Long‐acting beta2‐agonists for chronic obstructive pulmonary disease
Abstract
Background
Chronic obstructive pulmonary disease (COPD) is a respiratory disease that causes progressive symptoms of breathlessness, cough and mucus build‐up. It is the fourth or fifth most common cause of death worldwide and is associated with significant healthcare costs.
Inhaled long‐acting beta2‐agonists (LABAs) are widely prescribed to manage the symptoms of COPD when short‐acting agents alone are no longer sufficient. Twice‐daily treatment with an inhaled LABA is aimed at relieving symptoms, improving exercise tolerance and quality of life, slowing decline and even improving lung function and preventing and treating exacerbations.
Objectives
To assess the effects of twice‐daily long‐acting beta2‐agonists compared with placebo for patients with COPD on the basis of clinically important endpoints, primarily quality of life and COPD exacerbations.
Search methods
We searched the Cochrane Airways Group trials register, ClinicalTrials.gov and manufacturers' websites in June 2013.
Selection criteria
Parallel, randomised controlled trials (RCTs) recruiting populations of patients with chronic obstructive pulmonary disease. Studies were required to be at least 12 weeks in duration and designed to assess the safety and efficacy of a long‐acting beta2‐agonist against placebo.
Data collection and analysis
Data and characteristics were extracted independently by two review authors, and each study was assessed for potential sources of bias. Data for all outcomes were pooled and subgrouped by LABA agent (formoterol 12 μg, formoterol 24 μg and salmeterol 50 μg) and then were separately analysed by LABA agent and subgrouped by trial duration. Sensitivity analyses were conducted for the proportion of participants taking inhaled corticosteroids and for studies with high or uneven rates of attrition.
Main results
Twenty‐six RCTs met the inclusion criteria, randomly assigning 14,939 people with COPD to receive twice‐daily LABA or placebo. Study duration ranged from three months to three years; the median duration was six months. Participants were more often male with moderate to severe symptoms at randomisation; mean forced expiratory volume in 1 second (FEV1) was between 33% and 55% predicted normal in the studies, and mean St George's Respiratory Questionnaire score (SGRQ) ranged from 44 to 55 when reported.
Moderate‐quality evidence showed that LABA treatment improved quality of life on the SGRQ (mean difference (MD) ‐2.32, 95% confidence interval (CI) ‐3.09 to ‐1.54; I2 = 50%; 17 trials including 11,397 people) and reduced the number of exacerbations requiring hospitalisation (odds ratio (OR) 0.73, 95% CI 0.56 to 0.95; I2 = 10%; seven trials including 3804 people). In absolute terms, 18 fewer people per 1000 were hospitalised as the result of an exacerbation while receiving LABA therapy over a weighted mean of 7 months (95% CI 3 to 31 fewer). Scores were also improved on the Chronic Respiratory Disease Questionnaire (CRQ), and more people receiving LABA treatment showed clinically important improvement of at least four points on the SGRQ.
The number of people who had exacerbations requiring a course of oral steroids or antibiotics was also lower among those taking LABA (52 fewer per 1000 treated over 8 months; 95% CI 24 to 78 fewer, moderate quality evidence).
Mortality was low, and combined findings of all studies showed that LABA therapy did not significantly affect mortality (OR 0.90, 95% CI 0.75 to 1.08; I2 = 21%; 23 trials including 14,079 people, moderate quality evidence). LABA therapy did not affect the rate of serious adverse events (OR 0.97, 95% CI 0.83 to 1.14; I2 = 34%, moderate quality evidence), although there was significant unexplained heterogeneity, especially between the two formoterol doses.
LABA therapy improved predose FEV1 by 73 mL more than placebo (95% CI 48 to 98; I2 = 71%, low quality evidence), and people were more likely to withdraw from placebo than from LABA therapy (OR 0.74, 95% CI 0.69 to 0.80; I2 = 0%). Higher rates of withdrawal in the placebo arm may reduce our confidence in some results, but the disparity is more likely to reduce the magnitude of difference between LABA and placebo than inflate the true effect; removing studies at highest risk of bias on the basis of high and unbalanced attrition did not change conclusions for the primary outcomes.
Authors' conclusions
Moderate‐quality evidence from 26 studies showed that inhaled long‐acting beta2‐agonists are effective over the medium and long term for patients with moderate to severe COPD. Their use is associated with improved quality of life and reduced exacerbations, including those requiring hospitalisation. Overall, findings showed that inhaled LABAs did not significantly reduce mortality or serious adverse events.
PICOs
Plain language summary
Long‐acting beta2‐agonists for people with COPD
We wanted to know whether twice‐daily treatment with an inhaled long‐acting beta2‐agonist was better than treatment with a dummy inhaler for people with chronic obstructive pulmonary disease (COPD).
Background to the review
COPD is a disease of the lungs that causes airways to narrow. As a result, people with COPD experience symptoms of breathlessness, cough and mucus buildup, which worsen over time. Cigarette smoking is the most common cause of COPD, and it is the fourth or fifth most common cause of death worldwide.
Inhaled salmeterol and formoterol, known as long‐acting beta2‐agonists (LABAs), are widely used to manage the symptoms of COPD, so it is important to understand their benefits and side effects. They are often introduced when inhaled treatments for quick relief from symptoms (e.g. salbutamol) are no longer helpful. LABAs are designed to be taken twice a day to control symptoms and reduce the likelihood of flare‐ups.
What did we find?
Twenty‐six studies (including 14,939 people with moderate to severe symptoms of COPD) compared twice‐daily salmeterol or formoterol with a dummy inhaler. The evidence gathered for this review is current up to June 2013. Results within studies were described most often after six months of treatment, but some were reported at three months and others after as long as three years. More men than women took part, and they had moderate to severe symptoms when they began treatment.
People who took LABA inhalers showed greater improvement on quality of life scales than those taking dummy inhalers, and they had fewer serious flare‐ups that resulted in a hospital stay (18 fewer per 1000). They also had better lung function than people who had taken placebo. LABA inhalers did not reduce the number of people who died, and no significant difference was noted in the number who had serious adverse events while taking the medication.
These studies were most often sponsored by drug companies and were generally well designed. People in the studies did not know which treatment they were getting, and neither did the people doing the research. Several studies did not describe flare‐ups, hospital stays or lung volume, so there is a chance that evidence obtained in future studies would change the strength of what has been concluded. Additionally, quite a lot of variation was noted between studies in the effects of LABA inhalers on quality of life, serious side effects and lung function. This may be explained in part by variation in study methods regarding what medications people could continue to take.
Authors' conclusions
Summary of findings
| Long‐acting beta2‐agonists compared with placebo for chronic obstructive pulmonary disease | ||||||
| Patient or population: people with chronic obstructive pulmonary disease Setting: community | ||||||
| Outcomes Follow‐up: weighted means presented for each outcome | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Placebo | Long‐acting beta2‐agonists | |||||
| Quality of life1 16 months | 45.3 | Mean score in the intervention groups was | MD ‐2.32 (‐3.09 to ‐1.54) | 11,397 | ⊕⊕⊕⊝ | A difference of 4 points is generally accepted to be of clinical significance |
| Severe exacerbations (hospitalisations) 7 months | 71 per 1000 | 53 per 1000 | OR 0.73 | 2859 | ⊕⊕⊕⊝ | I2 = 10%, P = 0.35 |
| Moderate exacerbations (course of antibiotics or oral steroids) 8 months | 238 per 1000 | 186 per 1000 | OR 0.73 | 3375 | ⊕⊕⊕⊝ | I2 = 8%, P = 0.37 |
| Severe/moderate exacerbations (hospitalisation or course of medication or ER visit) 8 months | 336 per 1000 | 308 per 1000 | OR 0.88 | 3968 | ⊕⊕⊕⊝ | I2 = 0%, P = 0.80 |
| Mortality (all‐cause) 14 months | 5 per 1000 | 5 per 1000 | OR 0.90 | 14,079 | ⊕⊕⊕⊝ | I2 = 21%, P = 0.21 |
| Participants with one or more serious adverse event (non‐fatal)1 15 months | 86 per 1000 | 84 per 1000 | OR 0.97 | 12,446 | ⊕⊕⊕⊝ | |
| Predose FEV1 (mL)1 7 months; higher is better | 1221 mL | Mean predose FEV1 in the intervention groups was 73 mL higher | MD 73 mL (48 to 98 mL) | 6125 | ⊕⊕⊝⊝ | Subgroup differences (I2 = 84%) discussed in high heterogeneity within subgroups, with potential baseline differences |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence. | ||||||
| n.b. Unless otherwise stated, significant subgroup differences were not found. 1 Quality of life, serious adverse events and FEV1 were analysed with random effects as the result of heterogeneity. 2 I2 = 50%, P < 0.01 (‐1 for inconsistency) 3 Several studies did not report exacerbations in a form that could be included in any of the three outcomes included in this review (‐1 for publication bias) 4 Confidence intervals include important benefit and potential harm (‐1 for imprecision) 5 I2 = 34%, P = 0.06; opposite direction of effect observed for the two formoterol doses (‐1 for inconsistency) 6 I2 = 71%, P < 0.01 (‐1 for inconsistency) | ||||||
Background
Description of the condition
Chronic obstructive pulmonary disease (COPD) is a respiratory disease characterised by chronic and progressive breathlessness, cough, sputum production and airflow obstruction, which leads to restricted activity and poor quality of life (GOLD 2013). The World Health Organization (WHO) has estimated that COPD, which includes emphysema, chronic bronchitis and small airways disease, is the fourth or fifth most common single cause of death worldwide, and that the treatment and management costs associated with COPD present a significant burden to public health. In the United Kingdom (UK), the annual cost of COPD for the National Health Service (NHS) is estimated to be £1.3 million per 100,000 people (NICE 2011). Furthermore, because of its slow onset and under‐recognition of the disease, it is heavily under‐diagnosed (GOLD 2013).
