A Prospective, Blinded, Randomized Trial Assessing the Yield of Endoscopic Ultrasound Guided Fine Needle Sampling (EUS-FNS) of Solid Lesions with Suction Versus No Suction

Background: Attempts to improve diagnostic yield with endoscopic ultrasound guided fine EUS-FNS include refinements in needle design, needle size, and whether to use suction vs. no suction. To date, only adenopathy has been rigorously studied prospectively to determine optimal technique with regards to the use of suction (negative pressure applied to needle). Aim: Determine if suction or no suction offers a diagnostic advantage in solid lesions undergoing EUS-FNS. We hypothesize that the diagnostic yield with suction will be superior to no suction. Methods: Pts undergoing EUS-FNS of solid lesions from 2/15/08 and 11/15/08 were prospectively randomized (blocked) to the order of suction/no-suction. All lesions underwent 4 passes (2 with 10ml suction, 2 without suction) using the 22 gauge EchoTip® Ultra needle (Cook Medical). On-site cytotechnician and pathologist were blinded to the sampling technique. Data on gross appearance, cellularity and diagnosis were recorded. Diagnostic yield was defined as the acquisition of adequate cells for the pathologist to render a diagnosis. Results: 129 pts have been enrolled. 29 were excluded as follows: no detectable lesion by EUS (n=19), use of alternate needle (n=4), cystic lesion (n=4), difficulty obtaining sample (n=1) and immediate bleeding (n=1). 100 pts were included in this interim analysis (mean age = 65 years, 55% male). 56% of the sampled lesions were from the pancreas followed by lymph nodes (28%). Cytopathology results are shown in Table 1. The most common diagnoses were adenocarcinoma (48%), reactive adenopathy (13%), lymphoma (5%), and GIST (4%). Suction or no suction used first did not alter the diagnostic yield (p=0.715). Conclusion: The diagnostic yield of EUS -FNS is similar with suction or no suction. Predictably, when no suction is used, the samples tended to be less bloody with a trend towards higher cellularity and diagnostic yield. When performing EUS-FNS we recommend using no suction for the initial passes. Suction should be reserved for cases with low cellularity. Tabled 1Table 1 OVERALL Suction No Suction p value Bloody gross appearance 72% 50% 0.002 High cellularity 50% 59% 0.592 Diagnostic yield 78% 81% 0.616 PANCREAS Bloody gross appearance 61% 40% 0.026 High cellularity 57% 51% 0.551 Diagnostic yield 81% 85% 0.636 LYMPH NODES Bloody gross appearance 85% 67% 0.111 High cellularity 56% 69% 0.305 Diagnostic yield 70% 74% 0.761 Open table in a new tab