Objectives: Endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) has become widely accepted as an effective modality for obtaining tissue for primary diagnosis and staging. We have been using EUS-FNA since July 2001 and herein we summarize our experience over a 5-year period. Methods: A computer-based search for in-house EUS-FNA was performed in the pathology database from July 2001 to October 2006. To calculate the sensitivity, specificity and accuracy of EUS-FNA, the cytology diagnosis was compared with the surgical follow-up. Results: A total of 951 EUS-FNAs were performed during the study period and included 279 pancreatic solid lesions, 186 pancreatic cyst lesions, 249 lymph node aspirations, 111 gastrointestinal (GI) tract submucosal lesions, and 126 miscellaneous lesions. EUS-FNA had a very high sensitivity and accuracy for solid pancreatic lesions (94.7 and 97.7%, respectively), low sensitivity and accuracy but high specificity (47, 64.8 and 95%, respectively) for cystic lesions. Cyst fluid carcinoembryonic (CEA) levels were significantly higher in mucinous neoplasms than non-neoplastic cysts. EUS-FNA also had very high sensitivity and specificity for detecting metastatic carcinoma in lymph nodes (95 and 100%, respectively). GI submucosal spindle cell tumours were further classified with immunohistochemical stains performed either on a cell block or a core biopsy obtained via EUS guidance. Conclusions: EUS-FNA has a very high sensitivity and accuracy for pancreatic solid lesions, but the sensitivity for cystic lesions is generally low. Cyst fluid chemical analysis for CEA is helpful, but the overlap between mucinous neoplasm and non-neoplastic cysts is significant. Recognizing GI contamination is important and immunohistochemical stains are useful for GI submucosal spindle cell lesions.
- Carcinoembryonic antigen
- Endoscopic ultrasound-guided fine needle aspiration
- Pancreatic cystic lesion
- Pancreatic solid tumour
- Submucosal spindle cell tumour
ASJC Scopus subject areas
- Pathology and Forensic Medicine