COPD comprises a combination of bronchitis and emphysema and involves chronic inflammation and structural changes in the lung. Cigarette smoking is the most important risk factor, but air pollution and occupational dust and chemicals are also recognised risk factors. COPD is a progressive disease that leads to decreased lung function over time, even with the best available care. No cure for COPD is known, although it is a preventable and treatable disease. As yet, apart from smoking cessation and non‐pharmacological treatments such as long‐term oxygen therapy in hypoxic patients, and pulmonary rehabilitation, no intervention has been shown to reduce mortality (GOLD 2013; Puhan 2011). Management of the disease is multifaceted and includes interventions for smoking cessation (van der Meer 2001), pharmacological treatments (GOLD 2013), education (Effing 2007), and pulmonary rehabilitation (Lacasse 2006; Puhan 2011). Pharmacological therapy is aimed at relieving symptoms, improving exercise tolerance and quality of life, slowing decline and even improving lung function, and preventing and treating exacerbations. COPD exacerbations impair patients' quality of life (GOLD 2013), and a large part of the economic burden of COPD is attributed to the cost of managing exacerbations, particularly those resulting in the use of acute care services, or hospitalisations (Hutchinson 2010). In the UK, one in eight emergency admissions to hospital is for COPD, which makes it the second largest cause of emergency admissions, and one of the most costly conditions treated by the NHS (NICE 2011). Appropriate pharmacological management of the disease is therefore important, particularly for reducing and preventing exacerbations.
Description of the intervention
Pharmacological management of COPD tends to begin with one treatment, and additional therapies are introduced as necessary to relieve symptoms and reduce the frequency and severity of exacerbations (ATS/ERS 2011; GOLD 2013). The first step often involves a short‐acting bronchodilator for control of breathlessness when needed: either a short‐acting beta2‐agonist (SABA), such as salbutamol, or the short‐acting muscarinic antagonist (SAMA) ipratropium. For persistent or worsening breathlessness associated with lung function decline, long‐acting bronchodilators may be introduced (ATS/ERS 2011; GOLD 2013). These comprise twice daily long‐acting beta2‐agonists (LABAs; duration of action 12 hours); once‐daily long‐acting beta2‐agonists (sometimes referred to as ultra long‐acting; duration of action 24 hours); and the once‐daily long‐acting anticholinergic agent tiotropium. Regular treatment with long‐acting bronchodilators is preferred over treatment with regular short‐acting bronchodilators on the basis of efficacy and side effects (Beeh 2010; GOLD 2013). Theophylline, an oral phosphodiesterase (PDE) inhibitor, is an alternative when bronchodilators are not available or affordable. However, theophylline is less effective and is less well tolerated than inhaled long‐acting bronchodilators. For patients with severe or very severe COPD (i.e. with forced expiratory volume in 1 second (FEV1) < 50% predicted) and with repeated exacerbations, GOLD 2013 recommends the addition of inhaled corticosteroids (ICS) to bronchodilator treatment. ICS are anti‐inflammatory drugs that are licenced as combination inhalers with LABAs. This group of patients with severe COPD may also benefit from treatment with the PDE4 inhibitor roflumilast, which may reduce the risk of exacerbations (GOLD 2013).
How the intervention might work
LABAs are widely used in the management of COPD, alone or in combination with other bronchodilators, ICS or both. Commonly used LABAs include twice‐daily salmeterol and formoterol and the new once‐daily preparation, indacaterol. All inhaled beta2‐agonists activate beta2‐receptors on different cells in the lung. Activation of the receptor on airway smooth muscle leads to a cascade of reactions that result in dilation of the airways. However, the exact mechanism of action differs between the various LABAs, and different efficacy and safety profiles can be expected between them. LABAs are commonly used to relieve symptoms and reduce exacerbations in stable COPD (Rodrigo 2008). Possible side effects of LABAs include cardiac effects such as arrhythmia and palpitations, muscle tremors, headache and metabolic imbalances such as hypokalaemia and increased glucose levels (Berger 2008).
Why it is important to do this review
LABAs are used widely and play a central role in the management of COPD (ATS/ERS 2011; GOLD 2013; NICE 2010). An earlier Cochrane systematic review has presented evidence on the effects of LABAs for patients with poorly reversible COPD (Appleton 2006). However, in this systematic review, we summarised the evidence regarding the efficacy and safety of twice‐daily LABAs (i.e. those with a 12‐hour duration of action) compared with placebo in all patients with COPD. The efficacy and safety of the once‐daily LABA indacaterol in comparison with other LABAs and placebo will be assessed in another Cochrane systematic review (Geake 2012).
This review forms part of a suite of reviews on long‐acting therapies for COPD, including long‐acting anticholinergics, long‐acting beta2‐agonists, ultra‐long‐acting beta2‐agonists, PDE4 inhibitors, and ICS (Cheyne 2012; Chong 2011; Chong 2012; Geake 2012; Karner 2011; Karner 2011a; Karner 2012; Karner 2012a; Karner 2012b; Nannini 2010; Nannini 2010a; Nannini 2012; Spencer 2011; Welsh 2011; Yang 2012). These reviews, which look at long‐term treatment for COPD, will ultimately be summarised in an overview.
Objectives
To assess the effects of twice‐daily long‐acting beta2‐agonists compared with placebo for patients with COPD on the basis of clinically important endpoints, primarily quality of life and COPD exacerbations.
Methods
Criteria for considering studies for this review
Types of studies
We included randomised controlled trials (RCTs) with a parallel‐group design, of at least 12 weeks' duration. We did not exclude studies on the basis of blinding. We excluded cross‐over trials, as we were looking at long‐term effects including adverse events.
Types of participants
We included RCTs that recruited participants with a clinical diagnosis of COPD based on the following (GOLD 2013).
-
Forced expiratory volume after one second (FEV1)/forced vital capacity (FVC) ratio < 0.7, which confirms the presence of persistent airflow limitation.
-
Several of the following key indicators:
-
Progressive and/or persistent dyspnoea (breathlessness);
-
Chronic cough;
-
Chronic sputum production; and
-
History of exposure to risk factors (tobacco smoke, smoke from home cooking and heating fuels, occupational dusts and chemicals).
-
We excluded RCTs in which participants had to have asthma as well as COPD to be included.
Types of interventions
We included studies in which participants were randomly assigned to receive the following.
-
Salmeterol 50 μg or placebo twice daily.
-
Formoterol 12 μg or placebo twice daily.
-
Formoterol 24 μg or placebo twice daily.
We included studies that allowed concomitant short‐acting bronchodilators, provided they were not part of the trial treatment under study. We did not include studies in which most participants were receiving other COPD treatments.
Types of outcome measures
Primary outcomes
-
Quality of life; mean difference and responders analysis (number of participants with clinically significant improvement or worsening); measured with a scale validated for COPD, such as St George's Respiratory Questionnaire (SGRQ) or the Chronic Respiratory Disease Questionnaire (CRQ).
-
Severe COPD exacerbations (COPD exacerbations leading to hospitalisation).
Secondary outcomes
-
Moderate COPD exacerbations.a
-
Mortality; all‐cause.
-
Non‐fatal serious adverse events; all‐cause.
-
Trough (predose) FEV1.
-
Withdrawals from study treatment.
aA sample definition of COPD exacerbation is "an acute event characterised by a worsening of the patient’s respiratory symptoms that is beyond normal day‐to‐day variations and leads to a change in medication" (GOLD 2013).
Search methods for identification of studies
Electronic searches
We identified trials from the Cochrane Airways Group Specialised Register of trials (CAGR), which is derived from systematic searches of bibliographic databases including the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, CINAHL, AMED and PsycINFO, and conference abstracts found through handsearching (see Appendix 1 for further details). We searched all records in the CAGR coded 'COPD' using the following terms:
(*formoterol or salmeterol or Serevent or Foradil or Oxis or LABA or long‐acting* or "long acting*")
We also searched ClinicalTrials.gov using search terms described in Appendix 2. We searched all databases up to June 2013, with no restriction on date or language of publication.
Searching other resources
We checked reference lists of all primary studies and review articles for additional references. We searched the manufacturers’ websites (GlaxoSmithKline and AstraZeneca) for additional information on studies identified through the electronic searches.
Data collection and analysis
Selection of studies
Two review authors independently screened the titles and abstracts of citations retrieved through literature searches and obtained in full text those deemed to be potentially relevant. We assigned each reference to a study identifier and assessed each against the inclusion criteria of this protocol (see Criteria for considering studies for this review). Any disagreements were resolved by consensus.
Data extraction and management
Two review authors independently extracted information from each study to record the following characteristics.
-
Design (design, total study duration, number of study centres and location).
-
Participants (number randomly assigned to each treatment, mean age, gender, baseline lung function, smoking history, reversibility, inclusion criteria and exclusion criteria).
-
Interventions (run‐in, intervention and control treatments including concentration and formulation).
-
Outcomes (definition of exacerbation and outcome data using end of study as time of analysis for all studies).
In the event that treatment arms of interest with different doses were included, these findings were combined when possible according to recommendations outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2009). If the data could not be combined, we used the trial arm with the dose offering greatest homogeneity with other trials regarding dose. We also subgrouped primary outcome data according to dose. Any discrepancies in the data were resolved by discussion, or by consultation with a third party when necessary.
Assessment of risk of bias in included studies
For the following items, two review authors independently assessed the risk of bias and graded each potential source of bias as high risk, low risk or unclear risk and resolved disagreements by consensus, according to recommendations outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2009).
-
Random sequence generation.
-
Allocation concealment.
-
Blinding of participants and personnel.
-
Blinding of outcome assessment.
-
Incomplete outcome data.
-
Selective outcome reporting.
Measures of treatment effect
Dichotomous data: We analysed dichotomous data variables (such as mortality and withdrawals) using Mantel‐Haenzsel odds ratios (ORs) with 95% confidence intervals (CIs). If events were rare, we employed the Peto odds ratio. If count data were not available as the number of participants experiencing an event but rather were reported as rate ratios, we transformed them into log rate ratios and analysed the data using generic inverse variance (GIV).
Continuous data: We analysed continuous outcome data as fixed‐effect mean differences (MDs) with 95% CIs unless excessive heterogeneity was found.
Unit of analysis issues
We analysed dichotomous data using participants as the unit of analysis. For continuous data, the MD based on change from baseline was preferred over the MD based on absolute values.
Dealing with missing data
If outcome data or key study characteristics were not reported in the primary publication, we searched clinical trial reports and contacted study authors and sponsors for additional information. We used intention‐to‐treat (ITT) analysis on outcomes from all randomly assigned participants when possible. We also considered, as part of the sensitivity analysis, the impact of the unknown status of participants who withdraw from the trials.
Assessment of heterogeneity
We assessed clinical and methodological heterogeneity by recording differences in study design and participant characteristics between individual studies. We assessed the extent of statistical variation among study results by using the I2 measurement.
Assessment of reporting biases
We tried to minimise reporting bias from non‐publication of studies or selective outcome reporting by using a broad search strategy, checking references of included studies and relevant systematic reviews and contacting study authors to request additional outcome data. We visually inspected funnel plots when 10 or more studies were included.
