Taha, A.M.I.*; Sharif, K.*; Johnson, T.†; Clarke, S.†; Murphy, M.S.‡; Gupte, G.L.*
Isolated liver transplantation (iLTx) in children with short bowel syndrome (SBS) and intestinal failure–associated liver disease (IFALD) has been reported from our centre in 14 children with proposed selection criteria (1). The key determining factor in proposing this option in SBS and IFALD is the predicted ability of the small bowel to adapt and allow weaning from parenteral nutrition (PN) to enteral nutrition (EN) (2,3) and the shortage of size-matched organs in the United Kingdom that precludes them from combined liver and small bowel transplantation (iSBTx). The major difficulties in the initial care and short to medium term follow-up have been outlined in the previous report (1). This report details the long-term follow-up of 8 survivors with regard to nutritional rehabilitation and long-term enteral adaptation, and describes the outcomes of the 3 children who later developed irreversible intestinal failure in further detail.
A retrospective study of the medical records of 8 of 14 children, who survived more than 3 years following iLTx for SBS and IFALD at our institution from 1998 to 2005, documenting the following variables: duration of survival, time to wean from PN onto enteral feeds, re-commencement of PN, need for further bowel transplantation or nontransplant bowel surgery, time to wean PN afterwards, and growth by anthropometric measurements and z scores. We also describe in detail the clinical course of the 3 children who developed irreversible intestinal failure.
SPSS version 13 (SPSS Inc, Chicago, IL) was used for statistical analysis. Comparisons of medians of continuous variables are made by Wilcoxon test for nonparametric comparisons of related samples. We did not encounter missing data because the children were closely followed up.
Eight of the 14 children were still alive with median follow-up of 107.5 months (range 89–153) and median age is 115.2 months (range 94–169). Five of the 8 surviving children (thus only approximately one-third of the original group of 14 children) were weaned from PN into EN within a median of 10 months after iLTx (range 3–32 months). Three of those 5 children were subsequently weaned from EN to full oral feeding within 13, 24, and 53 months, respectively, of stopping PN (23, 56, and 57 months after the transplant, respectively), whereas the other 2, despite being weaned from PN successfully (3 and 15 months after the iLTx), still required EN by nasogastric tubes 118 and 74 months after stopping PN (121 and 89 months after iLTx, respectively). After initial weaning, 1 of the latter 2 subsequently required PN for 4 weeks when she underwent surgery for stoma closure. The other also required recommencement of PN 9 months after the iLTx. He underwent a surgical procedure to taper a dilated loop of bowel to promote bowel adaptation, and subsequently was successfully weaned from PN 6 months later. Both of these 2 children are being weaned progressively from EN to full oral feeding trying to circumvent their feeding problems as vomiting and less desire to feed. The cause of SBS in the former 3 children who are on oral feeding was necrotising enterocolitis and it was gastroschisis in the latter 2 children.
These 5 children, who successfully discontinued PN, maintained their weight median z score through the years (Fig. 1). The weight median z score in any single year is significantly (P < 0.001) more than the pretransplant weight median z score. The median increases in z scores for weight is 1.59 (range 1.24–1.79) compared with pretransplant z score. There is no statistically significant change in z score on comparing any 2 successive years beyond the third year. The height median z scores of these 5 children show fluctuations through the years (Fig. 2). Only the seventh year has z score significantly more than the pretransplant z score (P = 0.036). Both the weight and height z scores did not approach the normal even after 9 years of follow-up and both were lagging with >1 standard deviation behind the general population.
The other 3 of 8 surviving children had established irreversible intestinal failure at some point after their iLTx, despite exhaustive trials to maintain the autonomy of their bowels. Two of them had undergone iSBTx and the third is presently receiving PN. The clinical courses of these 3 children are described below (cases 1 to 3).
The overall 5-year survival in the whole cohort of 14 children is 57%. The liver graft survival of the whole cohort is also the same as the overall survival because the transplanted grafts had not failed in any of the survivors.
A child with gastroschisis and SBS (19 cm of jejunum with ileocecal valve [ICV] remaining) had recurrent central line infections and developed progressive IFALD. He was listed for liver and small bowel transplant, but because of acute deterioration in his liver function tests, he underwent an iLTx at the age of 6 months. He was discharged from the hospital on home PN with partial enteral feeding. He had Bianchi's isoperistaltic intestinal lengthening procedure 18 months after the transplant (following the procedure, the small bowel length was 55 cm), and he was weaned from PN several months later (4). One year after the lengthening procedure, he developed intestinal obstruction from a kink in the sigmoid colon, which was surgically corrected. He was subsequently followed up by the local team and remained independent of PN for a further 4 years. Once follow-up was established with the liver transplant team, he was recognised to have periods of being unwell with lethargy, poor weight gain, and clinical signs consistent with intermittent intestinal obstruction. Further attempts to enable him to gain weight with EN manipulation were attempted. He underwent a laparotomy (8 years after iLTx) to improve his intestinal autonomy. At laparotomy, a narrowing at the anastomosis of the lengthened bowel to the distal large bowel was identified and corrected. Subsequently, he experienced a gradual decompensation in intestinal function and required recommencement of PN with irreversible intestinal failure. Then, he experienced recurrent central venous catheter infections and because of the risk to his transplanted liver he underwent an iSBTx 112 months after the iLTx. After the iSBTx, he was weaned from PN within 4 weeks and discharged from the hospital within 6 weeks. Now, 41 months after the iSBTx, he is well and his growth and nutrition are satisfactory.
