• Eosinophilic esophagitis
• dysphagia
• atopy
• bolus impaction

Backgrounds and aims:
Eosinophilic esophagitis (EoE) is a chronic disease of the esophagus characterized by symptoms of esophageal dysfunction, including dysphagia and food impaction, which frequently presents in combination with atopy. The diagnosis requires at least 15 eosinophils per high-power field (eos/HPF) on at least one esophageal biopsy collected during an esophagogastroduodenoscopy (EGDS). Screening strategies to reduce the diagnostic delay of EoE are lacking. We investigated if clinical characteristics could be used for the screening of EoE in patients with dysphagia.

Methods: We included patients referred for dysphagia to the Esophageal Diseases Clinic of Pisa and Padua who had undergone an EGDS to rule out an obstructive neoplasia within six months. All patients were screened for EoE with an EGDS with at least four esophageal biopsies. Patients without EoE (eos/HPF<15) underwent esophageal high-resolution manometry (HRM) with or without reflux monitoring to investigate dysmotility or gastroesophageal reflux disease (GERD). Consecutive patients with EoE complaining of dysphagia and controls with non-EoE dysphagia (NED) were enrolled between October 2021 – April 2022 in a 1:1 ratio. For each patient, a detailed clinical history including age, sex, interval between symptoms onset and conclusive diagnosis (i.e., diagnostic delay), history of food impaction, allergic rhinitis, asthma, atopic dermatitis, and autoimmunity were recorded. The differences in clinical characteristics between EoE and controls were assessed using the Student’s t-test, Mann-Whitney U test, or Chi-squared test when appropriate. Normality and heterogeneity of variances were investigated with Shapiro-Wilk and Levene’s test. A Monte-Carlo approach was used to investigate the optimal combination of relevant clinical features that provided a diagnosis of EoE. Significance threshold was set at p <0.05.

Results:
Twenty patients with EoE (mean age 47±17 years, 16M) and 20 NED (mean age 54±18 years, 10M) were enrolled. The median diagnostic delay was 6 years (1.5-10.5) for EoE, and 3 years (1.3-6) for NED. Among patients with NED, six had achalasia, five had a normal HRM (two GERD, one either myasthenia gravis, myotonic dystrophy, or no found cause), three had esophagogastric junction outflow obstruction, three hypercontractile esophagus, two absent peristalsis, and one distal esophageal spasm. Patients with EoE were significantly younger (p<0.01), more frequently male (p=0.04), more frequently reported a history of allergic rhinitis (p<0.001) and food bolus impaction (p=0.004) compared to NED. In contrast the two groups did not differ in terms of asthma, atopic dermatitis, and autoimmunity history. A Monte-Carlo analysis of the clinical characteristics that differed significantly between the two groups generated over 50 predictive models for the screening of EoE. Among these, the model investigating male sex and/or the presence of allergic rhinitis performed best , with 100% sensitivity and 50% specificity for a diagnosis of EoE.

Conclusion:
Performing an EGDS with multiple esophageal biopsies in a patient referred for non-neoplastic dysphagia, who is male or has allergic rhinitis, identifies 100% of EoE cases and avoids unnecessary biopsies in 50% of NED. Despite this was a small cohort, the homogeneity of the two groups and the significance obtained strongly suggest that age, gender, history of rhinitis and bolus impaction segregate EoE from NED. Studies with larger cohorts of patients are needed to validate a clinical history-based screening for EoE.