Data synthesis
For all outcomes, we analysed each LABA and dose separately; however for our primary outcomes, we also pooled the three comparisons. We analysed data using a fixed‐effect model, but when heterogeneity was noted (I2 > 30%), we used a random‐effects model and explored the heterogeneity (see Subgroup analysis and investigation of heterogeneity). We presented the findings of our primary outcomes in a 'Summary of findings' table generated with the use of GradePro software and according to recommendations provided in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2009).
Subgroup analysis and investigation of heterogeneity
When substantial heterogeneity was identified among the studies (I2 > 30%), we explored it by analysing the data by the following subgroups.
-
Duration of LABA therapy (≤ 1 year; > 1 year).
-
Disease severity at baseline (FEV1 < 50% predicted; FEV1 ≥ 50% predicted).
Sensitivity analysis
We assessed the robustness of our analyses by performing sensitivity analyses and systematically excluding the following studies from the overall analysis.
-
Those at high risk of bias.
-
Those with high and/or uneven withdrawal rates
-
Concurrent use of inhaled steroids*
*Studies where more than fifty percent of patients continued taking ICS or other COPD medications were excluded. However, included studies in which a significant proportion (but not the majority) of patients continued taking ICS or other COPD medications were removed in a sensitivity analysis.
Results
Description of studies
Results of the search
Nine hundred and seventy‐two references were identified through searches of online databases, and 63 additional references were found by searching other resources (drug company websites, reference lists of systematic reviews, clinicaltrials.gov). After 16 duplicates were removed, 802 of the remaining 1019 references were excluded by sifting titles and abstracts. The most common reasons for exclusion were that no LABA versus placebo comparison was performed (n = 359) and that the study used a cross‐over design (n = 184). Full reasons for exclusion are listed in Figure 1. Full texts were consulted for the remaining 217 references, and 191 were found to meet all of the inclusion criteria, representing 26 studies. The main reason for exclusion at this stage was that more than 50% of participants in the trials were taking concomitant COPD medications (n = 22). Full details of the search history can be found in Appendix 1 and Appendix 2.
Study flow diagram.
Included studies
Twenty‐six studies met the inclusion criteria, randomly assigning 14,939 people with a diagnosis of COPD to LABA or placebo. All but one of the trials (Watkins 2002) contributed data to at least one analysis. Calverley 2007 [TORCH] contributed the greatest number of people to the analyses, with 3087 people randomly assigned to the two groups of interest. A small industry‐funded trial, SLMF4010 2005 included the smallest number of people, with 17 people randomly assigned to each group.
Design and duration
All twenty‐six studies were randomised, double‐blind, parallel‐group controlled trials. Nine trials lasted for three months, 10 for six months and six for a year.The remaining trial (Calverley 2007 [TORCH]) was a three‐year study. In accordance with the protocol, separate analyses of the three included LABA doses were subgrouped according to length of trial (three, six, 12 and 36 months). Distribution of studies among the four duration categories and descriptive statistics of age and sex are given in Table 1.
| Trial length | Study IDs | Mean age, years, | Male, %, |
| 3 months | Aalbers 2002; Bogdan 2011; Dahl 2001; Hanrahan 2008; | 63.5 (62 to 67) | 66.5 (52 to 87) |
| 6 months | Brusasco 2003; Campbell 2005; Doherty 2012; Hanania 2003; | 63 (59 to 65) | 75 (63 to 88) |
| 12 months | Calverley 2003a; Calverley 2003b [TRISTAN]; Dahl 2010; | 63.5 (63 to 64) | 77.5 (75 to 80) |
| 36 months | 65 (N/A) | 76 |
Participant inclusion and exclusion criteria
Full details of the inclusion and exclusion criteria for each trial can be found in Characteristics of included studies. Inclusion and exclusion criteria were largely similar across trials. Participants were required to be over the age of 40 (35 in three studies) and to have a smoking history of at least 10 pack‐years.
Baseline characteristics of participants
Full details of the baseline characteristics of participants in each study can be found in Characteristics of included studies. Participants' mean age was similar across trials, ranging from 58.8 to 67.2 years in individual trial arms. Trial participants were most often male (range 50% to 94%, median 75%) and Caucasian (range 46% to 100%, median 92%). Seventeen trials reported mean or median smoking pack‐years; medians ranged from 40 to 60, and means ranged from 35.4 to 52.5, pack‐years. Twenty studies reporting percentage predicted FEV1 indicated that the populations were of moderate to high disease severity, with baseline means ranging from 32.6% to 54.7%. Distribution of mean percentage predicted FEV1 at baseline within studies and mean baseline SGRQ (when reported) are presented in Table 2.
| FEV1 % predicted at baseline | Study IDs | Mean baseline SGRQ, |
| 30% ≤ mean < 40% | Brusasco 2003; Calverley 2003a; Doherty 2012; | 50 (47 to 53) |
| 40% ≤ mean < 50% | Calverley 2003b [TRISTAN]; Calverley 2007 [TORCH]; | 50 (48 to 55) |
| 50% ≤ mean < 60% | Aalbers 2002; Bogdan 2011; Campbell 2005; | 45 (44 to 49) |
| Five studies did not report mean % predicted FEV1 at baseline; 10 of the studies presented here did not | ||
Descriptive statistics were analysed separately for the formoterol and salmeterol studies to check for systematic differences between trials of drugs made by different manufacturers. Percentage reversibility was reported in only 12 studies (46%), of which only two were studying salmeterol. Within the ten formoterol studies (both doses), reversibility reported for each arm ranged from 4.7% to 19.5%, with a mean value of 8.8. For the salmeterol studies, the mean was 7.9 and ranged from 4 to 13 in individual trial arms. Mean percentage predicted FEV1 was better reported (77% of studies), and no differences in baseline severity were apparent between formoterol and salmeterol; the formoterol studies had a mean of 44.7, ranging from 32.6 to 54.7, and the salmeterol studies a mean of 43.0, ranging from 37.7 to 53. Given the relative homogeneity of these metrics between subgroups and between the individual trials, it was not deemed appropriate to subgroup results on the basis of severity.
Characteristics of the interventions
Fourteen studies included the comparison of formoterol 12 μg twice daily (or metered dose equivalent) and placebo. Three of these studies (Aalbers 2002; Dahl 2001; Rossi 2002) also included a formoterol 24 μg twice‐daily arm. One additional study, Wadbo 2002, used only the comparison of formoterol 24 μg. Most formoterol studies used matching dry powder Turbuhaler devices to deliver blinded study medication. The remaining 11 studies compared salmeterol 50 μg twice daily (or equivalent) with placebo, primarily using blinded Diskus dry powder inhalers. Studies allowed the use of a short‐acting beta2‐agonist as reliever medication during the study period (salbutamol or terbutaline).
Outcomes and analysis structure
Health‐related quality of life measured on the St George's Respiratory Questionnaire (SGRQ) was reported in most of the studies. The SGRQ is a well‐validated and widely used measure of health status in patients with chronic airflow limitation; the total score ranges from 0 (perfect health) to 100 (most severe status) and includes three components of symptoms, activity and impacts. Only three of these studies (Bogdan 2011; Brusasco 2003; Kornmann 2011) reported the number of people improving by four or more points on the SGRQ, which is generally accepted to be the minimal clinically important difference. In two studies (Doherty 2012 and Szafranski 2003), variance was not given for the comparison of interest. In these cases, standard deviations of 14.5 were imputed on the basis of population variance and on that of other arms within the studies to increase the number of studies in the analysis. Four additional studies reported quality of life data using the Chronic Respiratory Diseases Questionnaire (CRQ) (Hanania 2003; Mahler 1999; Mahler 2002; Rennard 2001), of which two reported mean change with no measure of variance. All four were analysed separately from the main SGRQ analysis to reduce possible sources of heterogeneity, with missing variances imputed from those reported in Mahler 1999 and Rennard 2001. Because these four studies all compared salmeterol with placebo and were of three months' and six months' duration, results for the CRQ were not part of the subgroup analyses. Therefore, the quality of life data could be presented in a more clinically meaningful format by using mean differences on the SGRQ scale. For this outcome and predose FEV1, mean change and endpoint data were analysed together, and footnotes were entered to clarify which data were available.
The second primary outcome, severe COPD exacerbations requiring hospitalisation, was quite poorly reported. Seven studies (nine comparisons) reported this as an outcome, of which five also reported the number of people with one or more moderate COPD exacerbations (those requiring a course of antibiotics or oral steroids). Rennard 2001 and Calverley 2003b [TRISTAN] reported data that could be included in the moderate exacerbations analyses but not hospitalisations. Two additional studies, Mahler 1999 and Nelson 2007, did not explicitly define exacerbations but were included in the moderate exacerbations analysis because rates were more consistent with these rates in other studies. Data for hospitalisations are collated in Analysis 1.4 and for moderate exacerbations in Analysis 1.6.
Seven additional studies reported the overall number of people who had either one or more moderate or severe exacerbations during the study, and these data are presented separately in Analysis 1.5. Overall, 18 studies reported exacerbation count data that could be included in at least one of these three analyses. Four studies presented no data related to exacerbations (Aalbers 2002; Kornmann 2011; Wadbo 2002; Watkins 2002), and four further studies reported data as rate ratios or yearly patient rates that could not be incorporated with data from the other studies (Calverley 2007 [TORCH]; SLMF4010 2005; Szafranski 2003; Tashkin 2008 [SHINE]).
Mortality was well reported and was missing only in Aalbers 2002, Wadbo 2002 and Watkins 2002. Watkins 2002 did not contribute data to any analysis. Calverley 2007 [TORCH], the largest and longest study, was unique in tracking the status of all participants at endpoint, regardless of how long they stayed in the study. For this reason, the nature and quality of mortality data from this study are likely to differ from those of other studies in the analysis. The number of participants experiencing one or more serious adverse events was reported in slightly fewer studies. Within the six studies that could not be included, one reported only the number of serious cardiac adverse events, and the others reported overall serious and non‐serious events combined.
Lung function as measured by predose FEV1 was not consistently reported. Half of the studies reported data that could be used in meta‐analysis as mean change or endpoint scores with corresponding variance. Two studies reported absolute means at endpoint, and all other studies reported mean change from baseline. Two studies reporting change did not report a measure of variance (Calverley 2003a; Tashkin 2012), so standard deviations were imputed by calculating the mean of the change variances from the remaining nine studies.The remaining studies either reported other lung function measures that could not be analysed with predose FEV1 (e.g. peak expiratory flow, postbronchodilator FEV1) or did not present the data in a format that could be incorporated in the meta‐analysis. Rennard 2001 aimed to detect differences of effect between participants with high or low reversibility, and so the merged results may represent a heterogeneous population that is different from that seen in other studies.
