Early Cholangitis after Portoenterostomy in Children with Biliary Atresia

Abstract

Aims and Objectives:

Biliary atresia (BA) is a cholangiodestructive disease of the biliary tree. The first line of treatment is a Kasai portoenterostomy (PE) following which patients may develop cholangitis. We studied the effect of early cholangitis on the outcome of PE, namely jaundice clearance and early native liver survival (NLS).

Methods:

We reviewed the data of all children who developed cholangitis after PE from our prospectively maintained database of children with BA. The standardized treatment of all children in the database is described. The frequency and nature of these episodes were characterized, and the outcome of PE and NLS 1 year after PE was calculated.

Results:

Of 62 children who underwent PE in our institutions, 27 developed cholangitis. All episodes of cholangitis occurred within 14 months of PE. Of 25 children who cleared jaundice in the overall series, 19 had cholangitis. The incidence of cholangitis was significantly higher in children who cleared jaundice. Nine children who had cholangitis are alive with native livers for more than 1 year after PE. Twelve children had intractable cholangitis. Three of these children are alive with native liver 1 year after PE.

Conclusion:

In our series, cholangitis occurred in most children who cleared jaundice. Furthermore, the 1-year NLS of children who developed cholangitis was 33%.

INTRODUCTION

In biliary atresia (BA), the intrahepatic and extrahepatic biliary ductular continuum is destroyed by an inflammatory and fibrotic process. A portoenterostomy (PE) described by Professor Morio Kasai is the first line of management in controlling this disease. However, more than half of the children require liver transplantation (LT) even after this procedure. Cholangitis is the most common complication that occurs after a PE. It has been postulated that cholangitis causes rapid progression of liver injury leading to cirrhosis. Recurrent attacks of cholangitis have been shown to be a prognostic marker for hastened liver failure leading to the early requirement of transplantation.[1] The effect of early cholangitis (within 1 year of PE) on jaundice clearance has not been reported. In this study, we aimed to assess the impact of early cholangitis on the outcome of PE with regard to jaundice clearance and short-term native liver survival (NLS).

METHODS

Since February 2013, we maintained a prospective database of children with BA who had undergone PE in our institution. All PEs were performed by a single surgeon, and the technique used was an extended Kasai PE. This technique involved creation of a 60 cm roux loop to facilitate biliary drainage. After dissection of the hepatic vasculature at the hilum and excision of the hilar plate, hepaticojejunostomy was performed as an anastomosis extending from the bifurcation of the right hepatic artery into anterior and posterior branches to the Rex confluence of the left portal vein with the obliterated umbilical vein. The postoperative care was standardized and all children received piperacillin/tazobactam intravenously at a dose of 100 mg/kg four times a day for 1 week. Oral steroids (prednisolone) were started on day 6 after the establishment of oral feeds at a dose of 5 mg/kg, which was tapered weekly to 0.5 mg/kg (6 weeks). Patients were discharged after 1 week and given cyclical antibiotics as prophylaxis, which included amoxicillin-clavulanic acid (40 mg/kg/day in two divided doses) and cefpodoxime (10 mg/kg/day in two divided doses). This was continued for 6 months. Cholangitis was diagnosed by (1) the occurrence of fever, pale stools, (2) elevated white cell count and C-reactive protein, and (3) with elevation of bilirubin and/or elevation of liver enzymes. Once the diagnosis of cholangitis was considered, we performed routine blood cultures in these patients after which they were admitted for administration of intravenous (IV) piperacillin–tazobactam. If there was no clinical improvement, we upgraded the antibiotics to meropenem. In children with intractable fever and persistent jaundice, even in the absence of a positive blood culture, we empirically added antifungals (amphotericin B). All children underwent endoscopy and/or variceal banding 1 year after PE.

For this study, data were collected on the incidence of cholangitis after PE, including number and nature of such episodes, jaundice clearance after PE, need for LT, and the duration of survival with native liver. Patients who presented with 1–2 episodes of cholangitis amenable to a course of IV antibiotics were characterized as having simple cholangitis. Intractable cholangitis involved multiple episodes of infection or an episode of infection requiring prolonged treatment with antibiotics and antifungals. The outcome of PE in children who had cholangitis was computed.

Continuous variables were expressed as means with standard deviation and analyzed with Student t-test. Discrete variables were expressed as percentages and were analyzed using the Chi-square test. P < 0.05 was considered statistically significant. All data were collected after approval from our institutional review board.

RESULTS

During a 5-year period, from 2013 to 2018, 62 children underwent PE in our institution for BA. There was no perioperative mortality in our series. In this overall group, 25 (40%) children cleared jaundice and 27 (43%) developed cholangitis after PE. On comparing the incidence of cholangitis among those who cleared jaundice to those who did not, the occurrence of cholangitis was significantly higher in those who cleared (n = 19, 76%) compared to those who did not (n = 8, 22%) clear jaundice (P < 0.001) [Table 1]. The average age at the time of surgery of children who developed cholangitis was 84.4 ± 30.7 days compared to 77.3 ± 23.2 days (P = 0.30). The follow-up ranged from 3 to 51 months. Blood culture was positive in two patients (Escherichia coli). Three children in this group had biliary atresia splenic malformation (BASM).

Table 1.

