Role of CT in detecting malignancy during follow-up of patients with branch-type IPMN of the pancreas

Hepatogastroenterology. 2009 Mar-Apr;56(90):515-8.

Abstract

Background/aims: This retrospective study evaluated the suitability of computed tomography (CT) to detect malignancy while following patients with branch-type IPMN, most of which are benign and may be treated with observation alone.

Methodology: Forty-two surgical specimens resected from patients with a diagnosis of branch-type IPMN were pathologically classified as benign (n=26), which included hyperplasia and adenoma, or malignant (n=16), including moderate dysplasia or adenocarcinoma. It was compared the differences in the sizes of the tumor and main pancreatic duct (MPD) and the presence of mural nodules on CT between the groups.

Results: In the malignant group, it was observed a larger tumor size (47.8 vs. 23.8 mm; p = 0.001) and increased dilation of the MPD (9.3 vs. 5.0 mm; p = 0.001) than those seen in the benign group. The accuracy of CT diagnosis of mural nodules, however, was only 62%. Tumor diameter > or =40 mm or MPD diameter >10 mm predicted malignancy with a sensitivity and negative predictive value of 93.8% and 95.7%, respectively.

Conclusions: Either tumor size or MPD dilation detected by CT could predict the majority of malignant branch-type IPMNs. Increases in these morphological characteristics on CT images during the follow-up period would be an accurate method to predict a diagnosis of malignancy.

MeSH terms

  • Adenocarcinoma, Mucinous / diagnostic imaging*
  • Adenocarcinoma, Mucinous / pathology
  • Aged
  • Carcinoma, Pancreatic Ductal / diagnostic imaging*
  • Carcinoma, Pancreatic Ductal / pathology
  • Carcinoma, Papillary / diagnostic imaging*
  • Carcinoma, Papillary / pathology
  • Female
  • Humans
  • Male
  • Pancreatic Neoplasms / diagnostic imaging*
  • Pancreatic Neoplasms / pathology
  • Retrospective Studies
  • Statistics, Nonparametric
  • Tomography, X-Ray Computed / methods*