Pediatric liver transplantation (PLT) sits at the pinnacle of pediatric surgical endeavors. In the early years, long-term survival rates after PLT were 11% to 39% and, since then, have improved to up to 90% with long-term graft survival rates of more than 80%. The need for PLT is estimated to be one to two per million population. With continuing improvements in surgical and interventional techniques as well as perioperative neonatal and pediatric intensive care medicine, the average age of pediatric transplant recipients has steadily declined, and there is a continuous increase of patients transplanted within the first year of life. Approximately 27% of PLTs are performed in recipients younger than 12 months. The main focus of care of children with end-stage liver disease has now shifted from achieving early survival after liver transplant to long-term follow-up, with prevention of immunosuppression-related complications and promotion of normal growth and development as possible through lifelong aftercare by a multidisciplinary team. This chapter focuses on the technical and surgical aspects that have enabled PLT to mature as a clinical therapy practiced worldwide in a large number of medical institutions.
The first attempt at liver transplant was in a child with biliary atresia (BA) in 1963. The first successful LT was also a PLT, performed by Thomas Starzl in 1967 on a 1-year-old child with hepatoma. The patient survived for over 12 months before dying from recurrence of the liver tumor. In Europe, Roy Calne performed his first PLT in Cambridge in 1968 on a 10-month-old child with BA. The child succumbed from an unexplained cardiac arrest 90 minutes after surgery. In 1988, Rudolf Pichlmayr performed the first split LT, offering one cadaveric liver to two recipients. After the first failed attempt by Raia et al. from Brazil and the first success obtained by Russel Strong in 1989, living donor liver transplantation (LDLT), mostly using the donor left lateral segment (LLS), has rapidly expanded worldwide because of the shortage of deceased donors.
Pediatric Liver Transplantation: Graft Types
Whole Liver Transplantation
This is the original technique used in PLT and is still being followed in centers with access to pediatric deceased donors. Here, the liver is retrieved from a size-matched pediatric donor and implanted as a whole graft. The procedure of whole liver procurement in pediatric donors is performed similar to adult donors, applying a technique that is a combination of the initial procurement technique described by Starzl et al. and the more recently described rapid flush technique. Whole liver transplantation can be performed with two different techniques: the classic technique with inferior vena cava (IVC) replacement, and the piggyback technique with preservation of the native IVC. Although it is technically easier, getting size-matched pediatric donors for whole liver transplantation, even in the West, is uncommon.
Reduced Size Liver Transplantation
This procedure was first described by Bismuth et al. and involves the procurement of the whole liver from an adult cadaver donor, which is then reduced in its size on the back table to match the recipient. In the original description, a right hepatectomy was performed on the back table: the right lobe of the liver was discarded, whereas the left lobe, including the vena cava, was transplanted in a child. This technique of parenchymal reduction, very seldom used today, allows surgeons to overcome differences in size between the donor and the recipient of up to 12 times. Its main limitation is that it withdraws organs from the larger adult recipient pool. For this reason, after the development of living donor and split liver transplantation, reduced size liver transplantation is rarely considered and should only be considered for large children and small adults who require a whole left lobe and full right-left split LT is not feasible.
Split Liver Transplantation
Pediatric patients used to be dependent solely on size- or age-matched donors, leading to a gross scarcity of organs and high waiting list mortality. Split liver transplantation, as described originally by Pichlmayr, involves procuring a whole liver from a cadaveric donor and dividing it ( ex situ ) into two sections along the round ligament, leaving the vascular structures for the two portions of hepatic parenchyma intact. In this way, two partial organs are obtained from a single liver: LLS (segments 2 and 3), transplanted in a child; and the extended right liver (segments 1 and 4–8), being transplanted into an adult. This procedure involves a much longer ischemic time, particularly when the two halves have to be shared by two centers. In small infants, even the LLS of the liver often is too large, and techniques to cut down LLSs may be necessary to prevent graft size mismatching and “large-for-size” syndrome. In situ splitting of the liver was developed as a consequence of LLS living donor operations. This technique involves dividing the donor liver in situ , enabling a better recognition of relevant anatomy, ease of sharing grafts between two centers, and securing hemostasis before implantation, thereby reducing intraoperative blood loss in the recipient. There is also a suggestion that biliary complication rates may be lower with in situ splits. However, in situ splitting requires additional resources and surgical expertise during the time of retrieval.
