- Email: [email protected]
Diagnostic impact of computed tomography cholangiography and magnetic resonance cholangiopancreatography on pancreaticobiliary maljunction
Journal of Pediatric Surgery (2011) 46, 1373–1378
www.elsevier.com/locate/jpedsurg
Diagnostic impact of computed tomography cholangiography and magnetic resonance cholangiopancreatography on pancreaticobiliary maljunction Shigehisa Fumino ⁎, Shigeru Ono, Osamu Kimura, Eiichi Deguchi, Naomi Iwai Department of Pediatric Surgery, Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kyoto 602-8566, Japan Received 9 August 2010; revised 6 January 2011; accepted 31 January 2011
Key words: Pancreaticobiliary maljunction; CT cholangiography; MRCP; Multidetector-row CT; Biliopancreatic reflux
Abstract Background/Purpose: The aim of this study was to investigate the diagnostic potential of computed tomography cholangiography (CTC) and magnetic resonance cholangiopancreatography (MRCP) in children with pancreaticobiliary maljunction (PBM). Methods: Fifty-three children with PBM were consecutively treated between 1997 and 2009. Among them, the patients who underwent CTC and/or MRCP preoperatively were enrolled in this study. Computed tomography cholangiography was examined after infusion of meglumine iodoxamate with subsequent 3-dimensional rendering. The visualization of the biliary and pancreatic duct systems was evaluated and compared with that visualized with MRCP. The findings of direct cholangiography were used as the standard of reference. Results: Of the 53 cases with PBM, 17 cases were examined by CTC, 10 cases by MRCP, and 17 with both. The extrahepatic bile tract was visualized in 32 (94.1%) of 34 patients in CTC and in all 27 patients in MRCP. The intrahepatic bile duct was more frequently demonstrated by MRCP than by CTC (96.3% vs 70.6%, P = .02). Pancreaticobiliary maljunction was noted in 13 (38.2%) of 34 with CTC and in 12 (44.4%) of 27 with MRCP. The minimum age for visualization of PBM was at 10 months in CTC and at 1 year and 11 months in MRCP, respectively. The main pancreatic duct was more frequently visualized by MRCP than by CTC (81.5% vs 8.8%, P b .001). Conclusions: Magnetic resonance cholangiopancreatography provides superior visualization of the intrahepatic duct and the pancreatic system when compared with CTC. However, it is still challenging to perform a good-quality examination in young infant. The great advantage of CTC is its ability to produce high-quality images without respiratory artifacts and that it allows accurate assessment of the presence of PBM equivalent to MRCP. © 2011 Elsevier Inc. All rights reserved.
⁎ Corresponding author. Tel.: +81 75 251 5809; fax: +81 75 251 5828. E-mail address: [email protected] (S. Fumino). 0022-3468/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2011.01.026
1374 Pancreaticobiliary maljunction (PBM) is a congenital biliary anomaly commonly associated with congenital biliary dilatation and defined as an anatomical maljunction of the bile and pancreatic ducts outside the duodenal wall forming a long, common channel [1,2]. Freed from the influence of the sphincter of Oddi, pancreatic juice can regurgitate into the biliary tree and bile juice can also enter the pancreatic duct. Overall, these regurgitations are responsible for the clinicopathology of PBM [3]. The definitive preoperative diagnosis of PBM has classically depended on direct cholangiopancreatography, such as endoscopic retrograde cholangiopancreatography (ERCP) and percutaneous transhepatic cholangiography. However, modern noninvasive imaging modalities are beginning to play an important role in the diagnostic workup of patients with suspected PBM. Magnetic resonance cholangiopancreatography (MRCP) is widely used for hepatobiliary and pancreatic disease even in pediatric population [4,5]. According to previous reports, MRCP allows for noninvasive and accurate detection of PBM without irradiation and avoids the lethal complications of unnecessary ERCP and percutaneous transhepatic cholangiography such as pancreatitis and cholangitis [6,7]. Meanwhile, CT cholangiography (CTC) is also a classical, noninvasive method for the evaluation of the biliary system aside from radiation. The contrast material, meglumine iodoxamate (Biliscopin; Schering AG, Berlin, Germany), has a biliary excretory profile. Bile duct imaging obtained from this material shows extensive anatomical detail along with reflection of physiologic bile flow in vivo [8,9]. Authors have previously reported biliopancreatic reflux on CTC and discussed the possibility of visualization of PBM and the pancreatic duct with CTC [10]. More recently, the emergence of multidetector-row CT (MDCT) was a major technological breakthrough that dramatically changed the practice of CT, and the spatial resolution on CTC with MDCT may be superior to that of MRCP. There are few reports of comparisons between CTC and MRCP for the preoperative assessment of PBM. The purposes of this study were to investigate the feasibility and accuracy of both modalities for the diagnosis of PBM in pediatric patients.
S. Fumino et al. noted. Fifty-one cases underwent definitive surgery, including intraoperative cholangiopancreatography and excision of the extrahepatic bile duct with Roux-en-Y biliary reconstruction. One of the remaining 2 patients without surgery, a 3-month-old girl developed severe liver failure despite external bile drainage, resulting in primary living-donor liver transplantation [11]. The other patient referenced above was on the waiting list for surgery. Patients undergoing CTC and/or MRCP, preoperatively, were enrolled in this study.