Three trials—Aalbers 2002, Dahl 2001 and Rossi 2002—included two LABA arms that met inclusion criteria for the review, and in these cases, all three arms were included and the placebo arm of each was split in comparison 1. In these cases, the power of the overall analysis is accurate, but the power to detect subgroup differences may be reduced. In comparisons 2, 3 and 4, where LABA doses are split into different analyses, no adjustments were made to the participant totals. When studies reported data at more than one time point of interest, we reported only the duration subgroup totals—not the pooled effect (i.e. data for the same participants were not double counted, but power to detect subgroup differences may be increased). For this reason, slight discrepancies may be evident between a subgroup result in comparison 1 and the corresponding effect in the following comparisons subgrouped by trial duration.
Sensitivity analyses
Use of inhaled corticosteroids (ICS) during the study period
As stated in the protocol, we intended to exclude trials in which more than 50% of participants were taking other COPD medications. Most available data in the study reports related to ICS use during the randomised period, and this was used as a proxy in cases where the preferred data were not available. Ten studies required that other medications, including ICS, were stopped during the run‐in. Eight further studies did not explicitly report the proportion of participants taking other medications, and inferences based on pharmacological exclusion criteria could not be made. Hanrahan 2008 reported that 28.6% and 23.9% of participants in the salmeterol and placebo groups, respectively, were taking regular ICS during the study period, and so the study was included. Similarly, Nelson 2007 reported that 33% and 38% of participants in the formoterol and placebo groups were taking COPD medications during the study period that were not part of the randomised interventions. Because ICS use fell just above or just below the predefined 50% threshold in the remaining six studies (Campbell 2005; Dahl 2001; Dahl 2010; Kornmann 2011; Rossi 2002; Vogelmeier 2008), these studies were removed from the primary outcomes in a sensitivity analysis.
Studies at high risk of bias/those with high or uneven withdrawal rates
Because no studies were rated at high risk of bias for either of the selection bias parameters, or for detection or performance bias, a sensitivity analysis was performed to remove the eight studies that were rated at high risk of bias due to attrition (Dahl 2010; Hanania 2003; Mahler 1999; Mahler 2002; Nelson 2007; Rennard 2009; SLMF4010 2005; Szafranski 2003). Given that high risk judgements in the reporting bias parameter usually were made because studies were already missing from analyses, it was not appropriate to perform a sensitivity analysis on selective outcome reporting. Reporting bias is reflected in the grade ratings of the affected outcomes.
Excluded studies
Details of why studies were excluded from the review, including exact percentages of participants taking additional COPD medications, can be found in Characteristics of excluded studies. Six studies (22 citations) were excluded because a clear majority of the participants were taking other COPD medications (Boyd 1995; Celli 2003; Chapman 2002; Dal Negro 2003; Rutten‐van Molken 1999; Stockley 2006). One additional trial (two publications) was excluded because the inclusion criteria allowed for participants with asthma or COPD, and data for diagnostic subgroups were not presented (Steffensen 1996).
Risk of bias in included studies
The methodological quality of the included trials was good. None of the studies were at high risk of selection, performance or detection bias, but there was high risk of bias for attrition in eight studies and selective reporting of outcomes in 11 studies. More detailed descriptions by domain (allocation generation, allocation concealment, blinding and incomplete data) are given below. Details of the risk of bias rating for each study and the reasons for each rating can be found in Characteristics of included studies and a summary of judgements by study and domain can be found in Figure 2. When studies funded by the same company were rated as unclear for one or more domain, we attempted to clarify study methods with the funder.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Allocation
None of the included studies were rated as having high risk of bias for either of the two allocation parameters (random sequence generation and allocation concealment).
With the exception of Watkins 2002 (which was not a full trial report and hence was rated as unclear), all of the trials were rated as having low risk of bias for random sequence generation. Although several of the trials did not adequately describe the methods of sequence generation used, the authors agreed that this was likely due to variations in reporting standards. All of the studies were industry sponsored; therefore the authors deemed it reasonable to assume that standardised drug company methods were used (i.e. computerised random list generators).
Nine studies (Aalbers 2002; Brusasco 2003; Calverley 2003b [TRISTAN]; Calverley 2007 [TORCH]Campbell 2005; Dahl 2010; Kornmann 2011; Rennard 2009; Tashkin 2012) fully described methods used for allocation concealment and were thus rated as having low risk of bias. Although some drug companies might have standard protocols for concealing allocation, the authors agreed that studies would be rated as unclear unless the methods used were properly described for this parameter. The remaining 17 studies were rated as unclear for this reason.
Blinding
None of the studies were rated as having high risk of bias for either of the two blinding parameters (participants and personnel, and outcome assessment).
When a study was described as double‐blind and no specific details were reported in the original report, trial registrations most often confirmed that the blind applied to participants and investigators. As a result, all trials were rated as having low risk of bias for this parameter. Twelve studies specifically described double‐dummy procedures and matched inhalers.
Studies were not rated as having low risk of bias for blinding of outcome assessors unless blinding was explicitly stated in the trial report or the protocol registration. This was the case for only nine of the studies; therefore most were rated as unclear.
Incomplete outcome data
More than half of the included studies were rated as having low risk of bias for this parameter (N = 16), either because dropout was deemed to be low and even between groups, or because dropout was considered acceptable given the methods of imputation described in the report. Two studies were rated as unclear: Calverley 2003b [TRISTAN] had high and even dropout in both arms but did not sufficiently describe the method of imputation to warrant a 'low' rating, and Watkins 2002 provided no information related to attrition. The remaining eight studies were considered to be at high risk of bias, either because dropout was very high in both groups, or because dropout was deemed excessive or uneven given the method of imputation or the analysis method used (i.e. per protocol or completers only).
Selective reporting
Twenty studies could be linked with their protocol registration on clinicaltrials.gov or with their industry report, allowing comparison of prespecified outcomes with published or unpublished results. Of these, ten reported all stated outcomes in a way that allowed them to be combined in meta‐analysis (Bogdan 2011; Calverley 2003a; Calverley 2003b [TRISTAN]; Calverley 2007 [TORCH]; Dahl 2010; Hanania 2003; Mahler 2002; Rennard 2001; SLMF4010 2005; Tashkin 2008 [SHINE]). Five studies that could not be linked to their trial registration documents but did not show clear evidence of selective outcome reporting were rated as unclear (Campbell 2005; Dahl 2001; Rossi 2002; Szafranski 2003; Tashkin 2012). Eleven studies were rated as having high risk of bias either because outcomes that were stated in the protocol were not reported in sufficient detail in the results, or because key outcomes expected in COPD trials were not included, regardless of whether they were named in a protocol (e.g. mortality, adverse events, exacerbations).
Funnel plots were constructed for the primary outcomes. Figure 3 is not suggestive of any serious publication bias of the SGRQ. However, some evidence suggests a small study effect on severe exacerbations (those resulting in hospitalisation), which might be due to publication bias, as shown in Figure 4.
![Funnel plot of comparison: 1 All LABA versus placebo [subgrouped by drug], outcome: 1.1 Quality of life (SGRQ total score).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-AFig-FIG03.png)
Funnel plot of comparison: 1 All LABA versus placebo [subgrouped by drug], outcome: 1.1 Quality of life (SGRQ total score).
![Funnel plot of comparison: 1 All LABA versus placebo [subgrouped by drug], outcome: 1.4 Severe exacerbations (hospitalisations).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-AFig-FIG04.png)
Funnel plot of comparison: 1 All LABA versus placebo [subgrouped by drug], outcome: 1.4 Severe exacerbations (hospitalisations).
Other potential sources of bias
No other sources of bias were identified in the included studies.
Effects of interventions
Full details of the analyses and their GRADE ratings can be found in Data and analyses and summary of findings Table for the main comparison. The first comparison presents results for all outcomes, with studies subgrouped according to the type and dose of LABA. Comparisons 2, 3 and 4 show results for formoterol 12 μg, formoterol 24 μg and salmeterol 50 μg, respectively (all twice daily), subgrouped according to study duration. The final comparison presents data for the two primary outcomes, excluding six studies in which relatively large proportions of the population were taking regular COPD therapies other than the study medication (see 'Use of inhaled corticosteroids (ICS) during the study period' in Included studies).
Primary outcomes
Health‐related quality of life
Improvement on the St George's Respiratory Questionnaire (SGRQ) was greater with LABA therapy than with placebo (MD ‐2.32, 95% CI ‐3.09 to ‐1.54; I2 = 50%, P = 0.007; Analysis 1.1), based on data from 11,397 people in 17 studies. Results were analysed using a random‐effects model because heterogeneity was high, and the outcome was downgraded from high to moderate quality for this reason. Heterogeneity could not be explained by differences between the effects of LABA drug and dose (test for subgroup difference, I2 = 0%, P = 0.45). The difference between LABA and placebo was significant for all three LABA agents separately, but important heterogeneity was noted within the formoterol 12 μg and salmeterol 50 μg subgroups (I2 = 57% and 48%, respectively).
More people taking a LABA showed clinically important improvement of at least four points on the SGRQ (OR 1.58, 95% CI 1.32 to 1.90; I2 = 86%, P = 0.0007; Analysis 1.2) based on 1871 people in three studies. Data were insufficient to allow conclusions regarding the difference between formoterol and salmeterol.
Four salmeterol studies reporting the Chronic Respiratory Disease Questionnaire (CRQ) were analysed separately and showed a significant benefit of LABA therapy (MD 3.10, 95% CI 1.22 to 4.98; I2 = 0%, P = 0.58; Analysis 1.3) with no heterogeneity. Because so few trials reported the CRQ and the number of people achieving a minimally clinically important difference (MCID) on the SGRQ of 4 units, we did subgroup and sensitivity analyses on the SGRQ continuous data only.
