Peroral pancreatoscopy and intraductal ultrasound for diagnosis of intraductal papillary mucinous tumors of the pancreas 1,2

Abstract

Intraductal papillary mucinous tumor (IPMT) of the pancreas is a rare disease. Preoperative determination of the nature (benign or malignant IPMT) and extent of the disease may be challenging. Experience in this regard is still limited. Hara et al. performed a retrospective review of their extended experience in evaluating patients with IPMT by means of peroral pancreatoscopy (POPS) and intraductal ultrasound (IDUS) over a 13-yr period. Sixty consecutive IPMT patients were included in this study (POPS performed in all and IDUS performed in 40 patients). The authors assessed tumor type (elevated vs excavated), tumor morphology as per POPS (type I: granular; type II: fish-egg like without vascular images; type III: fish-egg like with vascular images; type IV: villous type; and type V: vegetative type), maximum tumor height as determined by IDUS, and tumor extent (head vs body vs tail; main pancreatic duct vs side branches). Results obtained with POPS and IDUS were correlated and compared with surgical pathology serving as the gold standard. The ability of CT, endoscopic ultrasound, and K-ras point mutations in pancreatic juice to distinguish benign (hyperplasia or adenoma) from malignant (carcinoma in situ or invasive carcinoma) IPMT were also studied. Relapse-free and overall survival of this retrospective cohort of patients in whom treatment was guided by POPS and IDUS findings were assessed. Forty of the 60 patients evaluated had protruding lesions (67%). Among them, most malignant tumors had a POPS morphology type III, IV, or V ( P < 0.0001), with a reported sensitivity, specificity, and accuracy of 68%, 87%, and 75% for differentiating benign (hyperplasia or adenoma) from malignant (carcinoma in situ or invasive carcinoma) IPMT. Maximum tumor height as measured by IDUS (2.27 ± 1.5 mm in the benign group, and 5.96 ± 4.03 in the malignant group) was able to discriminate benign from malignant tumors ( p < 0.001). Lesions protruding 4 mm or more on IDUS were malignant in 88% of cases, and the sensitivity, specificity, and accuracy of IDUS at that cutoff was 68%, 89%, and 78%, respectively. CT and endoscopic ultrasound had a sensitivity and accuracy ranging from 32% to 65%. When positive K-ras point mutation was considered as a malignant finding, sensitivity, specificity, and accuracy reached 87%, 15%, and 61%, respectively. Only one of the 60 patients resected (1.6%) had positive margins on surgical resection after POPS and IDUS had been performed. The 3-yr relapse-free and overall survival was 93% and 95%, respectively. Based on these results, Hara et al. concluded POPS and IDUS can reliably distinguish benign from malignant IPMT, determine tumor extent, and guide therapy. These new techniques may contribute to improvement in postoperative results.