Cholestatic Syndromes

Children with BA constitute approximately 50% of the pediatric liver transplant population. Hepatoportoenterostomy (the Kasai procedure) should be the primary surgical intervention for all infants with BA unless the initial presentation is late in infancy (>120 days of age), the liver biopsy shows advanced cirrhosis, or the clinical course is unfavorable. In these rare patients, primary LTx is indicated [3]. Approximately 20% of patients with BA who undergo the Kasai procedure may not require LTx during childhood [4]. The sequential use of the Kasai procedure and LTx optimizes overall survival and organ utilization [5].

Clinical failure of the Kasai procedure can manifest as persistent bilirubin >2 mg/dL three months post-Kasai or various complications such as portal hypertension with refractory ascites or variceal bleeding even in patients with successful biliary drainage. Malnutrition, and/or progressive hepatic synthetic failure, or recurrent bacterial cholangitis are also indications to consider transplantation. Approximately 50% of all infants with BA will have one or more of these complications and will require LTx within the first two years of life. Later complications such as hypersplenism with variceal hemorrhage or complications of portal hypertension such as hepatopulmonary syndrome typically lead to LTx after the child is over two years of age.

Other cholestatic conditions, such as PFIC or Alagille syndrome may also cause cirrhosis with synthetic liver dysfunction and/or portal hypertension causing variceal bleeding. In children afflicted with these conditions, indications for transplant also include intractable pruritus, marked osteodystrophy with recurrent fractures recalcitrant to medical therapy or rarely hepatocellular carcinoma [6].

Hepatic-Based Metabolic Disease

A leading indication for liver transplantation in children is hepatic-based metabolic disease of which alpha-1 antitrypsin deficiency is most common. Please refer to relevant chapters for additional information on each topic. The frequency of LTx for urea cycle defects (UCD) and organic acidemias has also increased substantially over the past decade [7]. In metabolic patients, LTx can be lifesaving when end-stage liver disease results, but can also accomplishes phenotypic and functional cure of the disease. Liver replacement corrects the metabolic defect and should be considered before other organ systems are significantly or irreversibly affected, or irreversible quality-of-life compromises or complications develop that would preclude transplantation. Patients with UCD, for example, who have repetitive hyperammonemic crises, sustain significant neurologic injury with resulting developmental disability. Early transplantation allows the potential for neurologic protection and recovery with preservation of neurologic function and quality of life. The use of living donors who are heterozygous carriers of UCD has also been successful and allows planned early transplantation [8].

Patients with tyrosinemia historically had a high risk of developing liver failure or hepatocellular carcinoma (HCC), requiring LTx before extrahepatic spread of the tumor occurred. Currently, 2-(2-nitro-4-trifluoromethylbenzoyl)-1, 3-cyclohexanedione (NTBC) or nitisinone is considered to be first-line therapy for tyrosinemia. Nitisinone, through inhibition of 4-hydroxyphenylpyruvate oxidase, prevents dysplasia and HCC, thus obviating the need for LTx.

Gestational alloimmune liver disease (GALD), also known as congenital alloimmune hepatitis and formerly known as neonatal iron storage disease [9], presents a challenge in diagnosis and management. Most patients present at birth or within weeks with acute liver failure with hypoglycemia, marked coagulopathy, but minimal elevation of liver enzymes [10]. Diagnosis is made by MRI demonstration of extrahepatic iron deposition or by buccal biopsy demonstrating hemosiderin deposition within the salivary glands. The use of exchange transfusion and IVIG therapy has shown promise by increasing survival and reducing the need for LTx [11]. Of the patients who do respond to medical therapies, hypoglycemia often improves within 24 hours of starting treatment, although coagulopathy does not normalize for several weeks. Nevertheless, survival is poor among live-born infants with liver failure, both among those medically treated as well as infants who have undergone liver transplantation.

Acute Liver Failure

Patients with acute liver failure (ALF) pose diagnostic and prognostic difficulties [12]. Rapid clinical deterioration frequently makes establishment of a primary diagnosis impossible before the need for urgent transplantation. The most common cause of ALF in children is a presumed, yet unidentified, viral illness; this is followed in relative incidence by drug toxicity, toxin exposure, and previously unrecognized metabolic disease. In infants, causes of ALF such as GALD or mitochondrial diseases must be considered.

