|Home | About | Journals | Submit | Contact Us | Français|
Biliary Atresia is a progressive, fibro-obliterative disorder of the intra and extrahepatic bile ducts in infancy. The majority of affected children will eventually develop end-stage liver disease and require liver transplantation. Indications for liver transplant in biliary atresia include failed Kasai portoenterostomy, significant and recalcitrant malnutrition, recurrent cholangitis, and the progressive manifestations of portal hypertension. Extra-hepatic complications of this disease, such as hepatopulmonary syndrome and portopulmonary hypertension, are also indications for liver transplantation. Optimal pre-transplant management of these potentially life threatening complications and maximizing nutrition and growth require the expertise of a multi-disciplinary team with experience caring for biliary atresia. The timing of transplant for biliary atresia requires careful consideration of the potential risk of transplant versus the survival benefit at any given stage of disease. Children with biliary atresia often experience long wait times for transplant unless exception points are granted to reflect severity of disease. Family preparedness for this arduous process is therefore critical.
Biliary Atresia (BA) is a progressive, fibro-obliterative disorder of the intra and extrahepatic bile ducts with onset in the first 3 months of life. It occurs worldwide, affecting an estimated 1 in 8,000–18,000 live births(1). At least four phenotypes of BA are recognized: isolated BA, BA associated with laterality defects (asplenia, polysplenia, abdominal situs inversus and intestinal malrotation), BA associated with other major congenital malformations, and BA associated with a bile duct cyst (cystic BA). The etiology and pathogenesis of each type of BA is uncertain, however, abnormal bile duct development, perinatal viral infection, prenatal toxin exposure, and a dysregulated immune response have been proposed(1). Early diagnosis of BA is critical as a surgical Kasai portoenterostomy (KPE) may restore bile flow if performed prior to age 3 months and help prevent rapid progression of liver injury and development of cirrhosis. Unfortunately, the vast majority of affected children will eventually develop end-stage liver disease, with BA being the leading indication for pediatric liver transplantation.
The natural history of children with BA who do not demonstrate bile flow after KPE is similar to that of children with unrepaired BA. Very few will survive beyond 24 to 36 months of age without undergoing liver transplantation (2–4). Success of the KPE can best be judged by restoration of bile flow and clearance of jaundice. At 3 months post KPE, there is a clear difference in 2 year transplant-free survival between children with total serum bilirubin <2 mg/dL and those with total bilirubin >6 mg/dL (84% vs. 16%, p<0.001) (5). Likewise, if jaundice has resolved by 3 months post KPE, the 10-year transplant-free survival rate ranges from 75% to 90%; conversely, if jaundice persists after KPE, the 3-year transplant-free survival rate is only 20%. Thus, children who do not demonstrate good bile flow and clearance of jaundice by 3 months post KPE should be evaluated early for transplantation, ideally by 6–9 months of age.
While the benefits of early performance of KPE are well established, the concept of a threshold age beyond which KPE is futile, thus requiring proceeding directly to liver transplantation, remains controversial. Patients undergoing a “late KPE”, variably defined as age >90, 100 or 120 days, have diminished but variable long-term survival with their native liver. Previous studies have reported 42% two year, 23–45% four to five year, 15–40% ten year, 29% fifteen year, and 13% twenty year survival with native liver (6–10). Neither advanced histologic fibrosis nor nodular appearance of the liver at the time of KPE reliably predict outcome after surgery (9, 11, 12). However, KPE in infants with cirrhosis and ascites may precipitate hepatic decompensation. Several studies have suggested that patients with a previous KPE are at increased risk for bowel perforations and biliary complications from liver transplant while others showed no increased length of the transplant operation, blood loss, operative complications, or intensive care unit and hospital length of stay (8, 13–18). While future research may help identify a reliable early biomarker that predicts which child should undergo late KPE versus move towards primary liver transplant, current practice varies by center.
