Parenteral nutrition supplementation in biliary atresia patients listed for liver transplantation

Authors

  • Jillian S. Sullivan,

    1. Department of Pediatrics, Children's Hospital Colorado and University of Colorado School of Medicine, Aurora, CO
    2. Digestive Health Institute, Children's Hospital Colorado and University of Colorado School of Medicine, Aurora, CO
    Search for more papers by this author
  • Shikha S. Sundaram,

    Corresponding author
    1. Department of Pediatrics, Children's Hospital Colorado and University of Colorado School of Medicine, Aurora, CO
    2. Digestive Health Institute, Children's Hospital Colorado and University of Colorado School of Medicine, Aurora, CO
    • Digestive Health Institute, Children's Hospital Colorado, 13123 East 16th Avenue, B290, Aurora, CO 80045
    Search for more papers by this author
    • This study was supported in part by a Ruth L. Kirschstein National Research Service Award (5T32DK067009), a clinical fellowship award from the Cystic Fibrosis Foundation (GEIDER08B0), and a Colorado Clinical and Translational Sciences Institute grant from the National Institutes of Health/National Center for Research Resources (UL1 RR025780). The contents of this article are solely the responsibility of the authors and do not necessarily represent official views of the National Institutes of Health.

  • Zhaoxing Pan,

    1. Department of Pediatrics, Children's Hospital Colorado and University of Colorado School of Medicine, Aurora, CO
    2. Department of Biostatistics and Informatics, Colorado School of Public Health, Aurora, CO
    3. Clinical and Translational Sciences Institute, University of Colorado Denver, Aurora, CO
    Search for more papers by this author
  • Ronald J. Sokol

    1. Department of Pediatrics, Children's Hospital Colorado and University of Colorado School of Medicine, Aurora, CO
    2. Digestive Health Institute, Children's Hospital Colorado and University of Colorado School of Medicine, Aurora, CO
    3. Clinical and Translational Sciences Institute, University of Colorado Denver, Aurora, CO
    Search for more papers by this author

  • Telephone: 720-777-6669; FAX: 720-777-7277

Abstract

The objective of this study was to determine the impact of parenteral nutrition (PN) on the outcomes of biliary atresia (BA) patients listed for liver transplantation (LT). We retrospectively reviewed the charts of all BA patients at our institution who underwent hepatoportoenterostomy and were listed for LT before the age of 36 months between 1990 and 2010. The initiation of PN was based on clinical indications. Twenty-five PN subjects and 22 non-PN subjects (74% female) were studied. The median PN initiation age was 7.7 months, the mean duration was 86 days, and the mean amount of energy supplied by PN was 77 kcal/kg/day. Before PN, the triceps skinfold thickness (TSF) and the mid-arm circumference (MAC) z scores were decreasing. After PN, TSF (P < 0.001) and MAC (P < 0.001) improved significantly. The PN group had lower MAC and TSF scores than the non-PN group at the time of LT listing. Between listing and LT, MAC and TSF improved in the PN group and worsened in the non-PN groups; as a result, the 2 groups had the same z scores at LT. The PN group had a higher incidence of gastrointestinal bleeding and ascites before LT, but there were no differences in the rates of pre-LT bacteremia, days in the intensive care unit after LT, or patient or graft survival. In conclusion, PN improves the nutritional status of malnourished BA patients awaiting LT, and this is associated with post-LT outcomes comparable to those of patients not requiring PN. Liver Transpl 18:121–129, 2012. © 2011 AASLD.

Patients with biliary atresia (BA), which is a progressive fibro-obliterative disease of the extrahepatic and intrahepatic bile ducts, present with obstructive jaundice within the first 3 months of life. BA accounts for 30% to 40% of cases of neonatal cholestasis.1 The initial management includes surgical intervention with a Kasai hepatoportoenterostomy (HPE), ideally before 45 to 60 days of age. Even when HPE is performed in a timely fashion, nearly 70% to 80% of patients with BA will eventually require liver transplantation (LT), accounting for nearly half of all children undergoing LT.1

