Predictors of Malignancies in Patients with Inconclusive or Negative Results of Endoscopic Ultrasound-guided Fine-needle Aspiration for Solid Pancreatic Masses

Background/Aims: This study analyzed the diagnostic accuracy of endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) for pancreatic solid masses in patients with or without chronic pancreatitis as well as the clinical parameters relevant to a malignancy when EUS-FNA was negative or inconclusive. Methods: A total of 97 patients, who underwent EUS-FNA for solid pancreatic masses over 2 years at a single institution, were evaluated. All patients underwent EUS-FNA for 3-5 passes with 22 or 25 G needles without an on-site cytopathologist. The final diagnosis was obtained by surgery or compatible clinical outcomes for a more than 12 month follow-up. The diagnostic yields in the patients with or without chronic pancreatitis were compared and the histories and laboratory data relevant to pancreatic ductal adenocarcinoma (PDAC) or pseudo-tumor were analyzed. Results: The final diagnoses were adenocarcinoma in 88 patients (90.7%) and inflammatory pseudo-tumor in 9 (9.3%). The results of EUS-FNA were adenocarcinoma (74), suspicious (7), atypical (5), negative (10), and inadequate specimen (1). The diagnostic accuracies were 76.9% and 91.6% in patients with or without chronic pancreatitis, respectively. Among the 23 cases with non-diagnostic results of EUS-FNA, PDAC was finally diagnosed in 5 out of 7 suspicious, 3 out of 5 atypical, and 5 out of 10 negative cytology cases. The clinical parameters related to a pseudo-tumor were a history of alcohol consumption and pancreatitis, and normal alkaline phosphatase levels. Conclusions: The diagnostic accuracy of pancreatic masses in the background of chronic pancreatitis was low. When EUS-FNA produced inconclusive results, the histories of alcohol consumption, pancreatitis, and serum levels of alkaline phosphatase are useful for making a final diagnosis.