Evaluation of long-term course in children with eosinophilic esophagitis reveals distinct histologic patterns and clinical characteristics.

BACKGROUND:Eosinophilic esophagitis (EoE) is a chronic and increasingly prevalent antigen-driven disease. There is a paucity of information on long-term course in children. OBJECTIVE:We sought to understand the longitudinal trajectory of pediatric EoE during routine clinical care. METHODS:We prospectively enrolled children into an EoE database and reviewed their medical and pathologic records over 13 years. RESULTS:From 2011 to 2015, 146 children with EoE seen for their first visit at our center had 2 or more years of follow-up and 3 or more endoscopies over an average follow-up period of 5.13 years (range, 2-13 years). Longitudinal eosinophilic inflammation during treatment demonstrated 3 patterns over time. Children with less than 15 eosinophils/high-power field (hpf) for greater than 75% of their follow-up period were termed continuous responders (CRs). Children with waxing and waning inflammation of less than 15 eosinophils/hpf for less than 75% but 25% or more of the follow-up period were termed intermittent responders (IRs). Nonresponders (NRs) were defined as having less than 15 eosinophils/hpf for less than 25% of their follow-up. Fifty-nine (40%) of 146 patients were CRs, 65 (45%) of 146 were IRs, and 22 (15%) of 146 were NRs. CRs differed from IRs and NRs on the parameter of male/female ratio (1:1 in CRs, 4:1 in IRs, and 6:1 in NRs; P < .001) and in their initial response to any therapy, including proton pump inhibitors (P < .001). Endoscopic severity correlated with esophageal eosinophilia (r = 0.73, P < .001). On multivariate analysis, female sex and initial therapeutic response to medications or elimination diet were associated with long-term control of esophageal eosinophilia. CONCLUSIONS:Long-term pediatric EoE followed 3 different longitudinal trajectories of inflammation. The long-term histologic groups differed significantly in biological sex and initial therapeutic response.