Biliary atresia (BA) is a progressive fibrosclerosing cholangiopathy of infancy with a yet unclear etiology. If untreated, it progresses to cirrhosis with portal hypertension and liver failure, leading to death within 2 to 3 years (1). The Kasai operation was first used for BA in 1959, with encouraging results; henceforth, it has become the first-line treatment for BA (2). With the Kasai operation, bile flow can be restored through biliary reconstruction via hepatic portoenterostomy to a jejunal loop. Ascending cholangitis, however, is a frequent and often recurrent complication after surgery (3). Recurrent cholangitis causes secondary failure of the restoration of bile flow and possible exacerbation of portal hypertension, resulting in an increase in the risk of mortality and worsened prognosis (4,5). Patients with multiple episodes of ascending cholangitis are more likely to require liver transplantation than those without multiple recurrences (5,6). Therefore, prevention of cholangitis is important for patients with BA who have undergone Kasai operation.
Although the precise mechanism of cholangitis has not been established, ascending bacterial infection originating from the enteric conduit was reported as a possible explanation (3). Previous studies (5,7,8) demonstrated some oral antibiotics such as trimethoprim-sulfamethoxazole and neomycin to be effective as prophylactic agents against ascending cholangitis. A long-term antibiotic use may, however, increase the potential to develop antibiotic resistance and the psychological burden of the patient's caregiver, suggesting the possibility of decreased compliance to antibiotic treatment. Therefore, other safe prophylactic agents, in addition to antibiotics, for ascending cholangitis are needed. Probiotics are live microorganisms that have been reported to have a suppressive effect on bacterial overgrowth and to prevent infection or inflammation of the gastrointestinal tract (9,10). Lactobacillus casei rhamnosus is a probiotics that is easily available in Taiwan. This also has been used for children with chronic constipation in our previous study (11). Accordingly, the present study aimed to investigate the possibility of using L casei rhamnosus in the prophylaxis of ascending cholangitis in patients with BA.
Between 2007 and 2011, all of our authors enrolled their patients with BA ages 0 to 3 years who underwent Kasai portoenterostomy in this open-label study. All of the participants had normal stool color and jaundice-free status (total serum bilirubin level <2.0 mg/dL). Under parental consent and institutional review board approval, the total 20 patients were assigned randomly to 2 groups by our study nurse to receive treatment as follows: oral neomycin at 25 mg · kg−1 · day−1 for 4 days/wk in 10 patients and L casei rhamnosus (Lcr35; manufactured by Laboratoires Lyocentre, Aurillac, France; according to the information of the company, the strain is not deposited in an international collection) at 8 × 108 colony-forming unit per day in another 10 patients. The first case enrolled in our study was randomly assigned to the neomycin group by throwing a dice (≤3, neomycin; >3, Lactobacillus); the second case was then assigned to the Lactobacillus group; the third was assigned to the neomycin group; and the fourth was in the Lactobacillus group in rotation by one-to-one ratio in sequence. Although this is an open-label study, our study staff and parents knew the patients would be in which group only after signing the consent. The treatment duration was 6 months. Bacterial culture of stool samples was performed before initiation of the treatment and at 1, 3, and 6 months after starting the treatment to evaluate the patterns of intestinal flora.
We sampled approximately 0.2 to 0.3 g of stool and diluted it to a 10 times lower concentration. Then, we collected the diluted samples for the aerobic bacterial culture with 5% trypticase soy agar sheep blood culture media, anaerobic bacterial culture with anaerobic 6-benzylaminopurine culture media, Escherichia coli culture with the MacConkey agar culture media, and lactobacilli culture with the de Man-Rogosa-Sharpe agar culture media. Thereafter, we observed the changing patterns of the E coli and lactobacilli colony numbers against the total aerobic and anaerobic bacteria, respectively, in the stool samples. Once signs of cholangitis were observed, the patient was admitted for complete evaluation of blood counts, differential counts, C-reactive protein level, alanine aminotransferase (ALT) level, aspartate aminotransferase (AST) level, bilirubin level, urine culture, and blood culture. A diagnosis of cholangitis was made on the basis of unexplained fever (body temperature >38°C), acholic stools, increased jaundice, or positive blood culture results. Abdominal sonography was performed to detect ascites, intrahepatic bile duct lake, or liver cirrhosis. Parenteral ceftriaxone administration was the first-line antibiotic treatment, with modifications according to culture results. If a patient with BA developed cholangitis, a 14-day treatment with intravenous antibiotics was completed. After hospital discharge, the patients continued receiving their original prophylactic agent, either neomycin or probiotics.
