Glucocorticoids are among the most studied interventions following HPE for biliary atresia (19–27). The concept of “blast therapy” originated with Karrer and Lilly in 1975: a four day taper of high dose methylprednisolone (starting dose 10 mg/kg) increased bile flow and decreased serum bilirubin in post-HPE patients with cholangitis or decreased bile flow (28). Glucocorticoids also have profound anti-inflammatory effects (29). Current evidence for glucocorticoids in post-HPE biliary atresia is weak, although the data suggest a possible benefit especially in clearance of jaundice.

The highest quality evidence consists of three randomized placebo-controlled trials (19, 21, 26). Davenport et al. conducted a randomized, double-blind, placebo-controlled trial at two centers in the United Kingdom (26). In the prednisolone group, 26% underwent transplant at 12 months compared to 35% in the placebo group (P = 0.47), while 50% had normal bilirubin in the prednisolone group compared to 40% in the placebo group (P = 0.35). A subgroup analysis of infants undergoing early portoenterostomy (age < 70 days) found decreased median bilirubin at one month (64 vs. 116 umol/L, P = 0.01), but this was not sustained at six months.

The START trial was a multi-center, double-blind, randomized, placebo-controlled trial conducted at 14 United States centers (21). Infants were randomized to a 13-week steroid course or placebo (n = 70). There was no significant difference between the steroid group and control group in the primary endpoint of successful bile drainage at six months post-HPE in either the intention-to-treat (58.6% vs. 48.6%) or per-protocol analysis. In infants <70 days old at time of HPE, there was similarly no significant difference between the steroid and control groups for successful bile drainage (71.8% vs. 56.8%). There was similar survival with native liver at 24 months post-HPE (58.7% vs. 59.4%, P = 0.99) and successful bile drainage at 24 months (49.4% vs. 39.8%, P = 0.29).

A recent randomized controlled trial at a single center in China randomized infants post-HPE (performed by a single surgeon) to steroid treatment or placebo (19). The primary outcome of clearance of jaundice with native liver at six months was significantly higher in the steroid group compared to the placebo group (54.1% vs. 31.0%, P = 0.0015) with higher native liver survival rate at 24 months (57.1% vs. 40.0%, P = 0.03). While this study is limited in generalizability due to its single center and single surgeon experience, these data are consistent in direction and magnitude with the previously described randomized controlled trials and provide weak support for post-HPE glucocorticoids in biliary atresia.

Cholangitis affects 40%–93% of post-HPE patients, most commonly in the first two years of life (30). Prophylactic antibiotics are often prescribed to prevent cholangitis and, hopefully, prevent the progression of liver disease. Surveys of primarily European and North American centers have found most utilize prophylactic antibiotics post-HPE, although there is minimal evidence to support a specific antibiotic regimen or duration of prophylaxis (30, 31).

There are no randomized placebo-controlled trials evaluating antibiotic prophylaxis. A small single center trial in Taiwan randomized patients with one episode of cholangitis post-HPE to prophylaxis with trimethoprim-sulfamethoxazole (TMP/SMZ; n = 9) or neomycin (n = 10), compared to a historical control with no prophylactic antibiotics (n = 18) (32). Patients in both antibiotic groups had decreased cholangitis recurrence compared to the historical control group (TMP/SMZ: RR 0.52 [95% CI 0.29, 0.98], neomycin: RR 0.42 [95% CI 0.22,0.82]). A recent randomized controlled trial at a single center in China randomized infants post-HPE to seven or 14 days of intravenous antibiotics (cefoperazone/sulbactam and ornidazole) prior to an oral antibiotic regimen for 6 months (alternating sulfamethoxazole and cefaclor every two weeks) (33). There was no difference in the primary outcome of cholangitis occurrence at 6 months between the short- and long-term groups (62% vs. 70%, P = 0.27), but there was decreased early onset cholangitis (within one month) in the long-term group (61% vs. 38%, P = 0.02). Older retrospective cohort studies have evaluated patients receiving antibiotic prophylaxis versus historical controls that did not receive antibiotics, with only Lally et al. (antibiotic n = 34, no antibiotic n = 7) reporting significantly decreased cholangitis incidence, while Wu et al. (antibiotic n = 16, no antibiotic n = 21) and de Vries et al. (antibiotic n = 124, no antibiotic n = 80) found no difference in cholangitis incidence (34–36).

Ursodeoxycholic acid (UDCA), a hydrophilic secondary bile acid, is used for a variety of cholestatic liver diseases (37). UDCA decreases progression to liver transplant and death in primary biliary cholangitis, and improves liver biochemistry (but not histologic progression, need for transplant, or mortality) in primary sclerosing cholangitis (38, 39). At very high doses (28–30 mg/kg/day) in an adult primary sclerosing cholangitis population, there is a doubling of severe adverse effects as well as an increase in a composite outcome of death, transplantation, or minimal listing criteria (40). This toxicity has been attributed to UDCA metabolism into lithocholic acid, which in animal models may lead to an obstructive cholangiopathy and bile duct destruction (41). UCDA has a choleretic effect through increased chloride and bicarbonate secretion, posttranscriptional modification of hepatobiliary protein levels of key ATP-binding cassette transporters, and cholangiocyte/hepatocyte protection (42–44). In vitro experiments also demonstrate that UDCA moderates the inflammatory response of peripheral blood mononuclear cells and human B cell lymphoma cells with decreased immunoglobulin and cytokine production (45).

