Background & aims: Endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) is indispensable for the diagnosis of solid pancreatic lesions (SPLs). However, initial attempt of EUS-FNA can be inconclusive. We retrospectively evaluated the ability of repeat EUS-FNA at a tertiary referral hospital to obtain a conclusive cytological diagnosis after the initial inconclusive EUS-FNA results at referring facilities.
Methods: We identified patients who had undergone EUS-FNA for SPLs and with inconclusive cytological diagnosis at referring facilities. The diagnostic ability of EUS-FNA was defined as the percentage of cases with conclusive cytological diagnoses out of the total included cases in which initial results had been inconclusive. As a secondary outcome, we conducted sub-group analysis to reveal factors which might have assisted conclusive results at the University of Texas MD Anderson Cancer Center (MDA).
Results: We found 913 patients who underwent EUS-FNA for SPLs between 2005 and 2011. Among them, we recruited 84 patients who met the above criteria. Repeat EUS-FNA had ability to clarify the diagnosis in 82.1 % (69/84). No statistical differences were observed in tumor size (n = 50, 29.8 versus 29.4 mm, P-value = 0.84) and the number of needles passes (n = 40, 4.0 versus 3.4, P-value = 0.1) between outside MDA and MDA. Rapid on-site evaluation was present in less than half of cases outside MDA (n = 33, 42.4 versus 100%, P-value = 0.0001). MDA endosonographers had more years of experience compared to outside MDA (n = 50, 12.9 versus 10 years, P-value = 0.015).
Conclusions: A repeat EUS-FNA for SPLs with initial inconclusive diagnosis at a tertiary referral center establishes a diagnosis in the majority of patients.