RESUMEN
BACKGROUND: Conflicting results have been recently reported for the accuracy of the Endoscopic Reference Score (EREFS), an standardised endoscopic classification, to predict the histological activity of eosinophilic oesophagitis (EoE). AIM: To evaluate the accuracy of the EREFS to predict either histological or clinical activity of EoE. METHODS: Prospective multicentre study conducted in eight Spanish centres evaluating adult EoE patients, either naïve or after treatment. Symptoms were evaluated before upper endoscopy through the Dysphagia Symptom Score, whereas researchers scored the EREFS immediately after the endoscopic procedure, unaware of the histological outcome. RESULTS: One hundred and forty-five EoE patients undergoing 240 consecutive endoscopic procedures were included. Exudates (P = 0.03), furrows (P = 0.03) and a composite score of inflammatory signs (exudates, furrows and oedema) (P < 0.001) accurately predicted histological activity. Exudates were the only endoscopic sign showing a good correlation with histological outcome after therapy. Furrows and oedema persisted in 50% and 70% of patients despite histological remission. No endoscopic feature exceeded 70% accuracy to predict histological activity. Likewise, no endoscopic finding could adequately predict dysphagia severity. Crepe paper mucosa, diffuse exudates and severe rings correlated with higher symptom scores. CONCLUSIONS: Endoscopic findings assessed by the Endoscopic Reference Score did not correlate with histological or clinical disease activity in adult EoE patients. Only exudates correlated with peak eosinophil count and histological outcome, whereas furrows and oedema persisted in over half of patients despite histological remission.