New Insights Into the Indications for Intestinal Transplantation: Consensus in the Year 2019

Abstract

In 2001, a Statement was published that described indications for intestinal transplantation in patients with intestinal failure expected to require parenteral nutrition indefinitely. Since 2001, advances in the management of intestinal failure including transplantation and patient survival, both on extended parenteral nutrition and after transplantation, have improved, leading to a reduction in the number of intestinal transplants worldwide from a peak of 270 per year in 2008 to 149 per year in 2017. These changes suggest that the original 2001 Statement requires reassessment. All patients with permanent intestinal failure should be managed by dedicated multidisciplinary intestinal rehabilitation teams. Under care of these teams, patients should be considered for intestinal transplantation in the event of progressive intestinal failure–associated liver disease, progressive loss of central vein access, and repeated life-threatening central venous catheter–associated infections requiring critical care. Additional indications for transplantation include large desmoid tumors and other intra-abdominal tumors with reasonable expectation of posttransplant cure, extensive mesenteric vein thrombosis and intestinal infarction, total intestinal aganglionosis, and nonrecoverable congenital secretory diarrhea. Quality of life typically improves after successful intestinal transplantation and may support the decision to proceed with transplantation when other indications are present. However, the requirement for life-long immunosuppression and its associated side effects preclude intestinal transplantation if motivated only by an expectation of improved quality of life. Increasing experience with intestinal transplantation and critical appraisal of transplant outcomes including graft survival and patient quality of life together with potential advances in immunosuppression can be expected to influence transplant practices in the future.

Intestinal transplantation (ITx) has been an accepted therapy for over 30 years and was originally offered to patients with intestinal failure (IF) in whom continuing treatment with intravenous nutrition had become impossible because of end-stage loss of venous access or end-stage liver disease. Patients were often referred late and in poor condition; not surprisingly, early results were poor. Unlike end-stage renal failure, wherein kidney transplantation has a significant survival advantage over hemodialysis,¹ a controlled trial of ITx versus continuing home parenteral nutrition (HPN) in management of chronic IF has not been performed. Reasons for the lack of a controlled trial in IF to date are many, most important of which is the relatively small size of the patient population: The Intestinal Transplant Registry lists the total number of ITx performed worldwide with and without a liver graft as 4103 since the Registry was started in 1985.² In comparison, the renal database cited by Wolfe et al¹ included 228 552 patients, of whom 46 164 received a renal allograft. Despite the challenges, including the relatively small number of patients with IF of widely diverse ages and causes for their disease, the international community of physicians, surgeons, and allied health professionals has broadly responded to achieve parallel advances in intestinal rehabilitation and ITx.³ Fewer individuals now present with end-stage liver disease or in crisis due to loss of secure venous access, which provides time to evaluate the full range of options for these patients.

FEATURES OF THE IF POPULATION

Key questions affecting the potential ITx recipient include how IF is defined and relevance of the definition to that patient and the very wide range of causes and morbidity risk factors that determine his or her prognosis. The term IF has been defined as the reduction of gut function below the minimum necessary for the absorption of macronutrients or water and electrolytes, such that intravenous supplementation is required to maintain health or growth.⁴ This simple and practical definition is applicable to all age groups⁵ but does not provide guidance as to the severity or likely duration of an individual patient’s IF. The most common cause of IF is anatomic short bowel syndrome. In pediatric patients, short bowel syndrome is usually due to complications of prematurity, viz. necrotizing enterocolitis, and congenital intestinal malformations. In older individuals, short bowel syndrome can result from devastating intestinal injuries of mechanical, vascular, inflammatory, and traumatic origin that may develop either abruptly or evolve over long periods. A smaller number of persons have IF not due to critical loss of intestinal length but to protracted and severe intestinal dysfunction. These functional forms of IF usually fall into 1 of 2 categories: (1) globally defective intestinal mucosal function that is typically congenital and hereditary and (2) severely impaired alimentary tract motility, the neuropathic and myopathic intestinal pseudo-obstruction syndromes.

