Background/aims: IPMT (intraductal papillary-mucinous tumor) of the pancreas has unique clinicopathological characteristics. The lesions which show characteristic clinical features of IPMT exhibit a wide spectrum of histological types ranging from atypical hyperplasia to invasive cancer. Therefore, surgical treatment cannot be recommended for all patients with IPMT. It is necessary to assess the malignant potential of IPMT in individual patients in order to select an appropriate approach. The aim of this study was to evaluate the effectiveness of endoscopic ultrasonography and intraductal ultrasonography as compared with ultrasonography and computed tomography for this purpose.
Methodology: Ultrasonography, computed tomography, endoscopic ultrasonography and intraductal ultrasonography were performed in 49 cases of IPMT (atypical hyperplasia 7, adenoma 23, noninvasive 7 and invasive adenocarcinoma 12). On the basis of the histopathological analysis of another 28 cases of resected IPMT specimens, criteria for differential diagnosis by imaging modalities were defined as follows: Nonneoplastic lesion (atypical hyperplasia): no wall thickening or nodule; noninvasive IPMT (adenoma and intraductal carcinoma): a nodule or wall thickening is present; and invasive IPMT with pancreatic parenchymal invasion: a mass with a heterogenous pattern or interruption of the pancreatic duct wall by the mass.
Results: The diagnostic accuracy rate for differentiating nonneoplastic lesion noninvasive IPMT, and invasive IPMT was 33% by ultrasonography, 38% by computed tomography, 77% by endoscopic ultrasonography, and 67% by intraductal ultrasonography. Sensitivity, specificity and accuracy rates for differentiating neoplastic and nonneoplastic IPMT by ultrasonography was 33%, 100%, 42%, by computed tomography 36%, 100%, 44%, by endoscopic ultrasonography 90%, 71%, 88%, by intraductal ultrasonography 94%, 29%, 84%, respectively. Sensitivity, specificity and accuracy rates for differentiating invasive and noninvasive IPMT by ultrasonography was 25%, 100%, 80%, by computed tomography 33%, 100%, 83%, by endoscopic ultrasonography 55%, 97%, 88%, by intraductal ultrasonography 56%, 91%, 84%, respectively. Diagnostic accuracy for invasive IPMT except minimally invasive cases by endoscopic ultrasonography and intraductal ultrasonography was 80%, based on the results of the examination which demonstrated a higher grade lesion.
Conclusions: With these criteria, ultrasonography and computed tomography showed high specificity, but low sensitivity for the differential diagnosis of neoplastic/nonneoplastic and invasive/noninvasive IPMT. However, endoscopic ultrasonography and intraductal ultrasonography had high sensitivity and diagnostic accuracy for the differential diagnosis of neoplastic/nonneoplastic lesions. Combination of endoscopic ultrasonography and intraductal ultrasonography showed a high accuracy rate in the diagnosis of invasive IPMT. Thus endoscopic ultrasonography and intraductal ultrasonography contributed significantly to the choice of the treatment for IPMT.