LIVING RELATED LIVER TRANSPLANTATION IN CHILDREN

R. Reding, MD PhD
Dept of Surgery
Saint-Luc University Clinics
10, Hippocrate Avenue
B-1200 Brussels, Belgium

Because of the natural history of their liver disease, the vast majority of children are very young at the time of liver transplantation (LT). In the Brussels series of 487 pediatric LT recipients (1984-1999), about two-thirds weighted less than 12 kg at the time of registration on the waiting list. On the other hand, the number of pediatric candidates exceeds the number of small-sized pediatric donors of similar weight allowing full-size LT to be performed. The result of this inadequacy has been a high mortality on the cadaveric waiting list, between 10 and 20%. Despite the introduction of alternative techniques allowing transplantation in a child of part of a liver graft procured from an adult donor (reduced-size and split liver grafts), the rising mortality on the cadaveric waiting list in 1993 (up to 17%) led us to develop intrafamilial living related liver transplantation (LRLT). Beside an increased availability of liver grafts for pediatric recipients, the introduction of LRLT was shown at our centre to improve the results after pediatric LT, through appropriate preparation of the recipient, better graft quality, and shorter total ischemic time.

Ethical issues

Before any clinical implementation, the ethical questions regarding LRLT should be thoroughly discussed in clinicians-ethicists consultations in order to prepare a protocol to be approved by the local Ethics Committee or Institutional Review Board. The ethical approach of pediatric LRLT, in the context of cadaveric organ shortage in Western Europe, has to consider nine items, which are carefully balanced, allowing medical and ethical assessment of the clinical practice, as summarized in the table:

Elective as well as highly urgent recipients may be currently considered for LRLT, the latter only after case by case approval by a consultation between clinicians and ethicists. Potential donors are only parents and grandparents, but other family members may be accepted after case by case clinicians-ethicists consultation. Parents of every potential recipient are provided with a full information regarding the possible options (cadaveric - full-size or segmental - and living related donor grafts). If the parents explicitely express their interest for living donation, the ABO-identical or -compatible potential donor and his/her spouse sign a first informed consent form before thorough medical and psychological evaluation. If no contraindication is evidenced, a second informed consent form is signed before liver donation. An analysis of the psycho-socio-economic familial context is also included in the global evaluation. Finally, in order to respect equity and parents’ autonomy, several strategic options are offered to the parents of a child candidate to liver transplantation: (1) listing on the cadaveric waiting list with possibility for secondary living donation if the child’s condition deteriorates; (2) evaluation of a potential living donor with registration of the recipient on the cadaveric list in case of any contraindication in the potential donor or of consent withdrawal ; (3) listing for living donation as confirmed option.

Living donor liver transplantation in children can be described as the combination of two well-established surgical techniques: (1) a left lateral segmentectomy in the living donor (segment II+III according to Couinaud’s classification), technically adapted in order to procure a graft with anastomosable vascular and biliary pedicles; (2) the implantation of a segmental graft, which is a technique similar to left split liver grafting. The main preoperative requirement is the determination of liver volume in the donor as well as of minimal volume requirement in the recipient, which can be carried out by means of a liver volumetry evaluation software using CT scan or magnetic resonance imaging. Whole liver volume generally corresponds to 2.5% of the body weight, the corresponding figure being approximately 25%x2.5% (0.6% of the body weigth) for the left lobe (left lateral segment or segments II+III), with variations between individuals. The minimal liver volume to be implanted in the recipient should at least reach 1% of the body weight. Accordingly, a 10 kg pediatric recipient should receive a liver graft of at least 100g and his living donor of 60 kg has a left lobe with a theoretical weight of 360g, to be controlled with volumetry assessments. On the other hand, a big for recipient size graft will require abdominal closure with a temporary silastic prosthesis, in order to avoid graft and vascular compression.

A total of 178 primary LT were performed in children (<15 years) at our center between July 1993 and December 1999. This series included 77 (43%) LRLT, which are reviewed.

Pediatric recipients. Between July 1993 (initiation of our LRLT program) and December 1999, a total of 77 children (<15 years old; 39 males, 38 females) received a LRLT at our center. Median (range) age and body weight at LT were 1.1 year (0.4-13.1), and 8.3 kg (4.3-31.0), respectively; 36/77 (47%) recipients were less than 1 year. The indications for LT were as follows: biliary atresia in 55 cases (71%), progressive familial intrahepatic cholestasis in six, primitive hepatic malignancy in five (including four hepatoblastomas and one hepatocarcinoma), cholestatic cirrhosis in four, fulminant hepatitis in two, and miscellaneous diseases in the five remaining instances. Regarding the clinical status at LT, 41/77 (53%) children were at home, 32/77 (42%) were hospitalized, and 4/77 (5%) in the intensive care unit.

Living-related donors. The intrafamilial donor was the mother (n=43), the father (n=30), and the adoptive father, the grandmother, the aunt and the uncle of the child, in one case each. All were medically and psychologically assessed according to a protocol described elsewhere. The donor selection and consent process took place according to the pediatric LRLT protocol approved in 1992 by the Ethics Committee of our institution, and revisited in 1996.

