Aim of the study

Combined magnetic resonance imaging and magnetic resonance cholangiopancreatography (MRI/MRCP) can identify biliary strictures and diagnose primary sclerosing cholangitis (PSC). Diagnosis of cholangiocarcinoma in patients with PSC remains challenging, and the accuracy of MRI/MRCP has not been completely established. We aimed to determine the sensitivity and specificity of MRI/MRCP in the diagnosis of cholangiocarcinoma among patients with PSC from the published literature.

Material and methods

We searched Embase, PubMed, Cochrane, Scopus, ClinicalTrials.gov, and abstracts from relevant scientific meetings and performed a systematic review and meta-analysis to estimate the diagnostic yield of MRI/MRCP in patients with PSC. Sensitivity and specificity were calculated from pooled estimates of cholangiocarcinoma cases identified and lesions missed. Modifying variables were included in a meta-regression model.

Results

Our literature search yielded 302 articles and 9 conference abstracts; 8 studies involving 846 liver patients from 5 countries were included in the final analysis. Of those, 531 had PSC and received MRI/MRCP. Thirty-six (6.8%) patients were diagnosed with cholangiocarcinoma (33 true positive, 3 false negative and 1 false positive). Pooled sensitivity was 98.9% (95% CI: 98.6-99.3%). Cholangiocarcinoma cases missed by MRI/MRCP were diagnosed as beading irregularities of the central hepatic ducts, or PSC-related diffuse stricture. Metaregression revealed that neither publication year, study design, nor sample size had a significant effect on observed cancer rates (p = 0.9, 0.3, and 0.3, respectively).

Conclusions

MRI/MRCP followed by endoscopic retrograde cholangiopancreatography (ERCP) is a sensitive and specific tool to diagnose cholangiocarcinoma among patients with PSC. Further research should estimate MRI/MRCP diagnostic accuracy for cholangiocarcinoma using prospective methodology and longer term outcomes.