See “Pediatric Intestine Transplantation: Are We Ready for the Prime Time?" by Avitzur and Silva Sandy on page 584.
What Is Known/What Is New
What Is Known
- Intestinal transplantation is indicated for life-threatening complications of permanent intestinal failure.
- Five-year graft survival at high-activity pediatric centers approximates 70%.
- Factors predicting successful outcomes are inconsistently defined.
What Is New
- Outcomes of intestinal transplantation for anatomic intestinal failure were superior in this study, 5-year and 10-year graft survival equaling 79% and 76% compared with 45% and 38% for functional intestinal failure of all causes, respectively.
- Post-transplant lymphoproliferative disorders and graft-versus-host disease were more common after transplantation for functional intestinal failure.
- Outcomes of intestinal transplantation alone were comparable to outcomes after combined liver and intestinal transplantation.
Intestinal transplantation (ITx) has been an option for patients with life-threatening complications of intestinal failure (IF) for about 25 years. Although initial results were often poor, refinements in surgical technique and postoperative management have improved early outcomes (1). In order to characterize current long-term graft survival after ITx in infants and children and potential explanatory factors, we compared cohorts of ITx recipients with full enteral autonomy for at least 5 and up to 10 years with contemporaneous groups who failed to achieve these milestones.
Outcomes of patients who underwent ITx ages18 years or younger from inception of the program at our center in 2003 until December 31, 2013 were reviewed.
Indications for Intestinal Transplantation, Variant Operations, and Postoperative Management
Isolated ITx was performed for irreversible IF in combination with either mild IF-associated liver disease (IFALD) expected to progress or with progressive central vein loss expected to compromise indefinite parenteral nutrition (PN) (2). En bloc liver-intestine-pancreas transplant was performed for advanced IFALD with portal hypertension (3). Multivisceral transplant (MVTx) including stomach, small intestine, liver, and pancreas plus splenectomy was performed for the same indication as en bloc ITx with an unsalvageable recipient foregut in addition to mid-gut.
Management was as previously described (2). In summary, immunosuppression was based on tacrolimus and supplemented with sirolimus (SIR) or mycophenolate mofetil (MMF); SIR was preferred because of perceived superior efficacy and bioavailability. Prednisone or prednisolone, 0.1 mg · kg−1 · day−1, was continued indefinitely. The most common reasons to discontinue or replace SIR or MMF with the other were high frequency of community-acquired infections and drug toxicity. After ITx, oral diet was supplemented by tube feeding, usually via gastrostomy, as necessary to maintain age-appropriate weight gain. Long-term monitoring included annual surveillance pan-endoscopy with biopsy. Acute graft rejection, chronic graft rejection, post-transplant lymphoproliferative disorder (PTLD), and graft-versus-host disease (GVHD) were evaluated by standard criteria (4,5), whereas acute humoral rejection was not routinely assessed. Full graft function was defined as maintenance of appropriate nutritional status without routine use of intravenous fluid of any type, uninterrupted resumption of which indicated graft failure.
Evolution of Clinical Practice During the Study Period
Relevant changes included: (1) beginning in 2008, screening of recipients for HLA antibodies by single-antigen assay, virtual crossmatching to detect pre-formed, donor-specific anti-HLA antibodies (DSA) (6), and protocol use of rabbit anti-thymocyte globulin (r-ATG) for induction immunosuppression in place of basiliximab for positive crossmatch (7); beginning in 2009, routine inclusion of graft ileocecal valve and graft ascending/transverse colon (8); beginning in 2012, treatment with high-dose intravenous immune globulin (IVIG), anti-CD20 monoclonal antibody rituximab, and plasmapheresis in HLA antibody-sensitized recipients immediately before and after transplant based on a panel reactive anti-HLA antibody level (PRA) >20% and protocol monitoring for appearance of de novo DSA after ITx followed by treatment with IVIG when present (7).
