The value of magnetic resonance cholangiopancreatography for the exclusion of choledocholithiasis

Abstract

Objective: To investigate the ability of Magnetic resonance cholangiopancreatography (MRCP) to exclude choledocholithiasis (CDL) in symptomatic patients.

Material and methods: Patients suspected of choledocholithiasis who underwent MRCP from 2008 through 2013 in a population based study at the National University Hospital of Iceland were retrospectively analysed, using ERCP and/or intraoperative cholangiography as a gold standard diagnosis for CDL.

Results: Overall 920 patients [66% women, mean age 55 years (SD 21)] underwent MRCP. A total of 392 patients had a normal MRCP of which 71 underwent an ERCP investigation demonstrating a CBD stone in 29 patients. A normal MRCP was found to have a 93% negative predictive value (NPV) and 89% probability of having no CBD stone demonstrated as well as no readmission due to gallstone disease within six months following MRCP. During a 6-month follow-up period of the 321 patients who did not undergo an ERCP nine (2.8%) patients were readmitted with right upper quadrant pain and elevated liver tests which later normalised with no CBD stone being demonstrated, three (0.9%) patients were readmitted with presumed gallstone pancreatitis, two (0.6%) patients were readmitted with cholecystitis and two (0.6%) patients were lost to follow-up. Seven patients of those 321 underwent an intraoperative cholangiography (IOC) and all were negative for CBD stones. For the sub-group requiring ERCP following a normal MRCP the NPV was 63%.

Conclusion: Our results support the use of MRCP as a tool for exclusion of choledocholithiasis with the potential to reduce the amount of unnecessary ERCP procedures.

Keywords: Choledocholithiasis; ERCP; MRCP; common bile duct; intraoperative cholangiogram; logistic regression.