Surgical technique and post-transplant management. From the technical point of view, the extent of liver resection in the living donor was a left lobectomy (n=67), or a left lobectomy extended to segment IV (not including the median hepatic vein) (n=10). All grafts were implanted orthotopically with a triangulation of the anastomosis between the left hepatic vein and the recipient vena cava, except in one instance. Portal reconstruction consisted of an end-to-end anastomosis (n=57), or a jump venous auto- or allograft implanted either on the splenomesenteric confluence (n=10), or on the superior mesenteric vein (n=10). In most cases, one or two end-to-end anastomoses were performed between the donor left hepatic artery and the recipient proper hepatic artery, using magnifying loupes (x5.5). A cadaveric iliac artery allograft preserved in Terasaki solution was interposed between the recipient infrarenal aorta and graft hepatic artery in four cases. Biliary drainage consisted of hepatico-Roux-en-Y anastomosis (n=75) or hepatico-hepaticostomy (n=2). The primary immunosuppressive protocol varied along the period, but consisted of tacrolimus and low dose steroids since 1997.

Morbidity in donors. No significant intraoperative complications were encountered. Only one donor required blood bank transfusion. Median (range) peak total bilirubin and GPT during the post-operative period reached 1.4 mg/dl (0.6-5.9), and 283 IU/l (54-1630), respectively; the corresponding values at hospital discharge were 1.0 mg/dl (0.9-1.0), and 170 IU/l (36-611), respectively. Overall, a total of six donors had a peak value of total bilirubin above 3 mg/dl. The post-operative complications were as follows: biliary leakage from the cut surface, with spontaneous resolution (n=5), eventration (n=3), pleural effusion (n=2), and cubital nerve compression (n=1). The median (range) post-operative day at discharge was day 6 (5-11). No differences in post-operative biochemistry or complications could be related to the additional resection of segment IV. Except for the cubital nerve compression, which required surgical correction, no long-term sequela was observed in the living donors.

Overall results in pediatric recipients. Actuarial patient survival rate was 92% at one year and 89% at five years post-LT. The causes of death were as follows: adenovirus hepatitis (POD 29), delayed graft function complicated by portal thrombosis (POD 32), recurrent neonatal hepatitis (POD 36), recurrent hepatoblastoma (POD 158), post-transplant lymphoproliferative (POD 228), unknown etiology (POD 243), recurrent hepatocarcinoma (POD 385), biliary sepsis (POD 500). Actuarial graft survival rate was 90% at one year and 86% at five year. Two children were retransplanted for chronic rejection secondary to immunosuppression cessation for a post-transplant lymphoproliferative disease (POD 210), and for primary chronic rejection (POD 538). The incidence of acute rejection was thoroughly analysed elsewhere, and was found to be not statistically different from the incidence encountered in pediatric recipients of cadaveric allografts transplanted at our center during the same time interval.

Technical complications of graft reimplantation. Technical complications were analysed systematically for the four types of anastomosis necessary for graft reimplantation. No complication was encountered regarding the hepatic vein anastomosis. The overall incidence of portal vein thrombosis was 8/77 (10.4%), including four late thromboses among the nine patients with a homologous iliac vein jump graft between the superior mesenteric vein and the graft left portal vein. These latter cases were subsequently reoperated using interposition of an autologous internal jugular vein between the superior mesenteric vein and the Rex recessus of the graft left portal vein. One early hepatic artery thrombosis was encountered in the series (1.3%), and this complication was successfully reconstructed. Finally, the overall incidence of reoperated biliary stenosis reached 14/75 (18.7%) among the patients with hepatico-Roux-en-Y anastomoses, whereas three postoperative biliary fistula from the cut surface could be successfully corrected.

In the context of cadaveric organ shortage, easily available living related liver donors constitute for their pediatric recipient a unique opportunity to escape the uncertainty of the waiting time on the cadaveric list. Previously reported data from our center showed a higher pretransplant mortality rate on the cadaveric list (15%), when compared with that in the LRLT program (2%). The introduction of intrafamilial living donor liver transplantation in our program of pediatric LT required, however, a profound analysis of its ethical implications for the donor, the recipient, and their family. The medical and psychological selection process for potential living related donors has been thoroughly described in the literature, and it should be strictly followed, to minimize post-transplant morbidity.

The analysis of post-transplant technical complications given in this paper showed a low incidence of hepatic artery thrombosis (1%), but a high rate of stricture of the biliary anastomosis on the graft (19%), obviously not related to the former occurrence. Variations of the local arterialization of the left hepatic duct could be hypothesized in order to explain the high incidence of biliary complications, in our series as well as others, including experiences with left split liver transplantation. It should be stressed that the bile duct stenoses were corrected by a surgical redo in the majority of the cases, and were not responsible for graft loss or patient death, except in one case associated with cholangitis due to vancomycin-resistant enterococcus.

In conclusion, LRLT in pediatric recipients constitutes nowadays a validated procedure from the point of view of the living donors as well as of the children with chronic or acute liver diseases. It provides a valuable alternative to cadaveric LT, particularly in the current context of organ shortage.