Primary study end-points were 5-year and 10-year survival with a functioning intestinal graft. The secondary end-points were identification of factors associated with 5 and 10-year graft survival. Potential explanatory variables related to patient demography, perioperative events, and postoperative complications were tested as detailed in Results. Patients were classified by original etiology of IF; anatomic for short bowel syndrome (SBS) mainly because of necrotizing enterocolitis and various congenital malformations, and functional IF because of congenital secretory diarrhea, mainly tufting enteropathy and microvillous inclusion disease, and primary motility disorders including near-total to total intestinal Hirschsprung disease. Comparisons of continuous variables utilized the Wilcoxon-Mann-Whitney and Kruskal-Wallis tests, and proportions were compared using Fisher's exact test; central tendencies were expressed as medians with interquartile ranges (first quartile, third quartile). Cox regression modeling was used to estimate hazards of graft loss and relative importance of potential outcome determinants expressed as hazard ratios (HR) with 95% confidence intervals (CI). Significance was taken as P < 0.05. Statistical calculations were performed using R 3.5.3 (The R Foundation; www.r-project.org). This study was approved by the Institutional Review Board of Georgetown University.
Intestinal Transplantation Recipients
A total of 88 transplants were performed in 86 patients at a median age of 1.7 [1.1–4.7] years, including 63 with anatomic IF, 22 with functional IF, and 1 with a benign but locally invasive tumor compromising a functioning gastrointestinal tract. Types and numbers of individual IF diagnoses are as indicated in Table 1. The annual percentage of liver-inclusive grafts fell over time, the median through 2008 equaling 75% [63, 89%] and from 2009 to 2013 equaling 38% [35, 53%] (P = 0.0159). In contrast, there was no change in either the yearly percentage of ITx indicated for complications of functional IF (P = 0.6538) or in the median age of ITx recipients (P = 0.7147).
Five-year Graft Survival and Interval Associations with Graft Loss and Death
Of the 86 patients, 61 (71%) retained a fully functioning graft for at least 5 years after ITx including the second graft of a patient whose first graft failed within 5 years because of chronic rejection. The 26 patients who experienced a graft loss and/or death within 5 years after ITx included the above-referenced patient; each graft was included in survival analysis. Two other patients experienced perioperative graft thrombosis prompting graft removal immediately thereafter. Of these 2, 1 patient received a second graft that failed within 5 years. The 2 transplants with perioperative thrombosis were censored, leaving 25 patients having a within-5-year graft loss for analysis. Of the 25, severe graft dysfunction per se was responsible for graft loss in 10 patients; 4 grafts were acutely or chronically rejected and 6 were primarily nonfunctional. Four of the 10 survived the graft failure, enabling re-transplantation in all before December 31, 2018. The remaining 15 patients experienced graft failure concomitant to death because of complications not directly related to graft function including GVHD and immune-mediated blood dyscrasias, EBV-associated PTLD (all diffuse large B-cell lymphoma [DLBCL]), and respiratory failure, typically infective. In addition to 2 patients dying as a direct result of PTLD, 4 patients developed DLBCL and recovered after treatment only to experience graft loss within 5 years after ITx because of chronic rejection or GVHD. Median interval from transplant either to graft loss from primary failure or from death with a functioning graft was 0.7 [0.1, 1.6] years and did not differ between these 2 categories (P = 0.3969).
Table 1 lists explanatory factors used to identify differences between patients who did and did not achieve 5-year graft survival. As shown in Table 1, the single recipient variable associated with 5-year graft survival was original cause of IF; functional IF was strongly associated with inferior outcome. There were no differences in graft survival among various diagnoses within either the anatomic (P = 0.2282) or functional IF sub-groups (P = 0.1619). Although 5-year graft survival was unaffected by either a recipient or donor NOD2 mutation, considerably more ITx recipients than donors had a NOD2 mutation, 24.7% versus 6.4% (P = 0.0019). There was no difference in incidence of recipient NOD2 mutation in patients with historical anatomic versus functional IF (P = 0.7704), and recipient NOD2 mutation was not associated with either graft rejection (P = 0.5949) or GVHD (P = 0.5674). The sole perioperative variable associated with reduced 5-year graft survival was high donor body weight relative to recipient body weight expressed as the donor-recipient weight ratio (DRWR). A receiver-operating characteristic curve indicated an optimal DRWR cut-off point of 0.71 that produced positive and negative predictive values for 5-year graft loss of 43.1% and 90.9%, respectively (P = 0.001).
