Endoscopic Ultrasound-Guided Fine-Needle Aspiration of the Pancreas: A Retrospective Study of 1,000 Cases
Ashley A Ibrahim, Harvey M Cramer, Howard H Wu. Indiana University, Indianapolis, IN
Background: Although endoscopic ultrasound (EUS)-guided fine-needle aspiration (FNA) is an established method for the diagnosis of pancreatic tumors, only a few large studies have appeared in the literature.
Design: A computerized search of our pathology LIS was performed and all EUS-guided FNAs of the pancreas performed for the 7-year period from 2004 through 2011 were identified. Each cytologic diagnosis was classified as cystic or solid and placed into one of 6 categories: benign (B), atypical (A), suspicious (S), malignant (M), tumor (T) and non-diagnostic (ND). A complementary search for all related surgical pathology reports was also performed and stratified in the same manner. For the purpose of determining the degree of correlation, all non-diagnostic and atypical diagnoses were excluded from consideration and suspicious, malignant, and tumor diagnoses were categorized as positive.
Results: A total of 1,000 EUS-guided FNAs of the pancreas were identified. Of the cases, 579 were solid lesions obtained from 301 female and 278 male patients with an age range of 19-95 years (mean 63). The FNA diagnoses of the solid lesions were: B 229 (39.5%), A 22 (3.8%), S 27 (4.7%), M 260 (44.9%), T 1 (0.2%), and ND 40 (6.9%). The malignant FNA diagnoses included 209 adenocarcinomas, 21 metastases, 10 pancreatic neuroendocrine tumor (PNET), 6 solid pseudopapillary tumors, 2 lymphomas, 1 malignant giant cell tumor, 1 anaplastic carcinoma and 1 squamous cell carcinoma. Of the 579 FNAs of solid pancreatic masses, 145 had corresponding surgical follow-up. The sensitivity and specificity for solid lesions were 91.5% and 100% respectively. There were 9 false negative cases due to sampling error and specimen hypocellularity. There were 421 cystic lesions from 254 female and 167 male patients with an age range of 20-96 years (mean 63). The FNAs of the cystic lesions were classified as follows: B 343 (81.5%), A 5 (1.2%), S 5 (1.2%), M 7 (1.7%), T 44 (10.5%), and ND 17 (4.0%). Of the 421 cystic FNA cases, 96 cases had corresponding surgical follow-up. There were 40 cystic neoplasms failed to be diagnosed by FNA including 1 adenocarcinoma, 2 PNETs, 6 MCN, 8 serous cystadenomas and 23 IPMN. The sensitivity and specificity to identify tumor for cystic lesions were 34.4% and 93.3% respectively.
Conclusions: At our institution, EUS-guided FNA of solid pancreatic masses is both sensitive and specific for the diagnosis of both primary and metastatic tumors. For cystic lesions, FNA is not sensitive in identifying low-grade IPMN, MCN or serous cystadenoma, but its specificity remains high.
Tuesday, March 5, 2013 1:00 PM
Poster Session IV # 81, Tuesday Afternoon