From 1991 to date, no systemic change in post–Kasai operative care has been made, except in the concept of prophylactic antibiotic usage. Since 1997, most of the patients have been prescribed prophylactic antibiotics after operation (5). Therefore, for a historical control group, we randomly selected (by Excel's RAND function) another 10 jaundice-free patients with BA among a total of 18 cases ages 0 to 3 years who underwent Kasai operation during 1991 to 1996 but did not receive prophylaxis of cholangitis. Ursodiol (ursodeoxycholic acid) is a naturally occurring bile acid found in normal human bile. All of our cases in treatment and control groups were prescribed with oral ursodiol. In addition, we suggested the semielemental formula with adequate calories intake for patients and educated parents for adding solid food when patients grew up. The dietary policy is the same in the 3 groups. After the 6-month study period, comparisons of the frequency of cholangitis, cholangitis-free survival time for primary outcome, and body weight (z score) for secondary outcome were performed between the 3 groups. Summary statistics of the demographic variables were presented using frequency, mean, and standard deviations. Comparisons of these variables were performed using the Student t test. The Kaplan-Meier method and log-rank test were also used to assess the different treatment groups affecting cholangitis-free survival time. A P < 0.05 was considered statistically significant.
Comparisons Between Neomycin and Lactobacillus Groups
All the subjects had no significant difference in background and were randomized into 2 groups at enrollment (Table 1). No significant difference in sex distribution was observed between the 2 groups (P = 0.40). The mean ages at Kasai operation were 40.8 ± 13.75 days in the neomycin group and 52.2 ± 19.94 days in the Lactobacillus group (P = 0.16). The mean ages at the time enrolled were 0.88 ± 0.75 years in the neomycin group and 1 ± 0.74 years in the Lactobacillus group (P = 0.74). The AST and ALT levels at enrollment were, respectively, 112 ± 52 and 117 ± 74 IU/L in the neomycin group, and 99 ± 59 and 99 ± 72 IU/L in the Lactobacillus group (P = 0.62 and P = 0.59, respectively). The total and direct bilirubin levels at enrollment were 1.05 ± 0.53 and 0.58 ± 0.42 mg/dL in the neomycin group, and 0.91 ± 0.60 and 0.55 ± 0.45 in the Lactobacillus group (P = 0.59 and P = 0.88, respectively). The mean body weight z scores at enrollment were − 0.71 ± 1.71 and − 0.19 ± 1.35 in the neomycin and Lactobacillus groups, respectively (P = 0.46). All of the aforementioned data at enrollment were not significantly different between the 2 groups.
After the 6-month treatment with the prophylactic protocol, the mean body weight z scores were 0.19 ± 1.57 and 0.78 ± 1.44 in the neomycin and Lactobacillus groups, respectively (P = 0.39). The mean changes in the body weight z scores were 0.90 ± 0.95 and 0.97 ± 0.59 in the neomycin and Lactobacillus groups, respectively (P = 0.85). Cholangitis occurred in 2 patients each in the neomycin and Lactobacillus groups (P = 1). Each of the 4 patients only had 1 episode of cholangitis during the 6-month study period. The mean frequency of cholangitis (episodes per month) was consequently the same in the neomycin and Lactobacillus groups. Most important is that no adverse reaction was recorded during the 6-month study period in both groups.
