Domino liver transplantation: How far can we push the paradigm?


  • Irinel Popescu,

    Corresponding author
    1. Center of Digestive Diseases and Liver Transplantation, Fundeni Clinical Institute, Bucharest, Romania
    2. Carol Davila University of Medicine and Pharmacy, Bucharest, Romania
    • Center of Digestive Diseases and Liver Transplantation, Fundeni Clinical Institute, 258 Fundeni Street, Sector 2, Bucharest, Romania 022328
    Search for more papers by this author
    • Telephone: +40 21 318 04 17; FAX: +40 21 318 04 17

  • Simona O. Dima

    1. Center of Digestive Diseases and Liver Transplantation, Fundeni Clinical Institute, Bucharest, Romania
    Search for more papers by this author


Domino liver transplantation (DLT) has emerged as a strategy for increasing the number of liver grafts available: morphologically normal livers from donors with metabolic diseases can be used for select recipients with hepatocellular carcinoma (usually outside the Milan criteria). Familial amyloidotic polyneuropathy (FAP) is the most common indication for DLT. When FAP patients are involved in DLT, the indications and outcomes are clear and good, although de novo FAP development within various periods of time has been described in DLT recipients of FAP livers. With the increasing need for organs, livers explanted from patients with rare metabolic diseases, such as primary hyperoxaluria (PH), acute intermittent porphyria (AIP), maple syrup urine disease (MSUD), and homozygous familial hypercholesterolemia (HFHC), are being used for DLT. However, insufficient data about the use of livers from patients with these rare metabolic diseases are available. In this review, we focus on the latter disorders. PH is not a good indication for DLT because recipients of PH livers develop hyperoxaluria and early acute renal failure. AIP also seems to be a debatable indication for DLT because of the rapid development of neurotoxicity in AIP liver recipients. However, the outcomes of DLT with HFHC and MSUD liver grafts (which include the risk of the de novo development of these genetic diseases) are promising. For rare metabolic liver diseases to be established as indications for DLT, more reports and studies are needed. Liver Transpl 18:22–28, 2012. © 2011 AASLD.

Because of the organ shortage and the increasing number of patients on the waiting list for liver transplantation (LT), alternative techniques such as split LT and living donor LT have been developed. Another possible way of increasing the pool of liver grafts is the use of marginal livers. Domino liver transplantation (DLT) has emerged as a strategy for increasing the number of liver grafts available: explanted organs are used for select patients. Except for the production of an abnormal protein or enzyme, these livers are morphologically normal and fully functional. Additionally, the metabolic diseases of the donors usually should not produce symptoms in the DLT recipients for many years. In comparison with deceased donor liver transplantation (DDLT), DLT should not expose the recipients or the donors to any significant additional operative risks. The survival of the recipients should be as good as it would be with DDLT, and the morbidity and mortality of the donors must be kept to a minimum.1

So that as much information as possible could be gathered about DLT, an international registry (the Domino Liver Transplant Registry) was created in 1999 as an extension of the already existing Familial Amyloidotic Polyneuropathy World Transplant Registry.2 According to this registry, DLT had been performed 790 times by December 31, 2009.3

This procedure raises ethical and surgical issues. The most important ethical principle concerns the informed consent of both the domino recipient and the donor; this includes the need to emphasize that the recipient may develop the domino donor's genetic disease. Additionally, the domino donors must be assured that the technical demands of the hepatectomy for this procedure will not expose them to any unusual risks.

When DLT first began to be used, the recipient indications were primary [especially hepatocellular carcinoma (HCC) and, rarely, cholangiocarcinoma] or even secondary liver malignancies. As experience with transplantation has accumulated, HCC has remained the main liver malignancy in DLT recipients and especially in those recipients who are beyond the Milan criteria.4

Today, DLT liver grafts are used for patients with alcohol- and virus-related liver cirrhosis. The prototypical candidates are patients whose condition will ensure a long time on the waiting list: elderly patients (55-60 years old) with a life expectancy shorter than the time needed to develop the symptoms of the domino donor's disease.

