Biliary atresia: A transplant perspective

Biliary atresia is the most common single pediatric liver disease leading to liver transplantation (LT) during childhood.1 Between 1995 and 2002, 42% of children undergoing their first liver transplant while enrolled in the Studies of Pediatric Liver Transplantation (SPLIT) did so for a primary diagnosis of biliary atresia. There have been a number of recent reviews and advances related to the detection, diagnosis, and pathogenesis of biliary atresia.2-6 The purpose of this review is to focus on aspects of biliary atresia that are directly relevant to LT as a modality for its treatment. In particular, this review will concentrate on (1) specific indications for LT in the setting of biliary atresia, (2) management issues in children awaiting LT, (3) special surgical approaches in transplantation for biliary atresia, and (4) outcomes after LT. Biliary atresia is a unique disease with a relatively well-characterized clinical course. Careful analysis of the clinical course permits identification of indications for and timing of LT (Table 1). Transplantation decisions should take into consideration therapeutic approaches to the specific indications and complications that have developed. Thus, the following sections will examine both indications for transplantation and the management of those clinical problems. The HPE (Kasai procedure) is an excellent surgical treatment for biliary atresia. In most cases, it is not a cure for this condition but provides a means of permitting bile flow and preventing rapid progression of disease from chronic high-grade cholestasis. The natural history of unrepaired biliary atresia is reasonably well documented; in the absence of LT, survival beyond 24 or 36 months of age is very unusual.7-10 Similarly, the natural history of children with biliary atresia who do not demonstrate bile flow after HPE (a “failed” Kasai) is very similar to the history of children with unrepaired biliary atresia.11-13 As such, the failed Kasai is a strong indication for early consideration of LT. Very few pediatric liver diseases have such a predictable prognosis. In general, these children have a brief period of relative well-being followed by progressive failure to thrive, development of ascites and/or spontaneous bacterial peritonitis, and variceal hemorrhage. Survival beyond 24 to 36 months of age without LT occurs in only a small minority of children. Quality of life is poor in the setting of a failed HPE as a result of repeated hospitalizations, failure to thrive, developmental delay, and intractable pruritus. Thus, an aggressive, pre-emptive approach to LT in this setting is justified in centers with technical expertise in the transplantation of infants. One of the caveats with this important indication for pediatric LT is operationally defining a lack of bile flow. This operational definition must include both a marker of bile flow and a time period beyond which it is unlikely that bile flow will develop. In prior surgical eras, when externalization ostomies were utilized early after the Kasai, bile flow could be directly observed. This surgical approach has largely been abandoned, and thus indirect markers of bile flow must be utilized. Persistently acholic stools are a reasonable indication of poor bile flow, although caution must be exercised in the visual assessment of stool color. Comparisons to standardized color samples may improve this method, although it must be acknowledged that these systems were developed as screening methods for biliary atresia and not for clinical follow-up.14 The finding of “pigmented” stools does not necessarily indicate adequate bile flow. Alternative techniques, such as hepatobiliary scintigraphy, to assess bile flow after HPE are not well described as a measure of outcome. The most commonly utilized marker of bile flow after HPE is measurement of serum bilirubin levels. Total bilirubin levels, as opposed to direct or conjugated bilirubin levels, are most commonly utilized because of the variability of methods for the measurement of the latter. Coexisting Gilbert's syndrome could complicate the use of total bilirubin levels in this setting, and thus clinicians should confirm that direct or conjugated bilirubin levels are significantly elevated in an infant who is suspected to have a poorly draining HPE. A total bilirubin in excess of 6 mg/dL (100 μM) at a time greater than 3 months after HPE has been associated with poor short-term outcome in biliary atresia and may be a reasonable marker of a failed Kasai.15 To date, there are no clearly described interventions to improve bile flow after the Kasai procedure. Controversy exists as to whether perioperative corticosteroid administration improves surgical outcome in biliary atresia. Reoperation is unlikely to be helpful in most cases of biliary atresia. In the United States, utilization of the failed Kasai as the sole indication for LT typically necessitates a live donor approach. These infants have a reduced priority for transplantation because they typically have low pediatric end-stage liver disease (PELD) scores. In the setting of adequate nutrition, they almost always have normal or