HPN has been the primary treatment for IF since the early 1980s. ITx became an additional option for persons with IF and life-threatening complications of their disease or severely impaired quality of life (QoL) since the mid-1990s.⁶ In 2001, the American Society of Transplantation endorsed a statement⁷ that described a set of indications for ITx with an emphasis on children who were receiving the majority of transplants at that time largely because of the particular vulnerability of young children to rapidly progressive liver disease caused by physiological immaturity and sepsis. Thus, one of the most common indications for transplantation was advanced and progressive IF-associated cholestatic liver disease (total serum bilirubin concentration of up to 6 mg/dL [100 μmol/L]) combined with portal hypertension and impaired hepatic synthetic dysfunction. Additionally, loss of 50% of available central vein access sites for extended catheter placement for parenteral nutrition delivery; ultrashort bowel syndrome per se; congenital intestinal diarrhea syndromes; recurrent life-threatening central line–associated bloodstream infections, particularly fungal; and frequent episodes of severe dehydration were also included.

DEVELOPMENTS THAT HAVE LED TO REEVALUATION OF CURRENT GUIDELINES

Since publication of the 2001 Position Paper,⁷ there have been many changes in the management of IF that have improved prognosis of the disease, especially for short bowel syndrome. Among these, multidisciplinary teams (gastroenterologists, surgeons, dieticians, social workers, and pharmacists) have proliferated to manage this complex disorder and have been instrumental in providing more integrated care, enabling many of the complications encountered through the 1990s to be reversed or at least ameliorated.⁸ For example, postresection intestinal reconstruction procedures have been found to improve absorptive function and challenged traditional concepts concerning relationships between immediate postresection intestinal length and parenteral nutrition (PN) dependence.^9,10 Newer intravenous lipid emulsions and modifications in their use (see below) have improved tolerance of PN, and recently available intestinal growth factors, specifically teduglutide, can increase intestinal adaptation beyond that achievable with enteral nutrition alone.^11–14 Furthermore, the natural history of IF in different patient populations has been clarified by better delineation of groups at high risk for severe morbidity and mortality while receiving extended PN therapy.^15–17

In addition to improvements in basic management of IF including HPN, increasing experience with ITx has provided insights into its benefits beyond the simple abrogation of death due to complications of IF. These insights have included (1) delineation of long-term survival after transplant that, like IF in general, has improved over time¹⁸; (2) magnitude of improved QoL after successful transplantation, that is, status without a central venous catheter and PN, compared with life quality with continued PN^19,20; and (3) cost of transplantation over time, also in comparison with continued PN.^21,22 The dataset regarding QoL and healthcare costs is still quite small but has expanded in the past 5 years. Despite increasing experience and improving outcomes, ITx has decreased worldwide over the past decade, from 270 during the peak year of 2008 to 149 in 2017.²

In recognition of the changes in practice and outcomes in IF and ITx since publication of the original 2001 Position Paper, a working group of the XIV International Small Bowel Transplant Symposium convening at Buenos Aires, Argentina in June 2015 was formed to reevaluate indications for ITx. A goal was to produce a single statement with equal emphasis on adult and pediatric patients in recognition of the increasing proportion of transplants performed in adults who now represent the majority of transplant recipients²³ and similarities in transplant practice irrespective of patient age. As with the original statement, the intended audience for this document remains all healthcare professionals who are actively involved in the care of patients with IF as well as other interested healthcare professionals and members of the lay community. The authors of this Statement acknowledge that the quality of care and outcome data for IF and transplantation will evolve over time; thus, current indications for ITx should be considered as tentative and subject to future revision as has been true for the original 2001 Position Paper.

QOL CONSIDERATIONS

QoL is defined by the World Health Organization as a state of complete physical, mental, and social wellbeing and not merely the absence of disease and infirmity²⁴ and, as such, is a subjective mindset belonging to the patient. Physicians are often surprised to learn that a wheelchair user with pressure sores may declare that his or her QoL is good, whereas an apparently able-bodied individual with comparatively lesser health impediments is made miserable by the condition. The evaluation of QoL is a relatively young field with a key philosophical concept at its heart: the autonomy of patients entitles them (or their advocates) to make treatment decisions independently.²⁵ In the field of IF, it is recognized that there are wide individual variations in QoL and that QoL research needs methodological improvement, with adequate QoL conceptualization, appropriate measurements, and implementation of prospective multicenter studies.²⁶ Tables 1 and 2 illustrate several studies and range of tools to measure QoL that have been deployed in the patients on HPN and after ITx, respectively.

TABLE 1.