Off-label administration of UDCA post-HPE is routine at many centers around the world, with surveys of European and Japanese centers finding near-universal use (31, 46). Despite this, a lack of high-quality literature exists to support a benefit or harm of UDCA. In a prospective single institution study of 16 children in France, after UDCA was discontinued at 18 months post-HPE, liver enzymes worsened in 13 children (47). Six months following resumption of UDCA, liver enzymes, bilirubin, and bile acids decreased, providing the most direct evidence for use of UDCA in biliary atresia. A retrospective cohort study at a single Egyptian hospital compared infants post-HPE who received UDCA (n = 108) versus those who did not (n = 33), suggesting a worse outcome in patients who received UDCA (48). A “successful” outcome, defined as absence of jaundice with aminotransferase levels below twice the upper limit of normal, was achieved in 24.2% that did not receive UDCA versus 10.2% of infants who did (P = 0.04), but there was no difference in the proportion that “failed” (75.7% vs. 77.8%, P = 0.49). The study was limited by treatment selection bias, unclear temporality of the UDCA initiation, and overall poor outcomes. The progression to transplantation as well as actual biochemical values were not reported. Given the paucity of investigation of UDCA in this population, well-designed multicenter randomized controlled trials are needed to both better understand the possible benefit or harm of this medication as well as determine the optimal dose.

Children with biliary atresia are at high risk for malnutrition, growth failure, and vitamin deficiencies (especially fat-soluble vitamins) due to malabsorption, anorexia, and altered nutrient metabolism (49). Growth failure is associated with increased pre- and post-transplant mortality and graft failure (50, 51). Close surveillance combined with aggressive nutritional supplementation is critical. Initial recommendations include high protein intake (3–4 g/kg/day) and up to 130%–150% of recommended calories for age, with sequential stepwise escalation of nutritional support to nasogastric feeds and ultimately parenteral nutrition (49). Medium chain triglyceride supplementation is under study as a source of caloric dense, easily absorbable lipids (NCT05072626). Critically, medium chain triglycerides do not contain the essential polyunsaturated long-chain fatty acids necessary for growth and development. As growth failure is common in biliary atresia and often leads to early liver transplantation, further research is critical in this area to optimize nutrition, as well as further investigating detoxification genomics in biliary atresia (52).

Various viral infections have been proposed as possible inciting events in a subset of biliary atresia pathogenesis, with the greatest investigation of cytomegalovirus (CMV). Between 10% and 74% of biliary atresia patients have positive CMV serology depending on series, and CMV-positive patients have worse outcomes (53). Limited investigation has evaluated the role of antiviral therapy in this population. One retrospective single center trial found that in 36 CMV IgM+ patients, treatment with ganciclovir and/or valganciclovir in the early postoperative period until negative CMV serology resulted in greatly improved clearance of jaundice (75 vs. 21%, P = 0.04) (54). Prospective investigation is critical to demonstrate efficacy of targeted antiviral therapy.

Cholestyramine and phenobarbital are used at some centers despite minimal evidence of effectiveness (31). Cholestyramine is a bile acid sequestrant, while phenobarbital is a choleretic that, in rats, increases the bile acid-independent portion of bile flow (55). The primary trial investigating these agents (from 1986) randomized infants with biliary atresia following surgical repair (primarily Kasai HPE) to no choleretic (n = 12), phenobarbital (n = 21), or cholestyramine (n = 23) for three months (56). There was no difference in total bilirubin or bile acid levels at three months in any group.

Odevixibat is an oral small molecule ileal bile acid transporter inhibitor FDA-approved for pruritus progressive familial intrahepatic cholestasis (PFIC). Inhibition of the ileal bile acid transporter prevents bile acid absorption in the terminal ileum, increasing fecal bile acid excretion. Importantly, this may lead to deficiency in fat-soluble vitamin absorption and the impact on growth and development is unknown. In the PEDFIC 1 trial, daily administration of odevixibat reduced serum bile acids and pruritis in children compared to placebo, and was well tolerated (57). Notably, the PEDFIC 1 trial only included type I and type II PFIC, which are distinct from type III and IV PFIC due to their low-GGT cholestasis. Odevixibat is currently under study in a randomized, double-blind, placebo-controlled phase III trial for biliary atresia (NCT04336722).