A child with gastroschisis and intestinal failure caused by SBS (40 cm of jejunum with loss of ICV) developed progressive IFALD associated with PN. He underwent an iLTx at the age of 9 months, and remained dependent on PN. Stoma closure was performed 15 months after iLTx, and several months later, he developed partial intestinal obstruction requiring laparotomy and adhesiolysis. A liver graft biopsy, done at the same time, demonstrated mild fibrosis of the graft secondary to IFALD. His clinical course was complicated by frequent admissions to hospital with life-threatening central venous catheter infections. Repeated attempts at weaning him from PN failed because of malabsorption. During an episode of presumed sepsis, he became profoundly ill with multiorgan failure, necessitating prolonged admission to the intensive care unit and tracheostomy. Assessment of his graft disease showed features of IFALD (clusters of large macrophages, showing golden pigmentation, within portal tracts) with progression of fibrosis from mild to at least a moderate degree and no evidence of rejection. He underwent a successful iSBTx, 84 months after the iLTx. Now, 11 months after the iSBTx he is well, with good growth and weight gain.
A child born with gastroschisis and intestinal failure associated with SBS (65 cm jejunum remaining with ICV) experienced recurrent central venous catheter infections, including life-threatening line infections at 2 and 7 months. Manifestations of advanced IFALD were evident at the age of 9 months and progressed rapidly, with 2 episodes of severe gastrointestinal bleeding that required blood transfusion. He underwent iLTx at 15 months because it was considered that intestinal adaptation was likely to allow eventual discontinuation of PN. One week later, he required a laparotomy for bile leak repair and then made a good recovery. PN was discontinued 14 months after the transplant. He was restarted on PN because of faltering growth 2 years after iLTx, and he then required home PN for 9 months, from which he was weaned to enteral feeds. He was tolerating enteral feeds and did not require PN for the next 4 years. He required PN for a 10-month period following an episode of bleeding from distal ileal ulceration (7 years from transplant), which required a complicated mini-laparotomy to assist endoscopy. He was tolerating enteral feeds for a period of 16 months but eventually his intestinal function deteriorated (9 years from transplant) with worsening diarrhoea and weight loss. He is now on home PN for 5 nights a week and is making good progress with satisfactory growth and nutrition.
Our study in a group of children undergoing iLTx for SBS and IFALD documents that they still need PN intermittently, has the potential to develop complications needing close follow-up, and many of these occur well beyond 5 years. The children need continuing active management by an experienced multidisciplinary intestinal failure team, including close dietetic follow-up, frequent hospitalisations, and nontransplant surgery. In our group of patients, 5 of 14 children (approximately one-third) could be weaned from PN and only 3 of 5 survivors could be weaned from EN into full oral feeding. The catch-up growth demonstrated within the first few years has not been sustained in the long term.
Dell-Olio et al (1) reported on 9 survivors of the same 14 children of the present report who had a median follow-up of 3 years after iLTx for IFALD. It was also reported that 8 children were weaned from PN and the ninth child was being weaned. The report also showed promising catch-up in weight and height, after an initial drop in the height in the first year. Of more relevance was that there was improvement in both weight and height year after year, implying that those children are going to maintain the trend of improvement and with speculation that they would continue to progress towards the normal expected weight and height for age. Examining our present results reveals that the child who was being weaned off PN could not be weaned at all and later received iSBTx (case 2). In addition, 1 child had died and 2 other children developed irreversible intestinal failure; one required an iSBTx (case 1) and the other was PN dependent (case 3). For the remaining 5 children, despite maintaining their weight median z score, there is no further significant improvement beyond the third year, with a small improvement after the seventh year. Despite the catchup in the height median z score noted in the first 3 years, the z score dropped again in the fourth and later in the eighth year. We do not understand the results of suboptimal growth. The patients were closely followed up by a multidisciplinary team and a senior dietitian experienced in the management of intestinal failure and weaning children from PN. We can only speculate that the cause for poor growth is multifactorial and includes many factors: prolonged hospitalisation and suboptimal growth before transplant, recurrent admissions to the hospital, and probably a much higher caloric requirement than is recommended for the SBS population.