Comparisons Between Lactobacillus and Control Groups
No significant differences in sex distribution; age at Kasai operation; age at the time of enrollment; AST, ALT, and total and direct bilirubin levels; and body weight z score at enrollment were observed between the Lactobacillus and control groups (Table 2). Six months later, however, the body weight z score in the Lactobacillus group was significantly higher than that in the control group (P = 0.005) and the change in body weight z score was significantly better in the Lactobacillus group than in the control group (P = 0.006). Eight patients in the control group (8/10) experienced cholangitis at least once during the 6-month study period, which was significantly more than that in the Lactobacillus group (2/10, P = 0.005). Among the 8 patients who experienced cholangitis in the control group, 3 patients had cholangitis twice and 1 patient had it 3 times. The mean frequency of cholangitis (episodes per month) was 0.22 ± 0.16 in the control group, indicating a significantly higher rate than that in the Lactobacillus group (0.03 ± 0.07) (P = 0.005).
Bacterial Stool Culture
In the neomycin group, the pretreatment ratio of E coli to the total aerobes was 51.7% ± 6.8%, which decreased significantly to 28.3% ± 8.7% after the 6-month treatment (Table 3). The baseline ratio of E coli to the total aerobes in the Lactobacillus group was 49.3% ± 11.2%, which also significantly decreased to 32.8% ± 5.6% after treatment completion. The ratio of E coli to the total aerobes was not significantly different between the 2 groups after 6 months of treatment (P = 0.17), implying that the strategy of substituting bad bacteria in the gut with probiotics may have worked and that the efficacy of decreasing the proportion of E coli is as good as the use of antibiotics (neomycin). In contrast, the initial ratio of Lactobacillus to the total anaerobes was 13.6% ± 5.8%, which significantly increased to 35.7% ± 6.5% after 6 months of treatment in the Lactobacillus group; however, it did not change significantly in the neomycin group. Furthermore, we also found that the significant difference in the ratio of Lactobacillus to the total anaerobes between the 2 groups was already evident since the first month of treatment.
Cholangitis-free survival in the Lactobacillus and neomycin groups is presented in Figure 1. Two patients in the Lactobacillus group had cholangitis on the 2nd and 112th days after starting treatment, respectively. Another 2 patients in the neomycin group had cholangitis on the 26th and 38th days after starting treatment, respectively. No difference in cholangitis-free survival was observed between the Lactobacillus and neomycin groups (P = 1.00). Eight patients in the control group experienced cholangitis at least once, occurring on the 24th, 32nd, 68th, 92nd, 152nd, 156th, 162nd, and 178th days, respectively. Figure 2 shows the significant difference in cholangitis-free survival between the Lactobacillus and control groups (P = 0.01).
BA is an obliterative cholangiopathy of unknown etiology. The Kasai operation targets the reconstruction of normal bile flow via the enteric conduit and is considered to be the first-line treatment for BA. Cholangitis, a common but serious complication after Kasai operation in patients with BA, may, however, result in bile flow obstruction with a subsequent deterioration of the hepatic function and ongoing cirrhosis (8). For patients who have had successful bile flow restoration with a timely portoenterostomy, the recurrence of ascending cholangitis is noted to be the single most significant variable pertaining to long-term prognosis (12). Nevertheless, the precise mechanism of cholangitis has not been well established yet. Reduced lymph drainage at the porta hepatis, portal venous infection, dehydration with reduced bile flow, ascending bacterial infection originating from the enteric conduit, insufficient bile flow volume, and partially obstructed biliary tree have all been reported as possible explanations in a previous study (3).
Given that cholangitis is a crucial event that influences the prognosis of patients with BA, prophylaxis of cholangitis in patients with BA who have undergone Kasai operation has become an important issue. Our previous studies have demonstrated that the use of prophylactic antibiotics can improve the 5-year jaundice-free survival of patients with BA with native livers (1) and that neomycin is an effective prophylactic agent (5). In this study, we have found that L casei rhamnosus is an alternative prophylactic agent.
