Objective To construct a prediction model for the recurrence risk of eosinophilic gastroenteritis (EGE) and analyze the clinical characteristics of EGE patients.

Methods Sixty-nine patients with EGE were selected as the research subjects and divided into recurrent patients and non-recurrent patients according to the recurrence status. Clinical data of EGE patients were collected. Logistic regression analysis was employed to screen factors influencing recurrence. A prediction model was then constructed and validated based on the identified influencing factors.

Results The most common symptoms and signs in EGE patients were abdominal pain and abdominal tenderness. The initial diagnosis age of relapsed patients was significantly lower than that of non-relapsed patients, and the incidence of nausea/vomiting and the utilization rate of glucocorticoids were significantly higher than those of non-relapsed patients (P < 0.05). The age at initial diagnosis was an influencing factor for recurrence in patients (OR=0.923, 95%CI, 0.877 to 0.971). The area under the curve for predicting recurrence by the age at initial diagnosis was 0.809 (95%CI, 0.690 to 0.928), the optimal cut-off value was 35.50 years old, the sensitivity was 66.70%, and the specificity was 88.20%.

Conclusion Young patients with EGPA have a high risk of recurrence. The age at initial diagnosis can be used as a predictive indicator, and glucocorticoid treatment may affect the risk of recurrence.