On further analysis of our patients who developed irreversible intestinal failure following iLTx, we identified that they did not have the criteria proposed by Dell-Olio et al for the success of iLTx (1), namely, established IFALD (serum bilirubin >200 μmol/L, moderate/severe fibrosis, portal hypertension), at least 50-cm functional small bowel remaining intact in the absence of ICV or 30 cm with ICV, at least 50% of the estimated daily energy requirement tolerated enterally for a few weeks before the development of liver disease and associated with an increase in weight, and finally no history of recurrent line infections in the presence of dilated dysmotile bowel. The child who died had had multiple recurrent line infections and had a bowel length of 45 cm without ICV before her iLTx. Case 1 had 19 cm of bowel in the presence of ICV, case 2 had 40 cm of bowel in the absence of ICV, and although case 3 had 65 cm of bowel with ICV, he had multiple episodes of recurrent line infections before iLTx.
None of the patients with gastroschisis in the whole series (8/14) could be weaned to full oral feeding. Three patients had early mortality, 3 developed irreversible intestinal failure (cases 1–3) and 2 still require EN by nasogastric tube. Conflicting reports exist in the literature about the potential for children with gastroschisis and SBS developing intestinal failure to be weaned from PN. Most reports describe the bowel length as the determining factor in weaning from PN rather than dysmotility. Casey et al (5), in a 10-year review, could not find an association between gastroschisis as a cause and failure of intestinal rehabilitation. In another 10-year review, Nucci et al (6) showed that increased mean bowel length for patients with gastroschisis (44 vs 23 cm, P < 0.05) is a condition to predict the possibility of weaning from PN without transplantation, but they did not specifically comment on motility in gastroschisis. The dysmotility in gastroschisis or other SBS is difficult to evaluate and this point should be taken into consideration when deciding among the treatment options for children with gastroschisis and irreversible intestinal failure.
A review of the English-language literature, searching MEDLINE and EMBASE, for children who received iLTx for IFALD using suitable combinations of search terms (Table 1) (1,7–15), reveals that >96 patients had undergone iLTx for IFALD secondary to SBS. They fall into 23 patients (8), 6 case series (range 3–14 cases) (1,9–13), and 2 case reports (14,15). Patients in the present series are reported in the study of Del-Olio et al (1). Few other case reports have been published and have been included in case series (4,16–18). Barshes et al (7) published the results of 73 paediatric patients identified through the Organ Procurement and Transplant Network/United Network for Organ Sharing liver transplant database, which contains data for patients who underwent orthotopic liver transplantation in the United States between January 1988 and December 31, 2003. This publication includes all of the patients in the published series from the United States until 2003. The maximum length of follow-up in the retrieved publications is 5 years, except for 1 case report (15).
Analysing and comparing these reports is difficult because of the small numbers of cases and differences in follow-up times, which range between as short as 25 days (10) to 8 years (15). The largest reported case series (8) show a 5-year survival of 72%. Interpreting the published results must be done with extreme caution because the published series are incomparable because the patients are extremely heterogeneous in different series and even in the same series, patients are being managed differently in different centres in addition to different referral patterns and availability of donor organs in different countries. Furthermore, there is a wide variability in patients’ numbers, lengths of follow-up, statistics used, and endpoint recorded, thus making any general recommendations difficult.
It is evident from these publications that not all of the patients, who were predicted before the iLTx that they would reach autonomy of their small bowels, could actually gain this autonomy and many are still dependent on PN.
Botha et al (8) described their experience with 23 children, who received iLTx for IFALD, 82% of survivors were weaned from PN, with median time to weaning of 3 months (range 1–72 months). Six of these patients met all of their nutritional requirements from oral intake; the remaining 8 still require supplemental or full tube feeds. Linear growth for 4 years is maintained in their experience with median change in weight and height of 0.42 (range −1.65 to 3.05) and 0.57 (range −4.47 to 2.68), respectively, for 4 years. One child developed recurrent IFALD and was listed for combined liver and iSBTx and 4 children have required nontransplant gut reconstruction (2 Bianchi procedures, 1 gastric outlet reconstruction, and 1 jejunocolostomy for colon stricture). Three of them have been successfully weaned from PN.
Diamond et al (13) reported on their experience with 3 cases. One case died from IFALD, and the other 2 were alive at 41 and 58 months; both are still PN dependent, despite full tolerance of enteral feeds, for poor weight gain.
Nathan et al (9) compared their results with 7 patients who received iLTx for IFALD and 5 patients who received multivisceral transplantation for the same indication. Four of the iLTx group were alive at median follow-up of 25.1 months; one of them is PN dependent. Two patients were alive in the multivisceral transplantation group at median follow-up of 13 months and both were PN independent.
The above centres’ reports are comparable to our results in which some children still encounter complications long time after their transplants; many of these occur well beyond the fifth year. We recommend that more studies be undertaken to address the long-term nutritional rehabilitation for this group of patients beyond the 5-year limit.