Malnutrition, a significant problem for infants with BA, is caused by early satiety from organomegaly and ascites, malabsorption of dietary lipids due to obstructed bile flow, and increased energy expenditure.2-4 Malnutrition places children at risk for poor clinical outcomes both before and after LT.5-7 Although the current literature suggests that a more normal nutritional status, as indicated by higher weight and/or length z scores, portends better clinical outcomes for patients with BA, the impact of parenteral nutrition (PN) supplementation has not been adequately addressed for patients with this disease. In a single case series, PN was initiated in 3 infants with BA after the failure of enteral therapy.4 Weight and length z scores have been used to describe the nutritional status of patients with BA; however, mid-arm circumference (MAC), a measurement of muscle mass, and triceps skinfold thickness (TSF), a measure of adipose tissue stores, are better measures of the nutritional status of children with chronic liver disease.2, 8-13 We performed a retrospective cohort study to examine the use of PN supplementation and its impact on the outcomes of BA patients treated at the Pediatric Liver Center of Children's Hospital Colorado.

PATIENTS AND METHODS

A comprehensive review of medical records was performed at our institution to identify all children with BA who underwent HPE and were listed for LT before the age of 36 months between January 1, 1990 and July 15, 2010. This group of BA patients was selected because they were at the highest risk for malnutrition and death while they were on the LT waiting list, were believed to be more difficult to nutritionally rehabilitate. The diagnostic criteria for BA were an age < 6 months with cholestatic jaundice and cholangiography, surgical exploration, and pathology findings demonstrating partial or complete obstruction of the biliary tree. This study was approved by the Colorado Multiple Institutional Review Board. The clinical endpoint for this study was defined as LT, death, or removal from the transplant waiting list.

Medical information was collected from medical records. Demographic data, including race, sex, date of birth, and age at HPE, were collected for each patient. Dietary intake data, nutritional indices, physical examination findings, and laboratory data were obtained from the records of each patient within 1 week of the time of HPE, transplant listing, clinical endpoint (LT, death, or removal from the transplant waiting list due to an improved clinical status), and initiation of PN. For each patient receiving PN, data including details about the composition of PN were collected at each monthly visit until the clinical endpoint.

Anthropometric data were abstracted from the medical records. The weight, length, and weight/length z scores were calculated according to the Centers for Disease Control and Prevention growth charts. TSF and MAC were measured as part of the standard of care at our institution. MAC and TSF were determined using standard techniques with Lange calipers by 5 clinical pediatric dieticians fully trained to measure anthropometrics.14 MAC was obtained with a single measurement and was documented in the medical record. TSF was obtained 3 times during each clinic visit, and the average was recorded in the medical record by the clinical pediatric dietician. TSF and MAC z scores were calculated with results from the US Ten-State Nutritional Survey.15 The use of nutritional supplementation (enteral or parenteral), the type and amount of the oral formula, and the intravenous PN formulation were obtained from the medical record, and the total amount of energy (kcal/kg/day) was calculated on the basis of the patient's measured weight and intake. We included the dietary intake in the analysis when it was completely documented in the medical record. Information on the enteral intake and the PN composition was collected. For each patient receiving PN, the mean glucose infusion rate, the mean daily dose of intravenous lipids, and the mean daily dose of amino acids were calculated. The median glucose infusion rate, dose of intravenous lipids, and dose of amino acids during the study period were calculated on the basis of aggregate averages for each patient receiving PN. A soy oil–based intravenous lipid emulsion (Intralipid, Fresenius Kabi, IL) was used in all patients; no patients in this cohort received fish oil–based intravenous lipid emulsions. The amino acid solution was TrophAmine (B. Braun, Bethlehem, PA) or Aminosyn (Hospira, Lake Forest, IL). Standard chemistries and complete blood counts were monitored monthly while patients were on PN. Adjustments in the PN formulation for cholestatic infants included removal of supplemental manganese and halving the dose of copper supplementation.

The decision to initiate nutritional supplementation (both enteral and parenteral) was based on the clinical judgment of the individual care providers. Enteral supplementation with nasogastric (NG) tube feeding was initiated when the TSF and MAC z scores were consistently falling despite maximal attempts to increase both the caloric density and the volume of the oral feedings. In some patients, z scores less than −2 standard deviations below normal were the indication for initiating NG tube feeding. In other cases, consistently falling z scores were the indication for initiating NG feeding to prevent severe malnutrition. PN was started when maximal enteral feeding (oral and NG), based on both the caloric density and volume, failed to achieve restoration of TSF and MAC z scores.