1.2. Computed tomography cholangiography Computed tomography cholangiography was preoperatively performed as we reported previously [10]. Briefly, 1 to 3 mL/kg Biliscopin was administered intravenously over 30 minutes. The CT scanning began 20 minutes after injection. Specific scan protocols varied depending on the CT scanner available at the time of examination. Between 1997 and 2005, a single-slice CT scanner (Asteion VR; Toshiba Medical, Tokyo, Japan) was used, with a 5 mm/s table feed and a pitch of 1.0 (120 kV [peak] and 50-100 mA). From 2006, CT scanning was performed using a 64-detector-row helical CT (LightSpeed VCTs: GE Medical Systems, Milwaukee, WI), with 55 mm per rotation table speed, a pitch of 1.375 (120 kV [peak] and 80-125 mA), and a reconstruction slice thickness of 0.625 mm. Multiplanar reformation images were reconstructed at the workstation, as well as volume rendering reconstructions if needed.
1.3. Magnetic resonance cholangiopancreatography
1. Materials and methods
Routine MRCP imaging was preoperatively performed using the Intera 1.5-T MRI scanner (Philips, Best, The Netherlands) with a body array wrap around coil without breath-holding. Patients were studied in the supine position with a thick-slab 2-dimensional turbo-spin echo, obtaining coronal and oblique coronal 40-mm-thick slices on a 320 × 256 matrix. Secretin (Secrepan; Eisai Corporation, Tokyo, Japan), a polypeptide hormone secreted by duodenal mucosa, was administered intravenously in a dose of 1 U/kg of body weight to increase the fluid signal in the pancreatic duct before 2005. These image sections were processed by the standard maximum-intensity-projection algorithm to obtain views of the entire pancreatobiliary system.
1.1. Patients
1.4. Sedation
From 1997 to 2009, a total of 53 children with PBM were consecutively diagnosed and treated at our institute. There were 17 boys and 36 girls, and their median age at diagnosis was 3.0 years (range, 0 months-17 years). Choledochal dilatation was cystic type (n = 30), fusiform type (n = 20), and nondilatation type (n = 3). Among 50 patients with the choledochal cyst, there were 21 children with Todani Ia, 20 with Ic, and 9 with IV-a. Twenty-three cases were examined by ERCP preoperatively under general anesthesia, and no major complications were
The patients who could not tolerate the examination were sedated with 30 to 50 mg/kg of body weight oral chlorate hydrate or with 30 to 50 mg/kg rectal thiopental sodium.
1.5. Imaging study evaluation The diagnostic criteria for PBM were as follows: (1) presence of a markedly long, common channel of the common bile duct and a major pancreatic duct more than 10 mm in length and (2) junction of the common bile duct and
Impact of CTC and MRCP on pancreaticobiliary maljunction the major pancreatic duct showing an apparently anomalous form [12]. A non–dilatation-type PBM was defined as a common bile duct with less than 8 mm in diameter [13]. The intrahepatic and extrahepatic bile duct, gallbladder, pancreatic duct, and common channel were visualized and evaluated. If available, the findings of direct cholangiography, such as ERCP and intraoperative cholangiopancreatography, were used as the standard of reference for the evaluation of the CTC and MRCP. The sensitivity of anatomical structures on CTC and MRCP was calculated. For statistical analysis, a paired t test was used to compare the parameters. A P value of less than .05 was considered to be significant.
1375 was successful in 1 patient. Although there was no statistical difference between those sensitivities, the minimum age for visualization of PBM was at 10 months of age in CTC (Fig. 1) and at 1 year and 11 months in MRCP (Fig. 2), respectively. The MPD was visualized in 3 (8.8%) of the 34 patients in CTC and in 22 (81.5%) of 27 patients in MRCP, and there was statistical difference between them (P b .001). Since 2006 when MDCT became available in our institute, the image reconstruction quality of CTC has been improving dramatically. Although the opacification of MPD in CTC occurred less frequently than in MRCP, CTC with MDCT could clearly demonstrate the ductal structures including the intrahepatic bile duct, PBM, and the main pancreatic duct with erasing artifacts and 3-dimensional (3D) reconstruction (Fig. 3).
2. Results Of 53 cases with PBM, 17 cases were examined by CTC, 10 cases by MRCP, and 17 using both CTC and MRCP. No major adverse reactions were noted during sedation and the infusion of contrast agent in any patients. The clinical results of those imaging studies are shown in Table 1. In CTC, the extrahepatic bile tract including the common hepatic and bile duct was visualized in 32 (94.1%) of 34 patients. One of the 2 negative cases included a 1-month-old infant with an elevated serum bilirubin level at inspection (total bilirubin, 10.28 mg/dL; direct bilirubin, 2.81 mg/dL). The other was a 1-year-old infant with an extremely large choledochal cyst. In MRCP, the extrahepatic bile duct was clearly demonstrated in all 27 patients. The gallbladder was shown in 24 (70.6%) of 34 patients with CTC and in 25 (92.6%) of 27 with MRCP. Regarding the intrahepatic bile duct, MRCP was significantly more sensitive than CTC (96.3% vs 70.6%, P = .02). In addition, the present series included 9 cases with Todani IV-a [14], in whom the dilated intrahepatic bile duct was visually more distinguishable in MRCP than in CTC (data not shown). In CTC, PBM was considered to be positive when the contrast medium was regurgitated into the main pancreatic duct (MPD) even if trivial. According to this definition, PBM was noted in 13 (38.2%) of 34 patients with CTC. Magnetic resonance cholangiopancreatography showed PBM in 12 (44.4%) of 27 patients. With regard to non–dilatation-type PBM, all 3 patients failed to show PBM in MRCP and CTC Table 1
3. Discussion The purposes of preoperative imaging studies for PBM are principally classified into 4 items: 1. First, the most important role for imaging allows for the total evaluation of the biliary system. This is important because the presence of a choledochal cyst must be confirmed, and these obtained images provide a road map for surgical planning. Moreover, the morphologic changes of the intrahepatic bile duct, such as stenosis and enlargement, must be assessed simultaneously. Imaging also allows for estimation of the confluent position of the cystic duct [15]. 2. Imaging also allows for the evaluation of the pancreatic system. Using imaging, one can visualize the changes of the pancreas parenchyma, the dilatation of the pancreatic duct, and the presence of a protein plug. This may be accompanied by pancreas divisum or an annular pancreas [16-18]. 3. Imaging also allows for the demonstration of PBM. Although the demonstration of PBM is necessary for the diagnosis of non–dilatation-type PBM to assess surgical indications, this diagnosis is often difficult to perform in pediatric patients [13]. 4. Imaging allows for functional evaluations. In PBM, 2way regurgitation can occur with pancreatic juice
Sensitivity of CTC and MRCP in children with PBM
Extrahepatic bile duct Intrahepatic bile duct Gallbladder PBM Main pancreatic duct Mean age (range) Minimum age for visualization of PBM
CTC (n = 34)
MRCP (n = 27)
32 (94.1%) 24 (70.6%) 24 (70.6%) 13 (38.2%) 3 (8.8%) 4 y 3 mo (0 m-17 y 3 mo) 10 mo
27 (100.0%) 26 (96.3%) 25 (92.6%) 12 (44.4%) 22 (81.5%) 4 y 2 mo (3 mo-16 y 5 mo) 1 y, 11 mo
