Identification of clinical and laboratory markers for predicting eosinophilic esophagitis in adults.

BACKGROUND: The diagnosis of eosinophilic esophagitis (EoE) and differentiation from gastroesophageal reflux disease (GERD) is potentially challenging and is based upon clinical signs and endoscopic and histological features. In order to alert the endoscopist to consider EoE in patients with esophageal symptoms before performing esophagogastroduodenoscopy, we aimed to identify a set of clinical and laboratory markers for predicting EoE.
METHODS: The study included 43 patients with either EoE (n = 23) or GERD (n = 20). The diagnosis of EoE was based on International Consensus Criteria. Age, gender, weight loss, history of atopy, dysphagia, history of food impaction, proton pump inhibitor (PPI) refractory heartburn, odynophagia, peripheral eosinophilia, and serum IgE were analyzed. Each symptom or sign was classified as '0' (absent, normal) or '1' (present, elevated), individually analyzed and statistically evaluated among the two groups of patients. Logistic regression analysis was carried out to identify a clinically applicable marker constellation to differentiate EoE from GERD.
RESULTS: Univariate analysis identified 6 out of the 10 variables to be significant between both groups. A stepwise procedure of logistic regression led to a model in which 3 out of the initial 10 items were found to be relevant for differentiating GERD and EoE. Derived from this model, an optimal differentiation was achieved by using the following simplified equation: peripheral eosinophilia + history of food impaction + PPI refractory heartburn leading to a maximal value of 3 (1 + 1 + 1). Based on a cut-off value of ≥2, sensitivity and specificity for diagnosing EoE were 91 and 100%, respectively.
CONCLUSION: A defined set of markers including two clinical features and one laboratory parameter is highly predictive of EoE and thus allows physicians to distinguish EoE from GERD even before upper gastrointestinal endoscopy is performed.