Quality of life studies in home parenteral nutrition

Authors and setting	Type of study	QoL tool used	Key outcomes
Tran et al,27 Pediatric	Single-center, cross-sectional, observational study; 14/17 families recruited.	Baxter questionnaire adapted for children.38	QoL impaired regarding items of daily living eating, dressing, washing, mobility—but was not affected in respect of school attendance, general fatigue, pain, and body image. QoL not affected by an enterostomy.
Gottrand et al,28 Pediatric	Multicenter, cross-sectional observational; N = 72 (27 = age <3 y completed by parents and doctors). French reference population 491 primary school children and 268 healthy adolescents.	Validated French questionnaires with 5-point scale denoting level of disturbance by indicating an expressive face.	Parents of infants with IF recorded lower scores than parents of healthy infants, and doctors scored lower than the parents of children with IF (P < 0.05). The QoL scores for children 3–11 y did not differ significantly from healthy pop; the QoL scores for adolescents with IF were higher than reference pop (P < 0.025).
Baxter et al,29 Adult	Multicenter cohort study 699 respondents on HPN.	HPN-QoL as per Baxter questionnaire.38 Internal consistency and item discriminant validity was satisfactory.	Multivariable linear regression showed QoL scores were significantly associated with: HPN duration (better in long term) underlying disease (better in Crohn’s disease and mesenteric ischemia) living status (worse if living alone) number of d of HPN infusion per wk (more d = worse QoL)
Bluthner et al,30 Adult	Single-center prospective observational study—90 respondents on HPN.	SF-36 Questionnaire.	SF-36 scores were improved with: oral intake early stoma closure.
Ballinger et al,31 Adult	Focus groups (100 members of the public) asked to match disease states with QoL.	8 clinical states from full health to PN required 7 d per wk matched against EuroQol-5 dimensions.	QoL was related to number of d per wk spent on PN.
Burden et al,32 Adult	Multicenter cohort study of 466 respondents on HPN from 9 clinics.	Parenteral Nutrition Impact Questionnaire.	The number of HPN infusions required per wk is inversely related to individuals’ needs-based QoL.
Samuel et al,33 Adult	Single-center, cross-sectional, observational study—196 patients recruited—94% of whom were employed before illness.	In house tool-structured questionnaire designed to probe employment both before and after starting HPN, intention to work, and social welfare status (benefits and pensions).	The number of d per wk on HPN and the desire of the patient to return to employment were significantly associated with returning to work.

HPN, home parenteral nutrition; IF, intestinal failure; pop, population; QoL, quality of life; SF-36, 36-Item Short Form Health Survey.

TABLE 2.

Quality of life studies in intestinal transplantation

Authors	Type of study	QoL tool used	Key outcomes
Sudan et al,34 Pediatric	Single-center, cross-sectional, observational study 20 children age over 5 y and at least 12 mo after ITx and 24 parents	CHQ (administered to adult caregivers and child)	Children with ITx reported QoL similar to their peer group  Parents reported lower levels than for healthy population in physical  function  health perception  emotional  time impact on the family.
Ngo et al,35 Pediatric	Single-center, cross-sectional, observational study 24 patients and their caregivers; QoL in ITx compared with normal population	CHQ, Pediatric quality of life (PedsQL4.0)	Patient responses  CHQ = same as normal population  PedsQL4 = lower than normal population for school functioning and psychosocial health summary score. Parent responses  CHQ = lower than normal population for physical health and social function  PedsQL4.0 = lower than normal population for physical health and social function.
Andres et al,36 Pediatric	Single-center, cross-sectional, cohort study 31/34 patients who underwent ITx from 1999 to 2012 and their caregivers	SF-36v2 for caregivers and patients >18 y TAPQOL for preschool children KINDL-R for children age 5–17 y CBI for caregivers	Patient responses  TAPQOL 78.2 ± 10.6 (n = 8), KINDL-R age 5–12 y = 77.7 ± 13.4 (n = 13)  KINDL-R age 13–17 y = 80.5 ± 12.4 (n = 7)  SF-36 = 82.2 ± 12.4 (n = 4)  CBI = stress in 52%. Parents  SF-36 was lower than general population worse perception of health than their children.
Abu-Elmagd et al,37 Pediatric and Adult	Single-center, cross-sectional, observational study Pediatric = 85 Adults = 92	Quality-of-life inventory 9-point Likert scale; 25 domains, selection of 31 clinically relevant questions	Psychological, emotional, and social QOL measures significantly improved after transplantation (P < 0.05). Global health post-ITx adversely affected in patients with  dysmotility (59%),  hypertension (37%),  osteoporosis (22%),  diabetes (11%). Higher incidence among adult recipients (P < 0.05).
Pither et al,19 Adult	Single-center, cross-sectional, observational study 16/26 patients	SF-36 VAS for performance KS ECOG scale	After ITx:  66% of patients had better VAS scores  75% of patients had better KS scores  Performance scores did not reach pre-TX levels  ECOG no overall change.
Pironi et al,20 Adult	Single-center, cross-sectional, observational study 33/33 HPN patients 18/22 post-ITx patients	HPN-QoL as per Baxter questionnaire3	ITx recipients showed a better score in following scales:  ability to holiday/travel (P < 0.001)  fatigue (P = 0.022)  intestinal symptoms (P < 0.001)  stoma management/bowel movements (P = 0.001)  global health status/quality of life (P = 0.012)  better score for ability to eat/drink (P = 0.070)  worse score for sleep pattern (P = 0.100).
Ceulemans et al,26 Review Adult	Adult Systematic review 9 studies identified	Structures data abstraction Standardized check list for methodological quality	Findings = suboptimal methodology:  retrospective (n =2) cross-sectional (n =7) study designs nonprobabilistic sampling with inadequate matching of TX subjects lack of operational definitions and variety in assessment instruments. Outcomes:  post-TX QoL improved vs pre-ITx (anxiety, sleep, social support, leisure)  post-TX QoL improved with longer follow-up (anxiety, impulsiveness/control)  QoL between ITx and HPN patients was similar for most domains yet TX patients excelled in energy, social functioning and travel ability.

