Accuracy of endoscopic ultrasonography and magnetic resonance cholangiopancreatography for the diagnosis of chronic pancreatitis: a prospective comparison study.

BACKGROUND: The diagnosis of chronic pancreatitis (CP) remains challenging. Endoscopic ultrasound (EUS) and magnetic resonance cholangiopancreatography (MRCP) have been proposed as highly accurate diagnostic modalities. Although endoscopic retrograde cholangiopancreatography (ERCP) has been previously used as a gold standard, it is associated with a small but significant risk. We aim to compare the accuracy of EUS and MRCP with the composite gold standard using ERCP, surgical pathology, and/or long-term clinical follow-up.

METHODS: Ninety-nine patients with clinical signs and/or symptoms suggestive of CP were prospectively enrolled. The diagnosis of CP by MRCP was established when one or more of these features were present: main duct dilation without obstruction, dilated side branches, intraductal stones, ductal irregularity, reduced T1-signal intensity, parenchymal atrophy, and reduced secretory response to secretin administration. The diagnosis of CP by EUS was made when 4 or more of the established criteria were present. Accuracy of all criteria used ("EUS only," "MRCP only," "either EUS or MRCP," and "both EUS and MRCP") was compared with the composite gold standard.

RESULTS: Forty patients were diagnosed with CP by the composite gold standard whereas the remaining 59 patients were controls. EUS only seemed more sensitive but equally specific compared with MRCP only to diagnose CP. The combination of EUS and MRCP had a sensitivity of 98% for either EUS or MRCP and a specificity of 100% for both EUS and MRCP.

CONCLUSIONS: EUS and MRCP are highly accurate modalities for the diagnosis of CP and are complementary when used together. If confirmed in larger series, the diagnosis of CP by both EUS and MRCP is highly predictive and ERCP is unlikely to add any useful information.