Auxiliary Liver Transplantation for Propionic Acidemia: A 10-Year Follow-Up


* Corresponding author: Mohamed Rela,


Orthotopic liver transplantation (OLT) is an established treatment for patients with liver-based metabolic disorders that produce structural and functional impairment. Auxiliary liver transplantation (ALT) has been proposed as an alternative approach due to the potential advantage of preserving the native liver that could be used for future gene therapy and also serves as a back-up should the graft fail. The aim of our study was to determine if ALT has the long-term potential to correct the underlying abnormality in propionic acidemia (PA). A retrospective analysis was performed on graft function, metabolic parameters and effects on development in a child who underwent ALT for PA at our institute. The clinical and biochemical parameters are near normal with no diet restrictions and with good graft survival. A normal growth and an acceptable neurological and psychomotor development were achieved in the child. ALT is feasible and provides adequate liver mass to prevent metabolic decompensation in PA.


Propionic acidemia (PA) is an autosomal recessive disorder caused by deficiency of propionyl-CoA carboxylase, the enzyme that converts propionyl-CoA to methymalonyl-CoA and subsequently to succinly-CoA that enters the krebs cycle and contributes to energy metabolism. This conversion occurs in the mitochondria and is part of the pathway for the degradation of the amino acids isoleucine, valine, threonine and methonine, odd chain fatty acids and cholesterol (1). Defects in this pathway result in the production of potentially toxic metabolites, such as propionic acid and methylmalonic acid.

The affected infants present with feeding intolerance, vomiting and lethargy progressing to coma from 16 h to weeks after birth, depending on the severity of the enzyme impairment. Metabolic acidosis with ketosis and hyperlactatemia is the biochemical hallmark with associated hyperammonaemia precipitated by protein ingestion or inter-current infection (2,3). Hamilton et al. suggested that the acid metabolites and high levels of ammonia may cause a selective damage to the basal ganglia (4).

Although dietary modifications avoiding essential amino acids are helpful in the initial management of these children, there is a well-documented increase in mortality and morbidity as these children grow up, particularly from the middle of the first decade even on treatment (5). In addition, management is expensive, stressful for family members and requires close medical and biochemical monitoring. Liver replacement only partially corrects the enzyme deficiency, but it was shown that it can offer protection from systemic metabolic decompensation and improves the quality of life (6–8). Auxiliary liver transplantation (ALT) has been reported as an alternative to whole liver replacement in noncirrhotic inborn errors of metabolism based in the liver (9). This technique has potential advantages over the conventional orthotopic liver transplantation (OLT) as part of the native liver is left in place and can function if graft function is lost and gene therapy is retained as a possibility in the future. ALT is technically and practically feasible for disorders such as Crigler–Najjar syndrome type-1(CNS1) as the required function of conjugating bilirubin requires relatively little functional liver mass (10,11). ALT for PA has not been reported. We report such a case with a 10-year follow-up.

Patients and Methods

A retrospective analysis of clinical, metabolic and developmental parameters in a child who underwent ALT for PA over a period of 10 years is as follows.

Case Report

A 3-week-old baby presented to a local hospital in 1993 with a history of failure to thrive, drowsiness and constipation. Blood and urine analysis suggested signs of PA, which was confirmed on skin fibroblast culture. She was started on a low protein diet of 1g/kg/day to reduce the load on amino acid catabolism, L-carnitine supplementation, which conjugates with propionate to aid its excretion via the kidney and metronidazole to reduce the bowel flora and to minimize further production of propionate from the gut. Her condition improved and she was discharged. Subsequently, she presented on eight occasions in the first year of life with metabolic decompensation requiring intensive care unit admissions.

In view of her unstable disease and that she was in the lower end of the normal development range with relatively good neurological function, she was referred to our institution for liver transplantation. ALT was performed in June 1995. At surgery the native liver was resected by an extended right hepatectomy and the extended right lobe including segment IV of the donor was implanted as shown in Figure 1. The donor supra hepatic vena cava was piggybacked onto the enlarged hepatic venous orifice, the infrahepatic vena cava was ligated and the donor portal vein was anastomosed end-to-side to the right portal venous stump of the recipient. The donor hepatic artery was anastamosed to a previously constructed infra-renal iliac arterial conduit. A prolene tie was used to narrow the portal vein by 60% and reduce the inflow into the native liver and favor flow to the graft. A Roux-en-Y hepatico-jejunostomy was performed.

Figure 1.

Illustration of ALT showing the native and transplanted liver.

The postoperative period was uneventful with no episodes of metabolic or hepatic decompensation. The immediate postoperative ammonia was within normal limits (28 umol/L), metronidazole and L-carnitine supplements were stopped, and she was discharged home after 3 weeks with a normal dietary protein intake. An episode of increased ammonia levels was seen in the year 1999 due to recurrent ear infections requiring grommet insertion—this shows that the underlying metabolic abnormality is not fully corrected. There were no other episodes of metabolic decompensations and she remains under long-term follow-up.