Children with poor bile drainage following KPE will uniformly develop significant malabsorption, protein-energy malnutrition, growth failure and developmental delay (19). A subset of those with normalization of serum bilirubin will also suffer from these consequences, albeit variably delayed, despite good bile flow. The resultant loss of lean body mass and subcutaneous adipose tissue are best demonstrated by triceps skinfold and mid-arm circumference measurements (20). In addition, biochemical and clinical deficiencies of fat-soluble vitamins, iron and zinc are common and require aggressive supplementation and frequent monitoring (21). Metabolic bone disease resulting in recurrent long-bone fractures with minimal trauma may also develop as end-stage liver disease progresses, even without vitamin D or calcium deficiency. The presence of failure-to-thrive requiring aggressive nutritional support (including nasogastric feedings or parenteral nutrition) or unremitting bone disease should prompt liver transplant evaluation.
Between 40–80% of patients with KPE experience at least one episode of bacterial cholangitis before the age of 2 years, and 25% of patients experience multiple episodes (22, 23). Post-operative cholangitis has been associated with decreasing rates of 1, 3, and 5-year survival with native liver compared to children without cholangitis (92%, 76% and 76% vs. 80%, 51% and 23%, respectively (P<0.01) (24). In addition, repeated cholangitis confers a 3-fold increased risk for early failure after KPE and 2-fold increased risk for failure 3 years after KPE (25). Liver transplant should be considered when a child develops recurrent cholangitis despite aggressive antibiotic therapy, multi-resistant bacterial organisms, episodes of life-threatening sepsis, or severely impaired quality of life due to recurrent hospitalizations for cholangitis.
Although 60% of infants with BA will initially have restored bile flow after KPE, hepatic fibrosis is progressive and portal hypertension (PHT) develops in the majority of children. Manifestations of PHT, including splenomegaly with hypersplenism, esophageal and gastrointestinal varices and ascites, are associated with significant morbidity and mortality. A study from the Childhood Liver Disease Research Network (ChiLDReN; United States and Canada) characterized PHT in 163 children with BA (mean age 9.2±5.6 years) with their native liver. Definite PHT (presence of complication of PHT or splenomegaly and thrombocytopenia) or possible PHT (presence of splenomegaly or thrombocytopenia only) was identified in 67% of subjects. The most common complication of PHT was variceal bleeding, occurring in 20% of subjects, although the majority (62%) of patients had only one episode of variceal bleeding (46). In addition, PHT in BA patients who survive into adulthood without liver transplantation is nearly universal. In Canada and Europe, 96% of adult patients with BA had features of PHT, with 65% having evidence of varices, 91% splenomegaly and 14% ascites (26). These findings corroborate a previous study from France, showing that 99% of BA survivors with their native liver in adulthood had evidence of cirrhosis and 70% had significant PHT (27). Thus, the complications of PHT can generate major morbidity in both children and adults with BA and requires frequent monitoring.
Although more commonly reported in patients with Alagille syndrome and progressive familial intrahepatic cholestasis, cholestasis-induced pruritus may occur in BA. In some patients, pruritus is significant enough to greatly impair quality of life for both patient and family. In these instances, one must confirm that the diagnosis of BA is correct, that other medical causes of pruritus (such as atopy, pediculosis or urticaria) have been excluded and that maximal medical management (including ursodiol, rifampin or naltrexone, Table 2) has failed (28). Liver transplant should be considered for patients with intractable pruritus and interference with sleep or normal activities unresponsive to medical management.
