Validation of European evidence‐based guidelines and American College of Gastroenterology guidelines as predictors of advanced neoplasia in patients with suspected mucinous pancreatic cystic neoplasms. Issue 9 (22nd January 2020)

• Record Type: Journal Article
• Authors: Sun, Liqi; Wang, Wei; Wang, Yang; Jiang, Fei; Peng, Lisi; Jin, Gang; Jin, Zhendong
• Is Part Of: Journal of gastroenterology and hepatology. Volume 35:Issue 9(2020)
• Page Start: 1644
• Page End: 1651
• Publication Date: 2020-01-22
• Subjects: imaging and advanced technology/applied therapeutics -- pancreas -- pancreatic neoplasms: biology and therapy; Gastroenterology -- Periodicals; Digestive organs -- Diseases -- Periodicals; Liver -- Diseases -- Periodicals; Gastroenterology -- Periodicals; Liver Diseases -- Periodicals; 616.33
• DOI: 10.1111/jgh.14973
• Languages: English
• ISSNs: 0815-9319

Abstract

Abstract: Background and Aim: The European evidence‐based guidelines (EEG) and American College of Gastroenterology Guidelines (ACGG) have been published to guide the management of pancreatic cystic lesions. We aim to evaluate the value of both guidelines in predicting advanced pancreatic cystic lesions (A‐PCLs) with preoperatively imaging‐suspected cystic mucinous pancreatic neoplasms (cMNs). Methods: One hundred ninety‐eight patients who underwent resections from 2013 to 2019 for suspected cMNs were retrospectively reviewed. Receiver operating characteristic curves were calculated and compared with measure diagnostic value. Results: Sixty‐two patients were diagnosed with A‐PCLs pathologically. Cross‐imaging modalities had comparable diagnostic accuracy to endoscopic ultrasound in type classification and A‐PCLs prediction. Receiver operating characteristic curve comparison analyses showed that EEG absolute + MCN (EEG AM ) and EEG relative + MCN (EEG RM ) having at least one indications criteria were comparable to the ACGG ( P = 0.21 and P = 0.45). For the criteria having at least two indications, ACGG was superior to EEG AM ( P = 0.001) but comparable to EEG RM ( P = 0.12). EEG AM ≥ 1 indication criteria was superior to ≥ 2 indications criteria ( P = 0.02). EEG RM ≥ 1 indication criteria had comparable diagnostic performance with ≥ 2 indications criteria ( P = 0.86). ACGG ≥ 2 indications criteria was superior to ≥ 1 indication criteria ( P = 0.02). Conclusion: On the basis of cross‐imaging evaluations, both sets of guidelines were found to be helpful in identifying A‐PCLs in suspected cMNs with comparable performance. EEG AM ≥ 1 indication criteria was superior to ≥ 2 indications criteria. ACGG ≥ 2 indications criteria was superior to ≥ 1 indication criteria.