Living Donor Liver Transplantation
Technical expertise in split LT led to the development of LDLT (see a more detailed discussion in Chapter 3 ). LDLT is now an established procedure and the main form of PLT in most Asian countries. The advantages of LDLT are the use of an optimal healthy donor, minimal ischemic time, and planned surgery. This is particularly relevant for pediatric patients because during waiting time for PLT, the underlying disease can cause significant somatic and psychosocial long-term morbidity in the child. It is also useful even in countries with large deceased donor LT programs, in situations where appropriate timing of transplantation positively affects outcome, such as in patients with hepatoblastoma on chemotherapy. Living-donor procurement for PLT usually involves performing a left lateral sectionectomy. This is sufficient for children of up to 20 kg weight.
Auxiliary Liver Transplantation
Auxiliary partial orthotopic liver transplantation (APOLT) is a type of partial liver transplantation where a partial native liver hepatectomy is performed to create space for implantation of a segmental liver graft. APOLT is technically complex and is used in selected indications. It has a well-defined role in the acute liver failure setting, where it acts as a bridge to native liver recovery, allowing for avoidance of lifelong immunosuppression. When sufficient native liver regeneration is achieved, the patient can undergo withdrawal of immunosuppression leading to graft atrophy. APOLT is an option in selected cases of noncirrhotic metabolic liver diseases (NCMLD) without primary hepatocellular dysfunction or cirrhosis. Although technically demanding, APOLT has several long-term advantages.
APOLT was initially offered for children with NCMLD with the hope that future development of less invasive therapies like gene therapy or hepatocyte transplantation would provide permanent therapeutic options in their adult life. It was argued that once gene therapy is clinically available, the native remnant liver can be treated, and these patients will be able to come off immunosuppression. Gene therapy still has not become a clinical reality, and this led to a dampening of enthusiasm in APOLT.
Patients with NCMLD undergoing LT are characterized by the absence of portal hypertension and portosystemic collaterals. Caval and portal vein (PV) clamping in a noncirrhotic setting cause significant hemodynamic changes and bowel edema. APOLT, being a form of partial liver transplant, does not require complete PV and IVC clamping. In addition, the systemic effects of the operation are minimal because of the small-volume graft and a nearly normally functioning native liver with intact blood flow throughout the operative period, providing intraoperative patient stability. In the early postoperative period, the normally functioning native liver reduces the systemic effect of graft dysfunction, which gives the patient and the graft an opportunity to recover. Poor graft function is also not a threat to the patient’s life as in the case of OLT.
Domino Liver Transplantation and the Emerging Concepts of Domino Auxiliary Partial Orthotopic Liver Transplantation and Cross-Domino Auxiliary Partial Orthotopic Liver Transplantation
Livers with certain NCMLD, such as maple syrup urine disease (MSUD) and familial amyloid neuropathy (FAP), have been used as domino grafts because they are noncirrhotic and functionally normal. In experienced hands, these domino grafts have shown to provide good immediate function without additional operative risks. Patients transplanted with MSUD livers do not develop the MSUD clinical phenotype, although subclinical decreased leucine oxidation has been reported. Nearly one-third of patients transplanted with FAP livers are at risk of developing de novo FAP-like disease after 5 years, and retransplantations have been described for patients who had undergone domino LT using FAP grafts. Use of these livers in children or younger recipients is hence not recommended. Similarly, livers from hyperoxaluria or urea cycle defects recipients cannot be used as orthotopic domino grafts because these patients will develop respective disease symptoms early.
The technique of APOLT has opened up a new and exciting area in the management of NCMLD. APOLT provides the ability to transplant the partial graft resected from one NCMLD recipient to a recipient with a different NCMLD so that the native remnant liver and the graft liver together provide the full complement of metabolic functions. In other words, each half of the liver cancels the metabolic defect of the other half ( Fig. 14.1 ) We reported the first such case of a domino auxiliary transplant where the resected left lobe of a child with propionic acidemia (PA) (blood group O +) undergoing an APOLT procedure with his mother as the donor was implanted as an auxiliary in a child with Crigler-Najjar syndrome type 1 (CNS1) (blood group B +) without a suitable family donor. The child rapidly cleared jaundice and is completely well 36 months post-transplant without any symptoms of PA or CNS1 ( Fig. 14.2 ). The conceivable next step would be a cross-domino APOLT where left hemilivers can be swapped between patients with different NCMLD and cure both without needing a donor. This was not possible in our reported case because the blood groups of the two children were not cross-compatible. The limiting factor to this technique would also be the number of patients with NCMLD who are awaiting LT in a center at any point of time. This is hence likely to be feasible in large centers specializing in the treatment of MLD and may also open up ways to coordinate transplants between multiple centers.