Laboratory data that were obtained from the medical records included the serum albumin, total and direct bilirubin, gamma-glutamyl transferase (GGT), aspartate aminotransferase (AST), alanine aminotransferase (ALT), and alkaline phosphatase levels; the white blood count; the hemoglobin level; the hematocrit; the platelet count; the prothrombin time; and the international normalized ratio. The Pediatric End-Stage Liver Disease (PELD) score was calculated at each time point on the basis of laboratory results and growth parameters.16

Medical complications of liver disease and portal hypertension were recorded; these included ascites, gastrointestinal bleeding (defined as hematemesis or melena requiring a blood transfusion, medical therapy, or endoscopic intervention), encephalopathy, and bacteremia (defined as 1 or more positive blood cultures). Portal hypertension was defined clinically as the presence of splenomegaly, thrombocytopenia (platelet count < 150 × 103/microliter), gastrointestinal bleeding, or ascites. For each patient who underwent LT, PELD score assigned at the time of transplantation, the type of transplant (deceased donor whole, split, or reduced liver), the number of postoperative days in the intensive care unit (ICU), the immunosuppressive drugs used in the first 6 months after transplantation, and patient and graft survival data were collected. All transplants came from deceased donors because our institution does not routinely perform living related LT. The immunosuppression medications included cyclosporine (US Adapted Name) from Novartis (International Nonproprietary Name: ciclosporin), tacrolimus (US Adapted Name/International Nonproprietary Name) from Astellas, corticosteroids, and mycophenolate mofetil (US Adapted Name) from Genentech (International Nonproprietary Name: mycophenolic acid).

Statistical Analysis

Fisher's exact tests and 2-sample t tests were used to examine the differences between the PN and non-PN groups with respect to categorical and continuous variables, respectively. Linear mixed effects models with an unstructured covariance structure were used to model serial continuous outcomes such as MAC and TSF z scores. Contrasts under this model were used to assess the changes in outcomes over time within a group and the differences between the 2 groups. Kaplan-Meier plots and log-rank tests were used to compare the survival of the 2 groups after LT. Values are expressed as means and standard errors. A P value < 0.05 was considered significant. Pearson correlations were used to determine correlations between laboratory data (albumin level, total bilirubin level, and prothrombin time) and weight, length, TSF, or MAC z score at the time of transplant listing.

Abbreviations:

ALT, alanine aminotransferase; AST, aspartate aminotransferase; BA, biliary atresia; GGT, gamma-glutamyl transferase; HPE, hepatoportoenterostomy; ICU, intensive care unit; LT, liver transplantation; MAC, mid-arm circumference; NG, nasogastric; NS, not significant; PELD, Pediatric End-Stage Liver Disease; PN, parenteral nutrition; TSF, triceps skinfold thickness.

RESULTS

One hundred fifty-nine patients underwent LT at our institution between January 1, 1990 and July 15, 2010, and 62 of these patients were listed for BA. Fifteen were excluded from this analysis: 9 did not undergo HPE, and 6 were older than 36 months at the time of listing for LT. Therefore, 47 patients with BA who underwent HPE and were listed for LT before the age of 36 months were included in this study. The patients were divided into 2 groups according to their exposure to PN. The PN group (n = 25) and the non-PN group (n = 22) were similar with respect to their demographics, age at HPE, age at listing for LT, and age at the clinical endpoint (LT, death, or removal from the transplant waiting list), as shown in Table 1.

Table 1. Demographics and Clinical Outcomes of BA Patients in the PN and Non-PN Groups
 All Patients
PN Group (n = 25)Non-PN Group (n = 22)P Value
Sex: female [n (%)]19 (76)16 (73)>0.99
Ethnicity [n (%)]   
Caucasian15 (60)12 (55)0.65
Hispanic8 (32)6 (27)
Other2 (8)4 (18)
Age at HPE (months)*2.1 ± 0.12.3 ± 0.20.48
Age at listing (months)*8.9 ± 1.09.4 ± 1.70.82
Age at clinical endpoint (months)*12.5 ± 1.213.6 ± 1.90.63
 Transplant Patients
PN Group (n = 22)Non-PN Group (n = 20)P Value
  • *

    The data are presented as means and standard errors.

  • Initial post-transplant calcineurin inhibitor.

  • During the first 6 months after transplantation.