near-normal prothrombin times and albumin levels. Thus, they accumulate PELD points only on the basis of their total bilirubin levels, which are routinely less than 12 mg/dL. On occasion in some regions in the United States, a split liver may become available for an infant with a failed Kasai. Otherwise, pre-emptive transplantation in this setting before the development of classical complications of liver disease can occur only if a live donor is available. Most children with a failed HPE will undergo transplantation after the development of complications including most commonly failure to thrive, ascites, and/or variceal hemorrhage, which are subsequently described. DDTV, deceased donor technical variant; HCC, hepatocellular carcinoma; HPE, hepatoportoenterostomy; HPS, hepatopulmonary syndrome; LT, liver transplantation; PELD, pediatric end-stage liver disease; SPLIT, Studies of Pediatric Liver Transplantation; UNOS, United Network for Organ Sharing. Failure to thrive, typically defined as poor somatic growth, is very common in biliary atresia and is one of the most prevalent indications present in children who undergo LT. Growth failure was reported in 40% of patients who underwent LT in the SPLIT registry, and the mean height (−1.3) and weight (−1.4) z scores were below average for this large cohort of children with biliary atresia.16 A comprehensive retrospective multicenter analysis of children with biliary atresia confirmed the prevalence of this problem and correlated failure to thrive with poor outcome.17 The importance of growth failure is acknowledged in the current allocation system in the United States as children receive higher wait-list priority when this complication is present.18 The pathogenesis of growth failure and malnutrition in biliary atresia is multifactorial, with major contributions from increased energy expenditure and malabsorption related to cholestasis.19, 20 Resting energy expenditure may be increased by as much as 30% in children with biliary atresia.21 Coupled with potential fat malabsorption, which correlates with the degree of cholestasis,22 biliary atresia puts children at tremendous risk for caloric deficits leading to failure to thrive. Clinical assessment of nutritional status in children with biliary atresia is less straightforward than it seems. Often the first visual impression of these patients is one of significant malnutrition, which is then not supported by routine height and weight measurements. Assessment of body weight is often quite misleading because of the effects of organomegaly and ascites. Diminishment of height and head circumference growth velocity is a late manifestation of poor nutritional status. Thus, a more accurate assessment of nutritional status in children with biliary atresia is dependent on more sensitive markers such as anthropometrics, including midarm muscle area and skinfold thickness.23, 24 Therefore, transplant assessment of a child with biliary atresia should include a comprehensive analysis of growth, permitting a more accurate appreciation of the child's clinical status. It is unusual for failure to thrive to be the sole indication for transplantation in biliary atresia, as it is typically accompanied by manifestations of poor bile flow. Accurate identification of failure to thrive may have implications for prioritization for transplant and permit important medical interventions. The need for nutritional supplementation (such as nasogastric tube feeding or intravenous administration) for failure to thrive in children with biliary atresia is a risk factor for both wait-list and posttransplant mortality.16 Similar findings have been found for failure to thrive in the more general population of children awaiting LT.18 At present, it is not clear whether one can reverse this mortality risk with aggressive nutritional support. Pretransplant growth retardation increases posttransplant length of stay and hospital costs.25 Long-term developmental outcome is likely to be adversely affected by malnutrition, especially if there is poor head growth. Therefore, nutritional support is a key part of the care of children with biliary atresia who await LT. A comprehensive description of this aspect of the care of these children is beyond the scope of this article, and interested readers are directed to recent reviews.19, 20, 26 Monitoring, including anthropometrics, fat-soluble vitamin levels, and prothrombin time, should be more frequent in children with evidence of more significant cholestasis. Fat-soluble vitamin supplementation should take advantage of the ability of tocopherol polyethylene glycol succinate to be absorbed independently of bile salts.27, 28 Medium chain triglycerides are a superior form of enteral fat administration in cholestasis as they are also absorbed independently of bile salts. Nasogastric tube feeding in biliary atresia is often utilized and may in fact be underutilized.17 The exact indications for commencing nasogastric tube feeding are not completely clear. Nasogastric tube feedings can enhance the nutritional status of children with biliary atresia.29, 30 It is obvious that nutritional supplementation is indicated and perhaps late in the patient