External validation of nomogram for predicting malignant intraductal papillary mucinous neoplasm (IPMN): from the theory to the clinical practice using the decision curve analysis model

The management of IPMNs is a challenging and controversial issue because the risk of malignancy is difficult to predict. The present study aimed to assess the clinical usefulness of two preoperative nomograms for predicting malignancy of IPMNs allowing their proper management. Retrospective study of patients affected by IPMNs. Two nomograms, regarding main (MD) and branch duct (BD) IPMN, respectively, were evaluated. Only patients who underwent pancreatic resection were collected to test the nomograms because a pathological diagnosis was available. The analysis included: 1-logistic regression analysis to calibrate the nomograms; 2-decision curve analysis (DCA) to test the nomograms concerning their clinical usefulness. 98 patients underwent pancreatic resection. The logistic regression showed that, increasing the score of both the MD-IPMN and BD-IPMN nomograms, significantly increases the probability of IPMN high grade or invasive carcinoma (P = 0.029 and P = 0.033, respectively). DCA of MD-IPMN nomogram showed that there were no net benefits with respect to surgical resection in all cases. DCA of BD-IPMN nomogram, showed a net benefit only for threshold probability between 40 and 60%. For these values, useless pancreatic resection should be avoided in 14.8%. The two nomograms allowed a reliable assessment of the malignancy rate. Their clinical usefulness is limited to BD-IPMN with threshold probability of malignancy of 40-60%, in which the patients can be selected better than the "treat all" strategy.