The significance of radiologic main duct dilatation in pancreatic intraductal papillary mucinous neoplasms

Abstract

Background: Intraductal papillary mucinous neoplasms are cystic neoplasms of the pancreas with a risk of malignant transformation. They are categorized based on morphology and ductal involvement. Compared with side-branch intraductal papillary mucinous neoplasms, main duct intraductal papillary mucinous neoplasms and mixed-type intraductal papillary mucinous neoplasms are generally considered high risk. Radiologic main pancreatic duct dilatation is a recognized risk factor for high-grade dysplasia and/or invasive carcinoma. Studies thus far have not explored the association between radiologic main pancreatic duct dilatation and dysplasia of the main pancreatic duct epithelium on surgical pathology.

Methods: An institutional database of intraductal papillary mucinous neoplasms was queried for patients from 1997 to 2023 undergoing resection of a main duct or mixed-type intraductal papillary mucinous neoplasm with radiologic diagnosis of main pancreatic duct dilatation before surgery. The χ² test, t test, logistic regression modeling, and Youden's index analysis were performed to test the association between "true" main pancreatic duct dysplasia involvement in surgical pathology reports and main pancreatic duct dilatation, as well as high-risk pathology (high-grade dysplasia or invasive carcinoma).

Results: A total of 91 patients were included in the final analysis. Radiologic main duct intraductal papillary mucinous neoplasms were more likely to demonstrate "true" main duct involvement versus mixed-type intraductal papillary mucinous neoplasms (79% vs 41%, P = .001). Among intraductal papillary mucinous neoplasms with true main pancreatic duct dysplasia, main duct intraductal papillary mucinous neoplasms were more likely than mixed-type intraductal papillary mucinous neoplasms to demonstrate invasive carcinoma on surgical pathology (27% vs 10%, P = .047). Cyst size was not predictive of main pancreatic duct dysplasia for mixed-type intraductal papillary mucinous neoplasms. Main pancreatic duct >9 mm at diagnosis was associated with increased likelihood of true main pancreatic duct dysplasia (odds ratio = 3.33, 95% confidence interval = 1.16-9.61 P = .026). Youden's index analysis demonstrated that a main pancreatic duct cutoff of 8.5 mm provided the greatest capability to predict main pancreatic duct dysplasia.

Conclusion: Mixed-type intraductal papillary mucinous neoplasm confers less dysplasia risk in the main pancreatic duct, supporting maintaining the distinction of mixed-type and main duct intraductal papillary mucinous neoplasm due to potentially different management. This study reaffirms the use of main pancreatic duct dilatation in intraductal papillary mucinous neoplasm risk stratification, because main pancreatic duct dilatation >8 mm was predictive of main pancreatic duct epithelial dysplasia and high-risk pathology.