HPS and POPH may occur as a consequence of BA and are indications for liver transplantation because of reversibility after, and high mortality without, transplantation (29–32). HPS is characterized by arterial hypoxemia caused by intrapulmonary vascular dilatations in patients with portal hypertension or congenital porto-systemic shunts. Hypoxia and fatigue are the main clinical symptoms and HPS is present in 3 to 20% of children with cirrhosis or being evaluated for liver transplantation (31, 33). HPS diagnosis is established by demonstrating increased right to left intrapulmonary shunting by transthoracic, saline contrast-enhanced echocardiography or macro-aggregated albumin scanning. POPH is defined as increased mean pulmonary artery pressure (mPAP) due to increased pulmonary vascular resistance identified on cardiac catheterization in a patient with normal pulmonary artery wedge pressure and portal hypertension. Exertional dyspnea, hypoxia and eventual right-sided heart failure are clinical features. POPH is rarer that HPS, present in <1% of children with cirrhosis or awaiting liver transplantation (30, 32, 34). There is no effective therapy to reverse HPS (although supplemental oxygen may improve PaO2) and although therapies for pulmonary artery hypertension may be useful to bridge a patient with POPH to transplantation, they should not substitute for liver replacement (34). BA patients should be screened at least annually and at the time of transplant evaluation with pulse oximetry and if hypoxia, dyspnea, a new heart murmur or fatigue is present, consideration should be given for transthoracic Doppler echocardiography using agitated saline (34). Severe hypoxemia (PaO2 < 45–50 mm Hg) in HPS has historically posed an increased risk for complications and mortality following liver transplantation, however more recent reports suggest transplant can be performed safely (34). POPH should be aggressively treated prior to liver transplant to reduce mPAP to < 35 mmHg. A mPAP>50 mm Hg remains a contraindication to liver transplantation at many centers because of high intra- and post-operative mortality. In summary, both HPS and POPH are indications for liver transplant in biliary atresia, however, the severity and pace of progression of these disorders dictates the urgency with which transplant evaluation should be performed.
Hepatorenal syndrome (HRS) is a rare complication of end-stage liver disease in which acute renal dysfunction secondary to diminished renal blood flow occurs in the absence of intrinsic renal disease(35). HRS generally resolves after liver transplantation(36); therefore, children who develop HRS should be evaluated and listed for liver transplantation.
Malignancy is a well-recognized complication of chronic liver disease. Although rare, hepatocellular carcinoma (HCC) may affect ~1% of children with BA and can occur as early as infancy (37, 38). Cholangiocarcinoma is even rarer. An elevated alpha fetoprotein level or concerning lesion on ultrasound should immediately lead to more definitive imaging to confirm HCC and evaluate for metastases. Complete surgical resection is the only curative option, although chemotherapy may be more effective in children than adults (39). Milan criteria may not be applicable in children and successful liver transplant outcomes have been achieved even in children who did not meet the more liberal University of California San Francisco criteria (single tumor <6.5 cm or maximum of three tumors with none >4.5 cm and cumulative tumor size <8 cm) or the “up to seven” criteria (absence of angioinvasion, number of nodules plus the maximum size of the largest nodule equal or lower than 7). Therefore, decision to list for transplant must be individualized for each child (40). Transplant should be considered in the absence of radiological evidence of extrahepatic disease or gross vascular invasion, irrespective of size of the lesion or number of lesions (41). In the most recent UNOS guidelines, children with HCC are listed with their calculated Pediatric End Stage Liver Disease (PELD) score for 6 months and then are upgraded to a PELD of 34.
Optimal pre-transplant management requires a multi-disciplinary team with experience in biliary atresia and chronic liver disease complications. The team includes pediatric hepatologists and surgeons, nurses, dieticians, social workers, feeding specialists, child psychologists or behavioral specialists, pharmacists and other specialists as needed. A number of critical issues must be addressed in every BA patient on a trajectory towards transplant (Tables 2 and and33).
One of the key predictors of successful liver transplantation in BA patients is nutritional status at the time of transplant (42) . Thus, maintaining the child’s nutritional homeostasis while awaiting liver transplantation is critical to transplant outcomes and the normal growth and development of the child. Nutritional assessment should be part of the standard of care from the time of BA diagnosis through the post-transplant period. In addition to length, weight and head circumference, routine measurement of triceps skinfold thickness and mid-arm circumference (MAC) every 3 months provides a much better indication of the child’s nutritional status and will dictate the aggressiveness with which nutritional support is initiated (Table 3). Weight gain alone can give the false impression of adequate nutrition as hepatosplenomegaly, ascites, and edema may confound this measurement. Triceps skinfold thickness (measure of adipose tissue) and MAC (measurement of lean body mass) give a better representation of the child’s status. Serum markers of nutritional status, such as visceral proteins, are problematic since liver synthetic failure may alter levels of albumin, retinol binding protein, and transferrin irrespective of nutritional status.
