes	Nonselective studied: labetalol

Fenster 1983.

Methods	Longer duration: 1 wk Tx, 1 wk placebo Cross‐over Single‐blind Dropout: 0%
Participants	n = 6 Mean age = 48.6 Inclusion: reversible COPD: < 60% predicted FEV1 + > 15% reversal Exclusion: angina, MI, hypertension, valvular dz, heart failure, LVH or clinically unstable
Interventions	Metoprolol 200/d Placebo
Outcomes	Increase in symptoms
Notes	FEV1 reported only as % of predicted

Fogari 1990.

Methods	Longer duration: 1 week treatment, 2 weeks placebo Cross‐over Double‐blind Dropout: 0%
Participants	n = 10 Mean age = 57 Inclusion: HTN + reversible airway dz, with > 20% increase FEV1 Exclusion: heart failure, A‐V block, valve dz, obstructive arteriopathy, renal insufficiency
Interventions	#1: Intervention: Atenolol 100/d Celiprolol 200/d Placebo #2: Inhaled salbutamol after Tx or placebo
Outcomes	FEV1 Symptoms
Notes	Nonselectives studied: propranolol, oxprenolol

Greefhorst 1984.

Methods	Single‐dose Cross‐over Double‐blind Dropout: 0%
Participants	n = 8 Mean age = 29 Inclusion: asthma with > 20% increase FEV1, all with intrinsic atopic asthma. All in stable respiratory state. Exclusion: cardiovascular dz
Interventions	#1: Metoprolol 100 mg Acebutolol 400 mg Placebo #2: IV terbutaline after Tx or placebo
Outcomes	FEV1 Symptoms
Notes

Johnsson 1975.

Methods	Single‐dose Cross‐over Single‐blind Dropout: 0%
Participants	n = 7 Mean age = 44 Inclusion: endogenous asthma longer than 2 years, >15% reversibility Exclusion: no acute exacerbation at time of study, no history or signs of heart disease
Interventions	Metoprolol 0.12 mg/kg IV Placebo
Outcomes	FEV1 Symptoms
Notes	1) IV isoprenaline given, but baseline FEV1 not reported 2) Nonselective studied: propranolol

Lammers 1984.

Methods	Single‐dose Cross‐over Double‐blind Dropout: 0%
Participants	n = 8 Mean age = 39 Inclusion: bronchial asthma, defined as paroxysmal reversible generalized airway obstruction, with FEV1 > 15 increase. Some with chronic bronchitis. Exclusion: no respiratory infection or increase in bronchoconstriction within 4 weeks of study
Interventions	#1: Bisoprolol 10 mg Bisoprolol 20 mg Metoprolol 100 mg Placebo #2: Inhaled terbutaline after Tx and placebo
Outcomes	FEV1 Symptoms
Notes

Lammers 1985a.

Methods	Longer duration: 4 weeks treatment and placebo Cross‐over Double‐blind Dropout: 0%
Participants	n = 8 Mean age = 52.7 Inclusion: COPD + hypertension Average reversal > 15% FEV1. All in stable respiratory state: no recent respiratory infection or event. Exclusion: none listed
Interventions	Metoprolol 100 mg BID Placebo
Outcomes	FEV1 Symptoms
Notes	Nonselective studied: pindolol

Lammers 1986a.

Methods	Single‐dose Cross‐over Single + double‐blind Dropout: 0%
Participants	n = 11 Mean age = 36.6 Inclusion: bronchial asthma as per ATS criteria, with > 15% increase. None with cardiovascular disease Exclusion: none listed
Interventions	#1: Xamoterol 200 mg Atenolol 100 mg Placebo #2: Inhaled terbutaline after Tx and placebo
Outcomes	FEV1 Symptoms
Notes	Single‐blind: placebo on first day Double‐blind: Tx drugs

Lammers 1988.

Methods	Single‐dose Cross‐over Single‐blind Dropout: 0%
Participants	n = 11 Age: 22 to 60 yrs Inclusion: asthma as defined by ATS, with > 15% increase FEV1. FEV1 ranged 40% to 74% predicted. All in stable phase without steroids or theophylline. Exclusion: none listed
Interventions	#1: Atenolol 50 mg Placebo #2: Inhaled terbutaline after Tx and placebo
Outcomes	FEV1 Symptoms
Notes

Lawrence 1982.

Methods	Single‐dose + longer duration: 3 weeks Tx and placebo Cross‐over Single‐blinded Dropout: 0%
Participants	n = 14 Mean age = +55.7 Inclusion: asthma + hypertension > 15% reversal Exclusion: pregnant, renal failure, heart failure, heart block
Interventions	Acute: #1: Intervention: Atenolol 100 mg Metoprolol 100 mg Placebo #2 Intervention: Inhaled salbutamol after Tx or placebo Longer duration: Atenolol 100 mg/d Metoprolol 100 mg BID
Outcomes	FEV1 Acute: symptoms Longer duration: Asthma attacks Inhaler use/wk
Notes

Lofdahl 1981.

