Background & aims: This study aimed to compare the accuracy of magnetic resonance imaging (MRI) with computed tomography (CT) in assessing acute pancreatitis (AP) and to explore the correlation between MRI findings and clinical outcome.
Methods: Patients with AP were investigated by contrast-enhanced CT and MRI on admission and 7 and 30 days thereafter. MRI was performed with intravenous secretin and contrast medium. Balthazar's grading system was used to measure CT and MRI severity indices (CTSI and MRSI, respectively).
Results: Thirty-nine patients (median age, 47 years; range, 15-86) were studied. AP was of biliary etiology in 19 patients (49%). On admission, AP was assessed clinically as severe in 7 patients (18%). A strong correlation was demonstrated between CTSI and MRSI on admission and 7 days later. MRSI on admission correlated with the following: the Ranson score, C-reactive protein levels 48 hours after admission, duration of hospitalization, and clinical outcome regarding morbidity, including local and systemic complications. Considering the Ranson score as the gold standard, MRI detected severe AP with 83% (58-96, 95% CI) sensitivity, 91% (68-98) specificity vs. 78% (52-93) and 86% (63-96) for CT. Magnetic resonance cholangiopancreatography after i.v. secretin injection showed pancreatic duct leakage in 3 patients (8%).
Conclusions: MRI is a reliable method of staging AP severity, has predictive value for the prognosis of the disease, and has fewer contraindications than CT. It can also detect pancreatic duct disruption, which may occur early in the course of AP.