CBI, Caregiver Burden Interview; CHQ, Child Health Questionnaire; ECOG, Eastern Cooperative Oncology Group; HPN, home parenteral nutrition; KINDL-R, Revidierter Kinder Lebensqualitätsfragebogen; KS, Karnofsky scale; QoL, quality of life; SF-36v2, 36-Item Short Form Version 2; TAPQOL, Preschool Children Quality of Life Questionnaire; VAS, visual analogue scale.

A QoL questionnaire developed for IF, validated by Baxter et al,³⁸ and adapted for children²⁷ has produced data from hundreds of patients on HPN.^20,29 In adults, QoL on HPN is generally good, especially in the absence of a stoma and if a moderate amount of oral intake is possible³⁰; however, several studies also report that living alone and the number of days spent on HPN per week adversely affect QoL.^29,31–33 In children, HPN adversely impacts activities of daily living, especially dressing, eating, washing, and mobility, although the effect on schooling is minimal.²⁷ It is important to note that children have a different social priority from their parents, viz. a desire to fit in with their peer group, that usually outweighs physical autonomy. This phenomenon has been shown in several studies, wherein parents/caregivers have reported a lower QoL than their children for example, a study in France showed that children aged 3–11 years on HPN had QoL scores that did not differ significantly from a healthy population; the QoL scores for adolescents with IF were, in fact, higher than for the reference population (P < 0.025).²⁸

Sudan et al³⁴ used the Child Health Questionnaire to compare QoL of pediatric ITx recipients (n = 21) at least 12 months beyond the operation with normal US school children (n = 232) and children with end-stage renal failure on dialysis (n = 18) and found that the ITx recipients assessed themselves as having QoL generally equal to a reference population, being superior in 1 domain; furthermore, QoL of the ITx recipients was significantly better than that of children on dialysis with respect to the mental health domain; parent responses in this study again suggested greater physical limitation in their transplanted children and a negative effect on family activities than perceived by the children themselves. This difference between pediatric patient perceptions of QoL and those of their parents was also seen in a study by Ngo et al³⁵ using the Child Health Questionnaire; patients rated their QoL as good as the healthy reference population, whereas parents scored their children below the reference population. A third pediatric study of QoL after ITx from Spain also showed that parents’ perception of the health of their children was worse than the healthy population (using the Short Form 36-item questionnaire), but self-scores by patients were well into the normal range using the TAPQOL and KINDL-R tools.³⁶

There are 2 large studies comparing home PN to ITx in adults reported in 2012.^20,37 Both indicated that psychological, emotional, and social QoL measures significantly improved after transplantation including ability to holiday/travel (P < 0.001), fatigue (P = 0.022), and gastrointestinal symptoms (P < 0.001) despite worsened sleep pattern and higher incidence of hypertension, osteoporosis, and diabetes.

