|Home | About | Journals | Submit | Contact Us | Français|
Intestinal transplantation is gradually becoming a therapeutic intervention rather than an experimental procedure. In the long term, the best outcome for patients with intestinal failure remains total parenteral nutrition, but where this is unlikely to allow long-term survival because of loss of venous access sites or severe cholestasis, intestinal transplantation should be considered. The technical aspects of the procedure are well described and advances in recent years in both immunosuppression and antimicrobial therapy have led to improved outcomes, particularly in the larger centers. Graft monitoring and the profound sepsis that accompanies graft dysfunction due to bacterial translocation remain major challenges, whereas the issues of denervation, lymphatic disruption, graft-versus-host disease (GVHD), and nonphysiological venous drainage have not proved to be major problems. Whether intestinal transplantation will become an alternative for the stable patient on total parenteral nutrition rather than a salvage procedure for when total parenteral nutrition fails remains to be seen.
Early attempts at intestinal transplantation made during the 1960s were uniformly unsuccessful. Failure was chiefly because the available immunosuppressive drugs and antibiotics available at the time were not sufficiently potent and because the surgery was performed in patients who were in very poor clinical condition, generally receiving transplants acutely immediately after massive bowel resection. With the advent of intravenous feeding in 1969, a safer alternative appeared for patients with intestinal failure and interest was lost for more than a decade. Only in the late 1980s with the advent of more powerful immunosuppressive drugs was the topic revisited, and several centers have developed active small bowel transplant programs. There is now a world experience of around 1000 cases.
Intestinal failure is defined as an inadequately functioning enterocyte mass to provide an individual with adequate nutritional support, and this condition represents the bulk of patients undergoing intestinal transplantation. The majority of patients with intestinal failure have short gut syndrome such as follows extensive bowel resection. A smaller proportion suffer from enterocyte dysfunction and motility disorders. The indications are listed in Table Table11.
Total parenteral nutrition (TPN) is still the treatment of choice for such patients, with 1-year survival of up to 96%.1 In the United Kingdom (UK) approximately 100 patients go on to long-term TPN each year, of whom half would be unsuitable for transplantation (e.g., elderly, comorbid diseases, malignancy).2 This gave an estimated case load of about 50 cases per year, and programs in the UK were established on this basis, although this has proved to be a gross overestimate of the true need.3
Intestinal transplantation is generally regarded as an alternative only in patients who are unlikely to survive in the long term on TPN because of TPN-related complications. The commonest causes of failure on TPN are line access problems from venous occlusions or catheter sepsis and cholestatic liver disease, which is generally multifactorial and results from factors such as reduced portal blood flow, sepsis, and inference with the enterohepatic circulation of bile salts.
Transplantation of the isolated small intestine is indicated for patients with line access problems and liver disease if it is not advanced and is potentially reversible. If the liver disease is more advanced and associated with significant fibrosis or has progressed to cirrhosis, a combined liver–small bowel graft is required. Liver failure is heralded by jaundice, depressed serum albumin, impaired clotting, and portal hypertension leading to ascites, encephalopathy, and variceal bleeding.
Approximately 60% of patients are pediatric and 40% adult, with roughly equal numbers requiring an isolated small bowel transplant and small bowel plus liver. Three UK centers are designated and funded by the Department of Health for intestinal transplantation—St. James's Hospital, Leeds and Addenbrooke's Hospital, Cambridge for adults and the Birmingham Children's Hospital for pediatric cases. A small proportion of patients (around 10%) require a multivisceral graft consisting of other organs in addition to the liver and bowel. This type of graft might be used for uncommon cases of low-grade but extensive intra-abdominal malignancy such as neuroendocrine tumors, requiring massive resection to establish control of the disease.
All alternative treatments must be explored when considering an intestinal transplant. Various bowel lengthening procedures such as the Bianchi procedure4 can improve enteral function and the adaptation process following massive bowel resection may continue for 2 to 3 years. After massive bowel resection, patients left with small intestine alone generally need at least a meter of bowel to maintain nutrition without TPN, whereas patients who have retained their ileocecal valve and colon can survive with as little as 30 cm of small intestine because of the ability of the colon to absorb salt and water and the role of the ileocecal valve in maintaining sterility of the small bowel contents.