Demographics and results

Variable	Total (n=62)	Cholangitis (n=27)	No cholangitis (n=35)	P
Age at KPE, days	80.4±26.7	84.4±30.7	77.3±23.2	0.30
Sex, male/female	32/30	15/12	17/18	0.62
Laboratory values at transplant
Total bilirubin	10.9±3.3	10.6±4.0	11.1±2.6	0.59
GGT (IU/L)	652 (55-2281)	695 (84-2281)	532 (55-1755)	0.59
AST	209 (65-790)	180 (136.5-487)	221 (114-790)	0.14
Platelet count (1000 cells/m3)	432.5±166.1	449.5±159.9	419.4±171.9	0.48
APRi	1.61 (0.30-8.72)	0.87 (0.30-8.72)	1.36 (0.47-7.77)	0.08
Posttransplant
Jaundice clearance, n (%)	25 (40.3)	19 (70.4)	6 (17.1)	<0.001

APRi: AST-to-platelet ratio, AST: Aspartate aminotransferase, GGT: Gamma-glutamyltransferase, KPE: Kasai portoenterostomy

Fourteen (51.8%) children had cholangitis within the 1^st month after PE. All episodes occurred within the 1^st year after PE except in one child who had an episode of cholangitis 14 months after PE. Twelve (44.4%) children had intractable cholangitis. All children responded to the treatment measures described above. Nine children who had PE more than 1 year ago and had cholangitis are alive with native liver. Thus, the overall NLS rate after 1 year in children who had cholangitis post PE was 33%. Three children with intractable cholangitis are alive with native liver at 1 year. Two children underwent LT for synthetic liver dysfunction. Both children did not clear jaundice after PE. All three children with BASM developed cholangitis. None of the children in this series had variceal bleeding.

DISCUSSION

Recurrent cholangitis is an indication for LT even in children who achieve jaundice clearance. Our study showed a good 1 year NLS for children who cleared jaundice and developed cholangitis. Since most episodes of cholangitis occurred within the 1^st year after PE, we feel that by managing these episodes effectively, we can improve the NLS and avoid or postpone LT.

The establishment of autoanastomosis between intrahepatic bile ducts and intestinal epithelium after a PE takes about 6 weeks. Hence, before this, the transected microscopic ducts in the hilum are susceptible to inflammation and closure.[2] If the intrahepatic bile ducts do not drain well, cholangitis may occur either until good drainage is established or until the bile ducts undergo fibrosis.[2] The incidence of cholangitis after a PE has been reported to range from 40% to 93%.[2] The pathogens usually involved include intestinal flora such as E. coli, Pseudomonas aeruginosa, Enterobacter cloacae, Acinetobacter baumani, and Salmonella typhi.[3] These pathogens translocate from the intestine into the liver from the enteric conduit and proliferate causing inflammation and obstruction to bile drainage.[4,5] This translocation is more common in the presence of obstructive jaundice probably due to sluggish bile flow.[2] In some cases, patients develop cystic dilation of the intrahepatic bile ducts and become prone to recurrent episodes of cholangitis due to bile stasis in the cavities.[6]

Multiple surgical and postoperative management approaches to prevent cholangitis have been reported. Surgical modifications to the original Kasai operation to prevent the occurrence of cholangitis have included exteriorization of the bilioenteric loop and creation of an intussusception type antireflux valve to prevent reflux. However, these alterations are not currently in use because they have not proven to be effective and result in complications such as electrolyte imbalance and bleeding due to portal hypertension.[2,7] Steroids have been known to reduce the inflammation and improve bile flow during episodes of cholangitis. A short burst of steroids as an adjunct to antibiotics in intractable cholangitis has been proved to be extremely successful.[8] Second- and third-generation cephalosporins have been used to treat cholangitis sometimes in conjunction with aminoglycosides. Meropenem has also been effectively used for treatment.[9] Prophylactic antibiotics prevent recurrent bouts of cholangitis.[10] Recently, IV immunoglobulin has also been used to treat intractable cholangitis.[11] Lactobacillus casei rhamnosus was effective in preventing cholangitis in patients after PE.[12]

Some studies have reported that good postoperative bile drainage and jaundice clearance is associated with a lower risk of cholangitis.[13] This has not been our experience. In our series, the incidence of cholangitis was significantly higher in children who cleared jaundice compared to those who did not. It is probable that the opening up of drainage channels during PE might cause sluggish bile fl ow initially which if coupled with translocation of intestinal pathogens into the liver leads to an infection. Fifty percent of children had early cholangitis within 1 month after PE which corroborates the fact that infection occurs before establishment of drainage channels which commonly takes about 6 weeks. Although it has been reported that cholangitis within 3 months of PE was a negative prognostic factor for NLS, 33% of children who developed cholangitis in our series are alive with native liver for more than 1 year after PE.[14]

Studies have reported that increased number of cholangitis episodes was predictive of a poor PE outcome.[15] It has been reported that recurrent cholangitis leads to a greater probability of cirrhosis and early transplantation.[1,16] In our series, 40% of children had intractable cholangitis and 3 of them are alive with native liver at 1 year.

Cholangitis as a late complication has been reported in long-term survivors of PE.[17] Further follow-up of our patients is needed to look at long-term NLS in children who develop early cholangitis after PE for BA as well as to chronicle the later episodes of cholangitis which might occur in long-term survivors of PE.

CONCLUSION

Cholangitis occurred predominantly in children who cleared jaundice in our series and these children obtained a 33% 1-year NLS. Since these episodes generally decrease 1 year after PE, we feel that if we optimize the management during this period, some of these children may survive with native livers and may never require an LT. Prevention of cholangitis with a long Roux-en-Y loop, routine use of postoperative steroids, prolonged antibiotic prophylaxis, and aggressive management of infection with antibiotics might lead to improved NLS.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.