In conclusion, ITx appears to achieve worthwhile gains in QoL for children and adult recipients in addition to its being a lifesaving treatment. However, limitations of immune suppression and the ever-present risk of rejection mean that QoL gains for HPN patients who are considering preemptive ITx (ie, before the onset of complications of therapy related to HPN) are not established. In the future, population-specific QoL tools such as the HPN-QOL combined with a well-validated generic tool like the EuroQol-5 will provide the guidance needed to inform discussions about proceeding with transplantation primarily based on QoL considerations.^29,38

REFERRAL TO A TRANSPLANT CENTER AND PLACEMENT ON A TRANSPLANT WAITING LIST

Early referral to a transplant center continues to be strongly endorsed—such a referral should not invariably lead to immediate placement on a transplant waiting list; rather, it facilitates collaborations between centers and expands the range of options available to the patient who can explore both rehabilitation and transplant treatment pathways. In this context, it is essential to recognize clinical scenarios that are higher risk for the development of complications such as liver disease and venous thrombosis as discussed below. These scenarios include ultrashort bowel (<10 cm in children and 20 cm in adults),³⁹ endduodenostomy, microvillus inclusion disease, multiple fistulae and frozen abdomen,⁴⁰ radiation enteritis,¹⁷ IF after bariatric surgery,⁴¹ and loss of 2 out of 4 upper body central veins. These subgroups of patients may not require a lifesaving intestinal transplant immediately, but referral to a large center with experience in rehabilitation is certainly justified. Furthermore, delaying ITx until a liver and pancreas graft are also required not only reduces early survival in comparison with an isolated ITx but also utilizes organs that could have been allocated to other equally deserving individuals. Delayed referral of patients for transplantation because of prolonged and ultimately unsuccessful attempts at intestinal rehabilitation also increase the probability of several complications that undermine transplantation if eventually performed. For example, repeated blood product transfusions over long periods contribute to formation of anti-HLA antibodies that significantly reduce intestinal graft survival.^42,43 Repeated catheter-related bloodstream infections (CRBSIs) and prolonged and repeated courses of enteral antibiotic therapy for intestinal bacterial over-growth lead to colonization with highly resistant microorganisms such as carbapenem-resistant Enterobacteriaceae, vancomycin-resistant enterococci, methicillin-resistant staphylococci, and extended-spectrum beta-lactamase–producing enterobacteriaceae.^44,45 In the event of multi-drug-resistant bacterial colonization, early posttransplant survival of these highly immunosuppressed patients is likely to be affected negatively, because sepsis remains the leading cause of intestinal graft loss.¹⁸ The decline in on–waiting list mortality for children since 2005⁴⁶ has probably resulted from a trend toward timelier referral and changes in organ allocation that favor children who require liver-inclusive intestinal allografts. Despite this decline, pretransplant mortality is currently highest for pediatric intestine-liver candidates and lowest for adult intestine candidates (respectively, 18.8 versus 1.9 deaths per 100 waitlist-years in 2016–2017) as depicted in Figure 1.²³

FIGURE 1.

Pretransplant mortality rates among candidates waitlisted for intestinal transplantation based on age and colisting for liver transplantation. IN, listing for isolated intestinal transplant; IN-LI, listing for liver and intestinal transplant. Adapted from Smith et al.²³

The revised indications for ITx are presented in Table 3. Since the year 2000, increasing numbers of adults relative to children have undergone ITx, which partly reflects the expanding indications for patients with motility disorders, tumors, and prothrombotic disorders resulting in mesenteric infarction.^47–49 Locally invasive and incompletely resectable intra-abdominal tumors treatable by ITx include desmoids associated with familial adenomatous polyposis. Malignant tumors including hepatoblastoma, hepatocellular carcinoma, neuroendocrine tumors, and gastrointestinal stromal tumors may also be considered for ITx, often in conjunction with chemotherapy.^50–52 ITx has been performed after detection of high-grade dysplastic polyps in Peutz-Jeghers syndrome,⁵³ and ITx may be considered for other multiple intestinal polyposis syndromes⁵⁴ when malignant and premalignant lesions are identified sufficiently early as give a reasonable chance for cure.

TABLE 3.

Criteria for placement on a waitlist for intestinal transplantation^a

Evidence of advanced or progressive intestinal failure–associated liver disease

Hyperbilirubinemia >75 μmol/Lb (4.5 mg/dL) despite intravenous lipid modification strategies that persists for >2 mo

Any combination of elevated serum bilirubin, reduced synthetic function (subnormal albumin or elevated international normalized ratio), and laboratory indications of portal hypertension and hypersplenism, especially low platelet count, persisting for >1 mo in the absence of a confounding infectious event(s)

Thrombosis of 3 out of 4 discrete upper body central veins (left subclavian and internal jugular, right subclavian and internal jugular) or occlusion of a brachiocephalic vein in children (in adults, this criterion should be evaluated in a case-by-case basis)

Life-threatening morbidity in the setting of indefinite parenteral nutrition dependence of either anatomical or functional cause, as suggested by:

In children, 2 admissions to an intensive care unit (after initial recovery from the event resulting in intestinal failure) because of cardiorespiratory failure (mechanical ventilation or inotrope infusion) due to sepsis or other complications of intestinal failure

In adults, on a case-by-case basis.