A test for subgroup differences suggested that some of the heterogeneity within the 12 trials which randomised people to either formoterol or placebo may be explained by differences in study duration (I2 = 75%, P = 0.02; Analysis 2.1). The effect of twice‐daily formoterol 24 μg was significant when analysed separately but lacked precision because only three studies compared the dose with placebo (MD ‐2.32, 95% CI ‐4.52 to ‐0.13; I2 = 0%, P = 0.76; Analysis 3.1). No difference was detected between the two three‐month trials (Dahl 2001; Wadbo 2002) and the year‐long Rossi 2002 study. With so few studies, no heterogeneity was evident between or within subgroups. Unlike the formoterol 12 μg analysis, trial duration did not appear to be an important source of heterogeneity in the salmeterol included trials (I2 = 0%, P = 0.41; Analysis 4.1).
The overall pooled effect and drug subgroup effects were largely consistent with the original analysis after removal of five studies identified as having a high proportion of ICS use during the trial (see Included studies—Use of inhaled corticosteroids (ICS) during the study period). The magnitude of improvement due to LABA therapy was reduced after the studies were removed, but somewhat tighter confidence intervals and less between‐study heterogeneity were noted (MD ‐1.53, 95% CI ‐2.08 to ‐0.98; I2 = 34%, P = 0.12; Analysis 5.1).
Four of the eight studies rated as having high risk of bias reported the SGRQ: three from the formoterol 12 subgroup and one from the salmeterol subgroup. After these studies were removed, the pooled difference between LABA and placebo remained significant but was again reduced (MD ‐1.60, 95% CI ‐2.15 to ‐1.05; I2 = 42%, P = 0.04; Analysis 6.1).
Severe COPD exacerbations (requiring hospitalisation)
When data from seven studies were combined (nine comparisons), the number of people hospitalised for COPD exacerbations was lower among LABA‐treated participants than in those receiving placebo (OR 0.73, 95% CI 0.56 to 0.95; Analysis 1.4). Although little evidence was found for explicit selective outcome reporting within trials, the outcome was downgraded for publication bias and was rated as having moderate quality; more than half of the studies did not report the outcome of high interest to patients and clinicians, and the estimate might have been different if all studies could have been included in the analysis (see Included studies—Outcomes and analysis structure). No heterogeneity was noted between trials, and the results of a test for subgroup differences between drugs/doses were not significant (I2 = 57.5%, P = 0.10).
No evidence of a statistically significant effect of trial duration was observed in the formoterol 12 μg trials (six studies across three duration subgroups). Similarly, no significant difference was observed between the three‐month Dahl 2001 study and the 12‐month Rossi 2002 study comparing formoterol 24 μg with placebo. Only one six‐month study comparing salmeterol and placebo reported hospitalisation rates, so no assumptions could be made regarding relative effects of trial duration.
The pooled effect of LABA versus placebo decreased in magnitude and was no longer significant after four studies with a high percentage of participants taking inhaled corticosteroids (six comparisons) were removed. This effect was based on only three remaining trials (OR 0.85, 95% CI 0.62 to 1.16; Analysis 5.2), and the change in effect could be explained by another variable. None of the eight studies rated as having high risk of bias reported the number of participants with exacerbations requiring hospitalisation, so the outcome does not appear in the sensitivity analysis.
Secondary outcomes
Moderate COPD exacerbations (requiring a course of antibiotics and/or oral steroids)
LABA treatment reduced the number of people who had one or more exacerbations requiring a course of antibiotics, oral steroids or both compared with placebo (OR 0.73, 95% CI 0.61 to 0.87; I2 = 8%, P = 0.37; Analysis 1.6), with minimal heterogeneity. The quantity of data for the outcome was similar to that provided for severe COPD exacerbations, so the outcome was downgraded for publication bias for the same reasons and was rated as having moderate quality. No evidence suggested important subgroup differences on the basis of which LABA/dose was used (I2 = 0%, P = 0.68). The three studies that compared formoterol 12 μg with placebo showed no significant benefit (OR 0.78, 95% CI 0.56 to 1.07; I2 = 7%, P = 0.34; Analysis 2.3) with little heterogeneity, and no evidence was found of differential effectiveness related to study duration. The higher‐dose formoterol 24 μg did show a significant reduction relative to placebo (OR 0.57, 95% CI 0.38 to 0.88; Analysis 3.3), although this was based on a total of 124 events in one study (Rossi 2002). Four studies that compared salmeterol with placebo showed an overall reduction in moderate exacerbations with the study drug (OR 0.75, 95% CI 0.60 to 0.94; I2 = 40%, P = 0.17; Analysis 4.3), with some non‐significant heterogeneity and no observed differences related to study duration.
Moderate/severe COPD exacerbations (requiring hospitalisation or a course of antibiotics/oral steroids or ER visit)
Around half of the studies did not report rates of moderate and severe COPD exacerbations as separate outcomes, as we had defined in the protocol, but rather reported a composite rate, which could not be combined with the other data; therefore we decided to present these data separately in the review. As with the other two exacerbation outcomes, the outcome was rated of moderate quality after it was downgraded for publication bias.
LABA treatment did not significantly reduce the number of people with moderate or severe exacerbations compared with placebo (OR 0.88, 95% CI 0.76 to 1.02; I2 = 0%, P = 0.80; Analysis 1.5), as determined on the basis of seven studies with 1142 participants. No between‐study or between‐drug heterogeneity was observed. Since the outcome was added, post hoc and subgroup analyses have been reported for the other two exacerbation outcomes, and we did not perform separate subgroup analyses of trial duration for this outcome.
Mortality; all‐cause
Study deaths were relatively uncommon; therefore the analyses were conducted using Peto odds ratio, as this method does not require adjustment for zero cells. LABA treatment did not significantly reduce mortality compared with placebo (OR 0.90, 95% 0.75 to 1.08; I2 = 21%, P = 0.21; Analysis 1.7), and a degree of between‐study heterogeneity was noted. The outcome was downgraded for imprecision and was rated as having moderate quality because the confidence intervals for the pooled effect included important benefit and potential harm. A test for subgroup differences between drugs was not statistically significant (I2 = 55%, P = 0.11). Results of a test for subgroup differences related to the duration of the formoterol trials were not significant (Analysis 2.4). In the two formoterol 24 μg trials that reported all‐cause mortality, only one death occurred in the LABA group of Rossi 2002; the confidence intervals were too wide to allow interpretation of direction or magnitude of effects (Analysis 3.4) or assumptions based on trial length. Four studies with no events did not contribute to the salmeterol analysis (Analysis 4.4), and the effect was largely influenced by the large Calverley 2007 [TORCH] study (93% of total weight). This large study was unique because it was much longer than the other trials (at three years), and investigators logged mortality for all participants, regardless of how long they stayed in the study. Between‐trial heterogeneity was not significant in the salmeterol analysis (I2 = 29%, P = 0.22); a test for subgroup differences regarding trial duration also was not significant (I2 = 0%).
Non‐fatal serious adverse events; all‐cause
All studies included in the analysis reported this outcome with participants as the level of analysis (i.e. number of people who had serious adverse events as opposed to the number of adverse events in total). When findings of all studies were pooled, no difference was observed between LABA and placebo (OR 0.97, 95% CI 0.83 to 1.14; I2 = 34%, P = 0.06; Analysis 1.8). Heterogeneity was significant at P = 0.1, and unexplained differences between the two formoterol doses showed opposite directions of effect. A test for subgroup differences indicated that some of the heterogeneity may be explained by these differences in individual drugs/doses, which were significant (I2 = 83%, P = 0.002). It is unclear whether formoterol 12 μg significantly increases rates of serious adverse events, as the confidence interval touched the line of no effect (OR 1.20, 95% CI 1.00 to 1.43; I2 = 14%, P = 0.32; Analysis 2.5), and differences in trial duration were not statistically significant. Three formoterol 24 μg studies showed that serious adverse events were lowered by LABA use (OR 0.53, 95% CI 0.36 to 0.79; I2 = 0%, P = 0.67; Analysis 3.5) with no significant heterogeneity, although confidence intervals were quite wide. Studies that compared salmeterol 50 μg with placebo showed no significant differences between groups (OR 0.94; 95% CI 0.83 to 1.06; I2 = 13%, P = 0.33; Analysis 4.5) based on nine studies with 1608 events. Some statistically insignificant heterogeneity between trial results was noted and could not be explained by reliable differences in trial duration.
Predose forced expiratory volume in one second (trough FEV1)
The predose FEV1 of participants taking LABA was 73 mL higher at the end of the trials than that of participants taking placebo inhalers (95% CI 48 to 98; I2 = 71%, P < 0.0001; Analysis 1.9); this finding was based on data from 6125 participants in thirteen studies that reported the outcome. A large degree of heterogeneity was noted, so the analysis was downgraded for inconsistency. Half of the studies did not report the outcome or reported the outcome in a way that could not be entered in meta‐analysis; therefore it was downgraded for publication bias and was rated as having low quality. A test for subgroup differences suggested that the heterogeneity may be accounted for by differences between formoterol 12 μg and salmeterol 50 μg (no formoterol 24 studies reported trough FEV1) (I2 = 84%, P = 0.01). Individually, formoterol 12 μg (MD 45 mL, 95% CI 29 to 60) and salmeterol 50 μg (MD 101 mL, 95% CI 60 to 142) were associated with improved predose FEV1 relative to placebo.
No heterogeneity was observed between the formoterol 12 μg studies (I2 = 0%, P = 0.57), and no observable differences related to trial duration were reported. In the salmeterol 50 μg studies, heterogeneity was substantial (I2 = 69%, P = 0.003), and a test for subgroup differences of study duration suggested that the benefits of LABA treatment over placebo become less distinct over time (I2 = 87%, P = 0.0005; Analysis 4.6). Some of the heterogeneity may have been introduced by within‐subgroup variation in the recruited populations. For example, Rennard 2001 split the population by high and low reversibility to compare the effects of salmeterol on different participant groups, and the data entered into the analysis represent the population as a whole.
Withdrawal from study treatment
Withdrawal rates were higher for placebo than for LABA treatment (OR 0.74, 95% CI 0.69 to 0.80; I2 = 0%, P = 0.75; Analysis 1.10), as determined on the basis of data from all studies except Watkins 2002. The formoterol 24 μg studies contributed the least data to the analysis and showed no difference between LABA and placebo, although subgroup differences between drugs and doses were not significant (I2 = 0%, P = 0.66). No heterogeneity between trials was noted, and no evidence showed significant subgroup differences between dose categories. No clear effects of trial duration were evident.
Discussion
Summary of main results
Twenty‐six studies including nearly 15,000 people with moderate to severe COPD were included in the review. Participants in the studies were more often male with mean baseline FEV1 between 33% and 55% predicted normal and mean SGRQ ranging from 44 to 55.