Time from listing to transplantation (months)*4 ± 1.14 ± 0.8>0.99
Calculated PELD score at transplantation*18 ± 2.614.2 ± 2.10.24
Age at transplantation (months)*12.9 ± 1.412.7 ± 2.00.91
Time in ICU after transplantation (days)*9.1 ± 1.98.9 ± 2.70.93
Donor organ type [n (%)]   
 Cadaveric whole7 (32)8 (40)0.60
 Cadaveric split5 (23)2 (10) 
 Cadaveric reduced10 (45)10 (50) 
Immunosuppression [n (%)]   
 Cyclosporine14 (64)16 (80)0.31
 Tacrolimus8 (36)4 (20)0.31
 Corticosteroids22 (100)20 (100)>0.99
 Mycophenolate mofetil10 (45)7 (35)0.54
Patients undergoing retransplantation [n (%)]1 (5)2 (10)0.49

Characteristics of the PN Group

Eighteen of the 25 patients who received PN (72%) were receiving NG tube feedings at the time of PN initiation and had failed to show significant improvement on NG supplementation despite the maximum tolerated enteral therapy. The mean duration of NG supplementation was 2.9 ± 0.5 months at the time of PN initiation, and NG supplementation accounted for a mean of 71% of the total energy intake. The mean energy density of the formula or fortified breast milk was 28 kcal/oz (range = 20-44 kcal/oz), and the mean enteral intake was 118 kcal/kg/day (range = 40-180 kcal/kg/day) at the time of listing for transplantation. One patient received nasojejunal supplementation at the time of transplant listing. Seven patients (28%) began PN without the initiation of NG supplementation because of severe malnutrition at the time of their referral to our center with a mean MAC z score of −2.5 and a mean TSF z score of −3.2 (4 patients), decreased skinfold thickness despite an adequate documented oral intake of 160 kcal/kg/day (1 patient), postsurgical ileus (1 patient), or hypoglycemia (1 patient). PN was initiated in all of these patients during an inpatient hospitalization and was then transitioned to home care.

The median age at PN initiation was 7.7 months (range = 1.9-26 months), and the mean duration of PN supplementation was 86 ± 18 days. At the clinical endpoint (LT, death, or removal from the transplant waiting list), the mean amount of energy supplied by PN was 77 ± 5.5 kcal/kg/day. The median glucose infusion rate was 13.2 mg/kg/min (range = 6.1-17.2 mg/kg/min), the median intravenous lipid dose was 2.7 g/kg/day (range = 1.7-3.9 g/kg/day), and the median intravenous amino acid dose was 2 g/kg/day (range = 0.7-3.2 g/kg/day). Two patients received intravenous lipid only (mean = 2.5 g/kg/day).

Although indices of nutritional status (weight, length, TSF, and MAC z scores) were decreasing before PN initiation, they stabilized or improved after PN administration. From the start of PN administration to the clinical endpoint, the mean TSF z score increased from −2.5 ± 0.2 to −1.8 ± 0.2 (P < 0.001), and the mean MAC z score increased from −2.2 ± 0.2 to −1.4 ± 0.2 (P < 0.001). Twenty-two of the patients in the PN group (88%) underwent LT. All transplants were deceased donor organs [whole (32%), split (23%), or reduced (45%)], and the mean age at the time of LT was 12.9 ± 1.4 months. Details of immunosuppression used during the first 6 months after transplantation are provided in Table 1. Three patients died before undergoing LT.

Characteristics of the Non-PN Group

Six of the 22 patients in the non-PN group (27%) received NG feeds. NG supplementation accounted for a mean of 56% of the total energy intake in these patients at the time of the clinical endpoint (LT, death, or removal from the transplant waiting list). The mean duration of NG supplementation was 3.4 ± 0.7 months. In 2 subjects, NG feeding was initiated before listing, and in 4 patients, NG feeding was initiated after listing but before the clinical endpoint. The weight z score (−1.0 ± 0.2 at the time of HPE, −1.4 ± 0.3 at the time of transplant listing, and −1.6 ± 0.4 at the clinical endpoint), the length z score (−1.0 ± 0.2 at the time of HPE, −1.2 ± 0.2 at the time of transplant listing, and −1.4 ± 0.3 at the clinical endpoint), and the TSF z score (−1.4 ± 0.2 at the time of HPE, −1.5 ± 0.3 at the time of transplant listing, and −1.6 ± 0.3 at the clinical endpoint) did not significantly change from the time of HPE to the clinical endpoint. The MAC scores significantly worsened in this group from the time of transplant listing to the clinical endpoint. The MAC z scores were −1.2 ± 0.3 at the time of HPE, −0.8 ± 0.3 at the time of transplant listing, and −1.3 ± 0.2 at the clinical endpoint (P = 0.03). Twenty of the patients in the non-PN group (91%) underwent LT. All transplants were deceased donor organs [whole (40%), split (10%), or reduced (50%)], and the mean age at the time of LT was 12.7 ± 2.0 months. Details of immunosuppression used during the first 6 months after transplantation are provided in Table 1. Two patients were removed from the transplant list because of significant improvements in their clinical status.