P .02
b.001
1376
S. Fumino et al.
Fig. 1 Computed tomography cholangiography images of the biliary duct of a 10-month-old boy with cystic-type choledochal dilatation. A, Opacification of the main pancreatic duct on CTC using MDCT. B, Intraoperative cholangiopancreatography. Magnetic resonance cholangiopancreatography was not done. He was the youngest patient showing PBM on CTC (arrows show the PBM).
reflux into the bile duct or with bile juice regurgitation into the pancreatic duct. These regurgitations can cause various pathologic conditions. Therefore, the physiologic dynamics of bile and pancreatic juice is critically important. For this purpose, contrast material–enhanced MRCP or dynamic MRCP with secretin was reported [5]. For these purposes, the preoperative investigation is performed in combination with several imaging modalities. Endoscopic retrograde cholangiopancreatography is regarded as the most definitive and reliable diagnostic method of revealing anatomical details of the biliary tree and PBM preoperatively [19]. However, limitations include the requirement of general anesthesia and an invasive nature
with possible complications, making physicians hesitant to choose this option. In addition, ERCP does not allow for the evaluation of the pancreatic parenchyma. For these reasons, we have not performed routine ERCP since 2001 in our institute, with the exception of non–dilatation-type PBM. Alternatively, CT and MRCP have become more common. Imaging modalities continue to advance, and these technologies can provide more detailed information on both biliary and pancreatic ductal anatomy. There are few reports concerning the comparison study between the sensitivity of MRCP and CTC in PBM [20], and we attempted to clarify the diagnostic impact of these modalities in this study. Magnetic resonance cholangiopancreatography has been an important diagnostic tool in the investigation of pediatric hepatobiliary pancreatic diseases because it requires neither
Fig. 2 Magnetic resonance cholangiopancreatography images of the biliary duct of a 1-year 11-month-old girl with cystic-type choledochal dilatation. Although MRCP shows PBM (A), CTC failed (B) to show this diagnosis. C, Intraoperative cholangiopancreatography. She was the youngest patient showing PBM on MRCP (arrows show the PBM).
Impact of CTC and MRCP on pancreaticobiliary maljunction
1377
Fig. 3 Computed tomography cholangiography images of the biliary duct of a 1-year-old boy with nondilated choledochus. Computed tomography cholangiography with MDCT clearly shows PBM including the whole length of the main pancreatic duct and long, common channel in 3D rendering images, whose quality was far superior to MRCP images (A, MRCP; B, CTC).
radiation exposures nor contrast agents. In PBM, there are many reports regarding its effectiveness [4-7]. In this study, we could obtain a diagnostic accuracy of almost 100% in the presence of a choledochal cyst and high sensitivities for the intrabile and extrabile duct in MRCP, which is particularly helpful for the type IV-a choledochal cyst. Moreover, the rate of successful MPD visualization was as high as 81.5%. Its greatest strength compared with CTC is its availability for pancreatobiliary obstructed patients such as patients with severe jaundice and protein plugs. However, it is still challenging to perform a quality examination in young children. Magnetic resonance cholangiopancreatography is limited by small-caliber ducts, poor signals, and patient motion, creating artifacts. Therefore, the minimum age for visualization of PBM in MRCP was 1 year and 11 months, more than 1 year greater than the minimum age requirement for CTC. Computed tomography cholangiography has been useful for preoperative biliary investigation using the administration of biliary contrast material. The great advantage of CTC is its ability to produce high-quality images without respiratory artifacts and to provide more anatomical detail of the biliary tract than conventional intravenous infusion cholangiography [21,22]. Contrast media in CTC is absorbed by hepatocytes and excreted into the bile after intravenous injection. Therefore, CTC is considered to reflect the dynamic and physiologic flow of the bile, and some cases with PBM showed biliopancreatic reflux [10], and it might be useful for the functional evaluation. The potential disadvantages of CTC are irradiation, allergic reactions caused by Biliscopin injections, and decreased efficacy with visualization in patients with hyperbilirubinemia (N3 mg/dL). The recent evolution of MDCT scanners has led to rapid true isotropic imaging, because thin-section axial images can be obtained at submillimeter slice thicknesses. Simultaneously, the increasing power of 3D applications has allowed for practical postprocessing of such data [23]. In this study, CTC
had some disadvantages for the visualization of the pancreas system compared with MRCP. However, the sensitivity for PBM was similar to MRCP, and CTC with MDCT could provide more elaborate images than MRCP, even in an infant younger than 1 year. Lam et al [20] showed that the sensitivity of the main pancreatic duct in drip infusion cholangiography–computed tomography and MRCP was 63.6% and 45.5%, respectively, in their series consisting of 11 children. The incidence of drip infusion cholangiography–computed tomography was higher than that in the present study and that fact implied that there is room for improvement in our method. To improve the sensitivity of MPD, we have used several different mechanisms including the slow administration of contrast agent and rolling the patient on the CT scanner; however, a truly reliable method is still unclear and further studies are necessary. Finally, MRCP is superior to CTC in the visualization of the intrahepatic duct and pancreatic system; however, it is still challenging to perform a good-quality examination in young infants. The great advantage of CTC is its ability to produce high-quality images without respiratory artifacts and to allow for an accurate assessment of the presence of PBM equivalent to MRCP. Although our series has a limited number of patients, CTC with MDCT seems to be a promising modality to diagnose PBM and navigate surgery especially in cases with non–dilatation type PBM. Neither CTC nor MRCP is an exclusive modality, and physicians should use them in combination. The common goal is to provide the most accurate and feasible preoperative plan for safety and treatment accuracy in affected children.