Invasive intra-abdominal desmoids in adolescents and adults

Acute diffuse intestinal infarction with hepatic failure

Failure of first intestinal transplant

^aPresuming that patients will have been assessed by a multidisciplinary team, rehabilitation options have been explored, and a state of permanent or life-limiting intestinal failure exists.

^bA total serum bilirubin of 62 or 70 μmol/L is also associated with increased mortality. A bilirubin level of 75 μmol/L is simply a useful, consensus marker of progressive liver disease to be taken in context with other parameters.^{66,67,69–71}

An ITx may incorporate colon, stomach, liver and pancreas, and kidney in various combinations depending on patient anatomy and comorbidities⁵⁵ and may also include a nonvascularized or vascularized abdominal wall.⁵⁶ Patients with end-stage liver disease complicated by complete portomesenteric vein thrombosis may require a multivisceral graft that, by definition, incorporates liver, stomach, duodenum, jejuno-ileum, and pancreas to correct portal hypertension fully as well as to replace a failing liver.⁴⁷ Splanchnic thrombosis constituted between 5% and 12% of indications for ITx during an 18-year center experience.⁴⁸ Of note, in the most recent cohort described in this report, 35 of 285 patient referrals were due to diffuse mesenteric thrombosis, primarily adults. Combined liver and ITx for severe portomesenteric vein thrombosis associated with demonstrable liver-based prothrombotic disorders can provide a definitive correction of the underlying hypercoagulable state.⁴⁹ For these reasons, complex mesenteric thrombosis even without HPN dependence may benefit from evaluation at an ITx center.⁵⁷

CONSIDERATIONS AT THE TIME OF LISTING BY A TRANSPLANT CENTER

IF caused by short bowel syndrome in infancy and childhood is now associated with extended survival on PN of varying duration in around 85%–100% of patients.^58–61 In adults, 5-year survival on HPN ranges from 58% to 83% with an increased risk of death related to the decade of age and type of underlying disease causing the IF.¹⁷ Survival of patients with PN-dependent functional IF due to pseudo-obstruction or other motility disorders is reported to be similarly high when aggressively managed with such interventions as feeding jejunostomy and decompressing ileostomy, permitting periods without PN support.^17,62

The results of ITx in high-volume centers (>5 patients per y for 5 y) indicate an overall 5-year survival with a functioning graft of around 70%,^16,63,64 which is similar to the 5-year survival obtained on HPN for adult patients. Although 5-year graft survival of pediatric patients after ITx is inferior to overall survival with IF, the latter group includes infants and children who have successfully ended PN and who are unlikely to have experienced the morbidities that would warrant consideration of transplantation.^58,61 At present neither treatment, ongoing home-based PN nor ITx, is clearly superior, although subsets of patients with excess morbidity on HPN will clearly do better with an ITx. Preemptive listing for transplantation immediately after diagnosis of IF remains controversial and is not generally considered to be an alternative to extended HPN, not least because preemptive ITx would negate the opportunity to attempt intestinal rehabilitation. However, it should be acknowledged that preemptive listing in the context of a controlled trial that includes QoL measures such as HPN-QOL^29,38 combined with a generic tool EuroQol-5 may be justified in the near future.⁶⁵

Progressive IF-associated liver disease as described in Table 3 remains the most common reason to abandon PN because of the high mortality of chronic liver disease associated with permanent IF in both adult and pediatric patients.^15–17 However, application of this recommendation may be difficult, reasons for which include uncertainty about (1) the stage at which evolving liver disease is independently life-threatening if not interdicted with a transplant and (2) the stage at which progressive IF-associated liver disease will independently increase perioperative mortality after isolated intestinal transplant enough to justify inclusion of a liver graft.