with inadequate intake and significant failure to thrive (for example, height or weight z score < −2). Interestingly, at initial presentation, the average child with biliary atresia has subnormal growth parameters.17 It is unknown whether nasogastric tube feedings at disease presentation can have an impact on overall long-term outcome. Between these two ends of the clinical spectrum of intervention, there are many questions as to the most advantageous approach to the use of nasogastric tube feeding. Approximately 40% to 60% of infants with biliary atresia will achieve good bile drainage after HPE. The subsequent course in patients with good bile drainage is highly variable and often involves the development of biliary cirrhosis and complications of portal hypertension (discussed later). Some long-term survivors will redevelop significant cholestasis later in life. This cholestasis is likely the result of ongoing intrahepatic biliary injury and is potentially related to recurrent episodes of cholangitis. In our experience, late-onset cholestasis, if not an acute manifestation of cholangitis, is a harbinger of a worsening in near-term outcome. Nearly all of the patients in a recent review of adult-to-adult living donor transplantation for biliary atresia were cholestatic.31 Thus, in the older child or young adult, reoccurrence of jaundice/cholestasis should trigger reassessment of the transplant candidacy of that patient. HPE is an excellent but imperfect surgical treatment for biliary atresia. There are numerous reports of long-term survival with native liver after HPE.32-35 Thus, in most circumstances, the concept of LT as the primary surgical approach to biliary atresia has been abandoned.36 Extensive evidence has been presented that indicates that age at HPE influences the response to surgery; in general, responses are better when surgery occurs at a younger age (Table 2). Despite this well-described phenomenon, there are some children who have a relatively successful postoperative outcome after HPE even when this procedure occurs after 100 or even 120 days of life. As such, a very reasonable question regarding HPE is when is it too late.37 Obviously, there is no simple answer to this question, and by corollary, there is no simple age at which LT is clearly preferred over HPE as the primary surgical therapy for biliary atresia. We have favored LT as the primary therapy for children who are greater than 120 days of age. Alternatively, if an infant has significant ascites or has had variceal hemorrhage as a complication of portal hypertension resulting from biliary atresia, it may be unwise to attempt HPE, but instead one might consider transplantation as the primary approach. Cholangitis is a common complication after HPE for biliary atresia, occurring in at least 30% to 60% of children, typically in the first 1 to 2 years after surgery.15, 38-40 Infection is presumed to be ascending and thus typically does not occur in children who have poor or no bile flow after HPE. Infection rates may approach 100% in children with evidence of good bile flow.41, 42 Cholangitis is a clinical syndrome, typically consisting of fever, leukocytosis, and evidence of worsening cholestasis, although there are no universally accepted criteria for the diagnosis. A positive blood culture with an appropriate organism can often clarify the diagnosis, although positive cultures are seen in fewer than 50% of cases of cholangitis. As fevers are not uncommon in children less than 2 years of age, it can be difficult to be certain of this diagnosis and/or its true prevalence. Typical means of trying to prevent cholangitis include prophylactic antibiotics (for example, trimethoprim sulfamethoxazole) and/or special surgical techniques (intussuscepted antirefluxing valve), although the efficacy of these interventions is unproven.39, 41, 43-45 Cholangitis can be recurrent and/or refractory to medical management, which typically includes a prolonged course of intravenous antibiotics. Two to four distinct episodes of cholangitis in the first 2 years after HPE are not unusual. In cases in which the episodes are more frequent and/or refractory to appropriate and prolonged courses of antibiotics, additional surgical intervention may be indicated. In selected cases, revision of the HPE may be useful in relieving partial and/or complete obstruction that may predispose to the refractory disease.46, 47 Hepatobiliary scintigraphy may identify circumstances under which revision may be successful.48 LT may be indicated as a means of treating highly recurrent and/or refractory disease. The clinician should be extremely careful about being relatively certain about the diagnosis of cholangitis, especially if this is the primary reason to consider LT. In many cases, recurrent or refractory disease is followed by the development of high-grade cholestasis, which evolves into another indication to consider LT. In the era of multiresistant bacterial organisms, the pattern of antibiotic resistance of positive blood cultures may also influence transplant decision-making. Variceal hemorrhage is a prominent and potentially fatal complication that develops in a significant number of