There is also a tendency to use these grafts (similarly to split grafts) as a bridging therapy for neonates; this allows them to grow until they can receive a normal and size-compatible liver.5


DLT with a liver from a patient with FAP was first performed in October 1995 in Portugal by Furtado et al.6 FAP is the most common indication for DLT, and DLT is performed more frequently in areas in which the incidence of this disease is higher (Portugal, Sweden, and Japan). DLT with a graft from a patient with FAP has been performed in more than 800 patients. Worldwide, many centers have reported successful DLT with FAP liver grafts.2

FAP is an autosomal dominant disease that is associated with a mutation of the transthyretin (TTR) gene, and it is characterized by the extracellular deposition of TTR amyloid fibrils, especially in the peripheral nervous system.

The clinical symptoms of FAP take 20 to 30 years to appear and another 10 to 14 years to become fatal.7

Because more than 95% of the circulating TTR is produced in the liver,1 LT represents a curative treatment for this disease.

In the case of FAP, the results of DLT are comparable to the results of DDLT.8 The recipient survival rates at 3 months and 1 year with DDLT and DLT are not significantly different.9 A recent study revealed that the domino procedure does not add any significant risk to either the FAP donor or the FAP liver recipient.1 One reason for expanding the indications for DLT is that this procedure is associated with decreased graft dysfunction and perioperative bleeding, possibly because of the protective role of the short ischemia time in DLT and the younger age of the donors.9

However, the de novo development of FAP within various periods of time (2-9 years) has been described in DLT recipients of FAP livers. Stangou et al.10 reported the first case of de novo FAP in a DLT recipient in 2005. They described the case of a 55-year-old patient with clinical polyneuropathy, which was confirmed by electroneuromyography and nerve biopsy 8 years after DLT. Lladó et al.11 identified 4 additional cases of de novo FAP; the diagnosis was based on sural nerve biopsy with electroneuromyography, rectal biopsy, and clinical correlations. The authors concluded that de novo systemic amyloidosis after DLT may be more frequent and may appear earlier than initially thought.

Mutations in the fibrinogen Aα-chain (Afib) gene are the main causes of hereditary renal amyloidosis. Combined liver-kidney transplantation has been performed in 22 patients with hereditary Afib amyloidosis and kidney impairment. Stangou et al.12 reported 4 cases of DLT with grafts from Afib patients. In 1 DLT case, there was no evidence of amyloid deposition after 5 years of follow-up.


Afib, fibrinogen Aα-chain; AGT, alanine-glyoxylate aminotransferase; AIP, acute intermittent porphyria; BCAA, branched-chain L-amino acid; DDLT, deceased donor liver transplantation; DLT, domino liver transplantation; ESRD, end-stage renal disease; FAP, familial amyloidotic polyneuropathy; FHC, familial hypercholesterolemia; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HFHC, homozygous familial hypercholesterolemia; HMG-CoA, 3-hydroxy-3-methyl-glutaryl-coenzyme A; KF, kidney failure; LDLR, low-density lipoprotein receptor; LT, liver transplantation; MSUD, maple syrup urine disease; NA, not available; PBC, primary biliary cirrhosis; PH, primary hyperoxaluria; POD, postoperative day; TTR, transthyretin.


With the increasing need for organs, livers explanted from patients with metabolic diseases other than amyloidosis are being used for DLT; primary hyperoxaluria (PH), acute intermittent porphyria (AIP), maple syrup urine disease (MSUD), and homozygous familial hypercholesterolemia (HFHC) are some of the metabolic diseases that represent rare indications for DLT. When DLT involves FAP patients, the indications and the results are clear. However, insufficient data on the use of livers from patients with other rare metabolic diseases are available.