References [1] Jona JZ, Babbitt DP, Starshak RJ, et al. Anatomic observations and etiologic and surgical considerations in choledochal cyst. J Pediatr Surg 1979;14:315-20.
1378 [2] Babbitt DP, Starshak RJ, Clemett AR. Choledochal cyst: a concept of etiology. Am J Roentgenol Radium Ther Nucl Med 1973;119:57-62. [3] Iwai N, Yanagihara J, Tokiwa K, et al. Congenital choledochal dilatation with emphasis on pathophysiology of the biliary tract. Ann Surg 1992;215:27-30. [4] Kamisawa T, Tu Y, Egawa N, et al. MRCP of congenital pancreaticobiliary malformation. Abdom Imaging 2007;32:129-33. [5] Chavhan GB, Babyn PS, Manson D, et al. Pediatric MR cholangiopancreatography: principles, technique, and clinical applications. Radiographics 2008;28:1951-62. [6] Hamada Y, Tanano A, Takada K, et al. Magnetic resonance cholangiopancreatography on postoperative work-up in children with choledochal cysts. Pediatr Surg Int 2004;20:43-6. [7] Kim MJ, Han SJ, Yoon CS, et al. Using MR cholangiopancreatography to reveal anomalous pancreaticobiliary ductal union in infants and children with choledochal cysts. AJR Am J Roentgenol 2002;179:209-14. [8] Persson A, Dahlström N, Smedby O, et al. Volume rendering of threedimensional drip infusion CT cholangiography in patients with suspected obstructive biliary disease: a retrospective study. Br J Radiol 2005;78:1078-85. [9] Okada M, Fukada J, Toya K, et al. The value of drip infusion cholangiography using multidetector-row helical CT in patients with choledocholithiasis. Eur Radiol 2005;15:2140-5. [10] Fumino S, Tokiwa K, Katoh T, et al. New insight into bile flow dynamics in anomalous arrangement of the pancreaticobiliary duct. Br J Surg 2002;89:865-9. [11] Fumino S, Ono S, Shimadera S, et al. Impact of age at diagnosis on clinical features in children with anomalous arrangement of the pancreaticobiliary duct. Eur J Pediatr Surg 2010;20:325-9. [12] Todani T, Akita E, Eto T, The Japanese Study Group on Pancreaticobiliary Maljunction. Diagnostic criteria of pancreaticobiliary maljunction. J Hepatobiliary Pancreat Surg 1994;1:219-21.
S. Fumino et al. [13] Iwai N, Fumino S, Tsuda T, et al. Surgical treatment for anomalous arrangement of the pancreaticobiliary duct with nondilatation of the common bile duct. J Pediatr Surg 2004;39:1794-6. [14] Todani T, Watanabe Y, Narusue M, et al. Congenital bile duct cysts: classification, operative procedures, and review of thirty-seven cases including cancer arising from choledochal cyst. Am J Surg 1977;134: 263-9. [15] Fumino S, Iwai N, Deguchi E, et al. Clinical significance of the confluence of the cystic duct in patients with anomalous arrangement of the pancreaticobiliary duct. J Pediatr Surg 2007;42: 1319-23. [16] Sugimoto T, Yamagiwa I, Obata K, et al. Choledochal cyst and duodenal atresia: a rare combination of malformations. Pediatr Surg Int 2004;20:724-6. [17] Komuro H, Makino S, Tahara K. Choledochal cyst associated with duodenal obstruction. J Pediatr Surg 2000;35:1259-62. [18] Dalvi AN, Pramesh CS, Prasanna GS, et al. Incomplete pancreas divisum with anomalous choledochopancreatic duct junction with choledochal cyst. Arch Surg 1999;134:1150-2. [19] Kim OH, Chung HJ, Choi BG. Imaging of the choledochal cyst. Radiographics 1995;15:69-88. [20] Lam WW, Lam TP, Saing H, et al. MR cholangiography and CT cholangiography of pediatric patients with choledochal cysts. AJR Am J Roentgenol 1999;73:401-5. [21] Hoglund M, Muren C, Boijsen MW. Computed tomography with intravenous cholangiography contrast: a method for visualizing choledochal cysts. Eur J Radiol 1990;10:159-63. [22] Hamada Y, Sato M, Sanada T, et al. Spiral computed tomography for biliary dilatation. J Pediatr Surg 1995;30:694-6. [23] Tamm EP, Balachandran A, Bhosale P, et al. Update on 3D and multiplanar MDCT in the assessment of biliary and pancreatic pathology. Abdom Imaging 2009;34:64-74.