Methods	Single‐dose Cross‐over Double‐blind Dropout: 0%
Participants	n = 8 Mean age = 52 Inclusion: intrinsic asthma, constant reversibility > 20% None had an acute exacerbation at time of study Exclusion: none listed
Interventions	#1: Atenolol 100 mg Metoprolol 100 mg Placebo #2: Terbutaline (IV then inhaled) after Tx or placebo
Outcomes	FEV1 Symptoms
Notes

Nicolaescu 1973.

Methods	Longer duration: 3 days treatment and placebo Cross‐over Double‐blind Dropout: 0%
Participants	n = 10 Mean age = 46.6 Inclusion: severe bronchial asthma for at least 5 yrs, average increase FEV1 > 15%. Steroids continued. Inhalers held prior to measurements. Exclusions: no asthma attack in past several days
Interventions	#1: Intervention: Practolol 200 mg/d Placebo #2: Inhaled orciprenaline after Tx or placebo
Outcomes	FEV1 Symptoms
Notes

Nicolescu 1972.

Methods	Longer duration: 3 days treatment and placebo Cross‐over Double‐blinded Dropout: 0%
Participants	n = 10 Mean age = 43.7 Inclusion: mild bronchial asthma for at least 5 yrs, defined according to WHO, with average decrease in FEV1 to acetylcholine 26% Exclusion: none listed
Interventions	#1: Practolol 50 mg QID Placebo #2: Inhaled orciprenaline after Tx and placebo
Outcomes	FEV1 Symptoms
Notes

Ruffin 1979.

Methods	Single‐dose Cross‐over Double‐blind Dropout: 0%
Participants	n = 12 Mean age = 30.8 Inclusion: asthma with episodic dyspnea + wheeze, > 20% drop in FEV1 with histamine Exclusion: FEV1 < 70% predicted at start of study or symptoms out of control
Interventions	Metoprolol 100 mg Placebo
Outcomes	FEV1 Symptoms
Notes	Nonselectives studied: propranolol, timolol

Skinner 1975b.

Methods	Single‐dose Cross‐over Double‐blind Dropout: 0%
Participants	n = 10 Mean age = 36.8 Inclusion: asthma, with > 15% increase FEV1 Exclusion: none given
Interventions	#1: Acebutolol 300 mg Practolol 300 mg Placebo #2: Inhaled isoprenaline after Tx or placebo
Outcomes	FEV1 Symptoms
Notes

Tantucci 1990.

Methods	Single‐dose Cross‐over Double‐blind Dropout: 0%
Participants	n = 12 Mean age = 37.3 Inclusion: reversible airway dz, stable, with > 15% improvement FEV1 Exclusion: hx atopy, recent respiratory tract infection, contraindications to beta‐blocker, pregnancy
Interventions	Metoprolol OROS 14/190 mg Atenolol 100 mg Placebo
Outcomes	FEV1 Symptoms
Notes

van den Bergh 1981.

Methods	Single‐dose Cross‐over Single + double‐blind Dropout: 0%
Participants	n = 8 Mean age = 24 Inclusion: bronchial asthma, with > 20% increase FEV1. Mild to moderate in severity. All in stable respiratory state Exclusion: none listed
Interventions	#1: Metoprolol OT 100 mg Metoprolol OT 200 mg Metoprolol SR 200 mg Placebo #2: IV terbutaline after Tx or placebo
Outcomes	FEV1 Symptoms
Notes	Single‐blind: metoprolol 100 + placebo Double‐blind: metoprolol 200 mg x 2 doses

van Zyl 1989.

Methods	Longer duration: 4 weeks treatment, 2 weeks placebo Cross‐over Single/double‐blind Placebo‐controlled Dropout: 16% with placebo, 0% with treatment
Participants	n = 12 Mean age = 45.9 Inclusion: asthma + hypertension Asthma: < 85% predicted FEV1, > 15% reversal Exclusion: none listed
Interventions	Atenolol 100 mg/d Celiprolol 400 mg/d Placebo
Outcomes	FEV1 Symptoms Inhaler use/wk
Notes	Single‐blind: placebo run in and cross‐over Double‐blind: Tx drugs

ATS: American Thoracic Society 
 BID: twice a day 
 COPD: chronic obstructive pulmonary disease 
 CVA: cerebrovascular accident 
 d: day 
 dz: disease 
 FEV1: forced expiratory volume in one second 
 FVC: forced vital capacity 
 HR: heart rate 
 HTN: hypertension 
 hx: history 
 IV: intravenous 
 LVH: Left ventricular hypertrophy 
 MI: myocardial infarction 
 QID: four times a day (from the Latin quarter in die) 
 Tx: treatment 
 WHO: World Health Organization 
 wk: week

