Improved survival rates in the pediatric intestinal transplant (ITx) population have occurred as a result of advances in surgical technique; improved immunosuppressive regimens; enhanced management of medical, surgical, and infectious complications; and expedient weaning from parenteral nutrition (PN) (1). Reducing mortality following ITx requires an increased recognition of important morbidities associated with post-ITx management. Although PN can be discontinued and enteral autonomy achieved in most patients following ITx (2–6), attaining sustained positive linear growth velocity remains a challenge for many children despite successful elimination of PN (7,8). Factors that affect linear growth in this population include pretransplant circumstances that involve their underlying condition, prematurity, being small-for-gestational age, end-stage liver disease, recurrent sepsis, and malabsorption. With a full complement of intestine and a healthy liver that is achieved after ITx, factors that may affect growth include immunosuppression, particularly the use of corticosteroids, intestinal infections, and graft rejection, which may also involve the use of corticosteroids.
The use of daily corticosteroids is associated with reduced growth velocity in a number of pediatric conditions, including asthma (9), inflammatory bowel disease (10), renal disease (11), and rheumatologic conditions (12). In 2002, we adopted an immunosuppression regimen for patients receiving ITx that involved induction immunosuppression with thymoglobulin and the elimination of corticosteroids from the management protocol with the exception of rejection episodes. This change in the management protocol provided us the opportunity to assess the effect of reduced corticosteroid exposure on linear growth in children following ITx. We hypothesized that this change in immunosuppressant treatment would also contribute to improved weight gain and linear growth in these children.
PATIENTS AND METHODS
The Intestinal Care and Rehabilitation Center at the Children's Hospital of Pittsburgh of the University of Pittsburgh Medical Center has served as a center for the evaluation and management of children with short bowel syndrome or intestinal failure for more than a decade. Using an established registry approved by the institutional review board of the University of Pittsburgh, we retrospectively reviewed all of the patients who had received a small bowel transplant between December 1996 and February 2007. Patients were categorized by their transplant induction status. Nutritional analyses included evaluation of achievement of nutritional autonomy, defined as the discontinuation of PN and intravenous fluids (IVFs), as well as trends in z scores for weight and height/length between transplantation and 2 years post-ITx.
Weight (kilograms) was measured with a digital medical scale. Height (centimeters) was determined using a stadiometer. Infants and toddlers (newborn to 18 months) had their weight and length measured using a digital infant scale and recumbent length board. Four measures including weight, weight z score, height, and height z score were recorded at 3 different time points: transplantation, 1 year posttransplant, and 2 years posttransplant. z Score statistics for weight and height/length were calculated using EpiInfo and the Centers for Disease Control and Prevention year 2000 growth charts (Department of Health and Human Services, Atlanta, GA).
The immunosuppression regimen between 1997 and 2001 was with TAC and tapering doses of steroids (steroid). TAC was titrated to whole-blood concentrations of 18 to 20 ng/mL during the first 2 months. By the end of the first year, target concentrations were 5 to 10 ng/mL. The immunomodulatory regimen that was adopted in 2002 included 2 to 3 mg/kg of a lymphocyte-depleting agent (rATG, thymoglobulin) just before ITx and 2 to 3 mg/kg postoperatively (total 5 mg/kg) (steroid-free) coupled with twice-daily doses of TAC provided enterally at a dose of 0.1 mg/kg that was titrated to achieve whole-blood trough concentrations of 12 to 18 ng/mL in the first month, 10 to 15 ng/mL during the second and third months, and 5 to 10 ng/mL by the end of the first year (13). Corticosteroids or other agents were provided only to treat breakthrough rejection.
Our protocol for nutritional support of postintestinal transplant recipients was unchanged between 1997 and 2007. Nutritional management included the continuation of PN until enteral nutrition (EN) was established. The progression of EN was dependent on the development of postoperative complications. Energy goals were adjusted based upon clinical condition and wound healing status using a formula with either isotonic hydrolyzed protein or peptide-based formula containing medium-chain triglycerides. The transition was made to an intact formula product when there were no signs of rejection by endoscopy and when stool output was maintained at <40 mL kg−1 day−1. Fluid replacement occasionally continued for some months after transplant.
