Biliary atresia continues to be the most common cause of cirrhosis and end-stage liver disease in the pediatric population in the United States and many other parts of the world. It is characterized by progressive inflammation and fibrosis of both the extrahepatic and intrahepatic bile ducts, leading to obstruction.1, 2 Early surgical intervention improves survival but liver transplantation is still required in most patients, and this condition remains the most common pediatric indication for liver transplantation.

What do we know about outcomes of patients with biliary atresia? When biliary atresia is detected early, surgical intervention begins with hepatoportoenterostomy (HPE; the Kasai procedure). Developed in 1959, this procedure has made a major improvement on the outcome of patients with this condition.3 The procedure achieves successful bile drainage in up to 80% of patients. If HPE is performed after 3 months of age, there is a significant decline in the success rate.4 However, even after a successful HPE, 70%-80% of patients will eventually go on to require a liver transplantation.4, 5 Significant research efforts have been made to identify reliable predictors for progressive disease and poor outcome, because identification of such markers may not only enhance our ability to target and determine the proper timing of currently available therapies but also may aid the development of novel treatment strategies to halt disease progression.

A breakthrough in predicting outcomes for pediatric patients with severe liver disease came in 2002. Using data from the Studies of Pediatric Liver Transplantation (SPLIT) registry, the pediatric end-stage liver disease (PELD) score was formulated to assess the risk of poor outcomes in patients awaiting liver transplantation. Components of the PELD score include growth failure (weight and height less than 2 standard deviations below the mean), chronological age less than one year, serum bilirubin, serum albumin, and international normalized ratio.6

Risk factors and outcomes specific for children with biliary atresia have also been examined. In their 2006 article, the Biliary Atresia Research Consortium (BARC) studied the outcome of 104 patients who underwent HPE at 9 clinical centers across the United States.7 Children were followed for 2 years or until loss of their native liver as a result of transplantation or death. At age 24 months, 52% of the children had a “good” outcome defined by the authors as survival without liver transplantation and with total serum bilirubin level < 6.0 mg/dL. Patients with serum bilirubin levels < 2 mg/dL at 3 months after HPE demonstrated markedly improved survival in comparison to those with total bilirubin > 6 mg/dL. Forty-six percent of the patients had a “poor outcome”, defined as death or liver transplantation before 24 months of age. Presence of biliary atresia splenic malformation (BASM), ascites, and cholangitis were predictors of poor outcome. The impact of nutrition and growth was decided to be analyzed separately.

In this issue of HEPATOLOGY, DeRusso and colleagues from the BARC address the effect of growth failure on the outcome of these infants.8 As noted by the authors, several previous studies have attempted to address this important issue. Growth failure is an established component of the PELD score, and a significant predictor of graft failure and death after liver transplantation.9 However, most of these studies have been limited by a single cross-sectional assessment of growth parameters. The authors of the current study followed 100 patients from the time of initial evaluation at the BARC center to age 2 years. Growth measurements were followed using age-adjusted z-scores for weight and length and obtained serially at different times allowing for assessment of the impact of growth velocity on patient outcomes. Although the average weight and length z-scores were similar in the good and poor outcome group (as defined previously) at the time of HPE, differences in growth velocity were apparent by 6 months following HPE. The mean weight z-score was better in the “good outcome” group compared to the “poor outcome” group. In fact, the weight z-score for the “good outcome” group reached pre-Kasai levels by 18 months of age. The difference in length z-score trended toward improvement for the “good outcome” group by 6 months post-Kasai and reached statistical significance at 18 months of age. The authors also noted that growth velocity may be an important predictor of outcome in those patients with intermediate levels of total bilirubin (between 2 and 6 mg/dL). These findings appear to be independent of previously recognized risk factors for “poor outcome” such as BASM and ascites.

Most patients (82%) in the “good outcome” group were receiving nutritional supplementation at the time of HPE versus only 36% of the patients in the “poor outcome” group. Unfortunately, the type of nutritional supplementation administered was not discussed in detail and data regarding the amount and duration of these supplementations were not available. Determining whether administering formula via nasogastric (NG)/gastrostomy tube is sufficient or administration of total parenteral nutrition is needed would be a clear next step in optimizing growth for these patients. Also, caloric requirement needed to produce optimal growth is unclear. Few previous small studies have attempted to address the role of supplemental feedings in this context.10, 11 Holt et al.10 examined the use of NG feeds in patients with biliary atresia and failure to thrive. Patients were given approximately 150% of the World Health Organization recommended daily intake per day. The patients were allowed to take calories perorally during the day with NG supplements administered at night. NG supplements consisted of a high-energy, high-protein medium chain triglyceride (MCT)-based formula. Six weeks after initiation of NG feeds, the patients had improvement in triceps skinfold thickness and height scores. Four months after initiation of NG feeds, weight scores were improved. Chin et al.11 showed similar positive results by NG supplementation with a high-energy, branched-chain amino acid (BCAA)-enriched semi-elemental formula but not with a standard formulation.

There have been several methods used to determine the presence of growth failure in patients with biliary atresia. Measurements of height and weight according to growth charts from the Centers for Disease Control and Prevention are the standard methods used in pediatrics. Mean weight z-scores reported in the literature for pediatric patients with chronic liver disease have ranged between −1.4 ± 1.8 to −2.02 ± 1.36 and mean height z-scores reported have ranged between −1.3 ± 1.8 to −2.53 ± 1.54.12 However, studies in patients with chronic liver disease have shown that height and weight measurements may underestimate the degree of growth failure. Measurements such as body mass index, triceps skinfold thickness, mid-arm circumference, subscapular skinfold thickness, and occipital-frontal circumference have all been investigated as tools to assess nutritional status. Previous studies have shown that triceps skinfold thickness may be a more accurate measurement of growth failure in children with chronic liver disease secondary to increased weight measurement due to ascites, edema, and organomegaly.13

Although DeRusso et al. attempted to adjust for these confounding factors, future studies focused on nutritional aspects of biliary atresia may benefit from using multiple modalities to assess growth and nutritional status in order to better identify those patients that should be targeted for supplemental therapy. Also determining the proper timing of supplementation should be addressed. Should it start at the diagnosis of biliary atresia regardless of nutritional status? Or should we target patients who are failing to thrive? How long after HPE should nutritional therapy be continued? What is the proper type and percentage of calories that should be administered? Answering these questions may allow for continuous improvements in morbidity and mortality for these patients.


  1. Top of page
  2. References