THE CLINICAL UTILITY OF ENDOSCOPIC ULTRASOUND-GUIDED FINE-NEEDLE ASPIRATION IN THE DIAGNOSIS AND STAGING OF PANCREATIC-CARCINOMA

Background: Endoscopic ultrasound (EUS) guided fine-needle aspiration (FNA) of pancreatic lesions is being increasingly used. Our aim was to determine the safety, accuracy, and clinical utility of EUS-guided FN A in both the diagnosis and staging of pancreatic cancer. Methods: For ty-four patients (24 men/20 women) had EUS-guided FNA of pancreatic le sions (39 head/neck, 5 body, 3 tail) and/or associated lymph nodes. Th e mean age was 61 (range, 28 to 88 years). The indication for EUS-guid ed FNA was a pancreatic lesion seen initially on CT (39%), ERCP (43%), or EUS (18%). Follow-up data were collected on all patients for mean of 14.5 months (range 1 to 33 months). Results: CT detected only 15 of 61 (25%) focal lesions seen by EUS. Adequate specimens were obtained by EUS-guided FNA in 44 of 47 (94%) pancreatic lesions and 14 of 14 (1 00%) associated lymph nodes (overall adequacy was 95%). Of the 46 lesi ons in which specimens were adequate and a final diagnosis was availab le (32 malignant, 14 benign), EUS-guided FNA had a sensitivity of 92%, specificity of 100%, and diagnostic accuracy of 95% for pancreatic le sions and 83%, 100%, and 88% for lymph nodes, respectively. Six percen t of pancreatic cases had inadequate specimens and, if included, lower ed the sensitivity to 83%, specificity to 80%, and diagnostic accuracy to 88% for pancreatic lesions. In 3 patients with enlarged celiac nod es on EUS, EUS-guided FNA was able to make a tissue diagnosis of metas tasis, which changed the preoperative staging and precluded surgery. E US in combination with EUS-guided FNA precluded surgery in 12 of 44 (2 7%) and may have Precluded surgery in an additional 6 of 44 (14%). EUS -guided FNA avoided the need for further diagnostic tests, thus expedi ting therapy in a total of 25 (57%) patients and influenced clinical d ecisions in 30 of 44 (68%) patients. The estimated cost savings based on surgeries avoided was approximately $3300 per patient. There was on ly one complication (2%), a post-FNA fever. Conclusion: EUS-guided FNA of the pancreas appears to be a safe and effective method that increa ses both the diagnostic and staging capability of EUS in pancreatic ca ncer. The clinical impact of EUS-guided FNA includes avoiding surgery and additional imaging studies with a substantial cost savings.