Clinical utility and diagnostic accuracy of endoscopic ultrasound-guided fine needle aspiration of pancreatic lesions: Saudi Arabian experience

Abstract

Background/aims: Endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) has emerged as a new and effective tool in the evaluation of pancreatic mass lesions. It has proven to be a safe, accurate, and reliable diagnostic procedure and should be performed in centers with experienced gastroenterologists and cytopathologists for optimum results. In this study, we elaborate on our experience of specimen adequacy criteria, overall efficiency of the technique, and the cytomorphologic features of various pancreatic lesions.

Patients and methods: EUS-FNA of the pancreas and subsequent surgical and/or clinical follow-up of all cases performed at King Khalid University Hospital, King Saud University, Riyadh, Saudi Arabia, from November 2006 to January 2010, were retrospectively reviewed. Diff-Quik and Papanicolaou staining methods were used. Immunohistochemistry stains on cell-block preparations were applied whenever available and required. A total of 43 aspirates from 42 patients were available for retrospective review, including 37 (86%) adequate/diagnostic and 6 (14%) inadequate/non-diagnostic cases. The diagnostic group included 17 (46%) malignant and 20 (54%) benign cases. Of the malignant category (17 cases), 11 (65%) were diagnosed as positive for adenocarcinoma, 4 (23%) as neuroendocrine tumor, 1 (6%) as solid pseudo-papillary neoplasm, and 1 (6%) as non-Hodgkin's lymphoma. The benign category (20 cases) included 5 (25%) cysts/pseudo-cysts, 8 (40%) non-specific pancreatitis, 4 (20%) granulomatous/tuberculous pancreatitis, 1 (5%) benign neoplasm, and 2 (10%) benign not otherwise specified, both were proven to be false-negative cases. The sensitivity, specificity and accuracy rate for the diagnosis of cancers were 74, 100, and 86%, respectively. No false-positive results were reported. All malignant cases showed characteristic cytomorphologic features that were sufficient for their diagnosis.

Conclusion: EUS-FNA in adequate samples is an efficient and accurate modality in the diagnosis of pancreatic lesions. Applying immunohistochemical studies on cell-block preparation can facilitate the final, definitive, and specific diagnosis of some difficult pancreatic neoplasms.