ABO-Incompatible Liver Transplantation
To overcome the shortage of organs for PLT, liver grafts from ABO-incompatible (ABOi) donors have been used to increase the possibilities of transplantation (see a more detailed discussion in Chapter 3 ). The pediatric population has a privileged immune system that helps support ABOi LT better than adults. Infants do not produce isohemagglutinins; therefore their anti-A and -B antibody titers remain at low levels even beyond 12 months of age. Additionally, the activation of the complement system is relatively suppressed in infants. Taken together, infants have fewer mediators in relation to antibody-mediated rejection. Several centers have reported no increase in rejection or graft loss as compared with ABO-compatible PLT. Desensitization protocols as used in adult ABOi LT are also not usually required in children under 2 years of age.
Abdominal incisions need to be lower in small children ( Fig. 14.3 ). When children grow, subcostal incisions tend to move up above the costal margin. Reexploration in later years may result in two separate, parallel scars. Lower incisions are more bowel friendly because removing bowel from within the abdominal cavity creates more space in the upper abdomen without causing angulation and drag on the small bowel mesentery. Care should be taken while filleting the liver from the vena cava. The vena cava can easily be narrowed or distorted, particularly when there are large caudate and accessory hepatic veins draining inferior to the right hepatic vein. Children with cirrhosis, particularly secondary to BA, often have dense adhesions. Diathermy should be used with caution. Serosal tears may often be unnoticed and, in combination with diathermy injuries, predispose to bowel perforations. Children tolerate both caval clamping and PV clamping well, and portocaval shunts or venovenous bypass have not been described in these patients. However, in the noncirrhotic patients with no portal hypertension, bowel congestion is often encountered, and the anhepatic phase should be kept to a minimum. This fluid sequestration in the bowel, in combination with cava clamping, can cause hemodynamic disturbance in children with compromised cardiovascular systems (primary hyperoxaluria, Alagille syndrome, etc.). When mobilizing the liver from the IVC at its cranial end, it is not uncommon to encounter the right phrenic nerve. Care should be taken while using diathermy in this region. Injuries to the phrenic nerve may lead to ventilation issues in the post-operative period. Blood loss should always be kept to a minimum. Because of a low blood volume in children, it is easy to lose track of the bleeding, and catchup transfusions are not ideal. To prevent hypothermia and consequent worsening coagulation, wet swabs should be changed periodically.
Size discrepancy is common in small children needing LT, even with an LLS graft. It is common to have a graft-to-recipient weight ratio (GRWR) greater than 3, and even GRWR greater than 5 is not uncommon. When GRWR is greater than 4, further reduction of the LLS graft is often required. However, a GRWR of greater than 4 does not always mandate graft reduction. Multiple additional factors, including the size of the child’s abdominal cavity, the anteroposterior thickness of the graft, presence of ascites, sarcopenia, and quality of portal inflow, have to be taken into account
The reduction of LLS can be performed in situ or ex situ . Ex situ liver graft reductions are usually segment III monosegment grafts, with the delineation based on the hepatic venous system. The transection line is dependent on the anatomical variation of the hepatic rather than the portal venous system, which is probed as segment II is excised from the graft.
In situ reductions of LLS graft are done using portal venous delineation. The segment II or III PV is identified in the Rex recess and ligated. This allows a clear demarcation between the segments II and III; segment II or III is then excised. It provides a thin graft, retaining the natural shape of the unreduced LLS, making it beneficial in children where the anteroposterior graft thickness is a problem. Graft reductions may also be performed nonanatomically ( Fig. 14.4 ). This could be either pre- or post-reperfusion and is a subjective decision based on the availability of space for thin grafts.