children with biliary atresia.49-53 Significant portal hypertension develops very early and can be present at the time of HPE.54 Hepatic synthetic function can be relatively intact at the time of variceal hemorrhage, so clinical decision-making and management approaches must take into consideration the near-term prognosis for an individual patient. Approaches to the management of variceal hemorrhage in children are largely anecdotal and derived from studies in adults.55 Primary prophylaxis of variceal hemorrhage in children is not the typical standard of care but is employed by some experienced clinicians.56 The approach to secondary prophylaxis is largely dependent on the presence or absence of bile flow, which has a significant impact on near-term mortality.57 Children with evidence of poor bile flow, as manifested by a total bilirubin greater than or equal to 10 mg/dL, have a high risk of near-term mortality, and thus LT is the approach of choice for these patients. In contrast, children with evidence of good bile flow, that is, total bilirubin less than 4 mg/dL, have a much lower risk of near-term mortality. As such, endoscopic band ligation may be the initial treatment of choice in these patients.58 Beta blockade has been utilized in children, although the physiologic parameters guiding its use are not well described in pediatrics.59 Children with refractory hemorrhage or varices that are not amenable to endoscopic approaches may be candidates for portosystemic shunting, especially distal splenorenal shunts.60-62 Portosystemic shunting in small children requires reasonable expertise in pediatric hepatobiliary and vascular surgery. Spontaneous bacterial peritonitis occurs in children with biliary atresia, although there is a surprising lack of comprehensive literature on this important complication in children with chronic liver disease.63-65 A high index of suspicion needs to be maintained for this entity in order to permit timely diagnosis and successful intervention. Prophylactic antibiotics are sometimes used in adults with chronic low albumin ascites, although there are no clear guidelines for this approach in children.66 In children with advanced liver disease and pharmacologically refractory ascites, LT is indicated, and prophylactic antibiotics with trimethoprim sulfamethoxyzole while they await LT are reasonable. In adults, development of spontaneous bacterial peritonitis is associated with increased near-term mortality.67-69 By corollary, the occurrence of an episode of spontaneous bacterial peritonitis in a child with biliary atresia is a reasonable indication to give strong consideration to LT and should also be managed with antibiotic prophylaxis to prevent recurrent disease.66 HPS has been described in a large number of children with biliary atresia. The prevalence of this complication varies between series in large part because of differences in methods used in screening and/or diagnosis. The most sensitive test for HPS involves contrast echocardiography using intravenous injection of agitated saline. In one study, more that 50% of children with biliary atresia had saline echocardiography suggestive of HPS.70 Right-to-left shunting can be quantified by macroaggregated albumin scanning, and in the aforementioned study, 20% of the children with positive echocardiograms had shunt fractions > 20%.70 An alternative means of screening for HPS is to measure transcutaneous upright oxygen saturations in room air.71 In an unpublished experience from Mount Sinai, fewer than 5% of children with cirrhosis and biliary atresia had upright oxygen saturations below 97%. Therefore, the prevalence of HPS is highly dependent on the method of screening. HPS is typically reversible after LT, so transplantation is clearly indicated in individuals with symptomatic HPS.70, 72 Optimally, transplantation should occur before hypoxia becomes too severe (that is, pO2 < 50 mm Hg) to avoid undue morbidity and mortality. The approach to asymptomatic HPS is less clear. At present, there are no clear guidelines on the most appropriate level of desaturation or shunting for LT in children. The natural history of HPS in biliary atresia is largely unknown. Thus, it is unclear how to proceed with a child with mild (for example, saturation between 94% and 97% and/or shunt fraction between 10% and 20%) and asymptomatic HPS. Close follow-up is clearly indicated, and consideration for transplantation should be given to children with saturations below 94% and/or a shunt fraction > There are no than transplantation for HPS, although there are some anecdotal the use of in hypertension is a more severe and less common complication that has been in children with biliary Most of the reports of hypertension are anecdotal and are often associated with a fatal outcome. In large clinical series from pediatric transplant only a few cases are reported from In of the severe associated with significant it can be that the prevalence of this complication in the pediatric age is and a are clinical that should the clinician to the of This entity is associated with hypoxia but not which is one to it from HPS. and do not for this there are findings on echocardiography