The immediate postoperative liver functions were normal with patent vessels—an episode of mild rejection was treated with steroids. Serial serum ammonia and carnitine levels are shown in Figure 2. A spike in propionyl carnitine was associated with upper respiratory tract infection. Clinical and biochemical parameters are near normal with no diet restrictions. There have been no further episodes of metabolic decompensation. Current immunosupression includes mycophenolate mofetil 360 mg twice a day, cyclosporin 30 mg twice a day and 1 mg of prednisolone. A recent CT scan showed good graft survival as shown in Figure 3. Although a delay in mental development with a general quotient of 84 was noted at 32 months of age, she had normal growth and acceptable neurological and psychomotor development with no further deterioration. A detailed neuropsychological assessment at a recent follow-up reported her mental age equivalence ranging from 6 to 8 1/2 years at 12 years and 4 months of physical age. She is attending a secondary school with one-toone help.

Figure 2.

Serial serum ammonia and carnitine levels with spikes corresponding to upper respiratory tract infections.

Figure 3.

A recent CT scan showing no signs of graft atrophy in the transplanted liver.


PA is a rare autosomal recessive disorder caused by deficiency of the enzyme propionyl-CoA carboxylase. Although the incidence is approximately 1 per 100,000 live births in the United States, it has been reported to be as high as 1 per 2,000 to 1 per 5,000 live births in Saudi Arabia (12). PA manifests with clinical signs and symptoms of acidosis, hyperammonia, lethargy, irritability, shock, coma and death. The mainstay of conservative management is dietary protein restriction, metronidazole and L-carnitine supplements. Despite the improvement in dietary therapy, conservative treatment can be associated with episodes of severe metabolic decompensation and the overall long-term outcome is disappointing. In a large case series involving 30 children with PA, Lehnert et al. reported that 70% had died before they reached the age of 8 years, and only 10% had a normal neurologic development (13). Liver transplantation is an established treatment for patients with liver-based metabolic disorders and correction of propionyl-CoA carboxylase deficiency by liver replacement seems to improve clinical and biochemical parameters (14). OLT for PA was first attempted in a 26-month child in 1992 in Birmingham, UK, but he died in the early postoperative period. Following this there were 10 reported cases of OLT for PA but only with partial success (15). The overall mortality rate following OLT was 36% (4/11). Among the children who survived for more than 30 days, three developed long-term graft dysfunction and died subsequently with metabolic decompensation. We hypothesize that poor results of OLT may be secondary to hepatic decompensation during the rejection episodes in the post-transplant period, which is complicated by an already existing metabolic decompensated state. During the recent years ALT was successfully used to correct metabolic liver disorders like adult-onset type II citrullinemia, methylmalonic or urea cycle defects and CNS1. It has been shown from animal experiments that only 1–2% of normal hepatocytes are needed for conjugation of bilirubin in CNS1 (16). The rationale of ALT in such patients is to provide a liver segment with normal enzyme activity to correct the metabolic abnormality, and removal of the whole native liver is not necessary. The advantage is that should the donor graft fail, it can be removed without endangering the life of the recipient and also the remnant liver acts as a back-up preventing the cumulative effect of hepatic decompensation with an existing metabolic abnormality. We performed the first ALT for PA in 1995 and the early success was briefly mentioned in a short report on our experiences with auxiliary liver transplantation. An extended right lobe was used to provide maximum liver volume to correct PA  as there is no evidence as in CNS1 with regard to the approximate volume of liver sufficient to correct the underlying metabolic defect. Following ALT, the metabolic parameters returned to near normal with no further episodes of decompensation and normal dietary protein intake without use of metronidazole or L-carnitine. Although she suffered a degree of developmental delay during the pretransplant period, currently she is able to care for herself and is attending secondary school and there is no worsening of the neurological condition. Graft atrophy has been reported to be a significant problem in patients who underwent ALT for non-cirrhotic inborn errors of metabolism. A preferential portal flow to the native liver or possible lack of hepatotrophic substances in the presence of the normal liver mass was proposed to be the possible reason for such a phenomenon (17). Narrowing or even complete division of the portal supply to the recipient liver appears to be the treatment option for the prevention of graft atrophy, but we have not divided the portal vein as it defeats the purpose of ALT preventing the future possibility of gene therapy. There were relatively few problems over the 10-year period with liver transplant with no signs of graft atrophy. The quality of life is improved and a near functional cure is achieved with ALT. Living donor liver transplantation (LDLT) for PA was performed in three cases in Japan; however, the metabolic abnormality was only partially corrected and patients received L-carnitine supplements routinely (18). In part this may be because heterozygous carriers were the donors, although the carriers are normal they may show half-normal enzyme levels as seen with hexosaminidase-A levels in the Tay–Sachs disease (19). PA is a mitochondrial-based metabolic disorder unlike CNS 1. Liver replacement cannot completely correct PA: the likelihood that gene therapy could be successful in future is a motivation for ALT. Our case proves that ALT can also provide adequate liver mass to partially correct the abnormality with an advantage of preserved native liver.


We conclude from our experience that ALT provides adequate liver mass to prevent metabolic decompensation in PA even though it only partially corrects the underlying metabolic disorder. Long-term graft atrophy was not a problem in our patient.