LABA treatment significantly improved quality of life and reduced hospitalisations relative to placebo, although unexplained variation was noted within the quality of life data. Exacerbations were not consistently defined in trials, but across three definitions (those leading to hospitalisation, requiring a course of antibiotics or steroids or either), good evidence suggested that LABA therapy was effective. In terms of adverse events, no significant difference was observed in rates of mortality or serious adverse events between LABA and placebo.
Most subgroup differences between drugs and between dose groups were not significant. Significant differences in serious adverse events data were complicated by the unexplained disparity of effect direction for the two formoterol doses, and the difference in FEV1 between formoterol and salmeterol is complicated by substantial heterogeneity within the salmeterol studies for that outcome. Systematic differences between drug company methodology and recruitment procedures may preclude comparisons between salmeterol and formoterol.
Participants were more likely to withdraw from placebo than from LABA therapy (OR 0.74, 95% CI 0.69 to 0.80; I2 = 0%), although sensitivity analyses removing studies at highest risk of bias for this reason did not change conclusions for the primary outcomes.
Overall completeness and applicability of evidence
The current review expanded and updated a previous Cochrane review that looked only at patients with poorly reversible COPD (Appleton 2006). Because the inclusion criteria were widened to include all patients with a diagnosis of COPD, findings are likely to apply to a larger group of patients for whom LABA therapy is indicated. Trials were largely consistent in their inclusion of moderately to severely affected participants, as confirmed through smoking and medication history and spirometric indices.
Because of variation in study protocols regarding the use of other COPD medications, it is difficult to accurately judge to which groups of patients the evidence accurately applies. However, the sensitivity analysis removing studies in which a large proportion of participants were taking other COPD medications goes some way to dispel this uncertainty, given that results were very similar to those of the original pooled analysis. Similarly, systematic differences may be noted between trials of salmeterol and those of formoterol because different manufacturers conducted most of the trials for these two drugs. However, analysis of descriptive statistics for the baseline measurements showed that the populations recruited to trials of salmeterol were similar to those enrolled in the formoterol trials (percentage reversibility and percentage predicted FEV1). Outlier studies (those recruiting unusually severe or mild participants) may still have introduced heterogeneity within analyses, and differences may have been masked by inconsistent reporting of severity metrics across the data set and the lack of individual participant data. Additionally, significantly higher withdrawal in the placebo groups is likely to be due primarily to lack of efficacy leading to protocol violations (Calverley 2003). This may have reduced the effectiveness of randomisation and decreased the magnitude of difference between groups by removing from the analysis participants with the most severe conditions who had been assigned to placebo.
Only three trials tested the higher dose of formoterol (24 μg twice daily) against placebo, of which two also included an arm that received the more commonly prescribed 12 μg dose. The higher dose is the maximum indicated by the British National Formulary (BNF) for additional symptom relief (BNF 2009), and, in this sense, the tendency for trials to use the lower dose is consistent with licensing and practice. Similarly, the use of 12 μg twice daily is consistent and hence comparable with the dose received through combination preparations of budesonide and formoterol. In addition, little variation in inhaler devices was seen within the formoterol or salmeterol studies, with most formoterol studies using masked Turbuhaler devices, and most salmeterol trials using the Diskus. However, fact that inhaler devices differed fairly systematically between the two drugs may have contributed to subgroup differences.
It is unclear whether unpublished trials are missing from the review or, in some cases, whether all data are reported in the available papers and unpublished industry summaries. Attempts to ascertain with GlaxoSmithKline, AstraZeneca and Novartis whether all conducted studies and measures within them were included were not fruitful in time for publication.
Quality of the evidence
The methodological quality of the included trials was good, and the single conference abstract of unknown quality did not contribute any data to the analysis. All studies were double‐blind, and so results are unlikely to be compromised by detection or performance biases. Similarly, most studies controlled adequately for selection bias or were presumed to do so, in accordance with industry protocol. Although evidence suggested bias from high and unbalanced attrition in some trials and significantly higher withdrawal in the placebo arm when analysed, sensitivity analyses showed that conclusions did not change when studies at highest risk of bias were removed from the primary outcomes. It is possible that higher attrition in the placebo group across many of the trials could have diluted the true difference between LABA and placebo for some outcomes; if we accept that people who drop out of studies have a less positive outcome than those who do not, the placebo estimate could appear more favourable than if everyone had contributed endpoint data, depending on the method used to impute values for missing data points (i.e. last observation carried forward or other imputation models).
As described above, some evidence of selective reporting bias was found, both within trials that failed to report key outcomes or those stated in the prospectively registered protocols, and in terms of the possibility that trials (industry funded or independent) may remain unpublished (Song 2010). Several studies did not report exacerbations or trough FEV1 in a way that could be included in the review, so additional evidence obtained in future studies might change our confidence in these results.
Most studies were sponsored by drug companies, and they generally were of good quality. A lot of variation between studies was noted in the effect of LABA inhalers on quality of life, serious side effects and lung function. Although studies with around half of people taking additional medications were removed in a sensitivity analysis, the variation in allowed co‐medications and the numbers taking them may explain some of the variation in reported findings.
Potential biases in the review process
Review authors made every effort to identify all relevant published and unpublished studies by using additional methods to catch anything that might not have been found in the main electronic search (e.g. searching drug company databases and clinical trial registration sites, checking reference lists). However, attempts to obtain data directly from drug companies were not successful, and we did not routinely contact individual trial authors for additional data unless outcomes were clearly selectively reported. Unpublished studies (industry funded or otherwise) may exist that might change our confidence in the conclusions. All authors adhered to the most recent best practice guidelines in terms of study selection, resolution of disagreements, data extraction and analysis to reduce bias and errors.
Agreements and disagreements with other studies or reviews
Five previous reviews were identified that specifically analysed LABA versus placebo. Results of this review were consistent with results for salmeterol 50 μg as reported in the previous Cochrane review specifically looking at patients with poorly reversible COPD (Appleton 2006). Specifically, benefit was associated with twice‐daily salmeterol 50 μg in terms of FEV1, quality of life and exacerbations. This review found that formoterol 12 μg also improves these outcomes—a fact that was previously unclear. The benefits of formoterol 24 μg remain unclear, and salmeterol 100 μg was not included in the current review, as this dose is not recommended in current clinical practice.
Kliber 2010 concluded that mortality was not significantly reduced by LABA therapy, and this conclusion was consistent with this review, despite some differences in trial inclusion criteria. Similarly, Rodrigo 2008 found comparable benefits of LABA on exacerbation frequency, spirometry and quality of life, and again, no significant effect on mortality.
Upon looking solely at salmeterol, Stockley 2006a found that study participants were less likely to withdraw when taking the study drug than when taking placebo, and that key efficacy measures were improved at three, six and 12 months.
Wang 2012 concluded that formoterol did not improve exacerbations unless used in conjunction with inhaled corticosteroids. However, Wang 2012 did not include trials of less than six months, excluded small trials and included trials of indacaterol, which may explain why the results differ.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Funnel plot of comparison: 1 All LABA versus placebo [subgrouped by drug], outcome: 1.1 Quality of life (SGRQ total score).
Funnel plot of comparison: 1 All LABA versus placebo [subgrouped by drug], outcome: 1.4 Severe exacerbations (hospitalisations).
![Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 1 Quality of life (SGRQ total score).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-001-01.png)
Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 1 Quality of life (SGRQ total score).
![Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 2 Quality of life (number of people achieving the MCID on the SGRQ).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-001-02.png)
Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 2 Quality of life (number of people achieving the MCID on the SGRQ).
![Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 3 Quality of life (CRQ).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-001-03.png)
Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 3 Quality of life (CRQ).
![Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 4 Severe exacerbations (hospitalisations).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-001-04.png)
Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 4 Severe exacerbations (hospitalisations).
![Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 5 Severe/moderate exacerbations (hospitalisation or course of meds or ER visit.](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-001-05.png)
Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 5 Severe/moderate exacerbations (hospitalisation or course of meds or ER visit.
![Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 6 Moderate exacerbations (course of antibiotics and/or steroids).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-001-06.png)
Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 6 Moderate exacerbations (course of antibiotics and/or steroids).
![Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 7 Mortality (all‐cause).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-001-07.png)
Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 7 Mortality (all‐cause).
![Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 8 People with one or more non‐fatal serious adverse event.](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-001-08.png)
Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 8 People with one or more non‐fatal serious adverse event.
![Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 9 Predose FEV1 (mL).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-001-09.png)
Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 9 Predose FEV1 (mL).
![Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 10 Withdrawal.](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-001-10.png)
Comparison 1 All LABA versus placebo [subgrouped by drug], Outcome 10 Withdrawal.
![Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 1 Quality of life (SGRQ).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-002-01.png)
Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 1 Quality of life (SGRQ).
![Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 2 Severe exacerbations (hospitalisations).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-002-02.png)
Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 2 Severe exacerbations (hospitalisations).
![Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 3 Moderate exacerbations (course of antibiotics and/or steroids).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-002-03.png)
Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 3 Moderate exacerbations (course of antibiotics and/or steroids).
![Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 4 Mortality (all‐cause).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-002-04.png)
Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 4 Mortality (all‐cause).
![Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 5 Patients with one or more serious adverse event.](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-002-05.png)
Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 5 Patients with one or more serious adverse event.
![Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 6 Predose FEV1 (mL).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-002-06.png)
Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 6 Predose FEV1 (mL).
![Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 7 Withdrawal.](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-002-07.png)
Comparison 2 Formoterol 12 μg versus placebo [subgrouped by trial duration], Outcome 7 Withdrawal.
![Comparison 3 Formoterol 24 μg versus placebo [subgrouped by trial duration], Outcome 1 Quality of life (SGRQ).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-003-01.png)
Comparison 3 Formoterol 24 μg versus placebo [subgrouped by trial duration], Outcome 1 Quality of life (SGRQ).
![Comparison 3 Formoterol 24 μg versus placebo [subgrouped by trial duration], Outcome 2 Severe exacerbations (hospitalisations).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-003-02.png)
Comparison 3 Formoterol 24 μg versus placebo [subgrouped by trial duration], Outcome 2 Severe exacerbations (hospitalisations).
![Comparison 3 Formoterol 24 μg versus placebo [subgrouped by trial duration], Outcome 3 Moderate exacerbations (course of antibiotics and/or steroids).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-003-03.png)
Comparison 3 Formoterol 24 μg versus placebo [subgrouped by trial duration], Outcome 3 Moderate exacerbations (course of antibiotics and/or steroids).