Comparisons of the PN and Non-PN Groups

The median time from transplant listing to PN initiation was 5 days. PN was initiated as early as 52 days before transplant listing and as late as 380 days after listing. There were no differences between the PN and non-PN subjects with respect to time from HPE to transplant listing, from transplant listing to the clinical endpoint (LT, death, or removal from the transplant waiting list), and from transplant listing to LT. There was no significant difference in the percentage of subjects who received PN before and after the implementation of the PELD scoring system in 2002.16 The calculated PELD scores were similar for the PN and non-PN groups at the time of HPE and at LT listing (Table 2).

Table 2. Laboratory, Nutritional, and PELD Score Data for the PN and Non-PN Groups
 HPELT ListingClinical Endpoint
PN GroupNon-PN GroupP ValuePN GroupNon-PN GroupP ValuePN GroupNon-PN GroupP Value
Patients (n)ValuePatients (n)ValuePatients (n)ValuePatients (n)ValuePatients (n)ValuePatients (n)Value
  • NOTE: Statistically significant values (P < 0.05) are bolded. The clinical endpoint was transplantation, death, or removal from the transplant waiting list.

  • *

    The data are presented as means and standard errors.

Albumin (g/dL)*193.3 ± 0.1223.6 ± 0.10.15242.8 ± 0.1223.2 ± 0.10.03242.8 ± 0.1183.1 ± 0.20.31
Total bilirubin (mg/dL)*199.7 ± 1.0228.7 ± 1.00.542412.1 ± 1.42211.8 ± 1.90.862421.9 ± 2.21813.7 ± 2.80.04
Direct bilirubin (mg/dL)*197.1 ± 0.9226.9 ± 0.80.79218.4 ± 1.0187.6 ± 1.30.822214.5 ± 1.71510.6 ± 2.60.50
AST (U/L)*19293 ± 6522186 ± 610.2524154 ± 1722181 ± 160.3824452 ± 2718189 ± 1960.92
ALT (U/L)*19194 ± 2922163 ± 270.4324112 ± 1922171 ± 180.0324153 ± 2418148 ± 340.73
GGT (U/L)*181032 ± 16117958 ± 1660.5621528 ± 15116940 ± 1670.0418210 ± 6011502 ± 66<0.001
White blood count (×103/μL)*1913.6 ± 1.22214.5 ± 1.20.682313.4 ± 1.22014.8 ± 1.30.262412.0 ± 1.61813.8 ± 2.20.57
Platelets (×103/μL)*18485 ± 3422515 ± 310.6524202 ± 2619330 ± 280.0012499 ± 1718207 ± 18<0.001
Hemoglobin*1716.8 ± 3.32011.4 ± 3.00.252410.4 ± 0.32011.3 ± 0.30.09239.5 ± 0.31611.7 ± 0.4<0.001
Prothrombin time (seconds)*1914.1 ± 0.72212.8 ± 0.60.142414.8 ± 0.72214.4 ± 0.90.602418.5 ± 0.91814.6 ± 1.10.01
Weight z score < −2 (%)25822180.4025402236>0.9925362232>0.99
Length z score < −2 (%)2542290.6025202218>0.99253622230.36
TSF z score < −2 (%)14760>0.99207014210.0113381030>0.99
MAC z score < −2 (%)1331617>0.99204014210.29132310100.60
Calculated PELD score*198.9 ± 0.9227.1 ± 0.90.122412.5 ± 1.32210.7 ± 1.80.222419.6 ± 2.01812.0 ± 2.70.04

