A Pathological Prediction Model and Scoring System Including Endoscopic Signs for Duodenal Papilla Neoplasms: A Retrospective Study.

Abstract

Aims/Background The application value of endoscopic ultrasound (EUS) in diagnosing duodenal papilla neoplasms (DPNs) remains underexplored. This study aims to evaluate the role of EUS and other clinical indicators in differentiating between benign and malignant DPNs and to establish a pathological prediction model for DPNs. Methods Clinical and imaging data of DPNs patients were collected. Least absolute shrinkage and selection operator (LASSO) regression was employed to screen independent predictors. Patients were divided into training and test cohorts. Univariate and multivariate logistic regression analyses were performed. A nomogram was developed alongside a scoring system, both of which were validated using the test cohort. Results A total of 56 benign and 95 malignant DPNs cases were included. Logistic regression analysis identified age, magnetic resonance imaging (MRI), EUS-measured size, echo intensity and papilla appearance as independent predictors of pathological diagnosis. The nomogram demonstrated a C-index of 0.876, with area under the curve (AUC) values of 0.88 and 0.82 in the training and test cohorts, respectively. The scoring system performed well, with an optimal cutoff value of 14 points. Conclusion Age, MRI, EUS size and papilla appearance are independent risk factors for malignant DPNs. EUS may have extraordinary effects in DPNs differential diagnosis.