In the presence of cholestasis and cirrhosis, the goal for energy intake should be between 125–140% of recommended caloric requirements based on ideal body weight. Additional calories may be needed to provide for catch-up growth if a significant deficit in weight is present, as detailed in Table 3. Increasing the caloric density of formula to 24 or 27 kcal/oz (100.5 or 113 kjoule/30 mL) by mixing with less water, using glucose polymers or MCT (medium-chain triglyceride) oil can also spur growth. Breast milk can be continued and supplemented with a breast milk fortifier, however, in our experience it is challenging to assure adequate weight gain on this regimen in the presence of severe cholestasis. Thus, an infant formula containing a substantial amount of MCT oil (which is less dependent on bile acids for absorption), but also containing adequate essential fatty acids as long chain triglycerides, is preferred. Protein intake should be preserved (at least 2–4 g/kg/d) and not restricted because of the presence of cirrhosis (Table 3).
Whenever possible, oral feeding is preferred, but with increasing anorexia, vomiting because of ascites and debilitation from progressive liver disease, supplemental nasogastric feedings may be required to meet caloric and fluid requirements and prevent or reverse inadequate weight gain in children awaiting liver transplantation. The use of narrow-bore, soft, weighted Silastic or polyurethane feeding tubes is generally well tolerated, with minimal risk of aspiration or upper gastrointestinal hemorrhage (43, 44). Compared with bolus gavage feeding techniques, continuous formula infusions lead to better energy balance (43). Nutritional rehabilitation using enteral drip feedings (140% of recommended caloric intake, 4 g/kg/d of protein) in children with BA and cirrhosis awaiting liver transplant leads to improved nutritional status without hyperammonemia or adverse clinical and biochemical effects (45, 46). Nocturnal or continuous 24 hours per day nasogastric feedings can be safely administered in the home. Because of portal hypertensive gastropathy and the risk for development of gastric varices, gastrostomy tubes are contra-indicated in this setting.
Occasionally, despite adequate energy intake or because of intolerance to enteral feedings, adequate growth is not achievable using enteral nutrition. Parenteral nutrition through an indwelling central venous catheter may be required (in addition to ad lib oral feedings) while a child is bridged to liver transplantation. Delivered energy may need to be increased and standard intravenous protein and lipid loads are generally well tolerated. Copper and manganese (both normally excreted in bile) should be reduced in the parenteral nutrition solution in BA patients and zinc may be increased. In our experience, this can be done safely at home, with excellent improvement in pre-transplant nutritional status and post-transplant outcomes (47).
Intestinal absorption of fat-soluble vitamins is severely impaired in BA patients who fail to drain bile into the intestinal lumen (21). Frequent assessment of these fat soluble vitamins (Table 3) is required, particularly in those awaiting liver transplantation. Standard vitamin supplements designed for children with fat malabsorption have inadequate levels of these vitamins for BA patients with elevated serum total bilirubin (43). Large doses of individual vitamins (Table 3) are often required to prevent deficiency in BA patients with a serum direct bilirubin > 2 mg/dl, thus avoiding potentially fatal consequences such as intracranial hemorrhage (48, 49). Bile acid binding resins and iron should not be administered simultaneously with fat-soluble vitamin supplements. Serum levels and ratios (Table 3) should be closely monitored at a minimum each 3 months while awaiting liver transplant and sooner if changes in the doses of vitamins are made. Occasional patients in whom the enteral route is unsuccessful may require parenteral administration of these vitamins. Status of iron and zinc should also be monitored periodically given that subclinical gastrointestinal blood loss may lead to iron deficiency and excessive fecal and urine zinc losses in BA may lead to deficiency; oral supplementation generally resolves these deficiencies (Table 3) (21).