The nutritional intake of the patients receiving PN was compared to that of the non-PN group. Sixty-nine percent of the cohort had complete documentation for their dietary intake and were included in this analysis. The remaining 31% had incomplete data available for the calculation of enteral intake (50% were breastfed, 20% were on a regular diet, and 30% received an undocumented amount of formula), and they were excluded from further analyses. The non-PN group was missing more dietary records than the PN group (40% versus 17%, P = 0.01). The mean total energy intake in the PN and non-PN groups at the time of HPE (143 ± 10 versus 150 ± 10 kcal/kg/day), at transplant listing (121 ± 7 versus 111 ± 7 kcal/kg/day), and at clinical endpoint (104 ± 7 versus 103 ± 8 kcal/kg/day) were similar [P = not significant (NS)]. Although the total energy intake (kcal/kg/day) at clinical endpoint was the same in the 2 groups, subjects in the PN group received a majority of their energy intake (63%) from PN and 37% of their energy intake from enteral nutrition (Fig. 1).

Figure 1.

Energy intake in the PN and non-PN groups. The total energy intake was similar in the 2 groups at the time of HPE, at the time of transplant listing, and at the clinical endpoint (LT, death, or removal from the transplant waiting list). There was a significant difference in the energy delivered by the parenteral route in the PN group versus the non-PN group at the clinical endpoint (*P = 0.002).

Measures of muscle mass (MAC) and adipose tissue stores (TSF) are believed to better reflect nutritional status than weight and weight/length measurements in infants and children with chronic liver disease because of the contributions of an enlarged liver and/or spleen and the presence of ascites and fluid retention, which may inflate the weight for calculation of growth parameters.2, 8-13 Therefore, MAC and TSF z scores were used as primary measures of the nutritional status. In the PN and non-PN groups at the time of HPE, MAC z scores (−1.4 ± 0.2 versus −1.2 ± 0.3) and TSF z scores (−1.4 ± 0.1 versus −1.4 ± 0.2) were similar (P = NS; Fig. 2). At the time of transplant listing, the PN group had lower MAC z scores than the non-PN group (−1.7 ± 0.2 versus −0.8 ± 0.3, P = 0.01). Between the time of listing and the clinical endpoint, MAC z scores improved in the PN group (−1.7 ± 0.2 to −1.4 ± 0.2, P = 0.02) but worsened in the non-PN group (−0.8 ± 0.3 to −1.3 ± 0.2, P = 0.03). After transplant listing, the PN group had an average monthly rate of change in MAC z score of +0.30 z score units per month, whereas the non-PN group had an average monthly rate of change in the score of −0.15 z score units per month (P < 0.001). At the time of transplant listing, the PN group had lower TSF z scores than the non-PN group (−2.2 ± 0.2 versus −1.5 ± 0.3, P = 0.001). Between the time of listing and the clinical endpoint, the TSF z scores improved significantly in the PN group (−2.2 ± 0.2 to −1.8 ± 0.2, P = 0.006), and they were unchanged in the non-PN group (−1.5 ± 0.2 to −1.6 ± 0.1, P = NS). The average monthly rates of change in the TSF z score from listing to the clinical endpoint were +0.25 and +0.02 z score units for the PN and non-PN groups, respectively (P = 0.04). At the clinical endpoint, there was no significant difference between the 2 groups with respect to MAC and TSF z scores. Similar patterns of change were seen in the weight, length, and weight/length z scores; however, the differences between the groups were not statistically significant.

Figure 2.

Comparison of MAC and TSF in the PN and non-PN groups. MAC and TSF were similar in the 2 groups at the time of HPE. However, by the time of listing for transplantation, MAC and TSF were significantly lower in the PN group versus the non-PN group (*P = 0.01 and **P = 0.001), respectively. At the clinical endpoint (LT, death, or removal from the transplant waiting list), because both MAC and TSF had improved in the PN group, there was no difference in MAC or TSF between the PN and non-PN groups.

At the time of HPE, the PN and non-PN groups had similar baseline laboratory studies (Table 2). At the time of transplant listing, patients on PN had a lower mean serum albumin (2.8 versus 3.2 g/dL, P = 0.03), GGT (528 versus 940 U/L, P = 0.04), and platelet count (202 versus 330 × 103/μL, P = 0.001). At the clinical endpoint, the mean serum albumins (2.8 versus 3.1 g/dL, P = NS) were similar in the PN and non-PN groups; however, the PN group had a higher serum total bilirubin level (21.9 versus 13.7 mg/dL, P = 0.04), a more prolonged prothrombin time (18.5 versus 14.6 seconds, P = 0.01), and a lower GGT (210 versus 502 U/L, P < 0.001).