In this review, we therefore focus on the latter disorders. Table 1 summarizes the characteristics of DLT donors with rare metabolic diseases (other than FAP) and their recipients.

Table 1. Review of the Worldwide Published Experience With DLT for Rare Metabolic Diseases Other Than FAP
StudyDisease/Domino Donor Age of Disease OnsetDomino Receptor Age/DiseaseSurvival Outcome at the Time of PublicationTime From DLT to Symptom Onset/Response to Medical Treatment
Donckier et al.13PH/7 years69 years/multifocal HCC and HCV-related Child A cirrhosisAlive8 months/hyperoxaluria and moderate renal insufficiency
Franchello et al.14PH/33 years50 years/cryptogenic cirrhosis and multifocal HCCAlive at 9 months2 months/nephrolithiasis and renal function deterioration treated with DDLT (retransplantation) 3 months after DLT
Farese et al.15PH/40 years64 years/locally advanced, nonresectable cholangiocarcinomaDead 8 months after DLT: bacterial pneumoniaRenal oxalate deposition and marked increase in hyperoxaluria with ESRD on POD 12 treated with high-dose pyroxidine and thiazide diuretics
Casas-Melley et al.5PH/NA23 days/hepatic veno-occlusive disease with liver failureAlivePOD 5/increased level of oxalic acid excretion treated with high fluid intake, pyroxidine, Neutra-Phos, and liver retransplantation
Saner et al.16PH/NA55 years/HCV and HCCDead at 5 months: intraoperative pulmonary embolismPOD 13/KF treated with dialysis
PH/NA68 years/PBC and HCCAlive at 20 monthsPOD 7/KF treated with liver retransplantation
PH/NA66 years/HCV and HCCDead at 3.3 months: malignant ventricular tachycardiaPOD 5/KF and initiation of dialysis on POD 6
PH/NA67 years/cryptogenic cirrhosis and HCCDead at 1.3 months: pericardial tamponadePOD 6/KF and initiation of dialysis on POD 13
PH/NA61 years/PBCDead at 11 monthsPOD 1/KF and initiation of dialysis on POD 1
Dowman et al.17AIP/NA65 yearsDead at 32 days: sepsisInitial decrease in urinary porphobilinogen excretion (but it remained elevated until the patient's death)
AIP/NA60 yearsDead at 24 hours: myocardial infarction
AIP/NA43 yearsAlive at 13 monthsWithin 3 weeks after DLT/treated with hemin infusions twice weekly
Khanna et al.18/ Mazariegos et al.19MSUD/infant53 years/HCC and HCVAlive at 42.4 monthsNearly normal levels of BCAA with an unrestricted diet
Mohan et al.20MSUD/7 days33 months/biliary cirrhosisAlive at 12 monthsNormal levels of leucine with an unrestricted diet
Mazariegos et al.19/ Gopasetty et al.21MSUD/NA65 years/primary sclerosing cholangitisAlive at 38.6 monthsBCAA homeostasis maintained with an unrestricted protein diet
MSUD/NA22 years/progressive familial intrahepatic cholestasisAlive at 25.8 monthsBCAA homeostasis maintained with an unrestricted protein diet
MSUD/NA18 years/cystic fibrosisAlive at 18.48 monthsBCAA homeostasis maintained with unrestricted protein diet
MSUD/NA20 years/congenital hepatic fibrosisAlive at 20.87 monthsBCAA homeostasis maintained with an unrestricted protein diet
Gopasetty et al.21MSUD/NAPBCAlive at 5.29 monthsBCAA homeostasis maintained with an unrestricted protein diet
MSUD/NAEmbryonal carcinomaAlive at 2.66 monthsBCAA homeostasis maintained with an unrestricted protein diet
Popescu et al.22, 23HFHC/7 years46 years/HCC and HBV cirrhosisAlive at 116 monthsProgressive increase in the serum cholesterol level treated with an HMG-CoA reductase inhibitor and a cholesterol absorption inhibitor
Liu et al.24HFHC/1 year60 years/multicentric HCC and HBV cirrhosisAlive at 24 monthsSlight increase in the serum cholesterol level treated with an HMG-CoA reductase inhibitor and a cholesterol absorption inhibitor