Complications of PHT that often require medical or procedural intervention include ascites, hypersplenism and variceal bleeding, by far the most common and potentially devastating complication of PHT. (50, 51). Up to 90% of children with BA have varices on endoscopic examination and ~30% will have at least one episode of variceal bleeding (52, 53). The decision to perform surveillance endoscopies looking for varices (primary prophylaxis) versus performing the first endoscopy at the time of a variceal bleed is quite controversial. A recent study from France characterized the incidence of esophageal and gastric varices and portal hypertensive gastropathy in 225 children with BA who had never had a variceal bleed (median age 16 months; range: 10–33 months) (48). These children underwent ≥2 repeat surveillance endoscopies within a 2.4 year time frame (range 1.5 months-10 years), of whom 72% had grade 1–2 esophageal varices at baseline and 34% had grade 3 varices or gastric cardia varices at follow-up. The authors concluded that progression of varices from the “low risk” group can be very rapid and estimated a 10% risk of emergence of a high risk endoscopic pattern over time, justifying the need for surveillance endoscopies and primary prophylaxis. In contrast, recent consensus from the Baveno VI symposia states that screening endoscopy is not universally recommended in children with cirrhosis (54). There are many arguments against performing surveillance endoscopy in BA patients who have never had a variceal bleed. First, ~50% of children with BA will require liver transplantation in the first 2 years of life, thus definitively treating the PHT and risk of GI bleeding. Secondly, in the study from France detailed above, despite primary prophylaxis with repeated sclerotherapy or band ligation, 25% of varices could not be eradicated and >30% had recurrence of varices after eradication. Thirdly, repeat endoscopies entail repeat exposure to general anesthesia. Recurrent anesthestic events have been associated with neurocognitive deficits in children (55, 56). Finally, the risk of mortality from a variceal bleed in children is rare, with mortality estimates of 0–5% based on the published literature(57–59). Despites the multiple arguments for not performing surveillance endoscopy, one potential exception would be the BA patient who lives in an extremely remote area and who would not have ready access to medical care in response to a variceal bleed.
Treatment of acute variceal bleeding includes the potential use of blood products, somatostatin analogs (e.g. octreotide) or terlipressin, sclerotherapy or band ligation, and, rarely, balloon tamponade or portosystemic shunting. Vitamin K deficient coagulopathy should be treated with intravenous Vitamin K. In addition, one should administer intravenous antibiotic therapy, in light of the high risk of potentially fatal infectious complications in cirrhotic patients with concurrent gastrointestinal bleeding (60).
Octreotide, which causes splanchnic vasoconstriction with subsequent decrease in portal venous pressure, allows for early control of variceal bleeding and hemodynamic stabilization prior to proceeding with sclerotherapy or band ligation. Esophageal variceal band ligation is the procedure of choice in children due to the low rate of re-bleeding with band ligation (4%) compared to sclerotherapy (25%), though may not be technically feasible in a child weighing less than 8–10 kilograms (61). Octreotide may be weaned off over the next 2–5 days, with the rate of weaning based on whether there is residual blood loss post-procedure. Rare complications from sclerotherapy or band ligation include esophageal perforation, ulceration or stricture formation.
Balloon tamponade of esophageal varices or portosystemic shunt surgery should be reserved for severe, refractory variceal bleeding. Balloon tamponade (Sengstaken-Blakemore tube) provides temporary hemostasis by direct compression of esophageal varices and is highly successful in stopping bleeding. However, recurrence of bleeding within 24 hours of deflating the balloon occurs in ~50% of patients (53). The risk of major complications, including esophageal rupture, has been reported in up to 20% of cases and therefore this procedure should only be performed by experienced personnel (62). Balloon tamponade should be considered a bridge to either liver transplantation or a portosystemic shunt procedure. Transjugular intrahepatic portosystemic shunt (TIPS) or, rarely, surgical portosystemic shunts (portocaval or distal splenorenal shunts) may be necessary to treat refractory variceal bleeding in BA patients (62).
The standard of care to prevent recurrence of esophageal variceal bleeding, secondary prophylaxis, includes repeat sessions of endoscopic variceal band ligation or sclerotherapy. Endoscopic therapy can be repeated every 2–6 weeks, with approximately 3–5 sessions necessary to eradicate all varices (51). Unfortunately there is a 30% recurrence rate of varices after eradication over time. Beta blockade therapy is an acceptable treatment for prevention of variceal bleeding in adults; however a paucity of data exists for any sustainable benefits in children (51, 63). Concerns regarding the use of beta-blockade in children include the lack of data to suggest appropriate dosing in order to decrease heart rate by the recommended 25% and the concern that infants may suffer poor outcomes from hypovolemic shock in the setting of variceal bleed while on beta-blockers, with the inability to mount a tachycardic response. Therefore, at present, the routine use of beta-blockade in children with significant PHT is not recommended.
