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


Review: Long-acting β-agonists increase severe asthma exacerbations and asthma-related deaths in children and adults


ACP J Club. 2007 Jan-Feb;146:17. doi:10.7326/ACPJC-2007-146-1-017

Clinical Impact Ratings

Emergency Med: 5 stars

GIM/FP/GP: 6 stars

Hospitalists: 6 stars

Allerg & Immunol: 6 stars

Pulmonology: 5 stars

Source Citation

Salpeter SR, Buckley NS, Ormiston TM, Salpeter EE. Meta-analysis: effect of long-acting beta-agonists on severe asthma exacerbations and asthma-related deaths. Ann Intern Med. 2006;144:904-12. [PubMed ID: 16754916]



In children and adults with asthma, do long-acting β-agonists (LABAs) increase severe asthma exacerbations requiring hospitalization, life-threatening asthma attacks, and asthma-related deaths?


Data sources: MEDLINE, EMBASE/Excerpta Medica, CINAHL, and Cochrane databases (all from 1966 to December 2005); references of selected reviews; and the U.S. Food and Drug Administration Web site.

Study selection and assessment: Randomized controlled trials (RCTs) in any language that compared LABAs (salmeterol, formoterol, or eformoterol) with placebo and had ≥ 3-month follow-up. Patients were allowed to use short- acting β-agonists if needed. 19 RCTs (n = 33 826; mean age 37 to 38 y, 51% men) met the selection criteria. Quality assessment of individual studies was based on randomization procedure, allocation concealment, blinding, dropouts and withdrawals, and intention-to-treat analysis.

Outcomes: Hospital admission for asthma exacerbations, life-threatening asthma exacerbations, and asthma-related death.

Main results

More patients in the LABA group than in the placebo group were hospitalized for asthma exacerbations, had life-threatening asthma exacerbations, or died from asthma (Table). Subgroup analyses showed that increased hospital admission occurred in both children (< 12 y) (odds ratio [OR] 3.9, 95% CI 1.7 to 8.8) and adults (OR 2.0, CI 1.0 to 3.9), and with salmeterol (OR 1.7, CI 1.1 to 2.7) or formoterol (OR 3.2, CI 1.7 to 6.0). RCTs with > 75% use of concomitant inhaled corticosteroids (ICSs) (mean 90% use) showed that LABAs also led to a greater rate of hospitalizations (OR 2.1, CI 1.3 to 3.4).


In children and adults with asthma, long-acting β-agonists increase hospital admissions for severe asthma exacerbations, life-threatening asthma attacks, and asthma-related death.

Source of funding: No external funding.

For correspondence: Dr. S.R. Salpeter, Santa Clara Valley Medical Center, San Jose, CA, USA. E-mail

Table. Long-acting β-agonists vs placebo in children and adults with asthma at mean 6 months*

Outcomes Number of trials (n) Weighted event rates RRI (95% CI) NNH (CI)
Long-acting β-agonists Placebo
Hospital admission for asthma exacerbations 12 (5091) 1.5% 0.6% 158% (59 to 322) 107 (53 to 282)
Life-threatening asthma exacerbations 7 (29 981) 0.31% 0.17% 80% (10 to 189) 730 (308 to 5827)
Asthma-related death† 1 (26 353) 0.08% 0.02% 250% (30 to 828) 1759 (531 to 14 650)

*Abbreviations defined in Glossary; RRI, NNH, and CI calculated from odds ratios and control event rates in article using a fixed-effects model.
†Data from the SMART study; 2 other trials reported 1 death in the β-agonist group and 0 deaths in the placebo group.


The meta-analysis by Salpeter and colleagues showed that treating patients with asthma using LABAs results in a small but significant increase in death and other major adverse outcomes. Given the high prevalence of asthma and chronic obstructive pulmonary disease (COPD) and the convenience of taking long-acting medication, these findings are important.

Salpeter and colleagues speculate that LABAs may mask symptoms when underlying obstructive airways disease worsens. Of particular concern is the subgroup analysis that examined studies in which ≥ 75% of patients also received ICSs. Even among these patients, the use of LABAs doubled risk for hospitalization, suggesting that ICSs do not fully protect against the adverse effects of LABAs. However, because the SMART study did not provide data on asthma-related hospitalization (1), the largest study was excluded from this subgroup analysis. Also, true use of concurrent ICSs is important in these studies because these adverse outcomes may be a consequence of LABA monotherapy that would be expected to be ameliorated by the concomitant use of ICSs. Furthermore, the review did not address combining LABAs and ICSs in one metered dose inhaler (MDI), which could be less expensive than separate MDIs, and hence more likely to be used. Limited data have been made available from the large multicentre TORCH study of mortality in COPD for participants receiving ICSs plus LABAs (2), but not for participants receiving LABAs alone. As with mortality data from the SMART study (3), these data from the TORCH study are awaited with considerable interest. Heterogeneity of results of relevant studies in relation to actual ICS use also requires further assessment.

The meta-analysis is unable to provide insights on whether the adverse outcomes may be related to racial or genetic factors (as has been suggested by post hoc analysis in the SMART study), co-treatments (including combination inhalers and methylxanthines), or baseline disease severity. The bottom line is that LABA monotherapy should not be used in patients with asthma. At the least, this meta-analysis underscores the need for LABAs to only be used as a component of combined ICS and LABA therapy.

Brian J. Smith, MBBS, FRACP, PhD
Anil Roy, MBBS, MD
Queen Elizabeth Hospital
Woodville South, South Australia, Australia


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. GlaxoSmithKline clinical trial register. Fluticasone propionate/salmeterol xinafoate studies: SCO30003. (accessed 9 Nov. 2006).

3. Lurie P, Wolfe S. Misleading data analyses in salmeterol (SMART) study. Lancet. 2005;366:1261-2. [PubMed ID: 16214589]