Diagnostic Yield of a Conventional Endoscopic Ultrasound-Fine Needle Aspiration (EUS-FNA) and Two Endoscopic Ultrasound-Fine Needle Biopsy (EUS-FNB) Needles: A Single-Center, Retrospective Study

Abstract

Introduction: EUS-FNA is considered to be the standard approach to sample solid lesions adjacent to the gastrointestinal tract. EUS-FNB needles have been gaining widespread popularity due to the increase in diagnostic yield associated with these needles. Aim: To evaluate and compare the diagnostic yield of the solid lesions sampled using an FNA (Echtotip Ultra® (EC), Cook Medical, IN, USA) and two FNB (Acquire ® (AC), Boston Scientific, MA, USA and Shark-Core (SC), Medtronic, CA, USA) needles. Methods: A retrospective review of consecutive patients undergoing an index EUS procedure for evaluation of a solid lesion by single provider with experience of over 1000 EUS procedures was performed. Demographic and procedure related data including location, lesion size, anesthesia type, needle type number of passes, presence of rapid onsite cytologic evaluation (ROSE), diagnostic yield, adverse events and final diagnosis was collected. Statistical analysis of the obtained data set was performed using Stata 15.1. Chi square and Logistic regression analysis was performed. Results: A total of 329 lesions were sampled between July 2016 and May 2018 (Table 1). The overall diagnostic yield was 93.6%. The diagnostic yield of the EC needle was significantly lower than both the SC and AC needles (86.2 vs 96 and 94.9% respectively). The diagnostic yield for lesions assessed using more than one needle in the same procedure (due to non-diagnostic yield with the first needle) was 100% however statistical significance. The odds of obtaining a higher yield with SC compared to EC was 3.7 (p=0.034) however the odds did not reach statistical significance for the AC needle compared to the EC needle. (OR 3.4, p=0.053). Conclusion: EUS-FNB needles had an overall better diagnostic yield than the EUS-FNA needle however only the diagnostic yield with the SharkCore ® EUS-FNB needle reached statistical significance while there was a trend towards a higher diagnostic yield with the Acquire ® EUS-FNB. There was no difference in diagnostic yield between the two EUS-FNB needles. Although not statistically significant due to a small sample size, an approach of switching to a second needle was associated with a diagnostic yield of 100%.796_A Figure 1. Demographic and procedure related data796_B Figure 2. Diagnostic yield based on needle type and site of lesion