Prospective assessment of the ability of endoscopic ultrasound to diagnose, exclude, or establish the severity of chronic pancreatitis found by endoscopic retrograde cholangiopancreatography.

BACKGROUND: Our aim was to verify endoscopic ultrasound (EUS) accuracy to diagnose, rule out, and establish the severity of chronic pancreatitis found by endoscopic retrograde cholangiopancreatography (ERCP).
METHODS: Patients undergoing ERCP for unexplained abdominal pain and/or suspected chronic pancreatitis underwent EUS. EUS was performed by experienced operators who were aware of the history but blinded to ERCP results. Chronic pancreatitis was defined using the Cambridge classification: 0 to 1 = "normal," 2 to 4 = "all chronic pancreatitis," 3 to 4 = "moderate to severe chronic pancreatitis." The number of EUS criteria required to obtain sensitivity, specificity, positive and negative predictive values > or = 85% was determined. EUS criteria for chronic pancreatitis are hyperechoic foci, hyperechoic strands, lobularity, hyperechoic duct, irregular duct, visible side-branches, ductal dilation, calcification, and cysts.
RESULTS: One hundred twenty-six patients underwent EUS and ERCP. EUS was highly sensitive and specific (> 85%) depending on the number of criteria present. Chronic pancreatitis is likely (PPV > 85%) when more than two criteria (for "all chronic pancreatitis") and more than six criteria (for "moderate to severe chronic pancreatitis") are present. "Moderate to severe chronic pancreatitis" is unlikely (NPV > 85%) when fewer than three criteria are present. Independent predictors of chronic pancreatitis were "calcification" (p = 0.000001), history of alcohol abuse (p = 0.002), and the total number of EUS criteria (p = 0.008).
CONCLUSIONS: EUS can accurately diagnose, rule out, and establish the severity of chronic pancreatitis found by ERCP.