Accuracy of Cytologic Interpretation of Pancreatic Neoplasms by Fine Needle Aspiration and Pancreatic Duct Brushings

Abstract

To investigate the accuracy of fine needle aspiration (FNA) specimens and pancreatic duct brushings in the detection of pancreatic lesions and to compare the results with follow-up biopsy and/or surgical interpretation. We reviewed a total of 57 specimens (37/20), 37 FNA specimens and 20 pancreatic duct brushings, from 45 patients treated at Froedtert Memorial Lutheran Hospital, affiliated with the Medical College of Wisconsin, Milwaukee, over a 4-year period. The FNA and brushing samples were categorized as follows: positive for malignancy (21/3 = 24), suspicious for malignancy (8/7 = 15) and atypical (8/10 = 18). The results were then correlated with the tissue diagnosis. The 24 cytologic samples positive for malignancy included 23 (20/3) pancreatic ductal carcinoma (CA) and 1 (1/0) neuroendocrine CA; in the suspicious category, 11 (6/5) were pancreatic ductal CA; 2 (0/2) mucinous neoplasms and (2/0) neuroendocrine neoplasms; in the atypical category; 2 (2/0) suggestive of mucinous neoplasia, 1 (1/0) suggestive of serous neoplasia and 9 (2/7) favor reactive; and 6 (3/3) without further categorization. Tissue diagnoses were available in 26 cases: 12 (10/2) cases positive for malignancy, 8 (5/3) suspicious for malignancy and 6 (5/1) atypical. The 12 cytologically positive cases confirmed by histology showed 10 ductal CA, 1 neuroendocrine CA and 1 negative. All 8 cases (100%) suspicious for malignancy revealed positive results, including 5 ductal CA, 1 neuroendocrine neoplasm, 1 mucinous cystic neoplasm and 1 lymphoma. Of the 6 atypical lesions, 1 showed ductal CA, 2 mucinous cystic neoplasm and 3 chronic pancreatitis. Pancreatic FNA and duct brushings [table: see text] are accurate methods in identifying pancreatic lesions, particularly ductal CA. Accuracy can be improved in the case of mucinous and other lesions with adequate cellularity of the smear and recognizing the limitations of brush samples in the case of mucinous cystic lesions. False negative results may occur in cases of poor representation of malignant cells or poor sampling.