Primary Hyperoxaluria

PH is a rare autosomal recessive metabolic disorder that is characterized by a defect in alanine-glyoxylate aminotransferase (AGT), which is encoded by a gene that is expressed only in the liver.25 This deficiency results in liver oxalate overproduction, hyperoxaluria, calcium oxalate deposition, nephrocalcinosis, and end-stage renal failure. Fifty percent of all patients reach end-stage renal disease (ESRD) by 25 years of age, and in these patients, conventional dialysis is ineffective because it does not clear sufficient amounts of oxalate.26

The treatment for these patients is usually combined liver-kidney transplantation.27 PH is a Model for End-Stage Liver Disease exception for organ allocation in Eurotransplant. We have found few reported cases of DLT with PH livers.5, 13-15 Saner et al.16 published 5 DLT cases with PH livers in the Eurotransplant region. All these domino recipients developed dialysis-dependent kidney failure (KF) despite good liver function within the first 4 weeks. Four of the 5 patients died, and the only survivor underwent retransplantation because of hepatic artery thrombosis.

Farese et al.15 reported a DLT case in which a compound heterozygous PH liver (with a G508A mutation in the AGXT gene, which encodes AGT) was used for a 64-year-old patient with locally advanced cholangiocarcinoma. Postoperatively, the recipient developed dialysis-dependent KF with hyperoxaluria.

Franchello et al.14 described a DLT case in which a PH type 1 liver was used for a 50-year-old recipient with HCC and cirrhosis; the result was a rapid onset of renal failure at 2 months requiring DDLT. The recovery of renal function was slow.

In a case reported by Casas-Melley et al.,5 a split graft was used in a 23-day-old boy with fulminant hepatic failure secondary to veno-occlusive disease as a bridge to definitive LT. The patient showed transient deterioration of renal function 3 months after DLT. He received segments II and III of a liver from a 7-year-old child with primary oxalosis. At 4 months of age (ie, when the recipient had reached an adequate weight for living donor transplantation), retransplantation was performed because of increasing renal injury from hyperoxaluria.28

Donckier et al.13 reported a 69-year-old patient with cirrhosis and multifocal HCC who received a PH liver. The recovery was uneventful, but by an 8-month follow-up visit, he had developed hyperoxaluria and moderate renal failure.

In studies by Azoulay et al.29 and Farese et al.,15 we found an unpublished case of DLT involving a patient with hereditary oxalosis who had also developed dialysis-dependent ESRD (G. Mentha, unpublished data, 2011). According to the Domino Liver Transplant Registry, DLT in this case was performed at the Cantonal Hospital of Geneva in 1998.2

A case report that was published as an abstract in the proceedings of the 6th European Workshop on Hyperoxaluria in 2002 described a DLT recipient of a PH liver who developed ESRD, and regular dialysis treatment was initiated 2 years later.30

The postoperative course of PH liver recipients is marked by early acute renal failure; the conclusion is that this metabolic disorder is not a good indication for DLT,15, 16 and these grafts are suggested only as a bridge to definitive transplantation. Hyperoxaluria in the donor should also be carefully considered because extrahepatic AGT in the domino recipient would be inadequate for detoxifying the de novo generated glyoxylate from a PH type 1 liver.31

Acute Intermittent Porphyria

AIP is the most common hepatic porphyria. AIP is an autosomal dominant disorder of the third enzyme in the heme biosynthetic pathway and results from a partial deficiency of hydroxymethylbilane synthase/porphobilinogen deaminase. The liver is the source of the heme precursor aminolevulinic acid, which is the major cause of neurological attack during AIP. LT represents a potentially effective treatment for severely affected patients with recurrent life-threatening neurovisceral attacks despite optimal medical therapy with human hemin.

