Biliary obstruction may be related to various aetiologies such as calculus, stricture, trauma, infection or tumour, the most common being choledocholithiasis. An accurate diagnosis is necessary for treatment planning.

Endoscopic retrograde cholangiopancreatography (ERCP) is the gold standard for visualisation of pancreaticobiliary ducts and provides an opportunity for interventional treatment procedures in the same setting.^1,2 The disadvantages of ERCP are that it is an invasive procedure, the patient is at risk for complications, the study involves radiation, contrast material is necessary and there is a requirement for sedation. Owing to these disadvantages, ERCP is preferred for therapeutic intervention whilst magnetic resonance cholangiopancreatography (MRCP) has gained wide acceptance for diagnostic evaluation.^3,4

Abdominal magnetic resonance imaging (MRI) and MRCP are non-invasive modalities that allow multiplanar evaluation with reformatting in the coronal and sagittal planes and three-dimensional (3D) reconstruction to provide detailed anatomical information. MRCP is favourable because there is a lack of radiation, contrast material is not required and it can be safely performed in patients with renal insufficiency or contrast allergy. MRCP is the best choice in patients who have contraindications for ERCP, such as patients with allergy to iodinated contrast material and those who cannot receive anaesthesia.^5,6 MRCP can prevent unnecessary ERCP interventions and, in certain cases, mortality and morbidity owing to ERCP can be avoided. The aims of our study were to analyse the diagnostic effectiveness of MRCP compared with ERCP and to evaluate the role of diffusion-weighted imaging (DWI) in obstructive biliary disease.

The Institutional Review Board approved the present retrospective, cross-sectional, analytic study, and informed consent was waived. Between January 2013 and December 2013, the patients (N=126) who underwent MRCP and then ERCP for suspected biliary obstruction were reviewed via the hospital’s picture archiving and communications system (PACS). Six cases with incomplete diagnostic workup were excluded. Three cases were also eliminated because of the long period (>3 months) between MRCP and ERCP. A total of 117 patients were enrolled in the present study. The findings on MRCP and, if available, DWI, were analysed. An experienced radiologist evaluated the MRCP and DWI findings without knowledge of the ERCP findings. The presence of calculus, biliary dilatation, malignant tumour, stricture and hydatid cyst communicating with the biliary system was evaluated. The largest diameter (<10 mm/≥10 mm) and number of stones (solitary/multiple) were also noted. Apparent diffusion coefficient (ADC) values of tumours were calculated on a workstation (Leonardo, Siemens Medical Solutions, Erlangen, Germany).

All MRCPs were performed on a 1.5 Tesla system (Magnetom Avanto, Siemens Medical Solutions), using a phased array body coil. Routine MRCP sequences were performed. The common bile duct was localised on coronal T2 half-Fourier-acquisition single-shot turbo spin-echo (HASTE) (TR/TE 1000/130, field of view (FOV) 420, slice thickness 7 mm, matrix 216×320) sequence. Axial T2 HASTE and fat-saturated T2- (TR/TE 800/162, FOV 440, flip angle 180°, slice thickness 4 mm) weighted images were achieved. Coronal thin slice (4mm) and thick slab (7mm) 3D respiratory-triggered heavily T2-weighted fast spin-echo (FSE) (TR/TE1580/594, FOV420, matrix 382×384) sequences were performed, focusing on the common bile duct. Reconstruction was done by using the Siemens-Advantage Windows programme to achieve MRCP images. Upper abdominal DWI was done by single-shot echo-planar imaging sequence (TR4200, TE83, FOV440, NEX2, matrix 144×192, slice thickness 6 mm) with b=0, b=400, b=800 s/mm² values.

A Fujinon ED530-XTduodenoscope with a side-view endoscope was used for ERCP by an experienced gastroenterologist. After cannulation of the ampulla of Vater, contrast was administered to illustrate the biliary system. During ERCP, procedures such as sphincterotomy, balloon dilatation, stone extraction, stent placement or biopsy were performed, where appropriate.

The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy of MRCP for diagnosis of the aetiology of obstructive biliary disease were estimated. The sensitivity, specificity, PPV, NPV and accuracy of DWI for diagnosis of choledocholithiasis, and malignant and benign strictures were calculated. Confidence intervals (CIs) for sensitivity and specificity are ‘exact’ Clopper-Pearson CIs (CI 95%).

Ethics committee approval was received for the study from the Ethics Committee of Health Sciences University Umraniye Training and Research Hospital (date of approval 14 January 2014; ref. number 2014/727).

A total of 117 patients (66 (56.4%) female and 51 (43.6%) male) were included in our study. Their mean age was 69 years (range 16–89 years). DWI was performed on 92 (78.6%) patients. The distribution of findings is shown in Table 1. The sensitivity, specificity and accuracy of MRCP for the detection of biliary dilatation compared with ERCP were 97%, 71% and 93%, respectively.