The weight, length, TSF, and MAC z scores were compared to serum albumin levels, total bilirubin levels, and prothrombin times at the time of transplant listing. The TSF z score was weakly associated with the serum albumin level (r2 = 0.34, P = 0.05), and the weight z score was weakly associated with the prothrombin time (r2 = 0.32, P = 0.03). All other associations were NS.

The medical outcomes of the PN and non-PN groups were compared from the time of HPE until the clinical endpoint. The PN group had a higher rate of gastrointestinal bleeding than the non-PN group (60% versus 14%, P = 0.002) and a higher rate of ascites development (76.0% versus 36.4%, P = 0.009). There was no significant association between the presence of a central line and the incidence of a positive blood culture (48% of the patients with a central line developed bacteremia, whereas 20% of the patients without a central line did; P = 0.09). There was no significant association between PN use and the incidence of a positive blood culture (52% of the PN patients developed bacteremia, whereas 41% of the non-PN patients did; P = 0.89).

There were no deaths in the non-PN group and 3 deaths in the PN group (1 from gastrointestinal bleeding, 1 from fungal sepsis, and 1 from respiratory arrest) before transplantation (P = NS). The mean age at LT, the time from listing to transplantation, the mean calculated PELD score at the time of transplantation, the graft types, and the types of immunosuppression used during the first 6 months after transplantation were similar between the PN and non-PN groups (Table 1). Among the transplant patients, there were no differences between the PN and non-PN groups in the days spent in the ICU after transplantation (9.1 versus 8.9 days) or in the graft survival, patient survival, and retransplantation rates (Fig. 3 and Table 1).

Figure 3.

Post-transplant patient and graft survival in the PN and non-PN groups. The patient survival rates [P = 0.37 (log-rank test)] and graft survival rates [P = 0.52 (log-rank test)] were similar in the PN and non-PN groups after LT.

DISCUSSION

This is the first study to report the impact of PN on the outcomes of BA patients with end-stage liver disease. The initiation of PN in malnourished BA patients restored many patients to the same nutritional status observed in those patients managed with enteral nutrition alone. The outcomes after LT were similar in the 2 groups despite more advanced liver disease at the time of transplantation in the PN group. These data suggest that the beneficial effects of PN on nutritional status may have contributed to better outcomes after LT than would have been expected had the patients remained in their severely compromised nutritional state.

Malnutrition, a significant problem for children with BA, can be difficult to assess because ascites, organomegaly, and peripheral edema may confound the interpretation of their weight and weight/length measurements.2, 13 Weight z scores and weight/length percentiles have been shown to overestimate the nutritional status in patients with BA. MAC (a measure of muscle mass) and TSF (a measure of adipose stores) more accurately assess the state of malnutrition in these patients and were used as the primary markers of nutritional status in this study.2, 8-13

Decreased oral intake, early satiety, fat malabsorption, and increased energy expenditure due to a hypermetabolic state all likely contribute to malnutrition in BA patients.17 Because the mean energy intake (111 ± 7 kcal/kg/day) in the PN group was well below the estimated caloric needs (131 kcal/kg/day) and there were signs of progressive malnutrition (lower MAC and TSF z scores), PN was initiated to improve nutritional status while patients awaited LT.3 Within the PN group, 72% had already failed to show improved nutritional status after the initiation of NG tube feeding supplementation, so PN was considered the only viable option. With the addition of PN, the total energy intake was maintained in the PN group at the same level found in the non-PN group at the clinical endpoint. The administration of PN reversed the trend of falling TSF z scores in the PN group and restored the values to those observed in the non-PN group by the time of the clinical endpoint. At transplant listing and at the clinical endpoint, although the total energy intake was similar in the 2 groups, the MAC and TSF z scores had improved only in the PN group. It is likely that the PN patients would have experienced continued deterioration of their nutritional status after transplant listing had PN not been instituted.

