BACKGROUND: It is controversial whether EUS-guided FNA by using 22-gauge (G) needles is useful for the diagnosis or evaluation of autoimmune pancreatitis (AIP). OBJECTIVE: To evaluate the usefulness of EUS-FNA by 22-G needles for the histopathological diagnosis of AIP. DESIGN: A retrospective study. SETTING: Single academic center. PATIENTS: A total of 273 patients, including 25 with AIP, underwent EUS-FNA and histological examinations. RESULTS: EUS-FNA by using 22-G needles provided adequate tissue samples for histopathological evaluation because more than 10 high-power fields were available for evaluation in 20 of 25 patients (80%). The mean immunoglobulin G4-positive plasma cell count was 13.7/high-power field. Obliterative phlebitis was observed in 10 of 25 patients (40%). In the context of the International Consensus Diagnostic Criteria for AIP, 14 and 6 of 25 patients were judged to have level 1 (positive for 3 or 4 items) and level 2 (positive for 2 items) histological findings, respectively, meaning that 20 of 25 patients were suggested to have lymphoplasmacytic sclerosing pancreatitis based on the International Consensus Diagnostic Criteria. The diagnosis in 1 patient was type 2 AIP because a granulocytic epithelial lesion was identified in this patient. LIMITATIONS: A retrospective study with a small number of patients. CONCLUSIONS: The results of this study suggest that EUS-FNA by using 22-G needles provides tissue samples adequate for histopathological evaluation and greatly contributes to the histological diagnosis of AIP.
BACKGROUND: It is controversial whether EUS-guided FNA by using 22-gauge (G) needles is useful for the diagnosis or evaluation of autoimmune pancreatitis (AIP). OBJECTIVE: To evaluate the usefulness of EUS-FNA by 22-G needles for the histopathological diagnosis of AIP. DESIGN: A retrospective study. SETTING: Single academic center. PATIENTS: A total of 273 patients, including 25 with AIP, underwent EUS-FNA and histological examinations. RESULTS: EUS-FNA by using 22-G needles provided adequate tissue samples for histopathological evaluation because more than 10 high-power fields were available for evaluation in 20 of 25 patients (80%). The mean immunoglobulin G4-positive plasma cell count was 13.7/high-power field. Obliterative phlebitis was observed in 10 of 25 patients (40%). In the context of the International Consensus Diagnostic Criteria for AIP, 14 and 6 of 25 patients were judged to have level 1 (positive for 3 or 4 items) and level 2 (positive for 2 items) histological findings, respectively, meaning that 20 of 25 patients were suggested to have lymphoplasmacytic sclerosing pancreatitis based on the International Consensus Diagnostic Criteria. The diagnosis in 1 patient was type 2 AIP because a granulocytic epithelial lesion was identified in this patient. LIMITATIONS: A retrospective study with a small number of patients. CONCLUSIONS: The results of this study suggest that EUS-FNA by using 22-G needles provides tissue samples adequate for histopathological evaluation and greatly contributes to the histological diagnosis of AIP.