We built linear models to evaluate growth trends after ITx (SPSS version 15.0, SPSS Inc, Chicago, IL). The models were built with either of the weight, height, and z scores versus the time (at transplantation, 1 year posttransplantation, and 2 years posttransplantation). The slope of the model indicated the change in growth pattern during the time period, with a positive slope meaning positive growth trend and a negative slope indicating a negative growth trend. An intercept of the model was the expected measure at ITx. The association between the numbers/percentages of children who were observed to have a positive growth trend and treatment category was detected by using the chi square or Fisher exact tests, if applicable. Linear regression was used to evaluate the relation between dependent nutrition variables (time to start of enteral feedings and time to full enteral/oral feedings) and confounding factors. For those who received the steroid-free regimen pretransplant, growth trends over time were evaluated using multivariate logistic regression to determine whether there is an effect of steroid use (<120, 121–365, and >365 days) on growth while taking into account confounding variables. To evaluate differences in demographic variables, liver disease status at the time of transplant, incidence of rejection and infection after transplant, and nutrition outcomes after transplant between the 2 cohorts, we used Mann-Whitney and chi square statistics.
A total of 109 children with intestinal failure received an isolated small bowel, small bowel/liver, or multiple organ transplant that included a small bowel between December 1996 and February 2007. Of these, 29 received a transplant before March 2002 and received TAC and steroid therapy, whereas the remaining 80 children received a steroid-free induction with rATG and TAC. The demographic characteristics for patients in each treatment group are shown in Table 1. No statistically significant differences were found between groups for sex, race, primary diagnosis, median age at transplant, and pretransplant total bilirubin level. However, a significantly greater percentage of patients receiving transplants before March 2002 required a liver to be part of the allograft (79% vs 57%, respectively; P < 0.05). One patient in the steroid cohort died in the immediate postoperative period, resulting in a steroid cohort of 28 for the purpose of the present investigation. Univariate analysis of the differences in episodes of rejection ≤90 days after transplant, the grade of severity and method of treatment of these episodes, and the incidence of tissue confirmed cytomegalovirus infection and incidence of Epstein-Barr virus infection or posttransplant lymphoproliferative disorder during the study period is shown in Table 2. No significant differences were observed between groups.
Although all of the patients were dependent on PN at the time of transplantation, 61 (45 of 80 in the steroid-free group and 16 of 29 in the steroid group or ∼56% in each) were receiving some form of EN before surgery. Table 3 summarizes the differences in nutritional outcomes by the type of immunosuppressant treatment received. For the 28 patients who underwent successful transplants before March 2002 and were treated with the steroid regimen, EN was initiated at a median of 11 days posttransplant. PN was continued a median of 27 days after surgery (range 10–289), and nutritional autonomy was achieved by 25 patients (89%) within 7 months postoperatively (range 2–66 months). In comparison, for the 80 patients who received steroid-free immunosuppression, enteral feedings were started more quickly or a median of 6 days after transplant (range 1–47 days; P < 0.001). PN support was continued a median of only 9 days postoperatively (range 0–188 days; P < 0.001), and full enteral and/or oral intake was achieved by 70 patients (88%) within a median of only 2 months (range 0–19 months; P < 0.001). Linear regression analysis was conducted to evaluate the relation between time to the start of enteral feedings and time to full enteral/oral feeding and confounding factors. Covariates included incidence of rejection within 90 days of transplant, episodes and severity of infection, sex, race, age at transplant, and type of immunosuppressant treatment received. After controlling for these confounding factors, we found a significant linear relation between feeding initiation and nutrition autonomy with the type of immunosuppression used (P < 0.001). The median number of days to ostomy closure was only slightly higher in those treated with the steroid-free regimen (n = 46, 194 days) than in those who received a transplant earlier (n = 16, 162 days) and was not significantly different.
Three patients (2 steroid-free group, 1 steroid group) did not have a weight or height value available at the time of transplant. Weight and/or height values were available at 1 and 2 years, respectively, for 25 and 17 patients who received steroid treatment and for 70 and 63 patients who received steroid-free immunosuppression. The percentage of patients with growth failure (<5th percentile for weight for age and height for age, or a z score of < −1.65) at transplant as well as 1 and 2 years after transplant by treatment status is shown in Figures 1 and 2, respectively. Although the percentage of children at the <5th percentile for height age in the steroid group remained relatively constant during the 2-year observation period (65% at transplant compared to 63% at 2 years postoperative), the percentage with a z score < −1.65 dropped somewhat in those who received the steroid-free immunosuppression regimen (55% at transplant vs 47% after 2 years).
Mean z scores for weight and height/length for patients with measures at the time of transplant (baseline) and at years 1 and 2 posttransplant by immunosuppressant treatment status are shown in Table 4. Although no statistically significant trend was observed, patients in the steroid-free immunosuppression group experienced less weight loss and a greater increase in linear growth than those receiving earlier transplant (weight −0.33 vs −0.40; height 0.26 vs 0.12, respectively). A positive trend in z scores for weight and height was observed in 41% and 44% of patients in the steroid group. In the population of patients who received steroid-free immunosuppression, the same percentage (41%) experienced a positive trend in weight over time. However, a numerically higher percentage (48%) of children receiving transplant since 2002 were observed to have a positive trend in linear growth.
