Laparoscopic Roux-en-Y Hepaticojejunostomy in Children with Long Common Pancreaticobiliary Channel: Surgical Technique and Functional Outcome

Introduction

Apancreaticobiliary maljunction (PBM), defined as a high union of pancreatic and bile duct—a long common channel (LCPC)—can lead to obstructive jaundice and acute pancreatitis. In addition, it has been reported that PBM is a risk factor for developing biliary carcinoma, depending on the degree of bile duct dilatation. ¹ The aim of treatment is the prevention of the reciprocal reflux of bile and pancreatic juice into the pancreas and the bile duct system. ² A resection of the common bile duct (CBD) and the gallbladder with hepaticojejunostomy is recommended in systems with dilated CBD. The standard procedure is a bile diversion with a Roux-en-Y hepaticojejunostomy. To date, this procedure has usually been performed through an open, conventional approach. In the literature, there is an increasing number of reports on the minimally invasive resection of choledochal cysts in children.^3–5 In this article, we report on our own experience with laparoscopic treatment by hepaticojejunostomy in 7 children with LCPC and their functional outcome.

Patients and Methods

PBM with LCPC was diagnosed by high-resolution magnetic resonance cholangiopancreatography (MRCP) with diaphragm navigation and three-dimensional (3D) visualization ⁶ preoperatively. According to the definition in the literature, all patients had a high junction of pancreatic and bile duct outside the duodenal wall (>15 mm) defined as an LCPC.^7,8 These findings were correlated with endoscopic retrograde cholangiopancreatography (ERCP) and intraoperative findings. ERCP was performed preoperatively for the removal of prepapillary gallstones therapeutically.

A four-trocar technique (Fig. 1) with 3- and 5-mm ports was used for the minimally invasive approach—one port for the laparoscope, two operative ports, and one port for liver retraction. The resection of a choledochal cyst, or dilated choledochal duct and anastomosis of hepatic ducts and bowel, was performed, via minimally invasive surgery (MIS), as a laparoscopic hepaticojejunostomy (LHJ). The first step included the preparation and dissection of the dilated extrahepatic bile ducts. The distal CBD was transected and a suture ligated by intracorporeally knotting. This procedure was performed under traction on the CBD to transect the CBD as close as possible to the duodenal wall with careful prevention of the pancreatic duct. The dilated portion of the CBD or choledochal cyst, respectively, was resected, then, in conjunction with the gallbladder. In cases of extended resection proximal to the conjunction of the hepatic ducts, the union of the remaining hepatic ducts was reconstructed in a semicircular manner with interrupted polydioxanone monofilament 5-0 sutures of the posterior wall of the ducts (Fig. 2).

OPEN IN VIEWER

FIG. 1.

Trocar setting. I: laparoscope; II + III: operative ports; IV: liver retraction.

OPEN IN VIEWER

FIG. 2.

Anastomosis of the posterior wall of right and left hepatic duct (patient 6).

A Roux-en-Y loop was constructed as follows: marking with suture of the jejunum 15 cm from Treitz; eventeration of the jejunal loop through the extended subumbilical port incision; transection of the gut 45 cm from Treitz; and closure of the distal part by extracororpeally suturing and end-to-side anastomosis of the proximal part of the jejunal limb in the level of the marking suture. The Roux limb was brought to the level of the common hepatic duct (CHD) or reconstructed hepatic ducts retrocolically. The anastomosis of the CHD and the Roux limb was performed laparoscopically with interrupted sutures. All children received a peritoneal drainage for 3–4 days. The drain tubing was brought out through an operative port side. All patients had a defined follow-up (Table 1). A hepatobiliary iminodiacetic-acid (HIDA) scan was used as a proof for tracer uptake in the liver and reestablished bile flow 8 weeks after the operation, in addition to the clinical course, repeated ultrasound, and liver-function tests.

Results

From 2004 to 2008, 7 children underwent MIS for LCPC. Mean age was 46 months (median, 41; range, 13 months to 10.3 years). All children showed clinical symptoms, such as abdominal pain, and changes in laboratory tests due to cholangitis, obstructive jaundice, and/or recurrent pancreatitis. In all patients, a LCPC (mean length, 17 mm; range, 15–24) could be detected by MRCP. Prepapillary gallstones leading to obstructive jaundice were found in 3 patients. In 5 children, the diagnosis of LCPC could be confirmed by ERCP performed therapeutically, indicated mainly for stone extraction. All children were operated on by laparoscopy after the normalization of laboratory parameters for cholestasis and pancreatitis.

In patient 2, the laparoscopic approach failed due to a large, vulnerable liver—a conversion to an open procedure became necessary for anastomosis of the bowel and CBD after laparoscopic cholecystectomy and dissection of the CBD. Two patients (patients 6 and 7) had a choledochal cyst type 4, according to Todani's classification. Therefore, an extended dissection of CBD, CHD, and of the right and left hepatic duct was necessary. The reunion of the separated hepatic ducts had to be reconstructed laparoscopically, as mentioned above. In patient 6, the separated right and left hepatic duct (Fig. 2), and in patient 7 the right hepatic duct and the duct of segments 2/3 and 4, were joined by sutures prior to bowel anastomosis. Mean operating time in all patients was 246 minutes (median, 191; range, 171–363).

