Eosinophilic Oesophagitis and Proton Pump Inhibitor-Responsive Oesophageal Eosinophilia Have Similar Clinical, Endoscopic and Histological Findings

F. J. Moawad; A. M. Schoepfer; E. Safroneeva‡, M. R. Ally; Y.-J. Chen; C. L. Maydonovitch; R. K. H. Wong


Aliment Pharmacol Ther. 2014;39(6):603-608. 

In This Article


EoE is an increasingly recognised entity, which has become the leading cause of dysphagia.[10] The clinical presentation and endoscopic features are often dramatic and gastroenterologists may initiate treatment with topical steroids prior to a course of PPIs,[11] a step necessary to establish the diagnosis of EoE.[2,5] Our study demonstrates that EoE and PPI-REE are indistinguishable without a PPI trial with regard to clinical presentation, endoscopic findings and histology based on data from two study sites. Furthermore, many patients with suspected EoE may respond to PPI, therefore obviating the need for topical steroids and specialised diets.[12]

PPI-REE is a newly described entity initially reported in 3 paediatric patients with suspected EoE who achieved clinical and histological response to an 8-week course of twice daily omeprazole.[13] Although PPI was not considered primary therapy for EoE in the initial guidelines,[14] it appeared that many children and adults with this condition reported improvement in symptoms. Additionally, histological response to PPI became more apparent in patients presenting with typical EoE symptoms and endoscopic findings, as well as having marked levels of eosinophils on biopsies. Therefore, the term PPI-REE was recognised in the latest expert consensus statement.[2] What remains unclear is whether this entity should be defined clinically, histologically, or requires both criteria.

Clinical response to PPI in EoE patients has been described in multiple studies; however, several of these studies were performed prior to the latest consensus statement and may have included PPI-REE patients. In one study, nearly 50% of adult patients presenting with a food impaction and having dense eosinophilia on oesophageal biopsies reported clinical improvement with acid suppression.[15] Similar results to a PPI were demonstrated in children with significant oesophageal eosinophilia despite a negative pH study.[16] Notably, in a recent study, baseline pH testing was not predictive of PPI response.[17]

Even more intriguing than achieving clinical response in suspected EoE patients is the presence of histological response, especially in the absence of classic reflux symptoms, negative pH testing and absence of erosive oesophagitis on endoscopy. To date, several prospective studies have reported such findings.[6,9,18,19] In one randomised controlled trial, histological response was achieved in approximately one-third of patients with suspected EoE with once-daily PPI for 8 weeks.[6] Another prospective study reported a 50% response in patients with an EoE phenotype after treatment with twice daily PPI.[18] In 60 adult patients presenting with symptoms of oesophageal dysfunction and dense oesophageal eosinophilia treated with twice-daily PPI for 8 weeks, histological improvement was demonstrated in more than 50% of the cohort to include more than one-third achieving complete resolution.[19] When stratified by symptom profile (GERD vs. EoE), there was no significant difference in response to PPI. In other words, patients with suspected EoE achieved a similar response to PPI compared with patients presenting with GERD symptoms. Most recently, in a prospective study evaluating the prevalence of PPI-REE, response to PPI was seen in one-third (24/66) of patients presenting with dysphagia and marked oesophageal eosinophilia.[9]

Several theories exist to explain the mechanism behind PPI response in oesophageal eosinophilia. First, PPIs can heal a disrupted epithelial barrier and normalise intercellular spaces, thereby preventing the permeability of food allergens.[20,21] This theory would support that PPI-REE may be a subset of GERD. It remains unclear, however, why patients with physiological oesophageal acid exposure on pH studies would develop such a degree of damage to promote antigen exposure. Another theory can be attributed to anti-inflammatory properties of PPI, independent of acid inhibition. PPIs have been shown to exhibit anti-inflammatory properties by acting directly on principal cytokines (IL-4 and IL-13) involved in the recruitment of eosinophils in the oesophagus.[22] Recently, omeparazole was shown to block the expression of TH2 cytokine-stimulated expression of eotaxin-3 in cultures from GERD and EoE patients.[23] Additionally, in one study, treatment with a PPI reduced levels of eotaxin-3 and TH2 cytokines in PPI-REE patients similar to what was observed in EoE following treatment with topical steroids.[24] This is suggestive of PPI-REE representing a subset of EoE patients who respond to PPI. The reason why some patients respond and others do not needs to be further explored. Interestingly, two studies in children raised questions whether PPI responsiveness may represent a transient phenomenon and dense eosinophilia may recur over time.[25] Larger studies are currently underway to address this observation.

Strengths of this study include combining data from two institutions which have comprehensive EoE registries. Patients were generally homogenous between the study sites and there was a similar proportion of PPI-REE patients between the two centres. However, there were some differences between patients of the two study sites. For instance, mean duration of symptoms was longer and was more variable in Walter Reed patients, while white plaques and strictures were more common among Swiss patients. These differences did not affect the results of the combined data as there were no significant differences between EoE and PPI-REE patients in duration of symptoms, white plaques and strictures within each study site. Limitations of this study include its retrospective design and the lack of a validated definition of PPI-REE. We chose to combine an absolute number of eosinophils as less than 15 eos/hpf and a decrease of 50% from baseline. This definition was to ensure that there was a significant change in eosinophils count from baseline.

In conclusion, the results of this study demonstrate that EoE and PPI-REE are similar in clinical, histological and endoscopic features and therefore cannot be distinguished without a PPI trial. Further studies are needed to determine why a subset of patients with oesophageal eosinophilia respond to PPI.

Explore product updates from manufacturers, available now
Information from Industry
Explore product updates from manufacturers, available now
Information from Industry
Explore product updates from manufacturers, available now
Information from Industry