It is essential to assess liver function carefully, usually with a liver biopsy, because this may affect the timing of the surgery and the choice of the type of graft. There is some evidence that the transplanted liver can impart a degree of donor-specific protection to a second organ (e.g., bowel) from the same donor, but the theoretical advantages of this are outweighed by the risks of not being able to remove the graft and withdraw immunosuppression if the patient suffers severe sepsis, and it is generally accepted that a combined liver-bowel graft should be performed only if the liver is failing or if the portal vein has thrombosed.
In patients with line access problems it is essential to assess the patency of all the great veins, both to ensure that all possible routes for TPN have been explored and to define for the anesthetists which central veins can be cannulated at the time of the transplant.
It is also essential to assess patients' cardiac, respiratory, and renal function to determine whether they are fit enough for the surgery, as well as their psychological suitability to ensure that they will comply with long-term follow-up and taking their immunosuppressive drugs. It is important to ensure that there is adequate renal reserve because the antirejection drugs and some antifungal agents that are frequently required are nephrotoxic.
The potential cadaveric organ donor must be free from sepsis, bowel disease, or any history of malignancy outside the central nervous system. Such donors will generally have died from intracerebral hemorrhage or a head injury and be declared brain dead while still ventilated. The circulation to the viscera must be maintained until the moment of perfusion with preservation solution. There is no proven benefit in matching donor and recipient tissue types, although they must be ABO blood group compatible.
Depending upon the need of the recipient, the graft can consist of intestine alone or be combined with the liver as a composite graft. Occasional cases require a multivisceral graft, which includes, for example, the pancreas in addition to the liver and small bowel. Because of graft swelling after transplantation and the possibility that the abdominal cavity is contracted following previous excision of the diseased intestine in the recipient, ideally a donor is selected who is smaller than the recipient.
Surgical techniques have been developed to establish an arterial supply to and venous drainage from the graft and restore intestinal continuity while allowing easy access to the lumen of the graft for mucosal biopsies.5
The donor retrieval is generally combined with retrieval of the thoracic organs and kidneys for transplantation. The isolated small bowel transplant is mobilized on a pedicle consisting of the superior mesenteric artery dissected to its aortic origin and the portal vein beyond the confluence of the splenic and superior mesenteric veins (Fig. 1). Branches arising from the right side of the superior mesenteric artery to supply the colon are ligated and divided.
Segmental intestinal grafts have been successfully taken from living related donors using a graft based on a pedicle of the distal superior mesenteric artery and vein.
Implantation is by anastomosis of the superior mesenteric artery of the graft to the infrarenal aorta of the recipient and anastomosis of the superior mesenteric vein–portal vein of the graft to the portal vein of the recipient in the free edge of the lesser omentum. If there is inadequate length of vein on the graft, as tends to occur with use of a live donor, the venous drainage can be to the inferior vena cava, which is nonphysiological but generally equally well tolerated.
The proximal end of the graft is anastomosed to the proximal end of the residual bowel, and the distal end is brought out as an ileostomy (for graft monitoring), with side-to-end anastomosis of the graft to the colon to simplify eventual closure of the stoma if the distal colon remains.
The peritoneal attachments of the liver and the vena cava above and below the liver are divided but the portal vein is kept in continuity with the superior mesenteric vein of the graft. Some centers divide the proximal jejunum, skeletonize the portal vein as it passes through the pancreas, and divide the bile duct at the upper border of the duodenum. In our unit, our preferred technique is to divide the first part of the duodenum and keep the bile duct within a small volume of pancreatic tissue within the concavity of the duodenum, which reduces the risk of the portal vein twisting and avoids the need for direct anastomosis of the bile duct. The graft is supplied with arterial blood via the superior mesenteric artery and the hepatic branch of the celiac artery, which can be kept together on an aortic segment or made into a single trunk by use of a Y-graft fashioned from the common iliac artery and its bifurcation taken from the donor (Fig. 2).
Following hepatectomy in the recipient, implantation is by anastomosis of the common arterial trunk of the graft to the recipient aorta and anastomosis of the inferior vena cava above and below the liver. The graft is then revascularized and the remaining foregut decompressed by anastomosis of the divided recipient portal vein to the side of the donor portal vein. The distal end of the remaining bowel in the recipient is anastomosed to the side of the proximal portion of the graft and the distal end of the graft brought out as an ileostomy as for the isolated bowel graft.
Early enthusiasm for including the colon with the intestinal transplant has waned because it does not impart any functional advantage and increases the risks of sepsis and lymphoma.
When the intestine is transplanted, it undergoes several pathological and physiological changes that might be expected to affect its function.
