BACKGROUND: Eosinophilic esophagitis (EoE) and gastroesophageal reflux (GERD) both cause esophageal eosinophilia. Reports show that esophageal eosinophilia meeting criteria for EoE may respond to acid suppression mono-therapy. Consensus guidelines have termed this entity "PPI-responsive esophageal eosinophilia" (PPIRee) and recommend a trial with proton-pump inhibitors (PPIs) prior to a definitive EoE diagnosis. The mechanisms of PPIRee and whether this represents a sub-phenotype of GERD, a sub-phenotype of EoE, or its own distinct entity remain unclear. METHODS: A database search revealed children who had an initial histologic response to PPI monotherapy but had recurrence of esophageal eosinophilia and symptoms despite continued PPI therapy. In order to understand the patterns of esophageal inflammatory cells during PPI therapy we performed quantitative immunohistochemistry for mast cells, CD1a positive antigen presenting cells, and CD45RO memory T cells. RESULTS: Four pediatric patients (mean age 9.5 years) had a mean peak eosinophil count of 52 eos/hpf which initially resolved completely during PPI mono-therapy. However, despite continued PPI therapy, endoscopic abnormalities and pan-esophageal eosinophilia recurred (mean peak eosinophil count of 64 eos/hpf). There was no seasonal variation or lack of PPI adherence that explained the return of eosinopihlia. Similar to eosinophilia, mastocytosis and CD45RO cells were transiently decreased during PPI therapy. CONCLUSION: PPIs appear to be capable of transiently resolving multiple inflammatory cell subsets including eosinophils, mast cells, and CD45RO cells. Our data suggest that patients with PPIRee should have continued monitoring for EoE during PPI monotherapy. The numbers of patients in whom PPIRee is a transient phenomenon and whether PPIRee represents a sub-phenotype of EoE in children merits further investigation.