Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: Anticholinergics, but not β2-agonists, reduce exacerbations requiring hospitalization and respiratory deaths in COPD


ACP J Club. 2007 Jan-Feb;146:19. doi:10.7326/ACPJC-2007-146-1-019

Related Content in this Issue
• Companion Abstract and Commentary: Review: Long-acting β2-adrenoceptor agonists are effective in poorly reversible chronic obstructive pulmonary disease

Clinical Impact Ratings

GIM/FP/GP: 6 stars

Pulmonology: 6 stars

Source Citation

Salpeter SR, Buckley NS, Salpeter EE. Meta-analysis: anticholinergics, but not beta-agonists, reduce severe exacerbations and respiratory mortality in COPD. J Gen Intern Med. 2006;21:1011-9. [PubMed ID: 16970553]



How effective are anticholinergics and β2-agonists for chronic obstructive pulmonary disease (COPD)?


Data sources: MEDLINE, EMBASE/Excerpta Medica, and Cochrane databases (to December 2005); U.S. Food and Drug Administration Web site; and references of identified reviews.

Study selection and assessment: Randomized controlled trials (RCTs) in any language that compared anticholinergics or β2-agonists with placebo or with each other, had ≥ 3-month follow-up, and reported COPD exacerbations requiring study withdrawal or hospitalization, or respiratory death. 22 RCTs (n = 15 276, mean age range 60 to 64 y) with mean 20-month follow-up (range 3 to 60 mo) met the selection criteria. Methodological quality of individual studies was based on randomization procedure and allocation concealment, blinding of patients and providers, reporting of withdrawals and dropouts, and intention-to-treat analysis.

Outcomes: Exacerbations causing withdrawal from the study, severe exacerbations requiring hospitalization, and respiratory death.

Main results

Anticholinergics used were ipratropium and tiotropium. β2-agonists used were albuterol, metaproterenol, formoterol, and salmeterol. Compared with placebo, anticholinergics reduced withdrawals, hospitalizations, and respiratory deaths (Table). β2-agonists reduced withdrawals, but did not have an effect on hospitalizations and increased respiratory deaths (Table). Compared with anticholinergics, β2-agonists increased withdrawals and hospitalizations, and were associated with a nonsignificant increase in respiratory deaths (Table).


In patients with chronic obstructive pulmonary disease, anticholinergics are superior to β2-agonists for reducing exacerbations. β2-agonists increase risk for respiratory deaths compared with placebo.

Source of funding: No external funding.

For correspondence: Dr. S.R. Salpeter, Stanford University School of Medicine, Stanford, CA, USA. E-mail

Table. Anticholinergics, β2-agonists, and placebo for chronic obstructive pulmonary disease at mean 20 months*

Comparisons Outcomes Number of trials (n) RRR (95% CI) NNT (CI)
Anticholinergics vs placebo Withdrawals 6 (4591) 40% (25 to 52) 29 (23 to 46)
Hospitalizations 3 (3552) 33% (14 to 47) 36 (26 to 85)
Respiratory death 5 (7881) 73% (19 to 91) 442 (355 to 1698)
β2-agonists vs placebo Withdrawals 11 (5333) 19% (5 to 32) 50 (30 to 189)
Hospitalizations 2 (911) 8.0% (−39 to 95) Not significant
Respiratory death 4 (2404) 147% (12 to 445) 92 (31 to 1127)
β2-agonists vs anticholinergics Withdrawals 7 (3044) 102% (39 to 193) 38 (20 to 98)
Hospitalizations 2 (1606) 95% (6 to 259) 54 (20 to 851)
Respiratory death 2 (1229) 591% (−15 to 5497) Not significant

*Abbreviations defined in Glossary; RRR, RRI, NNT, NNH, and CI calculated from control event rates in article using a fixed-effects model.


COPD is a major cause of morbidity and mortality with a rising incidence worldwide. Large RCTs and meta-analyses show that currently available pharmacotherapy may improve symptoms and perhaps even survival.

Appleton and colleagues provided an in-depth systematic review of the role of LABAs in the management of stable COPD. LABAs led to small statistically significant increases in lung function (FEV1) and improvement in some measures of health status compared with placebo. These findings further support current guideline recommendations for use of LABAs in stable moderate-to-severe COPD. However, the story does not end here. In light of recent concerns and the continued debate about LABAs and increased asthma-related deaths (1), is it possible that LABAs also increase deaths in patients with COPD?

The meta-analysis by Salpeter and colleagues shows the effectiveness of anticholinergics in reducing COPD exacerbations and hospitalizations. Interestingly, compared with placebo, anticholinergics reduced respiratory-related deaths while β2-agonists increased mortality in patients with COPD. A closer look at the data, however, reveals major limitations of the combined respiratory mortality results. No trials were designed to study mortality as a primary endpoint and the numbers of respiratory deaths reported in individual trials were small, rendering the pooling of respiratory deaths less certain. In addition, differential dropout rates in the treatment and placebo groups (more patients withdrawing from the placebo group) may have biased the accurate accounting of deaths. Therefore, any firm conclusions about the negative role of LABAs on COPD survival or on the positive survival effect of anticholinergics are premature. Encouragingly, preliminary reports from a large multicenter RCT (2) of 6112 patients with moderate-to-severe COPD who received combination therapy (salmeterol and fluticasone) or each component alone do not portend increased all-cause mortality in patients receiving salmeterol.

Salpeter and colleagues chose to combine results from both short- and long-acting anticholinergics (tiotropium) in their analysis. Several RCTs have shown that, compared with ipratropium, tiotropium leads to greater improvement in symptoms and fewer COPD exacerbations and hospitalizations (3). Perhaps even greater improvements in outcomes would have been noted if only the results for the tiotropium studies were combined. A recent meta-analysis by Barr and colleagues (3) showed reductions in COPD symptoms, exacerbations, and hospitalizations with tiotropium compared with placebo and ipratropium. Unlike Salpeter and colleagues, Barr and colleagues did not show a reduction in all-cause or respiratory-specific mortality in patients using tiotropium, nor did they show a difference between tiotropium and LABAs for any of the outcomes studied. The latter comparisons were limited to 2 trials.

In conclusion, we have good evidence for the use of anticholinergics and LABAs in reducing symptoms and exacerbations in patients with COPD. More studies are needed to determine whether anticholinergics are better than LABAs. The possibility of increased mortality in patients given LABAs is important to note, but at present, insufficient evidence exists to confirm this hypothesis. The effect of combination therapies is not yet fully understood but represents hope for the future of patients with COPD.

Shirin Shafazand, MD, MS, FRCPC
University of Miami, Miller School of Medicine
Miami, Florida, USA


1. Nelson HS, Weiss ST, Bleecker ER, et al. The Salmeterol Multicenter Asthma Research Trial: a comparison of usual pharmacotherapy for asthma or usual pharmacotherapy plus salmeterol. Chest. 2006;129:15-26. [PubMed ID: 16424409]

2. Vestbo J. The TORCH (towards a revolution in COPD health) survival study protocol. Eur Respir J. 2004;24:206-10. [PubMed ID: 15332386]

3. Barr RG, Bourbeau J, Camargo CA, et al. Tiotropium for stable chronic obstructive pulmonary disease: A meta-analysis. Thorax. 2006;61:854-62. [PubMed ID: 16844726]