Probiotics are live microorganisms and believed to be beneficial for individuals with inflammatory or infectious diseases of the gastrointestinal tract. These positive effects are generally attributed to the ability of probiotics to normalize host intestinal flora, followed by regulating intestinal permeability, improving gut immune barrier function, and equilibrating the balance between proinflammatory and anti-inflammatory cytokines (13). The proposed mechanisms of restoring the normal flora and preventing the overgrowth of pathogenic microorganisms include the competition for nutrients, stimulation of immunity, inhibition of mucosal adherence, and production of antimicrobial substances (14). For instance, the enterotoxigenic E coli is the most common cause for traveler's diarrhea, for which Lactobacillus was proven to have prophylactic efficacy in some studies (14). Thus, in our study, we chose L casei rhamnosus and neomycin as prophylactic agents for patients with BA who underwent Kasai operation and compared their outcomes between 2 groups of patients. After 6 months of treatment, the 2 groups did not significantly differ in body weight, change in body weight, number of cholangitis-free patients, and frequency of cholangitis. The lactobacillus showed similar good prophylactic efficacy as neomycin. In addition, we retrospectively recruited 10 patients with BA with similar statuses at enrollment as the control group. Based on the comparison, the outcomes (body weight, change in body weight, ratio of cholangitis-free patients, and frequency of cholangitis) were all significantly better in the Lactobacillus group than in the control group.
Based on the results of the stool bacterial culture for the patients, neomycin significantly decreased the ratio of E coli to the total aerobes after the 6-month treatment as expected. It is interesting that a significant decrease in the number of E coli bacteria also occurred in the Lactobacillus group after treatment completion. Considering the significant increase in the proportion of Lactobacillus in the Lactobacillus group, we speculate that the increase in the proportion of Lactobacillus will reduce the environmental growth resource of the E coli bacteria. Thus, we suppose that the same competitive relation is established between Lactobacillus and other ascending cholangitis-causing pathogens. Through the diminishment of the pathogenic bacterial concentration within the bilioenteric conduit, the L casei rhamnosus elaborated its prophylactic effects. Moreover, the significant difference in the proportion of Lactobacillus between the neomycin and Lactobacillus groups was already evident after 1 month of treatment.
Both in the neomycin and Lactobacillus groups, 80% of the patients were cholangitis-free during the 6-month treatment. We observed that 1 patient in the Lactobacillus group had cholangitis on the second day of treatment. We believe that L casei rhamnosus had not yet changed the gut environment during the short period. Thus, we may consider that actually only 1 patient had cholangitis in the Lactobacillus group. In contrast, only 20% of the patients in the control group were cholangitis-free during the 6-month observation period. According to the statistical analysis, the intervention with either neomycin or lactobacillus was beneficial in terms of prolonging the cholangitis-free survival time of the patients with BA who underwent Kasai operation.
Recurrent cholangitis worsen the prognosis of patients with BA. One study (15) even mentioned that patients with 1 or more episodes of early cholangitis (occurring within 6 months after Kasai operation) were at increased risk for developing episodes of late cholangitis (occurring beyond 6 months). Therefore, we strongly suggest that once the health status of patients with BA becomes stable after Kasai operation, the prophylaxis of cholangitis should be performed as soon as possible. Ascending cholangitis often occurred within first year after Kasai operation in patients with BA, but it also happened in those older patients. So in our experience, prophylactic antibiotics (neomycin) were usually maintained until patients reach the age of at least 3 years. Certainly the differences in patient tolerance and compliance for antibiotics prophylaxis did exist between each case in the past. In the present study, we enrolled 30 patients ages <3 years in 3 groups for competition and then demonstrated that L casei rhamnosus had good efficacy for prophylaxis of cholangitis and can therefore be considered for clinical use as an alternative choice for patients who are intolerant to antibiotics.
Furthermore, no adverse effects or complaints were documented in all of the patients and parents in our trial. Moreover, the mean body weight z score of the patients with BA in our study was lower than that of the general population at enrollment but increased after 6 months of treatment with either neomycin or lactobacillus. Conversely, the mean body weight z score slightly decreased after 6 months in the control group. We believe that by decreasing the episodes of cholangitis with our prophylactic protocol, patients with BA may return to stable growth development and catch up with body weight gain.
