Aims:  This analysis was designed to provide a comparison between budesonide/formoterol and salmeterol/fluticasone for the relative incidence of pneumonia adverse events, pneumonia serious adverse events and pneumonia-related mortality in patients being treated for chronic obstructive pulmonary disease.Methods:  An initial literature search revealed no suitable head-to-head trials between budesonide/formoterol and salmeterol/fluticasone and therefore a systematic review was conducted to find randomised controlled trials providing data for input into an adjusted indirect comparison of the two combination treatments using placebo as a common comparator. The Bucher adjusted indirect comparison method was used to calculate odds ratios and 95% confidence intervals.Results:  Eight salmeterol/fluticasone trials and four budesonide/formoterol trials were identified as being relevant for the analyses. The proportion of patients experiencing a pneumonia adverse event was significantly lower with budesonide/formoterol than salmeterol/fluticasone (odds ratio, 0.47; 95% confidence interval, 0.28–0.80). The proportion of patients experiencing a pneumonia serious adverse event was also significantly lower with budesonide/formoterol than salmeterol/fluticasone (odds ratio, 0.41; 95% confidence interval, 0.19–0.86). However, there were too few events to draw any firm conclusions on pneumonia-related mortality.Conclusions:  The results of the indirect comparison support the hypothesis that budesonide/formoterol is associated with fewer pneumonia events than salmeterol/fluticasone in chronic obstructive pulmonary disease. The limitations of the analysis are that the results from a single study, TORCH, have a large bearing on the overall findings of the analysis, and that there is heterogeneity in the length and the dosing of the included studies, although it does not appear that heterogeneity affected the reported results. Another important limitation is the lack of predefined diagnostic standards for pneumonia in these studies.