![Comparison 3 Formoterol 24 μg versus placebo [subgrouped by trial duration], Outcome 4 Mortality (all‐cause).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-003-04.png)
Comparison 3 Formoterol 24 μg versus placebo [subgrouped by trial duration], Outcome 4 Mortality (all‐cause).
![Comparison 3 Formoterol 24 μg versus placebo [subgrouped by trial duration], Outcome 5 People with one or more non‐fatal serious adverse events.](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-003-05.png)
Comparison 3 Formoterol 24 μg versus placebo [subgrouped by trial duration], Outcome 5 People with one or more non‐fatal serious adverse events.
![Comparison 3 Formoterol 24 μg versus placebo [subgrouped by trial duration], Outcome 6 Withdrawal.](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-003-06.png)
Comparison 3 Formoterol 24 μg versus placebo [subgrouped by trial duration], Outcome 6 Withdrawal.
![Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 1 Quality of life (SGRQ).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-004-01.png)
Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 1 Quality of life (SGRQ).
![Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 2 Severe exacerbations (hospitalisations).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-004-02.png)
Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 2 Severe exacerbations (hospitalisations).
![Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 3 Moderate exacerbations (course of antibiotics and/or steroids).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-004-03.png)
Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 3 Moderate exacerbations (course of antibiotics and/or steroids).
![Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 4 Mortality (all‐cause).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-004-04.png)
Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 4 Mortality (all‐cause).
![Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 5 People with one or more non‐fatal serious adverse events.](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-004-05.png)
Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 5 People with one or more non‐fatal serious adverse events.
![Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 6 Predose FEV1 (mL).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-004-06.png)
Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 6 Predose FEV1 (mL).
![Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 7 Withdrawal.](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-004-07.png)
Comparison 4 Salmeterol 50 μg versus placebo [subgrouped by trial duration], Outcome 7 Withdrawal.
![Comparison 5 [Sensitivity analysis—ICS use] All LABA versus placebo, Outcome 1 Quality of life (SGRQ).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-005-01.png)
Comparison 5 [Sensitivity analysis—ICS use] All LABA versus placebo, Outcome 1 Quality of life (SGRQ).
![Comparison 5 [Sensitivity analysis—ICS use] All LABA versus placebo, Outcome 2 Severe exacerbations (hospitalisations).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-005-02.png)
Comparison 5 [Sensitivity analysis—ICS use] All LABA versus placebo, Outcome 2 Severe exacerbations (hospitalisations).
![Comparison 6 [Sensitivity analysis—attrition] All LABA versus placebo, Outcome 1 Quality of life (SGRQ).](/cdsr/doi/10.1002/14651858.CD010177.pub2/media/CDSR/CD010177/image_n/nCD010177-CMP-006-01.png)
Comparison 6 [Sensitivity analysis—attrition] All LABA versus placebo, Outcome 1 Quality of life (SGRQ).
| Long‐acting beta2‐agonists compared with placebo for chronic obstructive pulmonary disease | ||||||
| Patient or population: people with chronic obstructive pulmonary disease Setting: community | ||||||
| Outcomes Follow‐up: weighted means presented for each outcome | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Placebo | Long‐acting beta2‐agonists | |||||
| Quality of life1 16 months | 45.3 | Mean score in the intervention groups was | MD ‐2.32 (‐3.09 to ‐1.54) | 11,397 | ⊕⊕⊕⊝ | A difference of 4 points is generally accepted to be of clinical significance |
| Severe exacerbations (hospitalisations) 7 months | 71 per 1000 | 53 per 1000 | OR 0.73 | 2859 | ⊕⊕⊕⊝ | I2 = 10%, P = 0.35 |
| Moderate exacerbations (course of antibiotics or oral steroids) 8 months | 238 per 1000 | 186 per 1000 | OR 0.73 | 3375 | ⊕⊕⊕⊝ | I2 = 8%, P = 0.37 |
| Severe/moderate exacerbations (hospitalisation or course of medication or ER visit) 8 months | 336 per 1000 | 308 per 1000 | OR 0.88 | 3968 | ⊕⊕⊕⊝ | I2 = 0%, P = 0.80 |
| Mortality (all‐cause) 14 months | 5 per 1000 | 5 per 1000 | OR 0.90 | 14,079 | ⊕⊕⊕⊝ | I2 = 21%, P = 0.21 |
| Participants with one or more serious adverse event (non‐fatal)1 15 months | 86 per 1000 | 84 per 1000 | OR 0.97 | 12,446 | ⊕⊕⊕⊝ | |
| Predose FEV1 (mL)1 7 months; higher is better | 1221 mL | Mean predose FEV1 in the intervention groups was 73 mL higher | MD 73 mL (48 to 98 mL) | 6125 | ⊕⊕⊝⊝ | Subgroup differences (I2 = 84%) discussed in high heterogeneity within subgroups, with potential baseline differences |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence. | ||||||
| n.b. Unless otherwise stated, significant subgroup differences were not found. 1 Quality of life, serious adverse events and FEV1 were analysed with random effects as the result of heterogeneity. 2 I2 = 50%, P < 0.01 (‐1 for inconsistency) 3 Several studies did not report exacerbations in a form that could be included in any of the three outcomes included in this review (‐1 for publication bias) 4 Confidence intervals include important benefit and potential harm (‐1 for imprecision) 5 I2 = 34%, P = 0.06; opposite direction of effect observed for the two formoterol doses (‐1 for inconsistency) 6 I2 = 71%, P < 0.01 (‐1 for inconsistency) | ||||||
| Trial length | Study IDs | Mean age, years, | Male, %, |
| 3 months | Aalbers 2002; Bogdan 2011; Dahl 2001; Hanrahan 2008; | 63.5 (62 to 67) | 66.5 (52 to 87) |
| 6 months | Brusasco 2003; Campbell 2005; Doherty 2012; Hanania 2003; | 63 (59 to 65) | 75 (63 to 88) |
| 12 months | Calverley 2003a; Calverley 2003b [TRISTAN]; Dahl 2010; | 63.5 (63 to 64) | 77.5 (75 to 80) |
| 36 months | 65 (N/A) | 76 |
| FEV1 % predicted at baseline | Study IDs | Mean baseline SGRQ, |
| 30% ≤ mean < 40% | Brusasco 2003; Calverley 2003a; Doherty 2012; | 50 (47 to 53) |
| 40% ≤ mean < 50% | Calverley 2003b [TRISTAN]; Calverley 2007 [TORCH]; | 50 (48 to 55) |
| 50% ≤ mean < 60% | Aalbers 2002; Bogdan 2011; Campbell 2005; | 45 (44 to 49) |
| Five studies did not report mean % predicted FEV1 at baseline; 10 of the studies presented here did not | ||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Quality of life (SGRQ total score) Show forest plot | 17 | 11397 | Mean Difference (IV, Random, 95% CI) | ‐2.32 [‐3.09, ‐1.54] |
| 1.1 Formoterol 12 μg twice daily | 11 | 5587 | Mean Difference (IV, Random, 95% CI) | ‐2.66 [‐3.84, ‐1.48] |
| 1.2 Formoterol 24 μg twice daily | 3 | 627 | Mean Difference (IV, Random, 95% CI) | ‐2.51 [‐4.51, ‐0.51] |
| 1.3 Salmeterol 50 μg twice daily | 5 | 5183 | Mean Difference (IV, Random, 95% CI) | ‐1.64 [‐2.78, ‐0.50] |
| 2 Quality of life (number of people achieving the MCID on the SGRQ) Show forest plot | 3 | 1871 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.58 [1.32, 1.90] |
| 2.1 Formoterol 12 μg twice daily | 1 | 397 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.11 [2.06, 4.70] |
| 2.2 Salmeterol 50 μg twice daily | 2 | 1474 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.33 [1.08, 1.64] |
| 3 Quality of life (CRQ) Show forest plot | 4 | 1193 | Mean Difference (IV, Fixed, 95% CI) | 3.10 [1.22, 4.98] |
| 3.1 Salmeterol 50 μg twice daily | 4 | 1193 | Mean Difference (IV, Fixed, 95% CI) | 3.10 [1.22, 4.98] |
| 4 Severe exacerbations (hospitalisations) Show forest plot | 7 | 3804 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.73 [0.56, 0.95] |
| 4.1 Formoterol 12 μg twice daily | 6 | 2418 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.75 [0.55, 1.02] |
| 4.2 Formoterol 24 μg twice daily | 2 | 581 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.28 [0.11, 0.73] |
| 4.3 Salmeterol 50 μg twice daily | 1 | 805 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.99 [0.52, 1.86] |
| 5 Severe/moderate exacerbations (hospitalisation or course of meds or ER visit Show forest plot | 7 | 3968 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.88 [0.76, 1.02] |
| 5.1 Formoterol 12 μg twice daily | 4 | 2682 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.87 [0.73, 1.04] |
| 5.2 Salmeterol 50 μg twice daily | 3 | 1286 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.90 [0.69, 1.16] |
| 6 Moderate exacerbations (course of antibiotics and/or steroids) Show forest plot | 7 | 3375 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.73 [0.61, 0.87] |
| 6.1 Formoterol 12 μg twice daily | 3 | 968 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.74 [0.51, 1.05] |
| 6.2 Formoterol 24 μg twice daily | 1 | 324 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.59 [0.35, 0.97] |
| 6.3 Salmeterol 50 μg twice daily | 4 | 2083 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.75 [0.60, 0.94] |
| 7 Mortality (all‐cause) Show forest plot | 23 | 14079 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.90 [0.75, 1.08] |
| 7.1 Formoterol 12 μg twice daily | 13 | 6343 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.41 [0.87, 2.30] |
| 7.2 Formoterol 24 μg twice daily | 2 | 615 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 4.54 [0.07, 285.14] |
| 7.3 Salmeterol 50 μg twice daily | 10 | 7121 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.83 [0.69, 1.01] |
| 8 People with one or more non‐fatal serious adverse event Show forest plot | 20 | 12446 | Odds Ratio (M‐H, Random, 95% CI) | 0.97 [0.83, 1.14] |
| 8.1 Formoterol 12 μg twice daily | 10 | 4829 | Odds Ratio (M‐H, Random, 95% CI) | 1.20 [0.99, 1.45] |
| 8.2 Formoterol 24 μg twice daily | 3 | 737 | Odds Ratio (M‐H, Random, 95% CI) | 0.42 [0.23, 0.77] |
