In order to provide a comparative evaluation of available pharmacologic treatments for eosinophilic esophagitis (EoE), we conducted a network meta-analysis. A variety of pharmacologic treatments for EoE have been reported, however there exists a paucity of direct comparisons. We searched randomized controlled trials using MEDLINE, EMBASE, and Cochrane Central Register of Controlled Trials database through December 2014. Studies were analyzed using a random-effects network meta-analysis to identify the most effective therapy. Subgroup analysis was performed among studies that excluded gastroesophageal reflux disease or proton-pump inhibitor responsive esophageal eosinophilia, and also among pediatric and adult populations. The ranking probability for the efficacy of each treatment was analyzed. Consistency of the included randomized controlled trials was checked by applying inconsistency and node-splitting models. Eleven studies of a total of 456 patients were identified. Six pharmacologic treatments (budesonide suspension and viscous, fluticasone, prednisone, esomeprazole, and mepolizumab) and placebo were included in our analysis. Meta-analysis showed superiority of budesonide viscous, budesonide suspension, and fluticasone over placebo. Network meta-analysis demonstrated the rank order of efficacy as budesonide viscous, esomeprazole, prednisone, budesonide suspension, fluticasone, mepolizumab, and placebo. The results were consistent from the inconsistency model analysis and node-splitting analysis. Subgroup analysis demonstrated prednisone, budesonide suspension, and esomeprazole were the most effective when network meta-analyses were performed among studies that excluded gastroesophageal reflux disease or proton-pump inhibitor responsive esophageal eosinophilia, and among pediatric and adult populations, respectively. On the basis of this network meta-analysis, viscous budesonide was shown to be the most effective pharmacologic therapy for EoE among the reported pharmacologic treatments.
Read full abstract