Various combinations and formulas of routine blood tests have been proposed to assist in assessment of liver disease severity and need for simultaneous liver and ITx,^66–68 although there is currently no consensus about the optimal combination of tests to perform or their interpretation. Endoscopic ultrasound and transient elastography may also be used to assess progression of chronic liver disease but await validation.^69,70 Liver biopsy remains the gold standard for determining the need for liver-inclusive ITx, although limited by the potential for sampling error, dangers inherent to the procedure itself,⁶⁶ and the possibility that precirrhotic hepatic fibrosis in IF may under-state mortality risk relative to chronic liver disease of other causes.⁶⁷ In general, the dictum remains that development of necro-inflammatory lesions (Metavir stage II) or early fibrosis (Scheuer I to II) should indicate isolated ITx, whereas stage Metavir III to IV or Scheuer III to IV should indicate a liver-inclusive graft.^64,71

Historically, liver graft-inclusive ITx have predominated in small children (based on Organ Procurement and Transplantation Network data as of August 10, 2015). As noted, within the past 5–10 years, there has been a substantial reduction in the incidence of established or threatened chronic liver failure associated with IF that would require combined liver and ITx¹⁸ such that only about half of all intestinal transplants now include a liver graft.²³ Reasons for the reduction include increasing recognition of the hepatotoxicity of excessive parenteral calories generally and soy-based intravenous lipid emulsions and the increased application of postresection intestinal reconstruction methods as noted.^72,73 This reduction has been particularly evident in North America where incidence of severe IF-associated liver disease has historically been greater than in Europe. However, use of liver-sparing, intravenous lipid modification strategies such as reduction of soy-based lipid intake to the equivalent of 1 g/kg/day or less, alone or in combination with total or partial substitution of fish oil for soy oil, may reduce if not normalize the serum bilirubin concentration without preventing or reversing hepatic fibrosis. Normalization of the serum bilirubin concentration following lipid modification, particularly introduction of fish oil, most probably complicates recognition of truly advanced chronic liver disease, increasing the need for liver biopsy when patients without obvious liver disease but who have received fish oil are considered for ITx because of other complications of IF.⁷⁴ Ultimately, the decision to include or not to include a liver graft depends on the experience and judgment of the transplant team.

The second major complication of irreversible IF that indicates listing for transplant is uncertainty about the very ability to continue PN because of threatened loss of all practical central venous access. The challenges inherent to implementing this guideline are (1) recognizing that IF is in fact permanent and (2) identifying the precise number and location of lost sites that define the threat. The 2001 Position Paper specified the loss of half of standard access sites consisting of the 2 internal jugular veins and 2 subclavian veins in infants and, in addition, the 2 femoral veins in older individuals in the setting of permanent IF.⁷ This criterion was based on the expert opinions of a working group of the VI International Small Bowel Transplant Symposium (Omaha, NE) in October 1999. A recent review of outcomes at a single large pediatric center (The Hospital for Sick Children, Toronto, ON, Canada) has supported this indication specifically when the femoral veins are included as potential sites of access.¹⁵ More controversially, this study also advocated foregoing conventional practice and deferral of ITx until 3 rather than 2 out of the 4 upper body central veins are lost. Also, in contrast with historical opinion, the recent European experience, encapsulated by the 5-year prospective survey of adult as well as pediatric patients receiving long-term PN at home by the European Society for Clinical Nutrition and Metabolism, concluded that the survival probability of patients with central vein catheter–related major complications committed to ITx was no better than that of patients remaining on PN.^16,17 Although not explicitly indicated by the survey, declining central vein access may be a stronger indication for placement on a transplant waiting list in children and young adults than in older individuals because of the presumably greater fraction of total life span that younger patients with permanent IF can be expected to require reliable central vein access. As such, occlusion of ≥2 central veins (internal jugular, subclavian, or femoral) in adults with permanent IF should be considered for transplant on a case-by-case basis.⁴⁰ Until more data that explicitly address the question of what constitutes adequate venous access in permanent IF, considerations for individual patients in transplant planning may include the apparent timing of thrombotic events, that is, recent or distant, that anticipate future vein occlusions, the impact of potential interval modifications in practice including use of antiseptic locks and antithrombotic drug prophylaxis, and the ability to monitor central vein access over time, noting that clear guidelines for surveillance imaging studies in permanent IF have not been provided.⁷⁵