Steroids were used to treat episodes of rejection after transplant in those patients who received the steroid-free immunosuppression regimen. Thirty-one received steroids for ≤120 days, 14 received steroids for 121 to 365 days, and 23 received >365 days of steroids within 2 years posttransplant (data missing for 12 patients). z Scores for weight and height/length at transplant, 1 year, and 2 years after transplantation by days of steroid use posttransplant are shown in Figures 3 and 4. Multivariate analysis that included demographic factors, total bilirubin pretransplant, incidence of rejection and infection posttransplant, and nutrition outcomes as confounding variables revealed a nonsignificant effect of steroid use on weight (P = 0.08). However, constituents of weight cannot easily be differentiated in this population because of fluid shifts experienced posttransplant. Incidence of rejection within 90 days after transplant had the greatest effect on linear growth, but it was also not statistically significant. Changes in z score values for height/length during the 2-year evaluation period were positive in the ≤120-day (0.55) and the 121- to 365-day group (0.27), but not in the group who received >365 days of steroids (−0.19). In addition, 62% of children who received steroids for ≤120 total days were observed to have positive linear growth, compared to 43% and 36% in the 121- to 365-day and >365-day groups, respectively.
Although achievement of positive growth after pediatric ITx remains a challenge for health care professionals, we have found more favorable growth trends in patients who received steroid-free immunosuppression as compared with those who received TAC and tapering doses of steroids. Growth trends are particularly encouraging among children in the steroid-free group who had limited exposure to steroids after transplant. Enteral feedings were initiated sooner and nutritional autonomy was achieved more rapidly in ITx patients who received the steroid-free regimen. Although our nutrition care philosophy and protocol remained essentially the same throughout the time period of the study, we cannot rule out the possibility of an effect of patient-care team experience and aggressiveness with feeding initiation on the present finding.
In 2002, with a cohort of 24 patients receiving transplants between 1996 and 2000, we reported that positive trends in weight and height/length z scores were observed during a 2-year follow-up period in 30% and 26% of patients, respectively (6). We concluded that achieving positive linear growth remained a challenge after ITx in children, despite successful elimination of PN and high survival rates. These findings were corroborated by Iyer et al (7) in a population of 46 children with an ITx, who had retained their graft for at least 1 year. Growth velocity was maintained for 2 years, but no catch-up growth had occurred in this series even though all of the survivors with a functioning graft had achieved full enteral feedings. Patients in both studies had received standard immunosuppression therapy with TAC and prednisone.
More recently, Ueno et al (3) described growth after ITx in 23 children who received transplants between 1999 and 2004. Baseline immunosuppression included TAC and corticosteroids with daclizumab and alemtuzumab used as an induction agent in 18 and 5 patients, respectively. Patients with the most significant growth retardation at the time of transplant (z score < −2.0) were found to have significant acceleration in both weight and linear growth after transplant. Lacaille et al (5) reported growth outcomes in 31 patients receiving transplants between 1988 and 2006 who were studied for between 2 and 18 years. The type of immunosuppressant regimen used included TAC plus prednisone from 1994 onward with azathioprine used between 1994 and 2004. Induction with basiliximab was used from 2001 onward, and sirolimus was used in 2002 for isolated small-bowel transplant only. Severe growth failure (z score < −2.0) was observed in only 23% of patients at the time of transplant, and 26 patients (84%) remained free of PN after ITx. During the follow-up period, two-thirds of patients achieved normal growth, and of the 6 patients who had completed their growth, 5 reached adult height within normal limits. However, catch-up growth was not observed, leading the authors to conclude that height lost before transplantation may not be regained. Venick et al (8) observed positive growth trends in 12 of 19 patients receiving transplants between 1991 and 2005 (63%), particularly in the first year after transplant, but no significant catch-up growth. Therefore, the effect of ITx on short-term growth parameters remains unclear. Despite the observation of positive growth trends after ITx in multiple studies, catch-up growth has not been consistently demonstrated in the majority of patients.
