Supplementary Material for: Assessing the accuracy of the international evidence-based Kyoto guidelines for detecting malignancy in intraductal papillary mucinous neoplasms of the pancreas

Introduction
Intraductal papillary mucinous neoplasms (IPMNs) are pancreatic tumours with an associated risk of malignant transformation. Due to the widespread use of imaging techniques, the diagnosis of IPMNs has been rising. The International Evidence-Based Kyoto Guidelines (IKG), the latest update of the International Guidelines, were developed to refine the management of IPMNs. IKG incorporate high-risk stigmata and worrisome features into the decision-making process. This study evaluates the accuracy of these guidelines in identifying patients requiring surgery.
Methods
Single-centre retrospective cohort study conducted at a University Hospital, including adult patients who underwent pancreatic surgery for IPMNs between 2010 and 2024. Data on demographic characteristics, imaging findings, surgical procedures, and histopathological outcomes were collected. Criteria for surgery, according to the IKG, were retrospectively applied. The diagnostic performance of IKG was assessed through sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy.
Results
One hundred and six patients (mean age: 67±10 years) were included, and 18.9% had histologically confirmed malignant IPMNs. The sensitivity of IKG in detecting malignancy was 100%, while the specificity was 19.8%. The positive negative predictive values were 22.5% and 100%, respectively. The most predictive factors for malignancy were the presence of lymphadenopathy, abrupt changes in main pancreatic duct calibre, and elevated serum carbohydrate antigen 19-9. An association was observed between the number of worrisome features and the malignancy rate.
Discussion/Conclusion
The IKG demonstrated excellent sensitivity in selecting patients for surgery, ensuring that all malignant cases were identified. However, the low specificity suggests the need for criteria revision to minimise overtreatment. Future guidelines should focus on integrating patient-specific factors, such as age and comorbidities, into decision-making algorithms to optimise management strategies and to consider the weight and number of worrisome features.