Acute rejection. The initial targets of the host T cells are the lymphocytes of the graft, leading to swelling of the Peyer's patches and mesenteric lymph nodes and a localized immunological engagement. Unchecked, this progresses to an immunological attack on the blood vessels of the transplant, and finally the enterocytes themselves are the target of the rejection. The combination of the enterocyte damage and the ischemic injury from the vascular invasion leads to bacterial translocation.
Chronic rejection. Chronic rejection probably results from an antibody-mediated immune response on the endothelial surface of the vessels of the graft. This causes local platelet aggregation, release of growth factors, and endothelial proliferation and results in obliteration of the medium-sized arterioles in the graft, leading to loss of function. Chronic rejection is more commonly observed in isolated bowel transplants rather than liver-bowel transplants.
Graft-versus-host disease. The substantial volume of donor lymphatic tissue in the graft led to concerns that graft-versus-host responses could be mounted by this lymphocyte population. Clinically, however, GVHD has rarely been a major problem, although donor lymphocytes have been identified in the recipient circulation in the presence of stable graft function. Donor pretreatment with antilymphocyte preparations was performed in an attempt to render the graft less immunogenic and to reduce the theoretical risks of GVHD, but there is no evidence for any clinical benefit from such treatment.
It is important for enterocyte recovery to institute enteral feeding as soon as the patient's condition allows. In addition, immunosuppressive drugs can be given enterally and may have some local action. Immunosuppression is generally based on tacrolimus (which inhibits cytokine transcription) combined with mycophenolate mofetil (which is an antiproliferative agent) and steroids with an increasing use of sirolimus (which blocks proliferative responses to activated T cells). Cyclosporin is generally considered not to be potent enough to protect the bowel graft, although the first successful case of the modern era was treated with cyclosporin. The rebirth of intestinal transplantation in the early 1990s was largely a result of the availability of tacrolimus, and very powerful immunosuppression was generally employed but led to a high incidence of sepsis and Epstein-Barr virus–related post-transplantation lymphoproliferative disorder (PTLD). This remains a significant risk following intestinal transplantation, although immunosuppression is generally less aggressive now. Some units use induction antibodies such as alemtuzumab (Campath), and most resort readily to antilymphocyte agents such as OKT3 and antithymocyte globulin (ATG) in the presence of rejection if it is not swiftly reversed with high-dose steroids, because of the fear of sepsis from translocation. These agents deplete or inhibit host lymphocytes by specific interaction with a surface receptor on the cell membrane that is specific to a particular class of lymphocyte. In the long term there is an increasing trend toward reducing and tailoring immunosuppression, although this cannot be withdrawn entirely without encountering refractory rejection. Technical failure due to graft thrombosis and technical problems such as anastomotic leakage do occur but are rare.
Graft monitoring is difficult as there are no conventional blood tests that reliably detect rejection, although some new tests of lymphocyte activation in peripheral blood are showing some promise. Tests of mucosal permeability with radioisotopes as a marker of loss of functional integrity have been disappointing, and mucosal biopsies taken via the stoma that is formed at the time of surgery in all patients are generally considered the “gold standard,” although the rejection process can be patchy and the biopsies can be misleading.
Grading erythema, friability, and villus height by magnification video endoscopy is being evaluated as a tool for graft monitoring.
The majority of the other agents given postoperatively are prophylactic against opportunistic infections—such as cotrimoxazole for pneumocystis, fluconazole for candida and aspergillus, and ganciclovir for cytomegalovirus and herpes simplex. Broad-spectrum luminal antibiotics, aimed at keeping the bowel lumen sterile, encourage overgrowth of resistant flora and are not generally used. The stomal effluent is cultured daily and monitored to direct treatment in the event of translocation.
The world experience is currently of around 1000 cases from 65 centers. All cases are reported to an international registry based in London, Ontario, which reports every 2 years. A small number of cases have used a segment of bowel taken from a living related donor, but 98% of cases are from cadaveric sources. Overall 5-year patient and graft survival is comparable for isolated small bowel and small bowel plus liver transplants and is around 50%, but the larger centers now report 80% survival at 1 year. The commonest causes of death are sepsis causing multiorgan failure and PTLD. The costs of intestinal transplantation actually compare favorably to those for TPN.11 The factors that affect survival most are the era of the transplant and the size of the center rather than any donor or recipient factors.12
After a disappointing early experience, results of intestinal transplantation are improving and although it should remain in the domain of highly specialized units, intestinal transplantation is gradually becoming accepted as a viable treatment for patients with intestinal failure rather than as an experimental procedure.