Pancreatic parenchymal volume measurement for predicting invasive intraductal papillary mucinous neoplasm.

Background: Surgical resection is strongly recommended for intraductal papillary mucinous neoplasm (IPMN) cases with high-risk stigmata (HRS). However, it is challenging to make this decision for elderly patients or those with multiple comorbidities. In such cases, it may be more appropriate to consider surgery for IPMN with higher malignancy potential, such as invasive IPMN. In this study, we investigated the significance of pancreatic parenchymal volume (PV) measurement in predicting invasive IPMN.

Methods: We retrospectively utilized a database comprising 162 consecutive resected IPMN cases between January 2000 and December 2018. PV, delineated through contrast-enhanced computed tomography scans, was quantified using Ziostation2 software and stratified into high and low categories based on the median value.

Results: The median PV was 36.40 cm3, ranging from 5.07 cm3 to 87.69 cm3. Low PV significantly correlated with invasive IPMN (multivariable odds ratio, 2.63; 95 % confidence interval, 1.22-5.68; P = 0.01). The specificity and accuracy of HRS for classifying invasive IPMN increased with the addition of PV measurement (specificity; HRS to HRS and low PV: 32.2 % [38/118] to 71.2 % [84/118], accuracy; HRS to HRS and low PV: 43.9 % [74/162] to 67.9 % [110/162]). Additionally, a significant decline in PV over time was observed in cases of invasive IPMN compared to those with low-grade and high-grade dysplasia (9.09 cm3/year vs. 2.79 cm3/year, P < 0.01).

Conclusions: Quantifying PV, particularly when combined with HRS, may improve diagnostic accuracy for invasive IPMN. Furthermore, changes in PV observed during surveillance may provide additional insight into the risk of progression to invasive IPMN.