including assessment of the is to confirm this diagnosis. LT can reverse hypertension if the procedure can be before 50 mm was used in one patient with severe hypertension as a means of the child to a successful liver it is a complication in biliary atresia, the poor prognosis and importance of early of hypertension support the for screening with echocardiography in all children with of significant portal The most appropriate timing for the of screening and the for follow-up studies are unknown. echocardiography be reasonable as part of the routine assessment of child being for of hypertension are and of the should be by In patients with lower than 50 mm prioritization for LT should be syndrome is an uncommon but severe complication of end-stage liver disease in children. It is to be a manifestation of a in and associated There is if specific literature to pediatric with this so must be from The diagnosis is when there is an acute worsening of function in the setting of portal and are present in the setting of no specific and there is no evidence of medical including and albumin have not been in children but are as syndrome typically after LT, so children with biliary atresia and this complication should be for is a common and very clinical in children with chronic cholestasis. In its severe it can have a significant impact on of life and/or The exact of the is not completely although its correlates with serum of bile It is a major in the clinical course of children with progressive intrahepatic cholestasis and In biliary atresia, it can be a major but is not typically as severe as that in these of intrahepatic cholestasis. In circumstances, intractable is the problem in a child with biliary atresia, and it becomes the leading indication for LT. In that certain are one should of chronic in children, such as and one should be certain that the initial diagnosis of biliary atresia was and/or was the only of cholestasis. syndrome and biliary atresia can be very similar in their early There are findings of 1 in children with well-characterized biliary The finding of syndrome has implications for the of as potential living An affected in this disease with variable could have bile that could one should be certain that appropriate have been in the treatment of the pruritus. are utilized for of pruritus. are in cholestasis, with the of in because of their or therapy can be in some children, by the bile has also been in some children with severe This is potentially so is for such as and are not well for use in children and should be employed with develops in a significant number of children with biliary It can in the perioperative period and can also be a manifestation of cirrhosis and portal In the setting of poor bile flow or major complications of end-stage liver it to the indications to proceed with LT. It is most by the of an in body is quite as a means of ascites. with normal serum levels is typically managed with and administration to a total In our experience with younger children, is not the major and ascites relatively to a of a and a more advanced the ascites is with and total body This is a finding and a poor near-term prognosis. On intractable ascites is the primary indication to consider LT in children. In this setting, it is important to that there is ascites. Children with organomegaly can give the of significant ascites when in fact there is in the It is important to that the patient is the and that there is not a history of can predispose to more refractory ascites, so is an appropriate part of the of these have children with and intractable ascites that developed on the basis of a bile These children have relatively normal bilirubin levels as bile flow is although it into the of ascites bilirubin levels and/or hepatobiliary scintigraphy can a bile which may be amenable to surgical revision of the In this LT can be for of Between and there were children who underwent LT for in the United States, although it is not clear how many of those children had an diagnosis of biliary is an uncommon but reported complication in children with biliary It has often been an finding and has been reported in an In almost all of the reported cases of biliary atresia, serum levels have been The of treatment of are relatively and thus the is to identify this early to permit complete surgical may be seen in children with biliary in the setting of the are related to the development of with contrast administration can be especially useful in that may have a high of being A child with suspected should undergo a if should be for for in adults with cirrhosis is an accepted standard of Typical approaches in adults include and measurement of serum levels 6 A similar standard of care does not for children, although almost all of the of reported cases of in biliary atresia have that in biliary atresia is The prevalence of in children with to adults the potential of in the pediatric age Thus, it is difficult to for in the setting of biliary atresia. On the basis of anecdotal it might be as a reasonable approach in those children with biliary atresia who with their native liver into In screening at the time of for LT might also be as a means of wait-list