Diagnosis of autoimmune pancreatitis by EUS-guided FNA using a 22-gauge needle: a prospective multicenter study

Background and Aims Histopathologic examination is critical for diagnosing autoimmune pancreatitis (AIP). However, specimens obtained using EUS-guided FNA (EUS-FNA) are not recommended for histopathologic diagnosis because of inadequate sample size volume. We evaluated EUS-FNA efficacy for AIP diagnosis using a 22G needle. Methods Seventy-eight patients exhibiting the imaging characteristics indicative of AIP in the pancreatic parenchyma and pancreatic duct underwent EUS-FNA with a 22G needle at 12 institutions between February 2013 and March 2014. Samples were evaluated for tissue sampling conditions, CD38- and IgG4-positive plasma cell counts, storiform fibrosis (SF), and obliterative phlebitis (OP). Results Tissue specimens containing >10, 5 to 10, and 1 to 4 high-power fields (HPFs) were obtained from 29 (37.2%), 18 (23.1%), and 15 (19.2%) of 78 patients, respectively. The mean ± standard deviation (SD) CD38- and IgG4-positive plasma cell counts were 23.2 ± 18.8/HPF and 5.1 ± 6.7/HPF, respectively. SF was detected in 49 of 78 patients (62.8%) and OP in 38 of 78 patients (48.7%). According to the International Consensus Diagnostic Criteria (ICDC), histopathologic levels corresponded to level 1 in 32, level 2 in 13, and unclassifiable in 17 patients. Hence, 45 of 78 patients (57.7%) could be diagnosed with lymphoplasmacytic sclerosing pancreatitis according to ICDC. Conclusions Pancreatic tissues with at least 1 HPF were obtained by EUS-FNA from approximately 80% of patients, and nearly 60% of patients were diagnosed with ICDC level 2 or higher. Our findings indicate that EUS-FNA with a 22G needle may be useful for the histopathologic diagnosis of AIP. (Clinical trial registration number: UMIN000010097.)