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Abraham 1981	Beta‐agonists not held: given 2 hrs before study Data analyzed separately in results
Astrom 1975	No FEV1 data No asthma definition Not blinded
Bernecker 1970	No FEV1 data No placebo
Beumer 1971	Not blinded No placebo
Beumer 1978	No FEV1 data (just % predicted) Data analyzed separately in results
Blaive 1988	No placebo Data analyzed separately in results
Boskabady 2000	Nonselective Not controlled
Boye 1977	Beta‐agonists not held 8 hours Data analyzed separately in results
Bruschi 1988	No FEV1 data: only in figure form
Cannon 1982	Nonselective Asthma not defined, ? reversal Agonists not held No FEV1 data
Carpentiere 1988	Nonselective Patients excluded for previous response to propranolol
Cazola 2000	Baseline FEV1 data not given Beta‐agonist not held
Christensen 1978	Nonselective Asthma not defined, ? reversal No FEV1 data
Clague 1984	No FEV1 data
Coleman 1974	Bronchial asthma: not defined
Connolly 1970	Nonselective No placebo No FEV1 data
Decalmer 1978	No FEV1 data: only in graph form
Devereux 1998	Nonselective
Dorow 1982a	No FEV1 data
Dorow 1982b	No FEV1 data
Dorow 1986a	No FEV1 data: only graph form
Dorow 1986b	No FEV1 data
Falliers 1985	Nonselective Inclusion based on prior response to propranolol Agonists not held 8 hours
Fleming 1978	No FEV1 data: only in graph form
Formgren 1976	No FEV1 data
Gayrard 1975	Nonselective No FEV1 data
George 1983	Nonselective No FEV1 data
George 1985	Nonselective Inclusion based on prior response to propranolol
Giulekas 1986	Nonselective
Greefhorst 1981	Patients excluded on basis of prior bronchoconstriction to propranolol
Greefhorst 1983	Nonselective
Groth 1986	No FEV1 data: only in graph form
Hamm 1970	Nonselective No FEV1 data
Hedner 1989	Nonselective No FEV1 data
Hugues 1978	Not blinded No placebo
Krauss 1984	Not blinded
Lammers 1985b	FEV1 measured during exercise provocation
Lammers 1986b	Duplicate patients to Lammers 1984
Larsson 1982	Asthma: < 15% increase in FEV1 (average 11%) Data analyzed separately in the results
Leary 1973	Not blinded
Light 1983	Nonselective For single‐dose study, agonists not held
Lofdahl 1982	Nonselective No FEV1 data
Lofdahl 1983b	No FEV1 data: graph form only
Lofdahl 1984	No FEV1 data: only in figure form
Lofdahl 1988	No FEV1 data: only in figure form
Macquin‐Mavier 1988	No FEV1 data: specific airway conductance
Matthys 1985	Patients included on basis of prior response to propranolol
Matthys 1986	Patients included on basis or prior response to propranolol
Meier 1966	Nonselective No FEV1 data
Mue 1979a	Asthma not defined ? reversal Data analyzed separately in the results
Mue 1979b	No FEV1 data: graph only Asthma with provocation test (? 15%)
Nair 1981	No FEV1 data: graph only
Oosterhoff 1995	Nonselective No FEV1 data
Palmer 1969	Asthma not defined ? agonist held No FEV1 data
Patakas 1983	Nonselective No FEV1 data
Philip‐Joet 1986	No FEV1 data: only in figure form
Pujet 1992	No placebo Data analyzed separately in results
Quan 1983	Beta‐agonists not held Data studied separately in results
Ranchod 1982	Chronic bronchitis: response to acetylcholine but ? > 15% No FEV1 data
Richardson 1969	Nonselective No FEV1 data
Ruffin 1982	No FEV1 data: only in figure form
Schindl 1986	No asthma definition (? reversal) Data analyzed separately in the results
Sheppard 1986	No FEV1 data: specific airway resistance
Singh 1976	No FEV1 data: specific airway resistance
Skinner 1975a	Nonselective Asthma criteria not defined
Skinner 1976	No FEV1 data No asthma definition
Sue 1981	Nonselective
Sue 1983	Nonselective No FEV1 data Only > 10% in FEV1
Szmidt 1999	No FEV1 data: only in figure form
Vilsvik 1976	No placebo Data analyzed separately in results
Vilsvik 1977	No FEV1 data
Von Graffenried 1978	Nonselective No FEV1 data
Wilcox 1986	No FEV1 data: only in figure form

FEV1: forced expiratory volume in one second

Differences between protocol and review

We changed the methods used to obtain standard deviations in the updated version of the review, to get a better assessment of the net treatment effect in these cross‐over trials(see Dealing with missing data).

Contributions of authors

Shelley Salpeter: developed review protocol, search strategy, trial selection, data extraction, manuscript preparation, management of RevMan (RevMan 2011) protocol.

Thomas Ormistion: search strategy, trial selection, data extraction, manuscript preparation.

Edwin Salpeter: Prof. Salpeter died November 2008, prior to the preparation of the 2011 update.  His contributions from 2001 to 2010 were: data analysis, statistical management, manuscript preparation.

Richard Wood‐Baker: editor.

Sources of support

Internal sources

• Santa Clara Valley Medical Center, USA.

External sources

• No sources of support supplied

Declarations of interest

None known.

New search for studies and content updated (no change to conclusions)