For significant variceal rebleeding despite frequent endoscopic ligation, options include TIPS, portosystemic shunt or liver transplantation. TIPS is usually a bridge to transplantation in BA, however it can be considered as a long-term alternative in children with a good overall prognosis of their liver disease (51). Portosystemic shunts are rarely considered for recurrent variceal bleeding in the setting of compensated cirrhosis. Based on primarily adult data, patients with decompensated cirrhosis and ligation-refractory variceal rebleeding should not undergo portosystemic shunt surgery due to the increased risk of hepatic decompensation (64), but rather proceed to liver transplantation.
In BA patients, the most common manifestation of hypersplenism is thrombocytopenia. Neutropenia is usually without consequence as white blood cells can be mobilized if needed. Platelet transfusions should be reserved for severe variceal bleeding or clinically significant bleeding from other sources in the setting of significant thrombocytopenia (<20–60 ×109/L) (56). For persistent bleeding related to PHT and thrombocytopenia, partial splenic embolization can be considered in patients with compensated cirrhosis (65).
Ascites occurs in approximately one-third of patients with PHT. Diuretic therapy and exclusion of excess sodium intake are the mainstays of treatment in children (Table 2). For mild ascites, a potassium sparing agent such as spironolactone is recommended. In cases of moderate to severe ascites, spironolactone plus furosemide should be used, along with avoidance of high sodium containing foods. Ascites associated with low serum albumin levels may benefit from intravenous infusions of 20% or 25% albumin followed by IV furosemide. In the setting of diuretic-refractory ascites associated with respiratory compromise, feeding difficulties or urinary retention, recurrent large-volume paracentesis may become necessary. It has been recommended that the volume removed be less than 200 ml/kg body weight and that the rate of removal be less than 680 ml/hour without albumin infusion. Larger or more rapid volumes removed require albumin infusion to prevent post paracentesis circulatory dysfunction (66). Hyponatremia (serum sodium <130 meq/L) is not uncommon in BA patients with ascites. Portal hypertension results in systemic and splanchnic vasodilation, which effectively reduces arterial blood volume (67). This triggers release of arginine vasopressin and anti-diuretic hormone (ADH), solute-free water renal tubular reabsorption and retention and resultant dilutional hyponatremia (67). Management of hyponatremia can be challenging. Patients with neurologic symptoms or severe hyponatremia (<115–120 mequ/L) may benefit from significant fluid restriction. Intravenous albumin may also be helpful for short-term management. However, patients imminently headed to liver transplant may require more aggressive correction of hyponatremia with normal saline or hypertonic saline, used with caution so as not to correct serum sodium more than 9 meq/L in 24 hours to avoid the risk of central pontine myelinolysis (67). While vaptans, selective arginine vasopressin V2 receptor antagonists in the kidney tubules, hold promise for pharmacologic therapy of hyponatremia, they are currently not approved by the Food and Drug Administration for use in liver disease and cirrhosis due to the increased risk of liver failure and death (68). The recent addition of serum sodium as a factor in the calculation of the MELD score may improve the prognostic accuracy of this score.
SBP is a serious complication of PHT and usually occurs in the setting of transient bacteremia. Infants and children may have vague symptoms including poor feeding, fatigue, fevers, increased abdominal distension, abdominal pain, vomiting or diarrhea. Studies in children suggest that the most common pathogen associated with SBP is Streptococcus pneumonia followed by gram negative organisms and rarely anaerobes(69). Prophylactic antibiotics targeting the above named organisms should be used in the setting of recurrent SBP. Furthermore, children with chronic liver disease should receive the pneumococcal vaccine.