www.elsevier.com/locate/jpedsurg
Diagnostic impact of computed tomography cholangiography and magnetic resonance cholangiopancreatography on pancreaticobiliary maljunction Shigehisa Fumino ⁎, Shigeru Ono, Osamu Kimura, Eiichi Deguchi, Naomi Iwai Department of Pediatric Surgery, Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kyoto 602-8566, Japan Received 9 August 2010; revised 6 January 2011; accepted 31 January 2011
Key words: Pancreaticobiliary maljunction; CT cholangiography; MRCP; Multidetector-row CT; Biliopancreatic reflux
Abstract Background/Purpose: The aim of this study was to investigate the diagnostic potential of computed tomography cholangiography (CTC) and magnetic resonance cholangiopancreatography (MRCP) in children with pancreaticobiliary maljunction (PBM). Methods: Fifty-three children with PBM were consecutively treated between 1997 and 2009. Among them, the patients who underwent CTC and/or MRCP preoperatively were enrolled in this study. Computed tomography cholangiography was examined after infusion of meglumine iodoxamate with subsequent 3-dimensional rendering. The visualization of the biliary and pancreatic duct systems was evaluated and compared with that visualized with MRCP. The findings of direct cholangiography were used as the standard of reference. Results: Of the 53 cases with PBM, 17 cases were examined by CTC, 10 cases by MRCP, and 17 with both. The extrahepatic bile tract was visualized in 32 (94.1%) of 34 patients in CTC and in all 27 patients in MRCP. The intrahepatic bile duct was more frequently demonstrated by MRCP than by CTC (96.3% vs 70.6%, P = .02). Pancreaticobiliary maljunction was noted in 13 (38.2%) of 34 with CTC and in 12 (44.4%) of 27 with MRCP. The minimum age for visualization of PBM was at 10 months in CTC and at 1 year and 11 months in MRCP, respectively. The main pancreatic duct was more frequently visualized by MRCP than by CTC (81.5% vs 8.8%, P b .001). Conclusions: Magnetic resonance cholangiopancreatography provides superior visualization of the intrahepatic duct and the pancreatic system when compared with CTC. However, it is still challenging to perform a good-quality examination in young infant. The great advantage of CTC is its ability to produce high-quality images without respiratory artifacts and that it allows accurate assessment of the presence of PBM equivalent to MRCP. © 2011 Elsevier Inc. All rights reserved.
⁎ Corresponding author. Tel.: +81 75 251 5809; fax: +81 75 251 5828. E-mail address: [email protected] (S. Fumino). 0022-3468/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2011.01.026
1374 Pancreaticobiliary maljunction (PBM) is a congenital biliary anomaly commonly associated with congenital biliary dilatation and defined as an anatomical maljunction of the bile and pancreatic ducts outside the duodenal wall forming a long, common channel [1,2]. Freed from the influence of the sphincter of Oddi, pancreatic juice can regurgitate into the biliary tree and bile juice can also enter the pancreatic duct. Overall, these regurgitations are responsible for the clinicopathology of PBM [3]. The definitive preoperative diagnosis of PBM has classically depended on direct cholangiopancreatography, such as endoscopic retrograde cholangiopancreatography (ERCP) and percutaneous transhepatic cholangiography. However, modern noninvasive imaging modalities are beginning to play an important role in the diagnostic workup of patients with suspected PBM. Magnetic resonance cholangiopancreatography (MRCP) is widely used for hepatobiliary and pancreatic disease even in pediatric population [4,5]. According to previous reports, MRCP allows for noninvasive and accurate detection of PBM without irradiation and avoids the lethal complications of unnecessary ERCP and percutaneous transhepatic cholangiography such as pancreatitis and cholangitis [6,7]. Meanwhile, CT cholangiography (CTC) is also a classical, noninvasive method for the evaluation of the biliary system aside from radiation. The contrast material, meglumine iodoxamate (Biliscopin; Schering AG, Berlin, Germany), has a biliary excretory profile. Bile duct imaging obtained from this material shows extensive anatomical detail along with reflection of physiologic bile flow in vivo [8,9]. Authors have previously reported biliopancreatic reflux on CTC and discussed the possibility of visualization of PBM and the pancreatic duct with CTC [10]. More recently, the emergence of multidetector-row CT (MDCT) was a major technological breakthrough that dramatically changed the practice of CT, and the spatial resolution on CTC with MDCT may be superior to that of MRCP. There are few reports of comparisons between CTC and MRCP for the preoperative assessment of PBM. The purposes of this study were to investigate the feasibility and accuracy of both modalities for the diagnosis of PBM in pediatric patients.
S. Fumino et al. noted. Fifty-one cases underwent definitive surgery, including intraoperative cholangiopancreatography and excision of the extrahepatic bile duct with Roux-en-Y biliary reconstruction. One of the remaining 2 patients without surgery, a 3-month-old girl developed severe liver failure despite external bile drainage, resulting in primary living-donor liver transplantation [11]. The other patient referenced above was on the waiting list for surgery. Patients undergoing CTC and/or MRCP, preoperatively, were enrolled in this study.
1.2. Computed tomography cholangiography Computed tomography cholangiography was preoperatively performed as we reported previously [10]. Briefly, 1 to 3 mL/kg Biliscopin was administered intravenously over 30 minutes. The CT scanning began 20 minutes after injection. Specific scan protocols varied depending on the CT scanner available at the time of examination. Between 1997 and 2005, a single-slice CT scanner (Asteion VR; Toshiba Medical, Tokyo, Japan) was used, with a 5 mm/s table feed and a pitch of 1.0 (120 kV [peak] and 50-100 mA). From 2006, CT scanning was performed using a 64-detector-row helical CT (LightSpeed VCTs: GE Medical Systems, Milwaukee, WI), with 55 mm per rotation table speed, a pitch of 1.375 (120 kV [peak] and 80-125 mA), and a reconstruction slice thickness of 0.625 mm. Multiplanar reformation images were reconstructed at the workstation, as well as volume rendering reconstructions if needed.
