Diagnostic Performance of Endoscopic Ultrasound-Guided Tissue Acquisition by EUS-FNA versus EUS-FNB for Solid Pancreatic Mass Without ROSE: A Retrospective Study.

Abstract Background: Endoscopic ultrasound-guided tissue acquisition (EUS-TA) is an established diagnostic procedure for solid pancreatic mass. However, the diagnostic yield between fineneedle aspiration (FNA) and fine-needle biopsy (FNB) remains unclear. Thus, we aimed to evaluate and compare the diagnostic yields between FNA and FNB using conventional FNA and Franseen needles of the same size (22 gauge), respectively, in patients with solid pancreatic mass who underwent EUS-TA without rapid onsite cytopathology evaluation (ROSE). Methods: All cases of EUS-TA by FNA or FNB for solid pancreatic mass between January 2017 and October 2020 in a single center university hospital were retrospectively reviewed. The procedure was performed without an onsite cytologist. After macroscopic onsite evaluation (MOSE), the endoscopist finished the procedure. The diagnostic yield and the average number of needle passes between FNB and FNA were then compared. Results: A total of 151 patients (FNA, n = 77; FNB, n = 74) with solid pancreatic mass detected by cross-sectional imaging underwent EUS-TA. The mean age was 62.3 ± 12.8 years, with 88 (58.3%) males. Age, gender, mass location, tumour size and disease stage from imaging were not significantly different. The diagnostic performance was dramatically higher in EUS-FNB (100%) than in EUS-FNA (89.6%). The mean number of needle passes was clearly fewer in FNB than FNA (2.8 vs. 3.8, p < 0.001). The total procedure time was less in FNB (34.7 minutes) than in FNA (41 minutes). The adverse event rate between FNB and FNA was not significantly different. Conclusions: The diagnostic performance for solid pancreatic mass without ROSE was significantly higher in FNB than in FNA. The number of needle passes and the total procedure time was also lesser in FNB.