BACKGROUND: Inhaled short acting beta2 adrenergic agonists and ipratropium bromide are both used in the treatment of acute exacerbations of chronic obstructive pulmonary disease. OBJECTIVES: In patients with acute exacerbations of COPD to: 1. To assess the efficacy of short-acting beta-2 agonists against placebo; 2. Compare the efficacy of short-acting beta-2 agonists and ipratropium. SEARCH STRATEGY: A comprehensive search of the literature was carried out of EMBASE, MEDLINE, CINAHL and the Cochrane COPD trials register was carried out using the terms: bronchodilator* OR albuterol OR metaproterenol OR terbutaline OR isoetharine OR pirbuterol OR salbutamol OR beta-2 agonist. SELECTION CRITERIA: All trials that appeared to be relevant were assessed by two reviewers who independently selected trials for inclusion. Differences were resolved by consensus. DATA COLLECTION AND ANALYSIS: All trials that appeared to be relevant were assessed by two reviewers who independently selected trials for inclusion. Differences were resolved by consensus. References listed in each included trial were searched for additional trial reports. Trials were combined using Review Manager using a fixed effects model. The size of the treatment effects were tested for heterogeneity. MAIN RESULTS: We identified no placebo-controlled comparisons of beta-2 agonists. Three studies permitted comparison of ipratropium to an inhaled beta-2 agonist. These studies included a total of 103 patients. The beta2-agonists used were: fenoterol and metaproterenol. One study was a parallel group trial of regular therapy for seven days. The other two were cross over studies of single dose treatments, with efficacy measured 90 min post dose. There was no washout period between treatments. Both treatments produced an improvement in forced expiratory volume (FEV1) after 90 min in the range 150-250 ml. The was no difference between treatments, mean difference in FEV1 10 ml; 95% CI -220, 230 ml. In one small crossover study (n=10) there was a significant improvement in arterial PaO2 after 30 minutes with ipratropium (+5.8 mm Hg +/- 3.0 (SEM)) compared to metaproterenol (-6.2 +/- 1.2 mm Hg), but this was not significant at 90 min. There were no data concerning respiratory symptoms. The crossover studies showed no evidence of an additive effect of the two treatments, although they were not designed specifically to test this. REVIEWER'S CONCLUSIONS: There are few controlled trial data concerning the use of inhaled beta2-agonist agents in acute exacerbations of COPD and none that have compared these agents directly with placebo. None of the studies used the more modern beta2-agonists used most widely in this setting (salbutamol and terbutaline). Beta2-agonists and ipratropium both produce small improvements in FEV1, but beta2-agonists may worsen PaO2 for a period. We could not draw conclusions concerning possible additive effects.