Comparison of Characteristics and Outcomes in Antenatally and Postnatally Detected Choledochal Cyst in Infants and Young Children in a Laparoscopic Surgery Center

Introduction

Choledochal cyst (CC) is characterized by congenital dilatation of the biliary system and is more common in female gender and in Orientals. ¹ Complications can be aroused from CC including obstructive jaundice, cholangitis, rupture, and even malignancy. ¹ Thus excision of CC and hepaticojejunostomy (HJ) is recommended to avoid the development of complications.^1,2

With the advancement in antenatal ultrasonography (USG), management of antenatally (AN) detected CC has been increasingly reported.^3–7 Laparoscopic management of CC has gained popularity since it was first reported in 1995.^8–12 Laparoscopic excision of CC and HJ is the preferred approach in the management of CC in our institute. ¹³ We have reported the use of laparoscopy in the management of AN detected CC. ¹⁴ In this study, we compared the characteristic and outcome in children with AN and postnatally (PN) detected CC that were operated before 36 months of age.

Materials and Methods

A retrospective study was conducted among all children who underwent excision of CC and HJ younger than 36 months of age between October 2004 and October 2019. The patients were divided into two groups: AN (patient with AN detected CC) and PN (patient with PN detected CC) group. In our institute, we aimed to perform surgery for AN detected CC at 3 months of age in asymptomatic patients. The protocol of the work-up of AN group had been described. ¹⁴ In PN group, we aimed to perform operation after acute complication had been settled by antibiotics or interventions including percutaneous transhepatic biliary drainage (PTBD) or endoscopic retrograde cholangiopancreatography (ERCP).

All patients had either magnetic resonance cholangiography (MRC) or computer tomography scan performed preoperatively. Our surgical technique in laparoscopic excision of CC and HJ was described previously.^8,9 In brief, the patient was placed in supine position and the surgeon operated on the right side of the patient. A 5 mm 30° laparoscope was inserted through a subumbilical 5 mm port. CO₂ pneumoperitoneum was established at 8–12 mmHg. Two 3.5–5 mm ports were inserted at right upper quadrant of the abdomen. Another 5 mm port was inserted over left upper quadrant of the abdomen. A liver retractor was inserted through a right subcostal incision. The gallbladder was freed from the liver and mobilized toward the cyst, and the cystic artery was cauterized. Duodenum and the pancreas were freed from the cyst. The most distal common duct was either cauterized with diathermy or ligated with suture. The cyst was then mobilized toward the common hepatic duct. The cyst was excised just distal to bifurcation of the left and right hepatic ducts. The common hepatic duct was flushed with normal saline. The umbilical wound was extended and a 40 cm Roux-en-Y jejunal loop was fashioned extracorporeally using hand suture technique. The Roux loop was then returned into peritoneal cavity and delivered to the hilum through a retrocolic route. An end-to-side HJ was fashioned with continuous 5-0 BIOSYNTM sutures, first on the posterior layer and then on the anterior layer. A drain was inserted at the subhepatic space in all cases.

Operation was performed by the same team of experienced pediatric surgeons. Postoperatively, MRC was arranged 6 months after the surgery and followed-up by yearly USG to assess any dilatation of the intrahepatic ducts (IHDs). Liver function test (LFT) was monitored regularly.

The demographic of patients, presenting symptoms, preoperative intervention, and type of CC were reviewed. Classification of CC was based on Todani's classification. As preoperative dilatation of IHD may be secondary to distal obstruction; patients with resolved dilatation of IHD in the postoperative 6-month MRC was classified as type I CC. Patients with persistent IHD dilatation postoperatively were classified as type IVa. The surgical approach, conversion to open procedure, operative time, complications, length of hospital stay, and follow-up time were studied. The results between the two groups were compared.

Results

Within the 15-year study period, 39 children (11 boys, 28 girls) younger than the age of 36 months underwent excision or CC and HJ in our institute. Twenty-one patients were in the AN group, whereas 18 patients were in the PN group. The overall median follow-up time was 71 months (range = 6–183 months). There was no lost to follow-up in this study.

The demographics of the two groups were given in Table 1. The median age of operation in the AN group was 3 months, which was our target age of operation. Preoperative imaging showed dilated IHD in 11 patients but the dilatation was resolved in 8 patients in postoperative MRC. Overall, 37 patients had Todani type I CC and only 3 patients had type IVa CC.