1.3. Magnetic resonance cholangiopancreatography
1. Materials and methods
Routine MRCP imaging was preoperatively performed using the Intera 1.5-T MRI scanner (Philips, Best, The Netherlands) with a body array wrap around coil without breath-holding. Patients were studied in the supine position with a thick-slab 2-dimensional turbo-spin echo, obtaining coronal and oblique coronal 40-mm-thick slices on a 320 × 256 matrix. Secretin (Secrepan; Eisai Corporation, Tokyo, Japan), a polypeptide hormone secreted by duodenal mucosa, was administered intravenously in a dose of 1 U/kg of body weight to increase the fluid signal in the pancreatic duct before 2005. These image sections were processed by the standard maximum-intensity-projection algorithm to obtain views of the entire pancreatobiliary system.
1.1. Patients
1.4. Sedation
From 1997 to 2009, a total of 53 children with PBM were consecutively diagnosed and treated at our institute. There were 17 boys and 36 girls, and their median age at diagnosis was 3.0 years (range, 0 months-17 years). Choledochal dilatation was cystic type (n = 30), fusiform type (n = 20), and nondilatation type (n = 3). Among 50 patients with the choledochal cyst, there were 21 children with Todani Ia, 20 with Ic, and 9 with IV-a. Twenty-three cases were examined by ERCP preoperatively under general anesthesia, and no major complications were
The patients who could not tolerate the examination were sedated with 30 to 50 mg/kg of body weight oral chlorate hydrate or with 30 to 50 mg/kg rectal thiopental sodium.
1.5. Imaging study evaluation The diagnostic criteria for PBM were as follows: (1) presence of a markedly long, common channel of the common bile duct and a major pancreatic duct more than 10 mm in length and (2) junction of the common bile duct and
Impact of CTC and MRCP on pancreaticobiliary maljunction the major pancreatic duct showing an apparently anomalous form [12]. A non–dilatation-type PBM was defined as a common bile duct with less than 8 mm in diameter [13]. The intrahepatic and extrahepatic bile duct, gallbladder, pancreatic duct, and common channel were visualized and evaluated. If available, the findings of direct cholangiography, such as ERCP and intraoperative cholangiopancreatography, were used as the standard of reference for the evaluation of the CTC and MRCP. The sensitivity of anatomical structures on CTC and MRCP was calculated. For statistical analysis, a paired t test was used to compare the parameters. A P value of less than .05 was considered to be significant.
1375 was successful in 1 patient. Although there was no statistical difference between those sensitivities, the minimum age for visualization of PBM was at 10 months of age in CTC (Fig. 1) and at 1 year and 11 months in MRCP (Fig. 2), respectively. The MPD was visualized in 3 (8.8%) of the 34 patients in CTC and in 22 (81.5%) of 27 patients in MRCP, and there was statistical difference between them (P b .001). Since 2006 when MDCT became available in our institute, the image reconstruction quality of CTC has been improving dramatically. Although the opacification of MPD in CTC occurred less frequently than in MRCP, CTC with MDCT could clearly demonstrate the ductal structures including the intrahepatic bile duct, PBM, and the main pancreatic duct with erasing artifacts and 3-dimensional (3D) reconstruction (Fig. 3).
2. Results Of 53 cases with PBM, 17 cases were examined by CTC, 10 cases by MRCP, and 17 using both CTC and MRCP. No major adverse reactions were noted during sedation and the infusion of contrast agent in any patients. The clinical results of those imaging studies are shown in Table 1. In CTC, the extrahepatic bile tract including the common hepatic and bile duct was visualized in 32 (94.1%) of 34 patients. One of the 2 negative cases included a 1-month-old infant with an elevated serum bilirubin level at inspection (total bilirubin, 10.28 mg/dL; direct bilirubin, 2.81 mg/dL). The other was a 1-year-old infant with an extremely large choledochal cyst. In MRCP, the extrahepatic bile duct was clearly demonstrated in all 27 patients. The gallbladder was shown in 24 (70.6%) of 34 patients with CTC and in 25 (92.6%) of 27 with MRCP. Regarding the intrahepatic bile duct, MRCP was significantly more sensitive than CTC (96.3% vs 70.6%, P = .02). In addition, the present series included 9 cases with Todani IV-a [14], in whom the dilated intrahepatic bile duct was visually more distinguishable in MRCP than in CTC (data not shown). In CTC, PBM was considered to be positive when the contrast medium was regurgitated into the main pancreatic duct (MPD) even if trivial. According to this definition, PBM was noted in 13 (38.2%) of 34 patients with CTC. Magnetic resonance cholangiopancreatography showed PBM in 12 (44.4%) of 27 patients. With regard to non–dilatation-type PBM, all 3 patients failed to show PBM in MRCP and CTC Table 1
3. Discussion The purposes of preoperative imaging studies for PBM are principally classified into 4 items: 1. First, the most important role for imaging allows for the total evaluation of the biliary system. This is important because the presence of a choledochal cyst must be confirmed, and these obtained images provide a road map for surgical planning. Moreover, the morphologic changes of the intrahepatic bile duct, such as stenosis and enlargement, must be assessed simultaneously. Imaging also allows for estimation of the confluent position of the cystic duct [15]. 2. Imaging also allows for the evaluation of the pancreatic system. Using imaging, one can visualize the changes of the pancreas parenchyma, the dilatation of the pancreatic duct, and the presence of a protein plug. This may be accompanied by pancreas divisum or an annular pancreas [16-18]. 3. Imaging also allows for the demonstration of PBM. Although the demonstration of PBM is necessary for the diagnosis of non–dilatation-type PBM to assess surgical indications, this diagnosis is often difficult to perform in pediatric patients [13]. 4. Imaging allows for functional evaluations. In PBM, 2way regurgitation can occur with pancreatic juice
Sensitivity of CTC and MRCP in children with PBM
Extrahepatic bile duct Intrahepatic bile duct Gallbladder PBM Main pancreatic duct Mean age (range) Minimum age for visualization of PBM
CTC (n = 34)
MRCP (n = 27)
32 (94.1%) 24 (70.6%) 24 (70.6%) 13 (38.2%) 3 (8.8%) 4 y 3 mo (0 m-17 y 3 mo) 10 mo
27 (100.0%) 26 (96.3%) 25 (92.6%) 12 (44.4%) 22 (81.5%) 4 y 2 mo (3 mo-16 y 5 mo) 1 y, 11 mo