After ITx, variables associated with within-5-year graft loss included GVHD, PTLD, and de novo appearance of DSA; DSA already formed at ITx were not associated with graft loss. Despite their association with graft loss, de novo DSA had no association with graft rejection (P = 0.2200), and, as noted in Table 1, rejection was not associated with inferior 5-year outcome; just 15% of intestinal graft rejection events resulted in graft loss and/or death. Only 3 of the 55 patients with a liver graft had a documented rejection of this organ alone, rejection being confined to the intestine in all other affected patients. As depicted in Table 2, Cox regression analysis confirmed that PTLD, GVHD, de novo DSA, high DRWR, and a functional cause of original IF were hazards for within 5-year graft loss and also indicated that splenectomy accompanying MVTx and parental consanguinity were also risks for graft loss that were not established by their comparative incidences in 5-year graft survivors and nonsurvivors (Table 1). No variables were significant in the multivariate Cox model.
Because of the unambiguous disparity in 5-year outcomes between patients receiving an ITx for complications of anatomic and functional IF, these groups were directly compared in search of explanations for the difference. As demonstrated in Table 3, a family history of consanguinity, a high DRWR, MVTx with splenectomy, and both PTLD and GVHD were present much more frequently in patients with histories of functional IF, the increases being enough to reduce graft survival within the entire patient cohort of 86, particularly as demonstrated by Cox modeling. Conversely, several factors more typical of ITx for functional IF were not associated with altered 5-year graft survival including older age at ITx, inclusion of a graft colon, and early use of SIR. Although de novo DSA were associated with within-5-year graft loss for the entire cohort, DSA were not more commonly observed in the functional IF sub-group.
Graft Survival Beyond 5 Years and Interval Associations with Graft Loss and Death
Of the 46 patients receiving an ITx before December 31, 2008, 33 (72%) retained a functioning graft after 5 years, declining to 30 patients (65%) after 10 years. All 3 graft losses in this group resulted from severe (grade 3) and ultimately fatal graft rejection; all 3 of these patients had had a history of functional IF, and none produced DSA within 4 months of initial rejection diagnosis. In contrast, only 1 of the 30 patients with 10-year graft survival experienced severe rejection between 5 and 10 years after ITx (p = 0.0007). As of December 31, 2018, 3 additional patients died more than 5 years after ITx including 1 due to complications of chronic rejection and graft failure in the context of recovered PTLD (DLBCL), 1 due to secondary malignancy after recovery from PTLD, and 1 due to infective complications of autoimmune hemolytic anemia.
In comparison with 5-year graft survival, Cox modeling of survival 10 years after ITx demonstrated continued significance of functional IF (HR 4.12 [CI 1.96--8.69], P < 0.001), consanguinity (HR 2.96 [CI 1.02--8.54], P = 0.045), high DRWR (HR 3.4 [CI 1.26--9.2], P = 0.016), GVHD (HR 3.83 [CI 1.31--11.24], P = 0.014), and PTLD (HR 3.97 [CI 1.59--9.9], P = 0.003). In the multivariate Cox model, only functional IF was unambiguously significant at 10 years, HR 3.38 [CI 1.29--8.82], (P = 0.013). Kaplan-Meier curves summarizing graft survival over 10 years according to pre-ITx history of anatomic versus functional IF are shown in Figure 1.
This review indicates that 71% of pediatric patients who received an ITx at our center between December 1, 2003 and December 31, 2013 maintained full intestinal graft function for no less than 5 years. This result approximates 5-year graft survival data recently reported by the international Intestinal Transplant Registry and indicates substantial improvement compared with the approximately 50% 5-year graft survival reported as recently as 5 to 10 years ago (1). To better gauge utility of ITx for pediatric patients, we also assessed 10-year graft outcomes; 10-year graft survival fell only to 65%, again considerably super-ior to previously reported outcomes.