Mitochondrial respiratory chain abnormalities, representing disorders in the electron transport proteins, present as ALF or as progressive liver disease with sudden decompensation [13]. Mitochondrial diseases are especially important to recognize during the evaluation process, if possible, as they represent multi-organ progressive diseases, which may not become evident until after LTx. Special attention should be given to patients with deoxyguanosine kinase (DGUOK) mutations, the most common type of mitochondrial DNA depletion associated with a hepatocerebral phenotype. Current literature would suggest that these individuals may be considered for transplantation if they do not manifest any neurologic abnormalities, however if abnormalities such as developmental delay or nystagmus exist, LTx is not appropriate since LTx does not improve survival for these patients [14]. A thorough evaluation of commonly affected, high energy utilizing organs (central nervous system, cardiac or skeletal muscle, etc.) must be carried out. In patients with multisystem involvement, albeit mild, transplantation is not curative, nor indicated. Additionally, patients with valproic acid-induced ALF are not transplant candidates, as one-year survival is significantly lower in this cohort of patients, even without a clear diagnosis of mitochondrial disease [15].

Patients may also present with primary or secondary hemophagocytic lymphohistiocytosis (HLH)-induced ALF. In children with HLH, inappropriate macrophage and natural killer cell activation causes severe hepatocyte injury. Epstein–Barr or other viruses may trigger the acquired form in an immunocompromised host. The primary treatment is hematologic, rather than organ replacement, as recurrence in the allograft frequently occurs, is suboptimal and data limited to small case reports [16].

Selection of ALF candidates for transplantation is difficult, as the natural history of each specific etiology is not clearly established. The King’s College Institute of Liver Studies has developed a scoring system for children with ALF, stratifying their risk [17]. Factors predictive of poor outcome included INR>4, serum bilirubin > 235 µmoles/L (13.8 mg/dL), age < 2 years, and WBC > 9,000/mm3. Sensitivity and predictive ability increase when multiple factors are present. Similarly, poor prognostic factors in other independent studies include the time to onset of encephalopathy >7 days, prothrombin time >55 sec, ALT < 2,384 IU/L on admission, grade 4 encephalopathy, infants <1 year, or the need for dialysis [18, 19]. Although these criteria are helpful, careful observation for progression and clinical change including worsening encephalopathy remains most valuable.

The role of liver biopsy is limited in ALF as these patients have marked coagulopathy necessitating a transjugular or open approach. Sampling error may also limit the accuracy of diagnosis or prognosis. When autoimmune hepatitis is considered, however, biopsy may be helpful as individual patients may respond to immunosuppression.

Patients who present with grade III–IV encephalopathy or those with progression to severe encephalopathy have worse outcomes [20]. When candidates undergo transplantation before developing irreversible neurologic abnormalities, survival improves greatly [20]. Patients who undergo LTx or recover spontaneously from ALF, however, may have suboptimal neurologic outcomes. Failure to maintain adequate cerebral perfusion pressure (mean arterial blood pressure (MAP) minus intracranial pressure [ICP]) of >50 mm Hg and an ICP <20 mm Hg) has been associated with very poor neurologic recovery post-transplantation [21].

For short-term stabilization, repetitive courses of plasmapheresis may be performed to ameliorate the clinical manifestations of ALF. Daily exchanges of volume equal to the extracellular volume (20% of body weight) are undertaken. Replacement fluids include fresh frozen plasma, platelets, and cryoprecipitate as needed. This will lead to temporary correction of coagulopathy and biochemical improvement. Plasmapheresis is also beneficial in preventing fluid overload that can result from factor replacement alone. Neurologic improvement is common but not sustained. There is no evidence that plasmapheresis enhances native liver recovery. Ultimately, LTx is the only effective treatment modality.

Liver Tumors

Liver tumors represent up to 10% of LTx cases. Liver malignancy can be primary, metastatic, or result from an underlying liver condition such as hepatocellular carcinoma in tyrosinemia or cholangiocarcinoma in sclerosing cholangitis. Children have undergone transplant for (1) hepatoblastoma; (2) hepatocellular carcinoma (HCC); (3) infantile hepatic hemangioma (IHH) with associated heart failure or Kasabach–Merritt syndrome; (4) metastatic neuroendocrine tumor (NET); (5) cholangiocarcinoma; or (6) sarcomas (biliary and vascular origin).

Hepatoblastoma (HBL) is the most common liver tumor requiring LTx. Children with biopsy-proven hepatoblastoma PRETEXT III or IV (SIOPEL classification) in which tumor size or central location prevents complete resection, or who have metastatic disease, should be referred to a transplant center for evaluation and multidisciplinary care. In the USA, these children, who have no lung metastases (before or after chemotherapy) qualify for additional priority (status 1B) on the waiting list. Children with metastatic pulmonary lesions may be considered for transplantation without 1B priority if pulmonary metastases remain after chemotherapy, although surgical resection of the metastases pre-LTx may be required.

