Development and validation of a risk score for incomplete resection during endoscopic papillectomy: PANETH score

Objectives: Endoscopic papillectomy (EP) is the gold standard treatment for ampullary adenomas. However, EP is still burdened by a nonnegligible rate of incomplete resections (IR). Different predictors have been linked to higher rates of IR, but the interaction between these factors is still unclear. The aim of the study was to develop a scoring system (hereby called PANETH score) able to quantify the risk of IR after EP. Methods: Patients who underwent EP in 19 Italian centers in 2016-2021 were included. IR was defined as the presence of residual tumor in lateral or endoampullary margins after EP. Predictors for IR were analyzed by logistic regression and were used to obtain an easy-to-use numeric score. The performance of the model was evaluated with a receiver operating characteristic curve analysis and tested by means of 10-fold cross-validation. Results: A total of 430 patients were included. On multivariate analysis, laterally spreading tumor (odds ratio [OR] 5.81, 3.21-7.65; P = 0.02), intraductal extension (OR 6.92, 3.33-9.87; P < 0.0001), and bile duct dilation (OR 2.61, 1.22-4.32; P = 0.004) were significant predictors of IR. The score was calculated by the sum of regression coefficients of each predictor. A ≥3 score indicated a 4-fold risk of IR (P < 0.0001). The internal validation resulted in an area under the curve of 0.83 and an overall error rate of 0.11. Conclusions: The proposed PANETH score may represent a reliable and easily applicable tool to predict the risk of IR after EP to optimize patient selection and risk stratification.