Postoperatively, feeding was started in all patients on day 2 or 3. In the postoperative course, a leakage of the hepatic anastomosis occurred in patient 7 and, therefore, an open revision was carried out. Mean follow-up was 34 months (median, 34; range, 7–58). A HIDA scan carried out 8 weeks after operation showed a normal uptake of the tracer in the liver and prompt passage into the bowel in all patients. No gallstones were detected by ultrasound in the time of follow-up. In 6 of 7 patients, the obstructive jaundice, cholangitis, or pancreatitis did not occur after the operation. Laboratory parameters and ultrasound were normal in the postoperative course. Intrahepatic distension of the bile ducts decreased in all children, if existing preoperatively. In patient 5, recurrent episodes of pancreatitis occurred 1 year after the LHJ. A 20-mm remnant of CBD was detected by MRCP and ERCP, which could be excised laparoscopically.

Discussion

The resection of the cyst or dilatated CBD and biliary diversion can be performed in the same manner as in an open approach by creating a Roux-en-Y limb and anastomosis of the common hepatic duct as a hepaticojejunostomy. The operative time in this series is comparable to other studies with laparoscopic-treated choledochal cysts: 1–7.5 hours, ⁴ 3.3–4 hours, ³ and 180–420 minutes. ⁹ The conversion rate with 1 of 7 patients is comparable with other series: 3 of 7. ⁴ The conversion in patient 2 was considered rather as an expression for a responsible handling of the minimally invasive technique to prevent further complications. One aim of the operation is to prevent recurrent pancreatitis, which may be difficult in treatment. ¹⁰ Episodes of pancreatitis after an operation occurred in only 1 of 7 patients in this series. A remnant of CBD (length, 2 cm) was found by MRCP an ERCP, which was mentioned as a cause of pancreatitis. The remnant was resected by laparoscopy successfully. Another aim is to create a sufficient bile flow and prevention of gallstone disease, developing in a dilated intrahepatic bile duct. Repeated operations or interventions are often required for the treatment of intrahepatic gallstones. ¹¹ In this series, the laboratory parameters and HIDA scan showed prompt bile drainage into the bowel and a decrease of the dilation of the bile ducts in ultrasound in all patients. In none of the patients could gallstones be seen in ultrasound investigations postoperatively. The risk of developing a biliary carcinoma can be minimized by excising the cystic part of the extrahepatic biliary tract. ¹² As shown in patients 6 and 7, an extended resection could be performed laparoscopically for choledochal cysts involving the hepatic ducts. The follow-up time has not been long enough to rule out the development of carcinoma in an age above 20 years. ¹³ Controversy exists in nondilated systems with LCPC, if an excision of the CBD or only of the gallbladder should be carried out. ¹⁴ However, there might be technical difficulties in performing an LHJ for biliary diversion in the nondilated CBD.

Advantages of the laparoscopic approach are an excellent overview of the extrahepatic structures and the porta hepatis and in the optical magnification of the structures for a safe dissection and suturing additionally. With respect to the distal bile duct, the surgical principle should be excision of a portion of the distal bile duct as near as possible to the duodenal wall with care to the pancreatic duct or an LCPC. As shown in patient 5, the laparoscopic procedure might also be a useful approach for the reresection of remnants of CBD, who perhaps can lead to recurrent pancreatitis.

Hospital stay in our series ranged from 8 to 22 days, as shown in Table 2, and was longer than in other series, ranging from 3 to 12 days. ⁴ This was due to prolonged hospital stay preoperatively for the treatment of acute pancreatitis prior to surgery. Complex situations, in which the hepatic ducts must be rejoined after the resection of a choledochal cyst, can also be managed by MIS safely. As shown in patient 7, this procedure is technically challenging, and there is a remarkable risk of leakage. Problems, such as leakage, occurred also in other studies. ⁴

In 5 of 7 patients, the postoperative course was completely uneventful. Although the same problems or complications as in the open procedure, such as leakage or ileus, can appear. ¹⁵ Advantages of the laparoscopic approach could also be seen in a reduction of postoperative pain and hospital stay and in a remarkablely better cosmesis. Despite there existing no data in the literature about reduced pain, shortened postoperative stay, and better cosmesis after minimal invasive correction of bile duct anomalies in children, these effects have been proven sufficiently in other, more frequent disorders, such as in pyeloplasty.^16,17

Conclusions

The laparoscopic approach in the maljunction of pancreaticobiliary ducts such as LCPC might be a new, safe surgical treatment option in infants and children—even in patients with extended choledochal cyst. We consider this procedure to be as safe and effective as the open operation.

Disclosure Statement

No competing financial interests exist.