P .02
b.001
1376
S. Fumino et al.
Fig. 1 Computed tomography cholangiography images of the biliary duct of a 10-month-old boy with cystic-type choledochal dilatation. A, Opacification of the main pancreatic duct on CTC using MDCT. B, Intraoperative cholangiopancreatography. Magnetic resonance cholangiopancreatography was not done. He was the youngest patient showing PBM on CTC (arrows show the PBM).
reflux into the bile duct or with bile juice regurgitation into the pancreatic duct. These regurgitations can cause various pathologic conditions. Therefore, the physiologic dynamics of bile and pancreatic juice is critically important. For this purpose, contrast material–enhanced MRCP or dynamic MRCP with secretin was reported [5]. For these purposes, the preoperative investigation is performed in combination with several imaging modalities. Endoscopic retrograde cholangiopancreatography is regarded as the most definitive and reliable diagnostic method of revealing anatomical details of the biliary tree and PBM preoperatively [19]. However, limitations include the requirement of general anesthesia and an invasive nature
with possible complications, making physicians hesitant to choose this option. In addition, ERCP does not allow for the evaluation of the pancreatic parenchyma. For these reasons, we have not performed routine ERCP since 2001 in our institute, with the exception of non–dilatation-type PBM. Alternatively, CT and MRCP have become more common. Imaging modalities continue to advance, and these technologies can provide more detailed information on both biliary and pancreatic ductal anatomy. There are few reports concerning the comparison study between the sensitivity of MRCP and CTC in PBM [20], and we attempted to clarify the diagnostic impact of these modalities in this study. Magnetic resonance cholangiopancreatography has been an important diagnostic tool in the investigation of pediatric hepatobiliary pancreatic diseases because it requires neither
Fig. 2 Magnetic resonance cholangiopancreatography images of the biliary duct of a 1-year 11-month-old girl with cystic-type choledochal dilatation. Although MRCP shows PBM (A), CTC failed (B) to show this diagnosis. C, Intraoperative cholangiopancreatography. She was the youngest patient showing PBM on MRCP (arrows show the PBM).
Impact of CTC and MRCP on pancreaticobiliary maljunction
1377
Fig. 3 Computed tomography cholangiography images of the biliary duct of a 1-year-old boy with nondilated choledochus. Computed tomography cholangiography with MDCT clearly shows PBM including the whole length of the main pancreatic duct and long, common channel in 3D rendering images, whose quality was far superior to MRCP images (A, MRCP; B, CTC).
radiation exposures nor contrast agents. In PBM, there are many reports regarding its effectiveness [4-7]. In this study, we could obtain a diagnostic accuracy of almost 100% in the presence of a choledochal cyst and high sensitivities for the intrabile and extrabile duct in MRCP, which is particularly helpful for the type IV-a choledochal cyst. Moreover, the rate of successful MPD visualization was as high as 81.5%. Its greatest strength compared with CTC is its availability for pancreatobiliary obstructed patients such as patients with severe jaundice and protein plugs. However, it is still challenging to perform a quality examination in young children. Magnetic resonance cholangiopancreatography is limited by small-caliber ducts, poor signals, and patient motion, creating artifacts. Therefore, the minimum age for visualization of PBM in MRCP was 1 year and 11 months, more than 1 year greater than the minimum age requirement for CTC. Computed tomography cholangiography has been useful for preoperative biliary investigation using the administration of biliary contrast material. The great advantage of CTC is its ability to produce high-quality images without respiratory artifacts and to provide more anatomical detail of the biliary tract than conventional intravenous infusion cholangiography [21,22]. Contrast media in CTC is absorbed by hepatocytes and excreted into the bile after intravenous injection. Therefore, CTC is considered to reflect the dynamic and physiologic flow of the bile, and some cases with PBM showed biliopancreatic reflux [10], and it might be useful for the functional evaluation. The potential disadvantages of CTC are irradiation, allergic reactions caused by Biliscopin injections, and decreased efficacy with visualization in patients with hyperbilirubinemia (N3 mg/dL). The recent evolution of MDCT scanners has led to rapid true isotropic imaging, because thin-section axial images can be obtained at submillimeter slice thicknesses. Simultaneously, the increasing power of 3D applications has allowed for practical postprocessing of such data [23]. In this study, CTC
had some disadvantages for the visualization of the pancreas system compared with MRCP. However, the sensitivity for PBM was similar to MRCP, and CTC with MDCT could provide more elaborate images than MRCP, even in an infant younger than 1 year. Lam et al [20] showed that the sensitivity of the main pancreatic duct in drip infusion cholangiography–computed tomography and MRCP was 63.6% and 45.5%, respectively, in their series consisting of 11 children. The incidence of drip infusion cholangiography–computed tomography was higher than that in the present study and that fact implied that there is room for improvement in our method. To improve the sensitivity of MPD, we have used several different mechanisms including the slow administration of contrast agent and rolling the patient on the CT scanner; however, a truly reliable method is still unclear and further studies are necessary. Finally, MRCP is superior to CTC in the visualization of the intrahepatic duct and pancreatic system; however, it is still challenging to perform a good-quality examination in young infants. The great advantage of CTC is its ability to produce high-quality images without respiratory artifacts and to allow for an accurate assessment of the presence of PBM equivalent to MRCP. Although our series has a limited number of patients, CTC with MDCT seems to be a promising modality to diagnose PBM and navigate surgery especially in cases with non–dilatation type PBM. Neither CTC nor MRCP is an exclusive modality, and physicians should use them in combination. The common goal is to provide the most accurate and feasible preoperative plan for safety and treatment accuracy in affected children.
