Background: The preoperative diagnosis of intraductal papillary mucinous tumors of the pancreas must be as certain as possible because removal of a large portion of the pancreas or even total pancreatectomy may be necessary. The value of cytologic and histopathologic analysis of specimens obtained by preoperative endoscopic investigations is unknown. The aim of this study was to assess the value of such analyses of specimens obtained by EUS-guided FNA and/or biopsy, or transpapillary biopsy specimens obtained during endoscopic retrograde pancreatography for the diagnosis of intraductal papillary mucinous tumors of the pancreas and for the detection of malignancy.
Methods: Between 1992 and 2001, 42 patients (22 men, 20 women; median age 64 years) underwent surgical resection for intraductal papillary mucinous tumors of the pancreas and had preoperative pancreatic tissue sampling. In the case of isolated dilatation of pancreatic ducts, pancreatic juice was obtained by EUS-guided FNA for cytologic analysis. In the presence of a solid lesion or main pancreatic duct stenosis, biopsy specimens were obtained by EUS-guided FNA biopsy or endoscopic retrograde pancreatography, which permitted histopathologic assessment. The accuracy of cytology and histopathology was evaluated for the following: (1) positive diagnosis of intraductal papillary mucinous tumors of the pancreas and (2) assessment of malignancy, by comparison with histopathologic examination of surgical resection specimens.
Results: Cytologic analysis was performed in 19 patients; it was positive in 4 (21%) and noninformative in 15 (79%). Histopathologic analysis was performed in 23 patients; it was positive in 21 (91%) and negative in 2 (9%). Histopathologic analysis yielded a positive result in 83% of patients who did not have extrusion of mucus from a patulous papilla. The sensitivity, specificity, and positive and negative predictive values of histopathologic analysis for the diagnosis of malignancy were, respectively, 44%, 100%, 100%, and 33%. When histopathologic analysis was positive, the tumor grade was similar to that determined by final histopathologic examination in 38% of patients, whereas the grade was underestimated in 62%. No complication occurred as a result of tissue sampling.
Conclusions: The sensitivity of histopathologic analysis of EUS-guided FNA biopsy specimens or transpapillary biopsy specimens is 91% for the positive diagnosis of intraductal papillary mucinous tumors of the pancreas with a solid component, which is of particular interest as extrusion mucus from the papilla was absent in most patients. Histopathologic analysis of biopsy specimens of malignant intraductal papillary mucinous tumors of the pancreas often underestimates tumor grade. The result for cytologic analysis of juice obtained from dilated pancreatic ducts is disappointing.