We and others have reported growth maintenance with limited catch-up growth in relatively small populations of ITx recipients. Up to two-thirds of children both with and without severe growth retardation at the time of transplant have exhibited growth maintenance. At the time of transplant, more than half of our patient population was observed to have growth failure. Therefore, if the same conditions of immunosuppressant therapy existed as in previous reports, we would have expected to observe growth maintenance in no more than 60% of our population and limited growth acceleration. Our study showed that a slightly higher percentage of patients receiving a steroid-free protocol experienced positive growth than those who had received TAC and tapering doses of steroids. However, our most encouraging finding was the observation of greater positive linear growth in almost two-thirds of patients on the steroid-free regimen who received <120 days of steroids during the 2-year follow-up period compared to patients who received steroids for longer periods.
Limited numbers of patients in the pediatric ITx population as well as the number of factors involved in the care and management of post-ITx patients continue to make the evaluation of the growth process somewhat challenging. The difficulties of data management from the rare ITx patient population are compounded further by the challenges of obtaining anthropometric measures after patients return home. Although our population was larger than previous investigations of this nature, missing anthropometric data resulted in a less than optimal number of comparisons between baseline and follow-up time periods. Growth parameters were not always obtained at the time of follow-up laboratory work once patients were medically stable. These measures are now being obtained regularly in an effort to more closely monitor growth progress.
We evaluated confounding factors that may affect growth, including the severity of illness before and after transplant, between immunosuppressant treatment groups and found a significant difference only in the percentage of patients requiring a liver to be part of the allograft in the steroid cohort. Although the groups appear to be comparable demographically, it is possible that the sickness of the patients in the earlier group resulted in the modest differences in growth that were observed. Nutritional autonomy was eventually achieved by the majority of patients. However, nutritional intake data were not collected to be part of the present evaluation. Although our goal was to provide adequate intake via a combination of parenteral and enteral routes at all times, it is possible that nutrition was suboptimal while PN and IVF were being weaned. Patients in the earlier group received PN for a longer period, but the transition period between discontinuation of PN and nutrition autonomy was also longer, which may explain the slower rate of growth that was observed.
Despite these limitations, improvement in immunosuppressant management in recent years has resulted in improved survival and complications as well as growth outcomes. We believe that regular nutritional follow-up with coordinated interdisciplinary management to address social, behavioral, and medical issues is essential to maintain nutrition status, to provide the base for growth acceleration, and to improve quality of life after ITx. Continuous assessment of strategies must continue as immunosuppressant treatment regimens are refined further.
1. Grant D, Abu-Elmagd K, Reyes J, et al
. 2003 report of the intestine transplant registry. Ann Surg 2005; 241:607–613.
2. Nucci AM, Barksdale EM, Yaworski JA, et al
. Enteral formula use in children after small bowel transplant. Nutr Clin Pract 2002; 17:113–117.
3. Ueno T, Kato T, Revas K, et al
. Growth after intestinal transplantation. Transplant Proc 2006; 38:1702–1704.
4. Sudan DL, Iverson A, Weseman RA, et al
. Assessment of function, growth and development, and long-term quality of life after small bowel transplantation. Transplant Proc 2000; 32:1211–1212.
5. Lacaille F, Vass N, Sauvat F, et al
. Long-term outcome, growth and digestive function in children 2 to 18 years after intestinal transplantation. Gut 2008; 57:455–461.
6. Nucci AM, Barksdale EM, Beserock N, et al
. Long-term nutritional outcome after pediatric intestinal transplantation. J Pediatr Surg 2002; 37:460–463.
7. Iyer K, Horslen S, Iverson A, et al
. Nutritional outcomes and growth of children after intestinal transplantation. J Pediatr Surg 2002; 37:464–466.
8. Venick RS, Farmer DG, Saikali D, et al. Nutritional outcomes following pediatric intestinal transplantation. Transpl Proc
9. Tamesis GP, Covar RA. Long-term effects of asthma medications in children. Curr Opin Allergy Immunol 2008; 8:163–167.
10. Conklin LS, Oliva-Hemker M. Nutritional considerations in pediatric inflammatory bowel disease. Expert Rev Gastroenterol Hepatol 2010; 4:305–317.
11. Vidhun JR, Sarwal MM. Corticosteroid avoidance in pediatric renal transplantation: can it be achieved? Pediatr Drugs 2004; 6:273–287.
12. Murray KJ, Lovell DJ. Advanced therapy for juvenile arthritis. Best Pract Res Clin Rheumatol 2002; 16:361–378.
13. Reyes J, Mazariegos GV, Abu-Elmagd K, et al
. Intestinal transplantation under tacrolimus monotherapy after perioperative lymphoid depletion with rabbit anti-thymocyte globulin (thymoglobulin). Am J Transplant 2005; 5:1430–1436.
Keywords:Copyright 2011 by ESPGHAN and NASPGHAN
intestinal failure; intestinal transplantation; parenteral nutrition