References [1] Jona JZ, Babbitt DP, Starshak RJ, et al. Anatomic observations and etiologic and surgical considerations in choledochal cyst. J Pediatr Surg 1979;14:315-20.
1378 [2] Babbitt DP, Starshak RJ, Clemett AR. Choledochal cyst: a concept of etiology. Am J Roentgenol Radium Ther Nucl Med 1973;119:57-62. [3] Iwai N, Yanagihara J, Tokiwa K, et al. Congenital choledochal dilatation with emphasis on pathophysiology of the biliary tract. Ann Surg 1992;215:27-30. [4] Kamisawa T, Tu Y, Egawa N, et al. MRCP of congenital pancreaticobiliary malformation. Abdom Imaging 2007;32:129-33. [5] Chavhan GB, Babyn PS, Manson D, et al. Pediatric MR cholangiopancreatography: principles, technique, and clinical applications. Radiographics 2008;28:1951-62. [6] Hamada Y, Tanano A, Takada K, et al. Magnetic resonance cholangiopancreatography on postoperative work-up in children with choledochal cysts. Pediatr Surg Int 2004;20:43-6. [7] Kim MJ, Han SJ, Yoon CS, et al. Using MR cholangiopancreatography to reveal anomalous pancreaticobiliary ductal union in infants and children with choledochal cysts. AJR Am J Roentgenol 2002;179:209-14. [8] Persson A, Dahlström N, Smedby O, et al. Volume rendering of threedimensional drip infusion CT cholangiography in patients with suspected obstructive biliary disease: a retrospective study. Br J Radiol 2005;78:1078-85. [9] Okada M, Fukada J, Toya K, et al. The value of drip infusion cholangiography using multidetector-row helical CT in patients with choledocholithiasis. Eur Radiol 2005;15:2140-5. [10] Fumino S, Tokiwa K, Katoh T, et al. New insight into bile flow dynamics in anomalous arrangement of the pancreaticobiliary duct. Br J Surg 2002;89:865-9. [11] Fumino S, Ono S, Shimadera S, et al. Impact of age at diagnosis on clinical features in children with anomalous arrangement of the pancreaticobiliary duct. Eur J Pediatr Surg 2010;20:325-9. [12] Todani T, Akita E, Eto T, The Japanese Study Group on Pancreaticobiliary Maljunction. Diagnostic criteria of pancreaticobiliary maljunction. J Hepatobiliary Pancreat Surg 1994;1:219-21.
S. Fumino et al. [13] Iwai N, Fumino S, Tsuda T, et al. Surgical treatment for anomalous arrangement of the pancreaticobiliary duct with nondilatation of the common bile duct. J Pediatr Surg 2004;39:1794-6. [14] Todani T, Watanabe Y, Narusue M, et al. Congenital bile duct cysts: classification, operative procedures, and review of thirty-seven cases including cancer arising from choledochal cyst. Am J Surg 1977;134: 263-9. [15] Fumino S, Iwai N, Deguchi E, et al. Clinical significance of the confluence of the cystic duct in patients with anomalous arrangement of the pancreaticobiliary duct. J Pediatr Surg 2007;42: 1319-23. [16] Sugimoto T, Yamagiwa I, Obata K, et al. Choledochal cyst and duodenal atresia: a rare combination of malformations. Pediatr Surg Int 2004;20:724-6. [17] Komuro H, Makino S, Tahara K. Choledochal cyst associated with duodenal obstruction. J Pediatr Surg 2000;35:1259-62. [18] Dalvi AN, Pramesh CS, Prasanna GS, et al. Incomplete pancreas divisum with anomalous choledochopancreatic duct junction with choledochal cyst. Arch Surg 1999;134:1150-2. [19] Kim OH, Chung HJ, Choi BG. Imaging of the choledochal cyst. Radiographics 1995;15:69-88. [20] Lam WW, Lam TP, Saing H, et al. MR cholangiography and CT cholangiography of pediatric patients with choledochal cysts. AJR Am J Roentgenol 1999;73:401-5. [21] Hoglund M, Muren C, Boijsen MW. Computed tomography with intravenous cholangiography contrast: a method for visualizing choledochal cysts. Eur J Radiol 1990;10:159-63. [22] Hamada Y, Sato M, Sanada T, et al. Spiral computed tomography for biliary dilatation. J Pediatr Surg 1995;30:694-6. [23] Tamm EP, Balachandran A, Bhosale P, et al. Update on 3D and multiplanar MDCT in the assessment of biliary and pancreatic pathology. Abdom Imaging 2009;34:64-74.