Our study, however, had some limitations and challenges. First, neomycin was prescribed in 4 days/wk, but L casei rhamnosus was prescribed everyday. Therefore, our study is open-labeled without blinding. Second, because prophylactic antibiotics were proved effective in the past study, prescribing placebo seemed not acceptable by parents; therefore, our control group is retrospective. Certainly we had tried our best to prove that all the patients in the retrospective control group had similar status to those in treatment groups. Third, all of our patients enrolled experienced a successful Kasai operation. All of them had stable health status and were jaundice free when enrolled in our study. It was difficult to persuade the parents of patients with BA who had unstable health status or were even waiting for liver transplantation to join the clinical trial. Thus, the prophylactic efficacy of probiotics in unstable patients with BA remains unclear. Finally, the sample size of this pilot study is small; therefore, the efficacy shown in this study is within the context of a small study. In the future, a larger and longer-term clinical trial with more patients with BA with different health statuses should be conducted to further investigate the prophylactic efficacy of probiotics.
The use of L casei rhamnosus significantly increased and decreased the number of Lactobacillus and E coli bacteria in the gut, respectively, indicating the efficacy of L casei rhamnosus for the prevention of cholangitis in the patients with BA who underwent Kasai operation. Moreover, no difference in prophylactic efficacy was observed between the lactobacillus and neomycin regimens.
The authors thank the National Center of Excellence for Clinical Trial and Research of the National Taiwan University Hospital (NTUH, project no NCTRC200709).
1. Lien TH, Chang MH, Wu JF, et al. Effects of the infant stool color card screening program on 5-year outcome of biliary atresia
in Taiwan. Hepatology
2. Oh M, Hobeldin M, Chen T, et al. The Kasai procedure in the treatment of biliary atresia
. J Pediatr Surg
3. Rothenberg SS, Schroter GPJ, Karrer FM, et al. Cholangitis
after the Kasai operation for biliary atresia
. J Pediatr Surg
4. Ohi R, Mochizuki I, Komatsu K, et al. Portal hypertension after successful hepatic portoenterostomy in biliary atresia
. J Pediatr Surg
5. Bu LN, Chen HL, Chang CJ, et al. Prophylactic oral antibiotics in prevention of recurrent cholangitis
after the Kasai portoenterostomy. J Pediatr Surg
6. Ohkohchi N, Chiba T, Ohi R, et al. Long-term follow-up study of patients with cholangitis
after successful Kasai operation in biliary atresia
: selection of recipients for liver transplantation. J Pediatr Gastroenterol Nutr
7. Mones RL, DeFelice AR, Homme DP. Use of neomycin as the prophylaxis
against recurrent cholangitis
after Kasai portoenterostomy. J Pediatr Surg
8. Wu ET, Chen HL, Ni YH, et al. Bacterial cholangitis
in patients with biliary atresia
: impact on short-term outcome. Pediatr Surg Int
9. Isolauri E, Kirjavainen PV, Salminen S. Probiotics: a role in the treatment of intestinal infection and inflammation? Gut
10. Reid G, Burton J. Use of lactobacillus
to prevent infection by pathogenic bacterial. Microbes Infect
11. Bu LN, Chang MH, Ni YH, et al. Lactobacillus casei rhamnosus
Lcr35 in children with chronic constipation. Pediatr Int
12. Houwen RHJ, Zwierstra RP, Severijnen RSVM, et al. Prognosis of extrahepatic biliary atresia
. Arch Dis Child
13. Girardin M, Seidman EG. Indications for the use of probiotics in gastrointestinal diseases. Dig Dis
14. Huebner ES, Surawicz CM. Probiotics in the prevention and treatment of gastrointestinal infections. Gastroenterol Clin N Am
15. Ernest van Heurn LW, Saing H, Tam PKH. Cholangitis
after hepatic portoenterostomy for biliary atresia
: a multivariate analysis of risk factors. J Pediatr