| 8.3 Salmeterol 50 μg twice daily | 9 | 6880 | Odds Ratio (M‐H, Random, 95% CI) | 0.93 [0.78, 1.10] |
| 9 Predose FEV1 (mL) Show forest plot | 13 | 6125 | Mean Difference (IV, Random, 95% CI) | 72.92 [48.02, 97.82] |
| 9.1 Formoterol 12 μg twice daily | 6 | 3222 | Mean Difference (IV, Random, 95% CI) | 44.68 [29.39, 59.97] |
| 9.2 Salmeterol 50 μg twice daily | 7 | 2903 | Mean Difference (IV, Random, 95% CI) | 101.01 [59.84, 142.18] |
| 10 Withdrawal Show forest plot | 25 | 14763 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.74 [0.69, 0.80] |
| 10.1 Formoterol 12 μg twice daily | 14 | 6597 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.76 [0.68, 0.86] |
| 10.2 Formoterol 24 μg twice daily | 4 | 1001 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.81 [0.58, 1.14] |
| 10.3 Salmeterol 50 μg twice daily | 10 | 7165 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.72 [0.65, 0.80] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Quality of life (SGRQ) Show forest plot | 11 | Mean Difference (IV, Random, 95% CI) | Subtotals only | |
| 1.1 3 months | 3 | 1627 | Mean Difference (IV, Random, 95% CI) | ‐3.86 [‐5.25, ‐2.46] |
| 1.2 6 months | 5 | 2427 | Mean Difference (IV, Random, 95% CI) | ‐1.14 [‐2.51, 0.23] |
| 1.3 12 months | 5 | 3079 | Mean Difference (IV, Random, 95% CI) | ‐3.24 [‐4.77, ‐1.71] |
| 2 Severe exacerbations (hospitalisations) Show forest plot | 6 | 2614 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.73 [0.54, 0.98] |
| 2.1 3 months | 1 | 353 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.47 [0.08, 2.60] |
| 2.2 6 months | 3 | 1319 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.53 [0.26, 1.08] |
| 2.3 12 months | 2 | 942 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.80 [0.57, 1.12] |
| 3 Moderate exacerbations (course of antibiotics and/or steroids) Show forest plot | 3 | 1078 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.78 [0.56, 1.07] |
| 3.1 3 months | 1 | 228 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.76 [0.27, 2.12] |
| 3.2 6 months | 1 | 419 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.53 [0.28, 0.99] |
| 3.3 12 months | 1 | 431 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.92 [0.62, 1.37] |
| 4 Mortality (all‐cause) Show forest plot | 13 | 6553 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.45 [0.89, 2.37] |
| 4.1 3 months | 5 | 1927 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 2.22 [0.64, 7.70] |
| 4.2 6 months | 4 | 1946 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 2.25 [0.98, 5.16] |
| 4.3 12 months | 4 | 2680 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.94 [0.47, 1.88] |
| 5 Patients with one or more serious adverse event Show forest plot | 10 | 5039 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.20 [1.00, 1.43] |
| 5.1 3 months | 2 | 799 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.21 [0.57, 2.58] |
| 5.2 6 months | 4 | 1916 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.10 [0.78, 1.55] |
| 5.3 12 months | 4 | 2324 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.23 [1.00, 1.52] |
| 6 Predose FEV1 (mL) Show forest plot | 6 | 3222 | Mean Difference (IV, Fixed, 95% CI) | 44.68 [29.39, 59.97] |
| 6.1 3 months | 1 | 407 | Mean Difference (IV, Fixed, 95% CI) | 40.0 [16.29, 63.71] |
| 6.2 6 months | 3 | 1437 | Mean Difference (IV, Fixed, 95% CI) | 44.41 [18.63, 70.18] |
| 6.3 12 months | 2 | 1378 | Mean Difference (IV, Fixed, 95% CI) | 53.46 [21.74, 85.17] |
| 7 Withdrawal Show forest plot | 14 | 6894 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.77 [0.69, 0.86] |
| 7.1 3 months | 4 | 1368 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.69 [0.50, 0.94] |
| 7.2 6 months | 5 | 2335 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.72 [0.59, 0.89] |
| 7.3 12 months | 5 | 3191 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.81 [0.70, 0.94] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Quality of life (SGRQ) Show forest plot | 3 | 707 | Mean Difference (IV, Fixed, 95% CI) | ‐2.32 [‐4.52, ‐0.13] |
| 1.1 3 months | 2 | 376 | Mean Difference (IV, Fixed, 95% CI) | ‐2.04 [‐4.65, 0.56] |
| 1.2 12 months | 1 | 331 | Mean Difference (IV, Fixed, 95% CI) | ‐3.0 [‐7.06, 1.06] |
| 2 Severe exacerbations (hospitalisations) Show forest plot | 2 | 777 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.28 [0.12, 0.67] |
| 2.1 3 months | 1 | 343 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.50 [0.09, 2.75] |
| 2.2 12 months | 1 | 434 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.24 [0.09, 0.65] |
| 3 Moderate exacerbations (course of antibiotics and/or steroids) Show forest plot | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 3.1 12 months | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 4 Mortality (all‐cause) Show forest plot | 2 | 825 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 7.60 [0.15, 383.12] |
| 4.1 3 months | 1 | 391 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 4.2 12 months | 1 | 434 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 7.60 [0.15, 383.12] |
| 5 People with one or more non‐fatal serious adverse events Show forest plot | 3 | 947 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.53 [0.36, 0.79] |
| 5.1 3 months | 2 | 513 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.36 [0.14, 0.94] |
| 5.2 12 months | 1 | 434 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.58 [0.38, 0.89] |
| 6 Withdrawal Show forest plot | 4 | 1297 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.79 [0.59, 1.05] |
| 6.1 3 months | 3 | 863 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.92 [0.64, 1.33] |
| 6.2 12 months | 1 | 434 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.63 [0.40, 0.99] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Quality of life (SGRQ) Show forest plot | 5 | 5183 | Mean Difference (IV, Random, 95% CI) | ‐1.64 [‐2.78, ‐0.50] |
| 1.1 6 months | 3 | 1405 | Mean Difference (IV, Random, 95% CI) | ‐3.10 [‐6.03, ‐0.17] |
| 1.2 12 months | 1 | 733 | Mean Difference (IV, Random, 95% CI) | ‐1.10 [‐2.35, 0.15] |
| 1.3 36 months | 1 | 3045 | Mean Difference (IV, Random, 95% CI) | ‐1.0 [‐2.00, 0.00] |
| 2 Severe exacerbations (hospitalisations) Show forest plot | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 2.1 6 months | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 3 Moderate exacerbations (course of antibiotics and/or steroids) Show forest plot | 4 | 2083 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.75 [0.60, 0.94] |
| 3.1 3 months | 2 | 545 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.71 [0.49, 1.03] |
| 3.2 6 months | 1 | 805 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.84 [0.63, 1.12] |
| 3.3 12 months | 1 | 733 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.40 [0.17, 0.92] |
| 4 Mortality (all‐cause) Show forest plot | 10 | 7121 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.83 [0.69, 1.01] |
| 4.1 3 months | 3 | 1132 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.14 [0.00, 6.98] |
| 4.2 6 months | 5 | 2211 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.56 [0.22, 1.46] |
| 4.3 12 months | 1 | 733 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.43 [0.12, 1.50] |
| 4.4 36 months | 1 | 3045 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.87 [0.71, 1.06] |
| 5 People with one or more non‐fatal serious adverse events Show forest plot | 9 | 6880 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.94 [0.83, 1.06] |
| 5.1 3 months | 2 | 850 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.81 [0.45, 1.46] |
| 5.2 6 months | 5 | 2211 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.74 [0.55, 1.01] |
| 5.3 12 months | 1 | 733 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.29 [0.88, 1.91] |
| 5.4 36 months | 1 | 3086 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.96 [0.83, 1.11] |
| 6 Predose FEV1 (mL) Show forest plot | 7 | 2903 | Mean Difference (IV, Random, 95% CI) | 101.01 [59.84, 142.18] |
| 6.1 3 months | 2 | 765 | Mean Difference (IV, Random, 95% CI) | 146.86 [114.38, 179.34] |
| 6.2 6 months | 4 | 1405 | Mean Difference (IV, Random, 95% CI) | 87.93 [39.05, 136.80] |
| 6.3 12 months | 1 | 733 | Mean Difference (IV, Random, 95% CI) | 59.0 [28.52, 89.48] |
| 7 Withdrawal Show forest plot | 10 | 7165 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.72 [0.65, 0.80] |
| 7.1 3 months | 3 | 1132 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.60 [0.44, 0.83] |
| 7.2 6 months | 5 | 2211 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.72 [0.59, 0.88] |
| 7.3 12 months | 1 | 733 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.74 [0.55, 1.01] |
| 7.4 36 months | 1 | 3089 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.74 [0.64, 0.86] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Quality of life (SGRQ) Show forest plot | 12 | 8520 | Mean Difference (IV, Fixed, 95% CI) | ‐1.53 [‐2.08, ‐0.98] |
| 1.1 Formoterol 12 μg twice daily | 7 | 3782 | Mean Difference (IV, Fixed, 95% CI) | ‐2.16 [‐3.04, ‐1.27] |
| 1.2 Formoterol 24 μg twice daily | 1 | 121 | Mean Difference (IV, Fixed, 95% CI) | ‐1.50 [‐4.62, 1.62] |
| 1.3 Salmeterol 50 μg twice daily | 4 | 4617 | Mean Difference (IV, Fixed, 95% CI) | ‐1.12 [‐1.84, ‐0.39] |
| 2 Severe exacerbations (hospitalisations) Show forest plot | 3 | 1795 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.85 [0.62, 1.16] |
| 2.1 Formoterol 12 μg twice daily | 2 | 990 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.81 [0.57, 1.15] |
| 2.2 Salmeterol 50 μg twice daily | 1 | 805 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.99 [0.52, 1.86] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Quality of life (SGRQ) Show forest plot | 13 | 9114 | Mean Difference (IV, Fixed, 95% CI) | ‐1.60 [‐2.15, ‐1.05] |
| 1.1 Formoterol 12 μg twice daily | 8 | 3338 | Mean Difference (IV, Fixed, 95% CI) | ‐2.07 [‐3.06, ‐1.08] |
| 1.2 Formoterol 24 μg twice daily | 3 | 627 | Mean Difference (IV, Fixed, 95% CI) | ‐2.25 [‐4.56, 0.06] |
| 1.3 Salmeterol 50 μg twice daily | 4 | 5149 | Mean Difference (IV, Fixed, 95% CI) | ‐1.31 [‐2.01, ‐0.61] |
