Introduction Magnetic Resonance Cholangiopancreatography (MRCP) is increasingly used in the diagnosis of biliary abnormalities, particularly in stone disease.¹However initial studies into its role in the absence of a dilated biliary system on initial imaging have demonstrated a low diagnostic yield.²The aim of this study was to further assess the role of MRCP when the biliary tree is non-dilated on index imaging. The factors associated with diagnostic yield in this setting are unclear.

Method We performed a retrospective observational study of consecutive MRCP studies (n = 375) performed between October 2010 and June 2013 at West Middlesex University Hospital using electronic medical records. All MRCPs were reported by a consultant gastrointestinal radiologist. MRCP findings were correlated with the presence of dilated or non-dilated biliary tree on initial imaging (USS/CT), jaundice (bilirubin > 21 μmol/L), transaminitis (alanine transaminase >45 iu/L, aspartate transaminase >34 iu/L), raised alkaline phosphatase (>120 iu/L) and abdominal pain, and demographics including age and gender. Multivariate logistic regression analysis was performed using SPSS software.

Results The study included 243 female and 132 male patients with a mean age of 62 (range 20–101) years. There were 153 patients with biochemical jaundice, 246 with a dilated biliary system and 129 with a non-dilated system. The sensitivity of MRCP was 81%, with a specificity of 80%. Multivariate analysis identified that age (p = 0.0253), the presence of jaundice (p = 0.0247), and abnormal index imaging (p = 0.0003), are statistically significant predictors of the finding of biliary tree abnormalities on subsequent MRCP (see attached Table 1). The c statistic for the multivariate analysis was only moderately high at 0.665 (95% CI 0.614–0.712, p < 0.001).

Conclusion These data confirm there is a low diagnostic yield of MRCP in the absence of dilated biliary system on previous USS/CT. We propose that there should be a specific rationale for performing MRCP when initial imaging is normal, especially when the bilirubin is normal. Other clinical indices such as pain or abnormal liver chemistry (other than hyper-bilirubinaemia) are not independently predictive of positive finding on MRCP in our analysis.

Disclosure of interest None Declared.

References

1. Shanmugam V, et al. Is magnetic resonance cholangiopancreatography the new gold standard in biliary imaging? Br J Radiol. 2005;78:888–893

2. Vadwhana B, et al. Is MRCP a useful investigation where the biliary tree is normal on previous imaging? Gut 2013;62(Suppl 1):A210