Cholangitis, one of the most common complications of BA, should be considered in any patient who develops fever, vomiting, right upper quadrant pain, pale stools, worsening jaundice, or rising aminotransferases with cholestatic laboratory values. A high index of suspicion is necessary, as fewer than 30% of patients with cholangitis will have a positive blood culture. Aggressive treatment with IV antibiotics that target gram negative and anaerobic bacteria should be administered. The use of prophylactic antibiotics such as Trimethoprim-Sulfamethoxazole (Table 2) to prevent cholangitis after KPE is common practice, however, studies of the utility of antibiotic prophylaxis have yielded conflicting results. In one study of 214 patients with BA in the Netherlands, antibiotic administration after KPE was independently associated with 4-year transplant-free survival; however, the incidence of cholangitis was not lower in the patients receiving antibiotic prophylaxis suggesting that there may have been other factors that contributed to improved survival (70), such as suppression of the intestinal microbiome and activation of innate inflammation in the liver. In a smaller study of 19 patients with BA who experienced an episode of cholangitis, there was a lower recurrence rate and higher survival rate in patients who received neomycin or trimethoprim/sulfamethoxazole prophylaxis compared to those children who received no antibiotics (71). There is a paucity of data regarding use of probiotics to prevent cholangitis. Multi-centered prospective randomized control studies are needed to better understand the utility of prophylactic antibiotics in preventing cholangitis and improving transplant-free survival after KPE (69).
In select non-cirrhotic patients who achieve sufficient bile drainage following the initial KPE, but then develop sudden onset bile flow cessation, redo/revision-KPE may be an option. In this procedure, surgical removal of fibrotic tissue is performed in an attempt to restore bile flow. Although redo KPE may theoretically make future transplantation more difficult (due to adhesions, increased blood loss, and increased surgical time), in select cases redo-KPE may offer patients long-term survival with their native liver (72–74).
Vaccine preventable infections are a recognized and potentially serious complication following pediatric liver transplant that lead to significant morbidity, mortality, and costs. The Infectious Diseases Society of America (IDSA) recommends that children with chronic liver disease receive all age appropriate vaccines based on the Centers for Disease Control (CDC) annual schedule for immunocompetent people, with a goal of completing the primary vaccine series and any necessary booster doses prior to transplantation. The CDC’s Minimal Ages and Intervals Schedule can be utilized to accelerate vaccinations for children in whom transplant is anticipated to occur in the near future(75). Live vaccines (MMR and Varicella) should be administered if transplantation is not anticipated within four weeks (76). In countries where tuberculosis (TB) is endemic, the tuberculin skin test (TST) should be part of the pre-transplant work-up; however, it should be noted that live vaccines can interfere with the TST response, therefore the TST must either be placed before/on the same day as live vaccines are given or the TST must be delayed 4–6 weeks after live vaccines are given. Despite recommendations, studies have shown as many as 71% of pediatric solid organ transplant recipients have not received their full set of standard pediatric immunizations at the time of transplant (77, 78). Future studies are needed to design and implement strategies to improve immunization rates in this population of children.
While BA is the most common indication for pediatric liver transplant, there is a paucity of data to guide optimal timing of transplant listing. Analysis of 2001–2004 UNOS data suggested that only patients with a PELD score of greater than 17 experienced considerable survival benefit in terms of life years gained from liver transplant (79). BA patients, however, are often listed with low PELD scores, primarily due to normal or near normal albumin and prothrombin time. In addition, weighting mechanisms designed to benefit infants, such as low weight z-score, are often not achievable because of artificially elevated weight due to ascites, organomegaly and aggressive nutritional interventions. Moreover, calculated PELD scores in BA may not accurately reflect true mortality risk reflected by complications of portal hypertension, variceal bleeding and refractory ascites. As such, common practice is to list BA patients when sick enough to assume the potential risk of transplant, while still well enough to both benefit and survive the surgery.