Six patients (29%) in the AN group were symptomatic before operation (Table 2). Four patients had painless obstructive jaundice. One patient had cholangitis with jaundice and 1 patient had cholangitis without jaundice. In the PN group, all patients were symptomatic. Thirteen patients (72%) presented with obstructive jaundice and 9 patients (50%) had concurrent cholangitis. One patient had cholangitis and ruptured CC at presentation. One patient with cholangitis has concurrent pancreatitis. Only 2 patients complained of abdominal pain on presentation, probably because of their young age. Four patients in the PN group had preoperative intervention including 2 patients, 27 and 32 months of age, who underwent ERCP with insertion of biliary stent to relieve the obstruction. One patient 32 months of age had failed ERCP and subsequently underwent PTBD. Another patient 25 months of age had PTBD for biliary diversion.

The perioperative outcomes are given in Table 3. As laparoscopic excision of CC and HJ is our standard approach, all patients in the AN group and 67% (12/18) patients in the PN group underwent laparoscopic surgery. The reasons for primary open surgery in the PN group were 2 patients had PTBD insertion, 2 patients had unfavorable anatomy on MRC, 1 patient had ruptured CC at presentation, and 1 patient had history of ileal perforation with operation performed in neonatal period. Four patients had conversion to open surgery in the AN group. Two patients with concurrent cholangitis had early conversion to open excision of CC, whereas 2 patients with small common hepatic duct had conversion to open HJ after laparoscopic excision of CC. Only 14% (3/21) patients in the AN group had ligation of most distal common bile duct (CBD). Cauterization was performed in the rest (86%) of the patients. On the contrary, cauterization was only performed in 33% (6/18) of patient in the PN group.

The overall median operative time was 300 minutes (range = 180–570 minutes) and the overall median length of hospital stay was 8 days (range = 6–25 days). In the PN group, the median operative time was 292.5 minutes (range = 180–330 minutes) in open surgery and was 330 minutes (230–570 minutes) in laparoscopic surgery. Despite smaller size of the patient in the AN group, there were no statistical differences on the operative time and the length of hospital stay. Three patients in the AN group had postoperative complications. Two patients had postoperative fever and required prolonged duration of antibiotics and 1 patient had minor anastomotic leak. The leak was managed successfully by percutaneous drainage. Two patients in the PN group who underwent laparoscopic excision of CC developed intestinal obstruction at postoperative 1 and 5 years, respectively. All patients had normal postoperative LFT in the latest follow-up and there was no feature of liver cirrhosis on USG. Neither cholangitis, stone complication, nor mortality was observed in both groups.

Discussion

This study demonstrated that laparoscopic management is safe and feasible in infants and young children with CC. CC is a congenital disease and with the prevalence of antenatal USG, CC is increasingly being diagnosed before birth.^1,7,15,16 In this study 54% (21/39) patients operated upon before 36 months of age presented AN.

As reported previously, 14 laparoscopic management of AN detected CC was technically feasible, but we do not know whether the outcome would be better if the operation was performed at a slight older age and when the size of patients were bigger. Our resulted showed that despite statistically significant differences on the age and body weight; the operative outcomes including the operative time, postoperative complication, and length of hospital stay were comparable between the two groups.

Patient with CC tended to develop complications and the complications tended to present early. ¹ It was not surprising that all patients in the PN group were symptomatic before operation. Even in the AN group, 29% (6/21) of patients were symptomatic before operation. However, only 2 patients (9.5%) in the AN group had cholangitis before operation, whereas half the patients in the PN group had cholangitis. This study highlighted the importance of early detection and early surgery. Patients with CC who presented with cholangitis required antibiotics treatment before definite surgery. However, if the cholangitis was not settled by antibiotics alone, decompression of the biliary system may be required. In older children, ERCP will be the surgical intervention of choice to relieve the obstruction.^17,18 On the contrary, ERCP in infants and young children was technically difficult and the procedure was likely to be performed under general anesthesia. ¹⁸ In this study, cholangitis in 5 patients in the PN group was settled by ERCP or antibiotics. We successfully performed the operation by laparoscopic approach without conversion to laparotomy.

Despite laparoscopic excision of CC and HJ being successfully performed in 74% patients (29/39), we do think there are limitations in laparoscopic approach in young infants. Regarding the 2 patients in the AN group with cholangitis, 1 patient had cholangitis few days before the scheduled date of operation. Another patient had cholangitis soon after birth and operation was performed at 20 days of age in a semi-emergency setting. In both cases, the operation was started with laparoscopy but soon was converted to open procedure because the CC was inflamed and oozy and the peritoneal space was very limited. Hence, in infants and young children with cholangitis that were not controlled by antibiotics or failed ERCP, early open surgery may be a better approach than laparoscopy.

