[Value of MRCP for detection of choledocholithiasis in symptomatic patients: one-year experience with a standardized high resolution breath-hold technique].

In our institution, MRCP is actually replacing diagnostic ERCP. This study is addressing its accuracy in the diagnosis of symptomatic choledocholithiasis. 137 MRCP were performed in patients with upper abdominal pain and biological anomalies (n = 48), pain with cholelithiasis at sonography (n = 69) or non-alcoholic acute pancreatitis (n = 20) on a 1.5 T system with high gradients using a standardized combination of breath-hold HASTE and RARE sequences. We included in our study 74 patients who underwent as second test ERCP (n = 36), peroperative cholangiography (n = 13), or extensive follow-up with laboratory tests and/or sonography (n = 25). All examinations were judged as diagnostic, discrepancies between direct cholangiography and MRCP being noted in 4 patients (5.4%). Lithiasis was misinterpreted on MRCP as bubbles in 2 patients after sphincterotomy or surgery. We had two false positive diagnoses on MRCP: in one patient peroperative cholangiography was negative and in the other one ERCP was negative. With a sensitivity and PPV of 92%, a specificity and NPV of 96% MRCP in our experience is shown as an accurate diagnostic tool for the detection of stone in the common bile duct replacing ERCP.