Letter: proton pump inhibitor-responsive oesophageal eosinophilia – more than just gastro-oesophageal reflux disease. Authors' reply

Sirs, we thank Molina-Infante et al. for taking the time to read our paper, and for their comments. The Authors criticized our selection of GERD patients and therefore the results of our study. We agree that GERD and EoE may co-exist, but before concluding in this manner the presence of both diseases should be demonstrated and this is not the case in our study. Indeed, our GERD patients were defined by the presence of proven reflux disease at upper endoscopy and/or abnormal pH-testing. In addition, we included in this group those subjects in whom an oesophageal eosinophilic infiltration in the distal oesophagus was found, although the minimal number of eosinophils to suggest the diagnosis of EoE or PPI-REE (i.e. > 15 eos/HPF) was not reached, as erroneously reported by Molina-Infante et al., and the biopsies taken at mid-proximal oesophagus for searching eosinophilic infiltration were negative. Thus, according to current guidelines and reviews, EoE or PPI-REE were excluded by both distal and proximal histologic assessment. More importantly, we have previously demonstrated that the presence of neutrophils/eosinophils in distal esophagus is very common in patients with GERD, particularly in case of erosive esophagitis or true NERD, as in our present investigation, and their detection confirms the existence of microscopic esophagitis as further demonstration of GERD. Accordingly, we decided to include as controls those patients with proven GERD and distal esophageal eosinophilia in order to have a group of individuals presenting as many similarities as possible with EoE and PPI-REE patients in terms of histological findings, although they had different and distinct diagnoses (EoE, PPI-REE, GERD). We strongly believe that our rigorously selected GERD patients represented the best control group for investigating the role of GERD and motility abnormalities in EoE and PPI-RRE pathogenesis. Furthermore, we would like to underline that one of the major strength of our study is its prospective design that allowed us to use validated instruments for investigating the presence of typical reflux symptoms in our EoE/PPI-REE patients, whereas we did not find any mention of the accuracy of reflux symptoms collection in the previous studies on this argument and this drawback has necessarily affected also the following systematic reviews. This defect may have underestimated the role of GERD symptoms in supporting the diagnosis of EoE or PPI-REE. Finally, we do agree with Molina-Infante et al. that GERD may play a role in PPI-REE by being one of the triggering events in an antigen-driven oesophageal eosinophilia, as clearly reported in our investigation. Therefore, a close relationship between PPI-REE and GERD has been rightly hypothesized and our findings seem to reinforce this possible link. Anyway, there is no doubt that more studies are needed to better clarify the relationship between GERD, PPI-REE and EoE. The authors' declarations of personal and financial interests are unchanged from those in the original article.