Study Subjects

The study protocol was approved by the Institutional Ethics Review Committee at Children’s Hospital of Soochow University. Informed consent was signed by subject guardians. All experiments were carried out in strict accordance with the institution guidelines regarding the acquisition and experimental use of human tissues. We retrospectively reviewed the medical records of 48 patients common bile duct dilation hospitalized between January 2009 and December 2014. Inclusion criteria of PBM was 1) the common channel is longer than 5 mm by MRCP and combined by intraoperative cholangiography(IOC); 2) patients with 5 mm or longer common channel had biliary amylase(obtained by IOC) >1,000 U/L. If the cases do not meet these criteria, then the cases were diagnosed as CCC.

MRCP

MRCP was performed before diagnosing of PBM in all cases. Before MRCP, all subjects were maintained in jejunitas for 4 h then MRCP was performed with sedation for subjects 10 years-of-age or younger. A Symphony 1.5 T scanner (Siemens, Erlangen, Germany) with an abdominal phased array coil was used as follows: T1-weighted and T1-weighted fast spin series (field of view 24–28 cm, repetition time [TR] 173 ms, echo time [TE] 2.64 ms, flip angle 70, matrix 256 * 128, radiofrequency (RF) bandwidth 260 Hz/Px) and a T2-weighted sequence (TR 1,000 ms, TE 60 ms, RF bandwidth 230 Hz/Px). For MRCP, half-Fourier acquisition single shot turbo spin echo (HASTE) was used with multilayer thin coronal and axial T2-weighted imaging (TR 1,200 ms, TE 80 ms, slice thickness 4 mm). Oblique thick slabs were acquired in the planes of the common bile and pancreatic ducts. For multi-angle imaging, TR was 4,500 ms, TE 950 ms and slice thickness 60 mm were used.

Two radiologists who were unaware of the pathological findings independently reviewed the images and reached consensus through discussion. A diagnosis of PBM was established if the common channel is longer than 5 mm. They also assess the shape of the intrahepatic bile duct and gallbladder, pancreatitis, surgical pathology, symptom profiles, operative notes and pathological records were compared with the imaging findings.

Factor Analysis

Forty-eight CCC patients were assessed for gender, age, biliary amylase and MRCP findings and two groups were established: PBM and non-PBM. Within these groups, age, gender, common bile duct shape, distal position of papilla of Vater, cyst stones and cyst size, gallbladder and pancreatic duct dilation and CCC were assessed as was stomach, duodenal, and small bowel fluid. MRCP helped classify cysts as being smaller or larger than 30 mm. Ages were grouped as infant (younger than one year-of-age) or pediatric (older than one year-of-age) as well. Stomach fluid was documented to occur in the gastric fundus or body and duodenal fluid was defined as occurring in the second, third, or fourth duodenal portion and small bowel fluid was defined as occurring in the lower left or the lower and middle abdominal area. Pancreatic duct dilation was defined as dilation greater than 1 mm. Descending cysts were obviously apparent in the introitus pelvis.

Statistical Analysis

Data are presented as number (n) and percentage. Univariate comparisons were made using a nonparametric one-way Wilcoxon rank sum, a χ², or Fisher’s exact test; depending on statistical distribution. To evaluate risk factors affecting diagnosing PBM by MRCP, a stepwise logistic regression analysis was performed with SAS 8. 0. P<0.05 were considered statistically significant.

General finding in MRCP

In clinical practice, radiographic detection of common channel length is key to diagnosing PBM but this is difficult. Routing CT cannot directly delineate the long common channel and ERCP; although invasive, it can confirm the common channel. Miyazaki’s group reported that MRCP with a half-Fourier acquisition single-shot spin-echo (HASTE) sequence is noninvasive and can evaluate biliary or pancreatic disease in young children [10,13,14]. However, detecting PBM with MRCP is challenging in young patients and cases of dilated common bile ducts. Reports suggest diagnosis of pediatric PBM ranges from 40–80% with MRCP [15,16]. Thus, we used HASTE to confirm 66.7% of PBM diagnosis by MRCP. In fourteen PBM cases, MRCP could not confirm the long common channel, a finding that agrees with previous studies [10,14]. We evaluated 38 PBM cases with distal position of papilla of Vater in the distal duodenum, and these cases were similar to those of Li [17]. PBM cases often associate with distal position of papilla of Vater in distal duodenal portion.

Related factors for detecting PBM in MRCP

To investigate potential factors that may affect identifying pancreatic ducts, we described MRCP and features for detecting pediatric PBM. Univariate analysis indicated that patient age, cyst size and cyst descent into the introitus pelvis may affect pancreaticobiliary junction delineation and PBM diagnoses. Suzuki’s group reported that cyst dilation and patient age affected pancreaticobiliary junction identification and PBM confirmation [18]. We found no significant difference among MRCP findings as detailed in the Results. Multiple logistic regression confirmed that only large cysts that descended into the introitus pelvis were independent factors for pancreaticobiliary junction identification and PBM diagnosis for pediatric cases.

In the present study, 8 cases with large and descended cysts had undetectable pancreatic ducts and common channels, and the duodenum could not be delineated. Thus, large and descended cysts occupies large intra-abdominal space and have modified the pelvic and duodenal anatomy. Displacement of vessels, severe inflammatory adhesion and severe biliary obstruction may be the reasons for the descended cysts[19]. And this made the pancreatic ducts and common channels can not be detected. Saito’s group [20] reported that for 16 pediatric CCCs, PBM was identified 56% by MRCP. Yamataka and colleagues [21] reported that cyst size and location prevented visualization of PBM with CCCs and Kamisawa and colleagues [22] agreed. Also, the sphincter of Oddi’s spasm can cause an artificially short common channel length [23,24]. For the 3 cases for which the common channel could not be identified and without biliary amylase level, CCC may be the cause.

For 8 of our infant patients, we could not detect the main pancreatic duct with MRCP due to small size (<1 mm). Siles’ group [25] reported 16 MRCP pediatric cases of children younger than 3 months-of-age and they also identified main pancreatic ducts in 2 cases. Fitoz’s group [26] reported that young age can hinder pancreaticobiliary maljunction identification.

Cyst diameter can complicate pancreatic duct identification by MRCP. With cysts larger than >30 mm detection decreased and distal common bile duct areas were not identified in 7 cases with MRCP. Also, dilated bile ducts exceeding 30 mm complicated identification by MRCP because the overlapping bottom edge of the large CCC obscured views [18,21]. Sugiyama’s group reported [27] similar data for pediatric and adult patients for which large and descended cysts were most prominent. Gwal’s group [12] reported cases of pancreatic and biliary disease in which main pancreatic ducts were identified in 80.5% cases suggesting that common bile duct dilation affects main pancreatic duct identification. We found no correlation between cyst size and common channel length or bile amylase, suggesting that cysts can occlude the distal areas of the common bile duct.

Limitations

This study is limited by being a small-sample retrospective study and selection bias may be a factor. Also, only HASTE MRCP was used and no controls were involved. A prospective study with more subjects is needed to further evaluate MRCP in pediatric cases.