Cause of underlying disease and related demographic factors may influence decisions about waiting list placement. In contrast with the variable clinical course of all but the most extreme forms of short bowel syndrome that may appropriately permit deferral of immediate transplantation, some disorders warrant early transplant listing because of their continued and consistently poor prognosis. Historically, patients with ultrashort small bowel, <10 cm in pediatric patients and <20 cm in adults, have been considered appropriate candidates for immediate placement on a transplant waiting list because of the low probability of ending PN in combination with a high risk of fatal infection and chronic liver failure, alone or in combination. In the modern era, patients with ultrashort small bowel can be sustained on PN for relatively long periods, thereby calling into question the appropriateness of traditional practice.^17,39 Although firm conclusions are limited by very small numbers of reported patients, that subgroup in whom there is no possibility of successfully restoring enterocolonic continuity appears to have a worse prognosis than those in whom enterocolonic continuity can be reestablished⁷⁶ and, in the absence of more definitive information, may still be considered case by case for preemptive placement on a transplant waiting list. Similarly, diagnosis of any congenital diarrhea syndrome that results from intrinsic mucosal dysfunction has been considered sufficient for immediate waiting list placement because of consistently poor outcomes with HPN-based management. Most consider this dictum still applicable to microvillus inclusion disease, especially because of the common development of disease-specific chronic liver failure.⁷⁷ In contrast, other disorders such as tufting enteropathy may improve during childhood and beyond, permitting marked reduction if not cessation of PN. In this event, listing for ITx may be postponed pending continued assessment of disease expression including severity and frequency of complications.^78,79

Long-term survival with Hirschsprung disease in which aganglionosis extends above the mid-small bowel remains relatively poor on long-term HPN, and immediate listing for transplantation can be considered in these cases.⁸⁰ As relates to adolescents and adults, the European Society for Clinical Nutrition and Metabolism survey concluded that development of large and invasive intra-abdominal desmoid tumors mostly as a sequela of familial adenomatous polyposis, particularly when >10–20 cm, also warrants immediate transplant listing based on demonstrable survival benefit.¹⁷ Finally, patients with acute splanchnic venous thrombosis complicated by diffuse intestinal infarction and liver failure can be placed on a transplant waitlist immediately if initial attempts at revascularization are unsuccessful.⁸¹

Given that there are few absolute indications for ITx, frequency and severity of ongoing morbidities will continue to have a major influence on the decision to proceed with placement on a transplant waiting list in individual patients.¹⁶ Whereas high rates of CRBSI alone may not be enough to justify waiting list placement, unusually high rates of life-threatening CRBSI unaffected by expert reappraisal of line care techniques in controlled settings can be a valid rationale to proceed with transplantation⁸²; occurrence of ≥2 admissions to an intensive care unit for management of septic shock has recently been proposed as a surrogate marker for this phenomenon.¹⁵ Similarly, recurring, life-threatening episodes of dehydration and electrolyte disturbances such as severe hyponatremia with potential for major central nervous system injury are not unusual with near-total intestinal loss, typically with end-enterostomy, and can warrant proceeding to transplantation.

A growing and challenging indication for ITx is retransplantation after failure of the first graft that now constitutes up to 8%–9% of current transplant activity worldwide.¹⁸ Optimal timing for intestinal retransplantation is controversial, as is the decision to remove the failed graft when technically feasible before the next transplant.⁸³ Another obstacle to successful retransplantation is the frequent sensitization of retransplant candidates, meaning that they tend to have high titers of circulating anti-HLA antibodies that are potentially associated with inferior transplant outcomes compared with outcomes of patients without antibodies.⁴² Various strategies to manage sensitization are under evaluation such as inclusion of a liver graft when retransplanting intestine and suppression of donor specific antibodies,^43,84,85 and this area of practice can be expected to improve.

CONCLUSION

IF is a complex disorder that has inspired collaboration between teams of clinicians incorporating meticulous monitoring and attention to hygiene combined with access to timely ITx and, in parallel, regular attempts to rehabilitate the bowel to minimize time on PN.⁸⁶ These endeavors have delivered drastically improved outcomes for patients with IF over the past 30 years. ITx is the ultimate alternative to PN for adult and pediatric patients with permanent IF. Solving the immunity challenge by inducing long-term tolerance to the intestinal allograft, a matter of ongoing work with exciting possibilities in prospect, will increase utilization of the operation.^{18,20,41–43,47,48,85} Future efforts that use new or existing multisite registries to generate outcome data including measures of QoL for IF with and without transplantation should provide answers to questions of timing for ITx including preemptive transplant, with less risk of biases inherent to single-center studies. Improving expertise in the management of IF has benefitted all patients including those who receive an intestinal allograft, but it remains essential that referral for transplantation should occur before the clinical problems that motivate referral become so serious that success of the operation is compromised.