Outcomes after ITx differ between adults and children (7). In this pediatric study, most remarkable was the difference in outcome between ITx for complicated anatomic IF, that is, SBS, who represented three-quarters of recipients, and the minority with functional IF including both motility disorders and congenital diarrhea. With a history of SBS, 5- and 10-year graft survivals were 79% and 76%, respectively, approaching survival of patients able to remain on PN indefinitely (9,10). In sharp contrast, for functional IF, 5- and 10-year graft survivals were only 45% and 38%, respectively, that are similar to historical outcomes after ITx irrespective of IF etiology (11). In fact, history of functional IF was the only significant predictor variable based on multivariate modeling at 10 years, recognizing the limitations of multivariate modeling with relatively small event numbers and total sample size. These results contrast with Intestinal Transplant Registry data (1) that have indicated superior outcomes in functional IF, although, similar to our findings, Ramisch et al (12) also recently found superior survival of patients who received an ITx for anatomic IF compared with other diagnoses.
We hypothesize that superior post-ITx survival with a background of anatomic IF is likely to be related to the paucity of co-associated extra-intestinal disorders in most pediatric patients with SBS, increasing probabilities that successful ITx will correct most pre-ITx morbidities. In contrast, severe functional gastrointestinal diseases are more likely to be components of inherited multisystem disorders not fully correctable with ITx alone; immunological abnormalities of varying severity may coexist (13–15). The increased incidence of consanguinity in our functional IF cohort is consistent with this theme. In this study, overall incidences of GVHD and PTLD, 6% and 9%, respectively, were similar to previous reports (5,16); however, incidence of GVHD was 11 times greater and PTLD, 8 times greater, in the functional compared with anatomic IF sub-group. As GVHD and PTLD are expressions of recipient immunological incompetence, their greatly increased frequency with functional IF implies that intrinsic immune dysregulation, however subtle, may be more pervasive in this group than previously appreciated and severe enough to undermine ITx. MVTx was more commonly needed for patients with motility disorders, and associated splenectomy may have amplified existing risks of dysfunctional recipient immune responses, including those eventuating in GVHD (5). Similarly, disproportionately greater use of SIR that increased cumulative IS could have contributed to reduced graft survival in the functional sub-group to a degree that precluded a survival benefit from more IS for the entire patient cohort. In the future, genetic testing may facilitate recognition of increased immunological and related risks to allow tailoring of immunosuppression to IF diagnosis.
This study again highlights the potentially severe consequences of GVHD and PTLD after solid organ transplantation. Early mortality after GVHD is high in the absence of any universally accepted treatment paradigm (5), although JAK inhibitor therapy may prove beneficial in the future (17). Similarly, successful early control of PTLD is qualified by the risk of late graft loss because of rejection (18), as was again found in this study. We did find evidence of increased graft loss associated with the appearance of DSA de novo after ITx that could not be directly attributed to graft rejection, although changes in screening for, and treatment of, DSA during the study increase the risk of incorrect conclusions. In contrast with de novo DSA, high levels of HLA antibodies including DSA before ITx had no impact on graft survival but the degree to which implementation of the desensitizing strategy summarized in the Methods section reduced risk of graft loss cannot be known with certainty. In the context of studies that have (19,20) and have not (21,22) demonstrated increased rejection and reduced graft survival in association with DSA, our experience implies that impact of DSA in ITx is complex and depends on other unknown factors (23,24).
