ABSTRACT
PURPOSE: Although endoscopic ultrasound (EUS) has been widely used for diagnosing chronic pancreatitis (CP), the assessment of fibrosis using the Rosemont criteria (RC) is generally subjective. Shear wave elastography using EUS (EUS-SWE) has been advocated as an objective approach to evaluating pancreatic fibrosis; however, it is unknown which pancreatic region should be selected for measurement. This study aims to compare the diagnostic accuracy in diagnosing CP by measurement site. METHODS: Fifty patients with CP or suspected CP who underwent EUS-SWE were retrospectively analyzed. As per the RC, they were classified into two groups: CP and non-CP. Pancreatic stiffness was evaluated by measuring the velocities of the shear wave (Vs) in addition to determining the relevant cutoff value of Vs for diagnosing CP. The correlation between Vs and RC, and the RC factors affecting pancreatic stiffness were evaluated. RESULTS: In the CP group, the Vs were notably higher in all regions (P < 0.001). The Vs for diagnostic accuracy of CP were highest in the body [area under the curve (AUC): 0.87]. A significant correlation was seen between the number of RC and Vs in all regions, with the correlation coefficient being highest in the pancreatic body (rs = 0.55). Multivariate analysis revealed that lobularity with honeycombing was an independent factor for pancreatic stiffness (P = 0.02). CONCLUSION: The pancreatic body is a suitable region for assessing pancreatic stiffness using EUS-SWE. Additionally, quantifying Vs is a valuable objective indicator for diagnosing CP.