Maralixibat (Mirum Pharmaceuticals) is a second oral small molecule ileal bile acid transporter inhibitor FDA-approved for treatment of cholestatic pruritis in Alagille Syndrome (58). In the phase IIb randomized placebo-controlled ICONIC trial, maralixibat treatment reduced serum bile acids and pruritis in children with Alagille syndrome complicated by cholestasis and intractable pruritis (59). A multinational, randomized, double-blind, placebo-controlled phase II trial is currently recruiting patients with biliary atresia to evaluate maralixibat (NCT04524390).

Obeticholic acid is a semi-synthetic bile acid that is a potent Farsenoid X receptor (FXR) agonist FDA-approved to treat UDCA-resistant primary biliary cholangitis (60). FXR is a key ligand-modulated nuclear regulator of hepatic lipogenesis, insulin resistance, bile metabolism, and inflammation (61). In a phase III trial for UDCA-resistant primary biliary cholangitis, obeticholic acid decreased alkaline phosphatase compared to placebo, although pruritis was a common side effect (62). A phase I multicenter open-label single/multiple dose study investigating the safety and tolerability of obeticholic acid in biliary atresia patients is underway in Europe (NCT05321524).

N-acetylcysteine is a hepatoprotective and antioxidant agent used for acetaminophen overdose that drives glutathione synthesis and increases bile flow (63, 64). Previous investigations of N-acetylcysteine in pediatric diseases such as parenteral nutrition-associated cholestasis and bronchopulmonary dysplasia demonstrated excellent safety, although it did not demonstrate efficacy (65–67). A phase II open-label single center trial (NCT03499249) is underway in the United States to evaluate if seven days intravenous infusion of N-acetylcysteine following HPE improves bile flow (68).

Intravenous immunoglobulin (IVIG) is a preparation of immunoglobulin G purified from donor plasma (69). Gamma globulins were first used for primary humoral immunodeficiency; IVIG was later introduced for autoimmune disease in idiopathic thrombocytopenic purpura (70, 71). The immunomodulatory and anti-inflammatory actions of IVIG involve both innate and adaptive immune cells, with effects extending beyond the half-life due to poorly understood mechanisms (72). Due to the proposed centrality of autoimmunity and inflammation in the pathogenesis of biliary atresia, IVIG has been investigated as a potential therapy.

The PRIME trial was a multicenter phase I/IIa trial performed at eight North American centers. Twenty-nine patients with biliary atresia received IVIG and were compared to a historical “placebo” control cohort (n = 64) from the START trial (73). Subjects received three 1 g/kg infusions of IVIG at days 3–5, 30, and 60 post-HPE. The primary outcome was safety: 89.7% of IVIG patients had a serious adverse event compared to 75% in the historical cohort, although no events were directly attributable to IVIG infusion. There was no difference in serum bilirubin, but there was a non-significant trend towards decreased transplant-free survival at 360 days in the IVIG group (58.6% vs. 70.5%, P > 0.05). Randomized controlled trials comparing IVIG and glucocorticoids are necessary.

Several small trials are underway to investigate other repurposed therapies for biliary atresia. Granulocyte-colony stimulating factor (GCSF) stimulates hematopoietic stem cell (HSC) and neutrophil production and mobilization. Typically used in myelosuppression, it recently has been investigated in various liver diseases due to the potential for HSCs to modulate hepatic inflammation and promote regeneration (74). GCSF is currently being investigated in post-HPE type 3 biliary atresia (NCT04373941). Autologous bone marrow mononuclear cell infusion has also been investigated in a small phase I/IIa open-label trial in Vietnam, although the primary outcomes assessed were safety-related and efficacy cannot be determined (75).

Rituximab is an anti-CD20 human/murine chimeric monoclonal antibody that targets B cells. Initially developed for B cell lymphoma, it under investigation for several inflammatory liver diseases. A small non-randomized clinical trial is currently enrolling patients in China (ChiCTR2000031738).

Bezafibrate is a peroxisome proliferator-activated receptor agonist that was found in a randomized, double-blind, placebo-controlled trial to decrease cholestatic pruritus in a combined population of primary sclerosing cholangitis and primary biliary cholangitis patients (76). A study at a single center in Japan for biliary atresia is underway (jRCTs031210066).

The installation of mitomycin C at the portal plate at time of HPE is being investigated in a randomized phase II trial due to its antifibrotic properties (CTRI/2018/12/016495).

The gut microbiome is altered in children with biliary atresia (77). Two small trials have investigated Lactobacillus casei rhamnosus (LGG) supplementation. A small randomized controlled trial found no difference in jaundice, cholangitis, or need for transplantation at two years between LGG (n = 14) and placebo (n = 16) (78). A second trial randomized patients to LGG (n = 10) or neomycin (n = 10) for cholangitis prophylaxis, finding a similar frequency of cholangitis episodes (although less than a small historical control with no prophylaxis) (79). Further research is necessary to evaluate the efficacy of probiotics for cholangitis prophylaxis and other outcomes.