The association of an original functional IF diagnosis with high DRWR in this study implies that use of larger grafts was primarily an accommodation to the large intra-abdominal volumes typical of this group. Periodic use of large grafts in small infants with SBS in advanced liver failure with severely reduced abdominal domain, particularly early in the experience, may have also contributed to generally inferior outcomes with higher DRWR. We, however, found no overall difference in survival of liver-inclusive versus liver-noninclusive intestinal grafts, as also recently reported by Norsa et al (10). These results contrast with historical ITx experience that has indicated superior long-term survival of liver-inclusive grafts (1) presumably owing to the propensity for the liver to promote immunological tolerance to any concurrently transplanted organ (25). In contrast with previous investigations by ourselves (26) and others (27), we found no increase in graft rejection or reduction in survival in the presence of recipient NOD2 polymorphisms that, reminiscent of inflammatory bowel disease, are more commonly encountered in SBS than in the general population (28); their similarly increased frequency with functional IF implies that these mutations adversely influence prognosis of a spectrum of intestinal disorders (29). The improved outcomes of ITx without liver inclusion and despite NOD polymorphisms demonstrated in this study in comparison with historical experience may be related to use of more potent IS incorporating SIR or MMF (11,30). However, absence of direct evidence that improved graft survival resulted from less rejection points to the trade-off of increased catastrophic infection and malignancy risks for reduced rejection under intensified IS (31). Although risk of graft rejection may fall over time, the potential for graft loss and death remains high should it occur (11).
Over the last 2 decades, outcomes in pediatric IF have improved under extended rehabilitation, particularly IF because of SBS (32), resulting in reduced ITx transplant activity; during the peak transplant year of 2007, 111 ITx were performed in patients up to 18 years old in the USA, dropping to just 37 by 2018 (based on OPTN data as of December 16, 2019). Many patients in this study with IFALD most probably would not have required ITx had they been managed according to current standards (33). Nonetheless, some patients will continue to experience life-threatening complications on PN that appropriately indicate ITx (34). As illustrated by this report, outcomes of ITx for complicated anatomic IF, still the most common indication (1), have also improved, enhancing the viability of ITx as a therapeutic option when long-term survival under extended PN is unlikely.
The original cause of IF strongly influences outcomes in pediatric ITx. With SBS, graft survival for 10 years was achieved in 76% of patients in this study whereas graft survival was lower by half in patients with IF due to motility disorders and secretory diarrhea. Inferior results in patients with functional IF may be related to their greater susceptibility to complications of immunosuppressive therapy.
We thank Mr. Sameer Desale, Senior Statistician, MedStar Health Research Institute, for statistical analysis.
1. Raghu VK, Beaumont JL, Everly MJ, et al. Pediatric intestinal transplantation: analysis of the intestinal transplant registry. Pediatr Transplant
2. Fishbein TM. Intestinal transplantation. N Engl J Med
3. Kaufman SS, Pehlivanova M, Fennelly EM, et al. Predicting liver failure in parenteral nutrition-dependent short bowel syndrome
of infancy. J Pediatr
4. Remotti H, Subramanian S, Martinez M, et al. Small-bowel allograft biopsies in the management of small-intestinal and multivisceral transplant recipients: histopathologic review and clinical correlations. Arch Pathol Lab Med
5. Ganoza A, Mazariegos GV, Khanna A. Current status of graft-versus-host disease after intestinal transplantation. Curr Opin Organ Transplant
6. Hawksworth JS, Rosen-Bronson S, Island E, et al. Successful isolated intestinal transplantation in sensitized recipients with the use of virtual crossmatching. Am J Transplant
2012; 12 Suppl 4:S33–42.