There is a single report describing DLT with AIP donors.17 DLT was performed with livers from 3 premenopausal women with AIP. The explanted livers were expected to be good grafts for older men, who would have a lower risk for clinical porphyria because the clinical expression of AIP is more common in premenopausal women.

Men who were 43, 60, or 65 years old, had chronic liver disease and associated hepatocarcinoma, and were outside the standard criteria for LT received these liver grafts. The outcomes showed that livers from donors with AIP are neurotoxic when they are transplanted into recipients without porphyria, and acute attacks can result from the production of neurotoxins by the liver.17

Maple Syrup Urine Disease

MSUD is an autosomal recessive metabolic disorder that is characterized by impaired activity of the branched-chain α-keto acid dehydrogenase complex, which results in an accumulation of branched-chain L-amino acids (BCAAs) and α-keto acids. The clinical course is marked by episodes of ketoacidosis with neurotoxic effects. The treatment consists of supplementation with thiamine and, in most patients, a strict diet with a reduced intake of protein and branched amino acids. However, even under these conditions, subsequent complications such as brain damage and death are often reported.32

LT has been performed for some patients with MSUD. LT corrects the BCAA levels, eliminates the metabolic crisis, and improves the long-term outcomes of patients with MSUD.33 The first 2 LT procedures for patients with MSUD were performed for nonmetabolic reasons: the development of acute liver failure secondary to a hepatitis A infection34 and hypervitaminosis A.35 The serum marker of MSUD is alloisoleucine, which is also used for treatment control.36 Strauss et al.33 reported 11 children with classical MSUD who had undergone DDLT. All 11 patients were alive after a median follow-up period of 13.8 months (range = 4 months to 8.6 years) and had normal BCAAs with unrestricted protein intake. In 2011, an updated series of 37 patients with MSUD who underwent DDLT was reported from the Children's Hospital of Pittsburgh along with 17 United Network for Organ Sharing cases.37 The follow-up period was 4.5 ± 2.2 years. Mazariegos et al. reported 5 DLT cases using livers from patients with MSUD in 200819 and more cases in 201021 (Table 1). The patient and graft survival rates were 100% at a mean of 25 months (range = 3-62 months), and all the livers functioned normally. BCAA homeostasis was maintained with an unrestricted protein diet. These recipients of MSUD livers maintained nearly normal levels of plasma amino acids, and this demonstrated that they did not develop the disease and exhibited a favorable evolution.18 This phenomenon can be explained by the fact that these domino recipients maintained their normal extrahepatic oxidation of leucine. Because branched-chain α-keto acid dehydrogenase is expressed in extrahepatic tissues, DLT with livers explanted from patients with MSUD is feasible, but more studies are required.18, 38

Mohan et al.20 reported DLT with a graft from a 22-month-old MSUD patient for a domino recipient who was 33 months old and had biliary cirrhosis caused by Langerhans cell histiocytosis (the world's youngest recipient in the International Domino Registry). One year after transplantation, both children were doing well on a normal diet and had normal leucine levels and liver function tests.

In summary, the use of MSUD liver explants may be a good option for patients on the waiting list who meet the criteria for DLT.33


FHC is an autosomal codominant metabolic disorder due to a mutation in the low-density lipoprotein gene39; this mutation causes a reduced number of functional low-density lipoprotein receptors (LDLRs) on the cell membrane. Defects in the genes for apolipoprotein B and proprotein convertase subtilisin/kexin type 9 have also been identified in patients with FHC.40 In homozygous cases of FHC, there are no or few LDLRs; the disease is present from birth, and the serum cholesterol levels often exceed 1000 mg/dL. Consequently, the child develops severe systemic atherosclerosis and dies from myocardial infarction before the age of 20 years.40, 41

