Original article	Endoscopy 2007; 39: 725-730 DOI: 10.1055/s-2007-966400
© Georg Thieme Verlag KG Stuttgart · New York
EUS-guided tissue sampling: comparison of ”dual sampling” (Trucut biopsy plus FNA) with ”sequential sampling” (Trucut biopsy and then FNA as required)
G.  P.  Aithal1, G.  K.  Anagnostopoulos1, W.  Tam2, J.  Dean3, A.  Zaitoun4, G.  Kocjan5, K.  Ragunath1, S.  P.  Pereira2
1 Wolfson Digestive Diseases Centre, Queen's Medical Centre University Hospital, Nottingham, United Kingdom 2 Department of Gastroenterology, University College Hospital, London, United Kingdom 3 Department of Gastroenterology, James Cook University Hospital, Middlesborough, United Kingdom 4 Department of Cellular Pathology, Queen's Medical Centre University Hospital, Nottingham, United Kingdom 5 Department of Histopathology, University College Hospital, London, United Kingdom

Background and study aims: Both endoscopic ultrasound- (EUS-) guided tissue sampling techniques, fine-needle aspiration (FNA) and Trucut biopsy, have advantages and limitations. The aim of this study was to develop a strategy of combining these two EUS-guided sampling techniques in order to maximize the diagnostic accuracy and minimize duplication.

Patients and methods: In this multicenter study we performed ”dual sampling” (i. e. with both FNA and Trucut biopsy) in 95 patients during phase 1 of the study and ”sequential sampling” (i. e. performing FNA only when Trucut biopsy tissue cores were macroscopically inadequate) in 72 patients during phase 2.

Results: During the study period, 167/401 patients referred for EUS-guided sampling were eligible for the study; only solid lesions were included. In 143/167 patients (86 %), sampling was performed via the transesophageal or transgastric routes. When the dual sampling strategy was used, an accurate diagnosis was achieved in 78/95 patients by FNA, compared with 85/95 by Trucut biopsy (P = 0.21). The combined accuracy of the dual sampling strategy was higher than FNA alone (88/95 vs. 78/95, P = 0.048), but was not significantly higher than Trucut biopsy alone (88/95 vs. 85/95, P = 0.61). Using the sequential sampling strategy, an accurate diagnosis was achieved in 66/72 patients (92 %) compared with 88/95 (93 %) for dual sampling (P = 1.0), and 8/72 patients (11 %) had to undergo FNA after Trucut biopsy failed to obtain an adequate sample. One patient with mediastinal tuberculosis developed a cold abscess following Trucut biopsy.

Conclusion: A sequential sampling strategy, in which EUS-guided Trucut biopsy is attempted first, and FNA performed only when Trucut biopsy fails to obtain a macroscopically adequate sample, achieves a diagnostic accuracy of 92 %, with 11 % of patients requiring both sampling procedures.