A mixed treatment meta-analysis of antibiotic treatment options for bovine respiratory disease - An update.

Bovine respiratory disease is the most economically important disease of feedlot cattle in North America. Choice of antibiotic is a critical factor for producers and veterinarians. We previously published a mixed-treatment comparison meta-analysis that combined evidence from published trials and published estimates of comparative efficacy for 12 antibiotics registered for use in the USA. Some of the comparative efficacy estimates were based only on indirect evidence. Since the original review was published, new studies that provide direct evidence of comparative efficacy have been published. We updated the original review to include the current evidence. We also compared the results from the indirect estimates from the prior model with the observed results from randomized control trials. We repeated the original search and found that five of the new studies met the criteria for inclusion in the updated review. Four of these studies provided new data on direct comparisons of active drugs. The results from one study (performed in 2002) that compared ceftiofur pinna and enrofloxacin were inconsistent with the network and were excluded from the analysis. Three new direct comparison studies examined gamithromycin compared with tulathromycin, florfenicol, and tilmicosin. The results of our analysis suggested that the indirect estimates from the prior model provided reasonable estimates of the risk ratios revealed by the primary studies. For example, for the comparison of gamithromycin (referent) with tulathromycin, the original model predicted a risk ratio of re-treatment of 0.54 (95% credible interval 0.27-0.87). The subsequent randomized controlled trial revealed that the observed risk ratio of re-treatment was 0.59 (95% confidence interval 0.45-0.78). The results of other comparisons were also similar. For the gamithromycin (referent) to florfenicol comparison, the observed randomized trial RR was 1.17 (95% confidence interval 0.83-1.64) and the indirect estimate of RR from the prior model was 0.84 (95% credibility interval 0.48-1.3). The gamithromycin to tilmicosin (referent) observed RR from the randomized trial was 0.99 (95% confidence interval 0.67-1.47) and the indirect estimate of RR from the prior model was 1.09 (95% credible interval 0.64-1.79). The results suggested that indirect estimates provided reasonable estimates of RR when direct data were not available.