Complete tumor resection remains the most crucial intervention to achieve long-term survival. According to the current Children’s Oncology Group algorithm, any tumor that remains unresectable after several rounds of chemotherapy requires complete hepatectomy with transplantation. Attempts at partial liver resection are ill-advised as survival rates after “rescue transplant” are far inferior. [22] For unresectable lesions, transplantation occurs ideally before completing chemotherapy, to allow an additional one to two cycles of chemotherapy post-LTx.

Hepatocellular carcinoma is the second most common hepatic malignancy in the pediatric population, primarily affecting children ten years or older who do not have metabolic liver disease [23]. Given the low incidence of HCC in children, experience with LTx for HCC was limited, but is rapidly increasing. HCC can present in patients with known liver disease and cirrhosis, but 20–60% of pediatric HCC occurs in non-cirrhotic livers. Important predisposing factors include prenatally acquired HBV and metabolic liver diseases such as tyrosinemia and PFIC II and IV. Glycogen storage disease (type 1), and portosystemic vascular (Abernethy) malformations each lead to adenomatous changes with infrequent transformation to HCC. Recent data showed HCC in Abernethy malformations typically develops later in life [24]. Screening for HCC is indicated in these conditions [24, 25]. Rare reports of HCC in patients with BA and Alagille also exist.

In children with HCC confined to the liver and associated metabolic disease, LTx is indicated with 1B prioritization. Although adults must fit within the Milan or UCSF criteria, cumulative data from the Pediatric Liver Unresectable Tumor Observatory registry suggests these criteria do not apply to pediatric HCC that is not associated with metabolic disease. Extrahepatic metastases remain an absolute contraindication to transplant. Chemotherapy alone does not appear to offer a survival advantage, unless complete tumor resection is also offered.

Additional liver tumors that may lead to LTx consideration include biopsy-proven embryonal sarcoma that has not metastasized or vascular malformations with Kasabach–Merritt syndrome or high-output cardiac failure requiring pressor or ventilatory support. Current policy suggests these children may receive 1B priority, but inadequate evidence exists currently to support additional priority for rhabdoid tumors or angiosarcoma.

Candidates with metastatic neuroendocrine tumor NET also benefit from LTx in select circumstances [26]. The primary malignancy must result from gastro-entero-pancreatic origin with portal system drainage and this and any extra-hepatic disease must be resected without recurrence for six months. Primary NET from other areas do not receive the same priority. Extra-hepatic solid organ metastases (i.e., bones, lungs) are permanent exclusion criteria [27].

Candidates with cholangiocarcinoma (CCA) must meet strict criteria as outlined in the OPTN policies [28]. This includes having an unresectable lesion after receiving chemotherapy followed by operative staging that excludes regional, intra- or extrahepatic metastases. Diagnosis of CCA requires a malignant appearing stricture on cholangiography and at least one of the following: biopsy or cytology with malignancy, CA-19–9 >100 U/mL in absence of cholangitis, or aneuploidy. Endoscopic ultrasound aspiration of regional hepatic lymph nodes may identify patients with obvious metastases, but percutaneous/transperitoneal biopsies of the primary tumor should be avoided due to seeding risk.

Cystic Fibrosis

Prolonged survival of patients with cystic fibrosis (CF) has increased consideration of selected individuals for liver transplantation. Liver disease occurs in up to one-third of children with CF depending on definition used. Significant portal hypertension develops in approximately 10% [29]. Direct management of portal hypertensive variceal bleeding by variceal banding and portosystemic shunting allows prolonged survival, but in the uncommon patient with hepatic synthetic decompensation, LTx should be considered. Signs of advanced portal hypertension, including hepatic encephalopathy (HE), diuretic refractory ascites, and recurrent variceal bleeding, are late complications that warrant referral for LTx evaluation. Traditionally, malnutrition and poor growth were treated as an indication for LTx. Data, however, does not show uniform improvements in this scenario, however, for which we recommend advanced portal hypertension as the primary LTx indication [30]. Combined liver-lung transplantation has been performed for patients with pulmonary compromise (FEV1 <50%) and because of the increased mortality rate without transplantation, exception points are available for patients with FEV1 <40% according to UNOS policy on liver allocation in patients with cystic fibrosis [31].

Special considerations include the impact of cystic fibrosis-related diabetes and infectious considerations which may require modifications of standard antimicrobial prophylaxis. A recent review also advocated for immediate treatment of chronic sinusitis, nasal polyps, or bronchoscopy in the setting of >6 months wait time or changes in pulmonary status to evaluate the respiratory flora [29]. Optimizing pulmonary status before and preparing for needed respiratory therapies after are vital. Postoperative long-term survival is compromised primarily by cardio-pulmonary events, but the increased risk of polymicrobial or fungal sepsis and potential malnutrition also influences survival.