The poor outcome reported in our case series (approximately one-third of the children were eventually weaned from PN) emphasises the importance of prevention of IFALD in children with SBS. It also raises questions about the ethical issues regarding allocation of scarce resources (liver grafts) to this group of children. The ethical issues are beyond the scope of this article.
iLTx should not become the standard of care in children with SBS and IFALD and should be considered with extreme caution. In most cases, end-stage liver failure is preventable and children should be evaluated in experienced centres long before this complication develops. In the rare circumstances that it needs to be exercised as an option, iLTx should be performed in experienced intestinal failure/rehabilitation centres with close links to intestinal transplant teams or centres.
We gratefully acknowledge Carla Lloyd (hepatology research project manager, Birmingham Children's Hospital [BCH], Birmingham, UK) for valuable contributions in the preparation of the manuscript. We acknowledge the help of the following in clinical management: Mr D. Mirza, Mr P. Muiesan, and Mr A.D. Mayer (transplant surgeons, Hepatobiliary Unit, Queen Elizabeth Hospital, Birmingham, UK); Mr Jean de Ville (paediatric transplant surgeon, 1998–2002, BCH); Mr Alastair Millar (paediatric transplant surgeon, 2002–2006, BCH); Professor D.A. Kelly, Dr I. Van Mourik, Dr P. McKiernan, and Dr S.V. Beath (consultant paediatric hepatologists, BCH); Dr D. Dell-Olio, Dr Sue Protheroe (consultant paediatric gastroenterologist, BCH); nutritional care nursing team; specialist nurses; and the nursing staff on wards 8 and 6 at BCH.
1. Dell-Olio D, Beath SV, de Ville de Goyet J, et al. Isolated liver transplant in infants with short bowel syndrome: insights into outcomes and prognostic factors. J Pediatr Gastroenterol Nutr
2. Mazariegos GV, Soltys KA, Bond GJ, et al. Isolated liver transplantation in infants with short gut syndrome: is less better? Liver Transpl
3. Spagnuolo MI, Iorio R, Vajro P. Isolated liver transplantation in children with intestinal failure-associated liver disease: a still-debated matter. J Pediatr Gastroenterol Nutr
4. Hassan K, Beath S, McKiernan P, et al. Difficult management choices for infants with short bowel syndrome and liver failure. J Pediatr Gastroenterol Nutr
5. Casey L, Lee KH, Rosychuk R, et al. 10-year review of pediatric intestinal failure: clinical factors associated with outcome. Nutr Clin Pract
6. Nucci A, Burns RC, Armah T, et al. Interdisciplinary management of pediatric intestinal failure: a 10-year review of rehabilitation and transplantation. J Gastrointest Surg
7. Barshes NR, Carter BA, Karpen SJ, et al. Isolated orthotopic liver transplantation for parenteral nutrition associated liver injury. J Parenter Enteral Nutr
8. Botha JF, Grant WJ, Torres C, et al. Isolated liver transplantation in infants with end-stage liver disease due to short bowel syndrome. Liver Transpl
9. Nathan JD, Rudolph JA, Kocoshis SA, et al. Isolated liver and multivisceral transplantation for total parenteral nutrition-related end-stage liver disease. J Pediatr Surg
10. Ventura K, Rekhtman Y, Jan D, et al. Isolated liver transplantation for decompensated end-stage liver disease in children with intestinal failure. Transplant Proc
11. Lawrence JP, Dunn SP, Billmire DF, et al. Isolated liver transplantation for liver failure in patients with short bowel syndrome. J Pediatr Surg
12. Muiesan P, Dhawan A, Novelli M, et al. Isolated liver transplant and sequential small bowel transplantation for intestinal failure and related liver disease in children. Transplantation
13. Diamond IR, Wales PW, Grant DR, et al. Isolated liver transplantation in pediatric short bowel syndrome: is there a role? J Pediatr Surg
14. Lezo A, Gennari F, Santini B, et al. Isolated liver transplantation in an infant with ultrashort gut. Transplant Proc
15. Pelissier-Delour L, Mention K, Michaud L, et al. Favorable long-term outcome after isolated liver transplantation in a child with short bowel syndrome. J Pediatr Gastroenterol Nutr
16. Gottrand F, Michaud L, Bonnevalle M, et al. Favorable nutritional outcome after isolated liver transplantation for liver failure in a child with short bowel syndrome. Transplantation
17. Horslen SP, Sudan DL, Iyer KR, et al. Isolated liver transplantation in infants with end-stage liver disease associated with short bowel syndrome. Ann Surg
18. Horslen SP, Kaufman SS, Sudan DL, et al. Isolated liver transplantation in infants with total parenteral nutrition-associated end-stage liver disease. Transplant Proc