Gastrointestinal lesions of eosinophilic granulomatosis with polyangiitis: a prediction model and clinical patterns

• Published in: Arthritis research & therapy Vol. 27; no. 1; pp. 3 - 13
• Main Authors: Liu, Suying, Yang, Yunjiao, Han, Linna, He, Chengmei, Li, Mengtao, Tian, Xinping, Meng, Juan, Wang, Li, Zhang, Fengchun
• Format: Journal Article
• Language: English
• Published: England BioMed Central Ltd 04.01.2025; BioMed Central; BMC
• Subjects: Brain-gut; Eosinophilic gastroenteritis; Eosinophilic granulomatosis with polyangiitis; Eosinophils; Gastrointestinal; Gastrointestinal diseases; Health aspects; Medical research; Medicine, Experimental; Physiological aspects; Prediction; Wegener's granulomatosis; China; Gastrointestinal; Eosinophilic granulomatosis with polyangiitis; Brain-gut; Prediction; Eosinophilic gastroenteritis
• ISSN: 1478-6362; 1478-6354; 1478-6362
• DOI: 10.1186/s13075-024-03467-7

Severe gastrointestinal lesions are associated with a poor prognosis in eosinophilic granulomatosis with polyangiitis (EGPA). The goal of this study was to develop an effective predictive model for gastrointestinal lesions and to examine clinical patterns, associated factors, treatment, and outcomes of gastrointestinal lesions in EGPA. We retrospectively enrolled 165 EGPA patients. The independent associated factors were analyzed using multivariate logistic regression. A nomogram was conducted to quantify the predictive factors. The correlation between different organ lesions was calculated to explore the clinical patterns. A total of 52 patients had gastrointestinal lesions, and 22 developed severe disorders. Common manifestations included abdominal pain (78%), diarrhea (40.4%), and nausea and/or vomiting (32.7%). Severe gastrointestinal lesions included hemorrhage (26.9%), ulcers (17.3%), obstruction (9.6%), and pancreatitis (5.8%). Eosinophilic tissue infiltration, weight loss, and myalgia were independently associated with gastrointestinal involvement. Patients with severe gastrointestinal lesions had a shorter duration from initial symptoms to EGPA diagnosis, less frequent asthma, and ear-nose-throat involvement, and were more likely to receive methylprednisolone pulse. Weight loss, central nervous system involvement, myalgia, and eosinophilic tissue infiltration were retained in the nomogram. An eosinophil ratio of over 19.2% identified gastrointestinal lesions. Significantly more patients with gastrointestinal involvement had a Five Factor Score ≥ 2. Five well-defined clinical models were identified, including the brain-gut pattern. Severe gastrointestinal lesions are common in EGPA and early detection is critical. Eosinophils are an important factor associated with gastrointestinal involvement of EGPA. We developed a model to predict the risk of gastrointestinal lesions. The brain-gut pattern might deserve further investigation in EGPA.