HFHC patients must undergo either portocaval shunting42 or LT.43-46 Low-density lipoprotein apheresis is another option for these patients, but it is marked by multiple complications (eg, difficulties with the long-term maintenance of vascular access and poor quality of life due to repeated procedures). LT, which was introduced as a curative procedure for HFHC by Starzl et al. in 1984,46 is based on the large proportion (50%-75%) of LDLRs in the liver.44, 45 LT must be performed as early as possible for HFHC before the development of severe atherosclerosis and coronary heart disease; otherwise, combined heart-liver transplantation must be performed.46 In a recent article,47 Maiorana et al. indicated that preemptive LT performed before the onset of cardiovascular disease is the only definitive therapy for this disease.48, 49 The indications for DLT with a graft from a patient with HFHC are rare; in fact, only 2 published cases have used a liver from a patient with HFHC. The first procedure was performed at our center and was reported by Popescu et al.22 A liver from a 25-year-old patient with HFHC was transplanted into a 46-year-old patient with HCC and hepatitis B virus (HBV)–related liver cirrhosis in 2001. The immediate postoperative course was uneventful. Three years after transplantation, an autologous CD34+ cell transplant was performed for better control over the patient's hypercholesterolemia. Seven years after LT, the donor was in excellent shape socially and professionally after kidney transplantation and stenting for coronary artery disease. Despite an elevated level of serum cholesterol (339 mg/dL) that was partially controlled by medication and diet, the recipient showed no signs of ischemic heart disease (this was verified by a normal coronary angiogram).23 Ten years after DLT, the patient was still free of disease with no recurrence of HCC or complications related to HFHC (I. Popescu, unpublished data, 2011).

The second case occurred in Taiwan in 2007.24 A liver graft from a 17-year-old patient with HFHC was transplanted into a 60-year-old patient with multicentric hepatocarcinoma and HBV-related liver cirrhosis.50 The donor and recipient follow-up at 10 months was satisfactory; the total serum cholesterol level of the DLT recipient was 300 mg/dL.

The positive outcomes of these patients can be explained by the activity of extrahepatic LDLRs. Therefore, each recipient candidate should ideally be screened for extrahepatic receptor activity; in reality, however, this is difficult and impractical. It is also important to use medications [eg, a 3-hydroxy-3-methyl-glutaryl-coenzyme A (HMG-CoA) reductase inhibitor and a cholesterol absorption inhibitor] and to modify the diet appropriately after DLT.

The encouraging outcomes of the 2 reported cases support the idea of using liver grafts from HFHC patients. Although only 2 cases of DLT with liver grafts from HFHC patients have been published worldwide in the English literature, the number of unpublished cases may be higher. There have been at least 4 additional cases in Turkey (Y. Tokat, unpublished data, 2011).

LT is an increasingly popular therapeutic option for many liver-based inborn errors of metabolism. These explanted, metabolically dysfunctional livers create new categories of potential donors and recipients.

In conclusion, the different outcomes of DLT depend on the underlying diseases of the donors. The risk of each metabolic disorder must be assessed in the context of each recipient. DLT is feasible but requires careful planning for the surgical procedures used in liver explantation and the proper selection of recipients. Although their use is limited so far, grafts from HFHC patients can offer satisfactory results to the appropriate recipients. The outcomes of DLT with HFHC and MSUD liver grafts (which include the risk of the de novo development of these genetic diseases) are promising. The results with livers from PH and AIP patients are still debatable, and it has been suggested that PH livers be used only as a bridge to LT or for stringently selected recipients who are excluded from regular allocation. The reason for this restriction is the onset of metabolic disease in DLT patients with PH or AIP liver grafts; this occurs much sooner in these patients versus those with HFHC or MSUD liver grafts. For rare metabolic liver diseases to be established as indications for DLT, larger numbers of patients with long-term follow-up must be studied. Until then, the decision to use this method will remain individualized and center-oriented.