In addition, the common hepatic duct in an infant is fragile and if the common hepatic duct (CHD) is small in caliber, we do not feel safe to perform HJ laparoscopically. Although ductoplasty has been described for small CHD, whether it is technically feasible depends on the friability of CHD.^19,20 In open HJ, we used interrupted sutures to share out the tension of the anastomosis and we believe it was a safer approach to decrease the risk of anastomotic leak.

Preoperative planning is important in the decision of surgical approach. As 1 patient had widely separated nondilated left and right IHD that joined into a dilated CHD, we decided to perform open biductal HJ. We believe complete excision of entire CC is the principle of surgery and not to leave a cuff of abnormal tissue just to facilitate laparoscopic approach. ²¹

A previous study suggested that there were two different pathologies in congenital CC. ²² AN detected CC had a stenotic distal CBD and PN detected CC had a nonstenotic distal CBD. In this study, our observation echoed this suggestion. Cauterization of distal CBD was performed in 86% patients in the AN group and only in 33% in the PN group. In those cases in whom cauterization was performed, we did not identify any sizable distal CBD and we did not observe any bile leak from CC after cauterization. In this study, the median size of CC in the AN group was 4 cm and 3 cm in the PN group despite the body size being smaller in the AN group. Our findings suggested that morphology of CC in the AN group was more cystic, whereas in the PN group it was more fusiform.

In recent years, hepaticoduodenostomy (HD) has been promoted over HJ in laparoscopic management of CC.^23–25 HD is technically easier to fashion and there is no need to fashion the enteroenterostomy. Studies have also reported that the risk of intestinal obstruction was higher after HJ than HD. ²⁴ We had two cases of intestinal obstruction in this series. In both cases, the adhesion and obstruction happened in the infracolic component and thus the HJ was left undisrupted. On the contrary, we did not experience any postoperative cholangitis or biliary gastritis. Well-known complications of HD include duodenogastric reflux and long-term risk of cholangiocarcinoma or gastric carcinoma. ²⁴ In one of the recent reports studying the long-term outcome of HD in laparoscopic management of CC in children, 48% of patients after HD had radiological-proven contrast reflux into the biliary tree. ²⁶ The median follow-up was only 65.0 months. This is a worrisome finding and a lifelong follow-up is required to answer the association between HD and malignancy. ²⁷ Comparing the potential risk of malignancy in HD and the increased risk of intestinal obstruction in HJ, we still think HJ is a safer method taking into consideration the young age of our patients.

Strictures in IHDs were reported to be associated with infants with CC. ¹⁶ Because all the patients in this series were younger than 36 months, we did not attempt to use any choledochoscope to inspect the IHD. ¹ Instead we flushed the IHDs thoroughly to remove any protein plug. Within the study period, we did not observe any progressive dilatation of IHD postoperatively on postoperative MRC and USG that signified anastomotic stricture. Eleven patients had IHD dilatation preoperatively and only three had residual dilatation postoperatively. A longer follow-up is required to look for possibility of stricture and stone formation in future in particular in patients with type IV CC.

In addition, we did not notice any feature of liver cirrhosis in USG and deranged liver function on the latest postoperative follow-up.^16,22 Infants with CC were reported to have higher risk of persistent liver fibrosis postoperatively.^16,22 The risk was attributed by the elevated intrabiliary pressure secondary to chronic biliary obstruction before the operation.^16,22 We believe our approach in early operation in all the infants and children with CC minimized the risk of ongoing liver fibrosis or cirrhosis. ²⁷

The strength of this study was that all operations were performed by the same team of experienced surgeons. The median follow-up time was >5 years. We also included patients who underwent primary open surgery that could make the clinical picture more complete. The limitation of this study included retrospective data collection. The number of cases was small but this reflected the rarity of this disease even in Asian countries. In addition, we did not perform any liver biopsy to confirm the absence of biliary cirrhosis as in other reports, ⁶ because postoperatively all patients had normal LFT without feature of cirrhosis on USG. We believe our patients did not warrant such invasive procedure.

In conclusion, this study demonstrated the operative outcome between AN detected and PN detected CC were comparable. Smaller size of the patients did not pose an increase in operative risk. In contrast, early detection of CC decreased the risk of preoperative surgical intervention. We still opted for early surgery of asymptomatic AN CC at 3 months of age to minimize the chance of cholangitis before surgery. Early laparoscopic management is safe and feasible for asymptomatic AN detected CC and for symptomatic CC in infants and young children without acute inflammation.