Once listed, BA patients have a median wait time in the United States of 90 days and a median calculated PELD score of 15 at the time of transplant (UNOS data). Unfortunately, the U.S. Department of Health and Human Services “Final Rule”, mandating a nationwide allocation system for deceased donor liver transplant designed to benefit patients based primarily on their medical status, has not uniformly benefited pediatric patients (80) . Recent studies demonstrate that ~15% of children with chronic liver disease have either died on the waiting list or been removed because they are too ill to transplant (81). The PELD system may be particularly tenuous for children less than 12 months of age, many of whom have BA, and in whom wait list mortality is higher (25%) than older pediatric patients (10%) and adults (82). Transplant centers have therefore increasingly appealed to Regional Review Boards for exception points on a case-by-case basis, resulting in a five- fold increase in PELD exception point use between 2002 and 2013 (83). In fact, PELD exception point utilization now occurs in the majority of pediatric liver transplants, with significant variability by region (84–86). Children < 12 months of age, including BA, have particularly high rates of exception score requests, with those listed by exception scores more likely to be white and privately insured (83). These data draw attention to shortcomings in the current allocation system for pediatric patients awaiting liver transplant and potential disadvantages for underserved populations.
Aggressive use of alternative surgical strategies can increase access to liver transplant for BA patients. Living donor transplants (LDT) may reduce not only pediatric waitlist time but also waitlist mortality (87). LDT can be timed to optimize outcomes, transplanting when the patient is in relatively good nutritional status and hopefully preserving brain and neurocognitive development. Some believe that LDT as young as 6 months of age, before patients show signs of significant malnutrition and hepatic decompensation, may avoid complications of cirrhosis and allow for even better post-transplant outcomes. In addition, LDT allows for procurement from a healthy donor with elective scheduling of the operation and shorter cold ischemia times (41, 87). Excellent patient survival, comparable to older children, is seen even in very young (< 90 days old) and small (< 10 kilograms) infants transplanted at experienced pediatric centers (88–90) . A Markov model simulation found an increase of 17.45 additional expected life years when a LDT was performed in a patient with a PELD of 15–25 with one or less systemic complication as compared to those with PELD scores >25 who had more than one systemic complications (91). LDT at experienced centers have excellent outcomes, for both donor and recipient, with similar surgical complication rates as cadaveric transplants (92–95). Unfortunately, access to LDT is not uniform as public insurers are less likely to cover donor surgeries. Increased utilization of split liver transplants, while more technically difficult, allows one donor to provide grafts for 2 recipients. Currently, only about 10% of eligible donor livers in the United States are split, missing an opportunity to expand access to transplant for BA patients (96). High pediatric waitlist mortalities emphasize the continued need to target novel approaches to increase the donor pool.
Family preparedness is critical to the long-term success of BA patients through the pre-and post-transplant periods. It is essential that the family have a clear understanding of their child’s medical condition, as well as resources available to them for support. Caregivers and patients are often plagued with feelings of inadequacy, guilt, stress, lack of control, anger and fear. Establishing a trusting relationship between the family, child, and care team can ease these burdens. Acknowledging and addressing these emotions can also positively impact both the long term family structure and disease management (97). Formal assessment tools, such as the Pediatric Transplant Rating Instrument (PTR-I), can help identify specific psychosocial risk factors for poor outcomes, including family motivation, knowledge of the transplant procedure, treatment adherence, and the quality of the relationship with the medical team (97). In addition, children with BA may have neurocognitive and expressive language developmental delays and gross motor weakness, thus benefitting from formal neurodevelopmental testing and therapeutic interventions (98, 99). Early identification of psychosocial and developmental vulnerabilities will allow for early interventions to maximize outcomes (98, 99).
Until the time arrives in which we are able to understand the exact pathophysiology of biliary atresia and, therefore, develop a primary cure, liver transplantation will remain a crucial treatment modality for this rare disease. Children with BA require liver transplant related to a variety of morbid complications, including severe malnutrition, portal hypertension and infection. Timing transplant for BA in order to maximize health, both pre and post-transplant, remains as much an art as science. The current organ allocation schema does not consistently allow for equitability of liver transplant in children affected by BA, particularly those who are young. Pediatric waitlist mortality also emphasizes the critical need for novel approaches to expand the donor pool. In the meantime, optimization of pre-transplant care as highlighted in this review will allow transplant teams to maximize transplant outcomes.
Supported in part by grants from NIH U01DK062453 and NIH/NCATS Colorado CTSA Grant UL1TR001082. Its contents are the authors’ sole responsibility and do not necessarily represent official views the National Institutes of Health.