Use of an Algo-Based Decision-Making Tool to Compare Real-Life Clinical Practice in a Single Tertiary Center with the...

Background/Objectives: Intraductal papillary mucinous neoplasms (IPMN) are the most common pancreatic cystic lesions and are established precancerous entities. Side-branch IPMN (SB-IPMN) is the most prevalent subtype and generally carries a low risk of malignant transformation. The revised 2024 Kyoto guidelines define management and surveillance strategies based on high-risk stigmata and worrisome features; however, real-life adherence to these recommendations remains variable. To compare real-world management of SB-IPMN at a tertiary medical center with Kyoto guideline-based recommendations using an AIgo-based decision-support tool. Methods: SB-IPMN cases were retrospectively analyzed. An algorithm implementing the Kyoto guidelines was used to generate recommended management strategies based on imaging, clinical, and laboratory data, and these recommendations were compared with actual clinical decisions. Long-term clinical and radiological follow-up data were collected, including development of pancreatic ductal adenocarcinoma (PDAC). Results: A total of 368 patients (69% male; median age 69.5 years) were followed for a median of 48.5 months radiologically and 64 months clinically. Median cyst size at presentation was 10 (6-14) mm. Only 58 patients (15.8%) were managed in accordance with the Kyoto guidelines; most underwent more intensive surveillance (60.3%), while 23.9% received less intensive monitoring (p = 0.04). Larger cyst size (>2 cm) was associated with higher concordance with current guidelines. Younger patients, including all patients under 50 years of age, were more frequently over-surveilled. Over-surveillance resulted in an excess of 0.42 MRI/MRCP examinations per patient-year. Only one PDAC case occurred, arising after more than five years of cyst stability. Conclusions: Fewer than 20% of patients with SB-IPMN were managed according to Kyoto guidelines. Over-surveillance was common, particularly in younger patients, without apparent oncologic benefit. AIgo-based decision-support tools may help standardize care and optimize resource utilization.