7. Elsabbagh AM, Hawksworth J, Khan KM, et al. Long-term survival in visceral transplant recipients in the new era: a single-center experience. Am J Transplant
8. Matsumoto CS, Kaufman SS, Fishbein TM. Inclusion of the colon in intestinal transplantation. Curr Opin Organ Transplant
9. Merras-Salmio L, Pakarinen MP. Refined multidisciplinary protocol-based approach to short bowel syndrome
improves outcomes. J Pediatr Gastroenterol Nutr
10. Norsa L, Artru S, Lambe C, et al. Long term outcomes of intestinal rehabilitation in children with neonatal very short bowel syndrome
: Parenteral nutrition or intestinal transplantation. Clin Nutr
11. Lacaille F, Irtan S, Dupic L, et al. Twenty-eight years of intestinal transplantation in Paris: experience of the oldest European center. Transpl Int
12. Ramisch D, Rumbo C, Echevarria C, et al. Long-term outcomes of intestinal and multivisceral transplantation at a single center in Argentina. Transplant Proc
13. Vakkilainen S, Taskinen M, Klemetti P, et al. A 30-year prospective follow-up study reveals risk factors for early death in cartilage-hair hypoplasia. Front Immunol
14. Vély F, Barlogis V, Marinier E, et al. Combined immunodeficiency in patients with trichohepatoenteric syndrome. Front Immunol
15. El-Daher MT, Lemale J, Bruneau J, et al. Chronic intestinal pseudo-obstruction and lymphoproliferative syndrome as a novel phenotype associated with tetratricopeptide repeat domain 7A deficiency. Front Immunol
16. Stanley K, Friehling E, Ranganathan S, et al. Post-transplant lymphoproliferative disorder in pediatric intestinal transplant recipients: a literature review. Pediatr Transplant
17. von Bubnoff N, Ihorst G, Grishina O, et al. Ruxolitinib in GvHD (RIG) study: a multicenter, randomized phase 2 trial to determine the response rate of Ruxolitinib and best available treatment (BAT) versus BAT in steroid-refractory acute graft-versus-host disease (aGvHD) (NCT02396628). BMC Cancer
18. Wozniak LJ, Mauer TL, Venick RS, et al. Clinical characteristics and outcomes of PTLD following intestinal transplantation. Clin Transplant
19. Cheng EY, Everly MJ, Kaneku H, et al. Prevalence and clinical impact of donor-specific alloantibody among intestinal transplant recipients. Transplantation
20. Abu-Elmagd KM, Wu G, Costa G, et al. Preformed and de novo donor specific antibodies in visceral transplantation: long-term outcome with special reference to the liver. Am J Transplant
21. Kubal C, Mangus R, Saxena R, et al. Prospective monitoring of donor-specific anti-HLA antibodies after intestine/multivisceral transplantation: significance of de novo antibodies. Transplantation
22. Petit LM, Rabant M, Canioni D, et al. Impacts of donor-specific anti-HLA antibodies and antibody-mediated rejection on outcomes after intestinal transplantation in children. Pediatr Transplant
23. Feng S, Demetris AJ, Spain KM, et al. Five-year histological and serological follow-up of operationally tolerant pediatric liver transplant recipients enrolled in WISP-R. Hepatology
24. Schmitz R, Fitch ZW, Schroder PM, et al. B cells in transplant tolerance and rejection: friends or foes? Transpl Int
25. Calne RY. Immunological tolerance--the liver effect. Immunol Rev
26. Fishbein T, Novitskiy G, Mishra L, et al. NOD2-expressing bone marrow-derived cells appear to regulate epithelial innate immunity of the transplanted human small intestine. Gut
27. Ningappa M, Higgs BW, Weeks DE, et al. NOD2 gene polymorphism rs2066844 associates with need for combined liver-intestine transplantation in children with short-gut syndrome. Am J Gastroenterol
28. Schäffler H, Schneider N, Hsieh CJ, et al. NOD2 mutations are associated with the development of intestinal failure in the absence of Crohn's disease. Clin Nutr
29. Tronstad RR, Polushina T, Brattbakk HR, et al. Genetic and transcriptional analysis of inflammatory bowel disease-associated pathways in patients with GUCY2C-linked familial diarrhea. Scand J Gastroenterol
30. Gupta P, Kaufman S, Fishbein TM. Sirolimus for solid organ transplantation in children. Pediatr Transplant
31. Amin A, Farmer DG. Current outcomes after pediatric and adult intestinal transplantation. Curr Opin Organ Transplant
32. Merritt RJ, Cohran V, Raphael BP, et al. Nutrition Committee of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. Intestinal rehabilitation programs in the management of pediatric intestinal failure and short bowel syndrome
. J Pediatr Gastroenterol Nutr
33. Norsa L, Nicastro E, Di Giorgio A, et al. Prevention and treatment of intestinal failure-associated liver disease in children. Nutrients
34. Kaufman SS, Avitzur Y, Beath SV, et al. New insights into the indications for intestinal transplantation: consensus in the year 2019. Transplantation