MRCP is a non-invasive modality for investigating the morphological features of the pancreaticobiliary system.^5,7 When T2-weighted sequences and MRCP images are evaluated together, detailed information can be obtained in the evaluation of extraductal causes of biliary obstruction, which provides an important advantage of MRCP over ERCP.⁸ MRCP has no known side effects, although it is contraindicated in patients with cardiac pacemakers, cochlear implants and other metal implants. In addition, there is no need for contrast and sedation with MRCP, as opposed to ERCP. Despite the risk of complications, ERCP is generally accepted as the gold standard because it allows demonstration of the biliary ducts as well as the performance of therapeutic interventions.^1,2

A number of studies have compared MRCP with ERCP findings. Ito et al.⁹ evaluated the clinical efficacy of MRCP in the diagnosis of biliary tract disease in comparison with ERCP in 311 patients. The sensitivity of MRCP for diagnosis of choledocholithiasis was lower than ours, at 95%, but the specificity of MRCP was higher than ours, at 97%. In our study, despite the high number of subcentimeter stones (30 out of 63), the specificity of MRCP was 94.7%. Ito et al.’s MRCP accuracy was similar to ours at 97%. Additionally, the accuracy of MRCP in differentiating benign strictures from malignant tumours was higher in our study (99% v. 100%).

Varghese et al.¹⁰ investigated 191 cases, comparing MRCP with ERCP, but most had normal findings. Only 31 cases with choledocholithiasis were included. Hekimoglu et al.¹ investigated 269 cases, with 228 normal findings. The present study comprised a higher number of patients (N=117) with various disorders (n=100). There were only 17 patients with normal findings in our series. The number of our patients with choledocholithiasis (n=63) was higher than that in similar studies in the literature.^1,3,10,11,12

In our series, 30 out of 63 patients had subcentimeter stones at ERCP. The specificity and PPV of MRCP in the diagnosis of calculi<10 mm was 91% and 90%, respectively. The specificity and PPV of MRCP for the diagnosis of stones ≥10 mm was 100%. In a previous study, Zidi et al.³ found a lesser sensitivity (57.1%) for MRCP in the detection of subcentimeter stones than in our study, where we found higher ratios of specificity and PPV at MRCP in the diagnosis of stones ≥10 mm. Nandalur et al.¹¹ also showed increased sensitivity in stones >3 mm (94%–100%).

To the best of our knowledge, the effect of the number of stones in the detection of choledocholithiasis at MRCP has not been investigated before. Amongst solitary stones (n=37) at ERCP, there were two false negative results for solitary stone at MRCP. Amongst patients with multiple stones (n=26), there were five false positive results for patients with multiple stones owing to underestimating the number of solitary stones (n=2) or misdiagnosis of subcentimeter stones (n=5) at MRCP.

DWI is based on the random motion of water molecules in tissue. The degree of restriction to water molecule diffusion is directly associated with tissue cellularity and cell membrane integrity. Tumours with increased cellularity show more restricted movement of water molecules than in less cellular tumours. The degree of restricted diffusion is proportional to the degree of signal intensity in DWI. DWI should always be confirmed with an ADC map. The ADC map is a quantitative parameter calculated from DWI. Higher signal intensities in DWI and lower ADC values are defined as the result of restricted diffusion. On the other hand, higher ADC values indicate less cellularity of lesions.^13,14

DWI is not a routine sequence used for the diagnosis of choledocholithiasis, but it is easily utilized because of the short scan time. At lower b values, the biliary system will be hyperintense, similar to MRCP, and stones can easily be seen. Whether DWI contributes to the diagnosis of choledocholithasis has not been investigated before. The sensitivity of DWI for stones <10 mm and ≥10 mm was estimated to be 56.6% and 94.4%, respectively. DWI was less sensitive to subcentimeter stones than MRCP. DWI is more sensitive to motion artifacts, which were common in our series, owing to difficulty in breath holding amongst elderly patients. The lower sensitivity of DWI in the diagnosis of choledocholithiasis was thought to be because of these artifacts.

MRCP was also found to be effective in differentiating benign strictures (n=15) from tumours (n=18). The sensitivity and specificity of MRCP was 100%. These results were higher than the findings of other studies in the differential diagnosis of benign and malignant strictures.^9,15 In our study, coronal and axial T2-weighted sequences and MRCP images were evaluated together; this could be the reason for the higher accuracy in our study.

The findings of previous studies on DWI and MRCP in diagnosis of tumours were similar to ours.^13,14 In our study, all tumours were diagnosed with DWI or MRCP with 100% sensitivity and specificity. All malignant tumours showed diffusion restriction.

Despite the small number of hydatid cysts communicating with the biliary system, the localisation of biliary rupture, anatomical variations of the biliary system, and daughter cysts inside the biliary system were accurately shown on MRCP.

MRCP is a non-invasive, alternative imaging modality to ERCP for the evaluation of various pancreaticobiliary disorders, especially calculi ≥10 mm, tumours, strictures and complicated hydatid cysts with biliary communication. DWI can be helpful in the diagnosis of choledocholithiasis and tumours. ERCP remains the modality of choice where intervention for the alleviation of cholestasis is required.