At transplant the albumin levels and platelet counts were significantly lower in the PN group versus the non-PN group, and this reflected malnutrition and more severe liver disease. Among the patients receiving PN, the albumin levels stabilized or improved after listing. The platelet counts, however, continued to fall, and the serum bilirubin levels and prothrombin times rose after transplant listing in the PN group; this suggested the progression of chronic liver failure and portal hypertension. Moreover, at the clinical endpoint, the serum bilirubin levels and the prothrombin times were higher and the platelet counts were significantly lower in the PN group versus the non-PN group. Thus, despite worsening liver function, PN administration was successful in improving or stabilizing the nutritional status of these infants with end-stage BA.

Malnourished BA patients are at higher risk for poor pre-transplant outcomes.5, 18 Length and weight z scores > −2 are associated with better outcomes in BA patients.5 Once they are listed for LT, malnutrition (length or weight z scores < −2) is associated with an increased need for pre-transplant ICU monitoring and with mortality.16 Decreasing weight and length z scores and serum albumin levels are also risk factors for death before transplantation.6 In our PN group, TSF, MAC, and albumin levels were decreasing before PN administration, and they stabilized or increased after PN. Before transplantation, the PN group had a higher incidence of gastrointestinal bleeding and ascites than the non-PN group. Although this most likely represents more advanced liver disease and portal hypertension in the PN group, PN and the subsequent expansion of the vascular volume may have also contributed. The administration of PN was not, however, associated with increased bacteremia or pre-transplant mortality. In summary, because of the expected worse pretransplant outcomes of BA patients who are severely malnourished,5, 6, 16 we speculate that the pre-transplant complications of liver disease would have been even worse in the PN group had they not achieved the improved nutritional status provided by PN administration.

Malnourished BA patients are also at increased risk for significant post-transplant complications. Growth failure at the time of LT confers an increased risk of graft failure and posttransplant death.6, 7, 18 In our study, PN improved MAC and TSF z scores to >−2 at the clinical endpoint. Although weight z scores also improved, these findings are difficult to interpret because of the frequency of ascites and fluid retention in this population. After transplantation, the PN and non-PN groups had similar clinical outcomes (days in the ICU after transplantation, need for retransplantation, and graft and patient survival). We postulate that without PN-driven improvements in nutritional status, this group would have had poorer posttransplant outcomes in comparison with the non-PN group.6, 7, 18

Although this study addresses the importance of nutrition in patients with BA, there are several limitations. The retrospective nature of this study resulted in incomplete nutritional data at all time points for some subjects (eg, those who were breast-fed). Although bacteremia rates were similar, we were unable to normalize bacteremia events to the number of days with intravenous catheters in place because of inadequate documentation of line removal dates in the medical records. In addition, although the pretransplant mortality rates were similar for the PN and non-PN groups, the possibility of a type 2 error exists. We acknowledge that the PN and non-PN groups were not clinically equivalent (the PN group had more advanced disease and more severe portal hypertension), and this limits to some extent the comparison of the outcomes of these 2 groups. However, the more advanced liver disease in the PN group would have biased this group toward worse clinical outcomes, which were not observed.

It should be noted that, despite the improved nutritional status of the patients receiving PN, their liver disease appeared to progress more rapidly after PN initiation (with higher bilirubin levels, lower platelet counts, longer prothrombin times, and higher calculated PELD scores) in the setting of significantly lower GGT levels in comparison with the non-PN group. PN-associated liver disease is well documented in premature infants and infants with intestinal failure requiring PN supplementation,19 but it has not been described in BA patients. Our PN patients received doses of soy-based intravenous lipids within the range implicated in the pathogenesis of liver injury in infants with intestinal failure.20-22 Thus, it is possible that PN may have contributed to the more rapid progression of cholestasis in the PN group. Lipid-sparing techniques and omega 3 fatty acid–based intravenous lipid preparations, which have been shown to reverse PN-associated cholestasis,20-22 might be considered for BA patients in the future to minimize this possibility.

In conclusion, malnutrition is a serious clinical problem in BA patients awaiting LT. This study provides evidence that PN can be effectively used in BA patients with advanced liver disease and portal hypertension to improve their nutritional status while they are awaiting LT. These data also suggest that PN may stabilize or improve the clinical outcomes both before and after LT by virtue of improvements in the nutritional status. The potential effect of PN on the pace of liver function worsening in BA patients requires additional investigation. On the basis of the outcomes documented in this study, we suggest that PN should be considered a useful option for infants and children with end-stage liver disease and severe malnutrition who are awaiting LT.

Ancillary