Primary Sclerosing Cholangitis

While the use of UDCA in primary sclerosing cholangitis (PSC) initially appeared promising, long-term controlled trials have not demonstrated any improvement in disease progression. In fact, high-dose UDCA (28–30 mg/kg daily) has been shown to increase the risk of adverse outcomes and increase the risk of colorectal neoplasia in patients with ulcerative colitis and PSC [8]. Additionally, UDCA has had little success in improving pruritus or fatigue associated with PSC. In a recent, retrospective study of UCDA use in children with PSC, less than one-half of the patients treated with UDCA had a complete GGT normalization in the first year after diagnosis, but this subset of patients had a more favorable five-year outcome than patients not improving to this extent. Long-term prospective pediatric trials to determine a potential benefit of UDCA in childhood PSC are lacking, and will be difficult to implement [9]. Nevertheless, UDCA is frequently prescribed in pediatric PSC patients.

Cystic Fibrosis

UCDA is commonly used to treat or prevent liver disease in cystic fibrosis (CF). Clinically significant liver disease develops in 10–20% of patients with CF and 5–10% develop cirrhosis by adolescence. The underlying defect in cholangiocytes leads to secretion of inspissated bile and the formation of bile plugs resulting in obstruction and biliary cirrhosis. This provides a logical rationale for the use of UDCA as a potential cytoprotective agent and stimulator of bicarbonate-rich bile flow. Prospective clinical trials of UDCA in children with CF liver disease at daily doses of 10–20 mg/kg for 6–12 months have shown significant improvement in alanine aminotransferase, alkaline phosphatase, and gamma-glutamyltransferase (GGT). A double-blind, multicenter trial demonstrated improved biochemical and clinical parameters (as measured by the Shwachman score) after one year of treatment with UDCA (15 mg/kg daily). Several studies have reported a dose–response effect for UDCA in CF liver disease with maximal effect at a daily dosage of 20 mg/kg, suggesting that higher doses of UDCA may be necessary in CF than in other forms of cholestasis. In addition to improvement of liver blood tests, UDCA improved hepatobiliary excretory function (as measured by radionuclide hepatobiliary scintigraphy), liver histology, and nutritional status of patients with CF-related liver disease. Despite the improved biochemical, biliary excretory, and perhaps histologic data, it remains to be seen whether UDCA alters the natural course of liver disease in patients with CF. While there is a need for long-term pediatric studies, it has seemed prudent to use UDCA in patients with CF and evidence of liver disease, as recommended by the Cystic Fibrosis Foundation Hepatobiliary Disease Consensus Group. However, the earlier and frequent use of UDCA over the last 20 years has not changed the incidence of severe CF liver disease, leading to questions about the use of this treatment in young children given its possible adverse effects [10, 11]. A more detailed discussion of the use of UDCA in cystic fibrosis is reviewed in Chapter 26.

Alagille Syndrome

Balistreri et al. [12], in a study of 31 patients with Alagille syndrome, demonstrated a decrease in serum alanine aminotransferase and cholesterol levels, and a marked improvement in pruritus during UDCA therapy: 15 of the 31 patients had an initial clinical response with a decrease in pruritus after one month of therapy (15–30 mg/kg daily). In addition, 11 of the 16 initial non-responders showed improvement with an increase in the dose of UDCA (45 mg/kg daily). Several case studies have also reported an improvement in pruritus, serum liver enzymes, cholesterol, triglyceride, phospholipid levels, and xanthomas. While further studies are desirable, these preliminary reports suggest that the use of UDCA in Alagille syndrome is warranted. However, there are no data available as to alteration of the natural history of Alagille syndrome by UDCA therapy.

Progressive Familial Intrahepatic Cholestasis

Progressive familial intrahepatic cholestasis (PFIC) is a group of childhood cholestatic diseases with at least three different subtypes: type 1 (Byler disease) is caused by mutations in ATP8B1 (coding FIC1) and is associated with normal serum GGT levels; type 2 results from a defect in the canalicular bile salt export pump (BSEP) (encoded by ABCB11) and is associated with normal GGT levels; and type 3 is caused by a deficiency in the multidrug resistance-associated protein 3 (MDR3, encoded by ABCB4) and is associated with elevated GGT levels. Jacquemin et al. used UDCA (20–30 mg/kg daily) in 39 patients divided into two groups, based on serum GGT levels (group 1 normal GGT and group 2 elevated GGT) [13]. After two to four years of therapy, liver tests normalized in 32%, improved in 20%, and worsened in 48% in group 1. In group 2, liver tests normalized in 50%, improved in 29%, and worsened in 21%. Children with PFIC type 3 and missense mutations had less severe disease and more often a beneficial response to UDCA therapy [11]. While long-term data are lacking, empiric therapy with UDCA appears worthwhile in some patients with PFIC, but low GGT familial intrahepatic cholestasis is generally refractory to medical treatment and UDCA has not been effective in patients with benign recurrent intrahepatic cholestasis.

Bile Acid Synthesis Defects

Several distinct abnormalities in primary bile salt synthesis have been described including Δ4-3-oxosteroid-5β-reductase deficiency and 3β-hydroxysteroid dehydrogenase/isomerase deficiency. In these inherited defects, primary bile acid synthesis is absent or markedly impaired. Δ4-3-oxosteroid-5β-reductase deficiency results in increased synthesis of abnormal oxo-bile acids and neonatal liver failure. Patients with this defect have demonstrated a dramatic response to UDCA combined with cholic acid therapy, with suppression of oxo-bile synthesis, normalization of liver blood tests, marked improvement in liver histology, and with long-term survival of a presumably fatal disorder. 3β-hydroxysteroid dehydrogenase/isomerase deficiency is clinically similar to PFIC, with low GGT; however, patients do not have pruritus. The failure of normal bile acid synthesis and the accumulation of atypical bile acids in this disorder presumably account for the progressive liver injury [14]. Cholic acid but not UDCA leads to suppression of bile acid synthesis at the level of 7α-hydroxylase and decreased production of toxic bile acid intermediates.

Biliary Atresia

Although UCDA is frequently used after portoenterostomy for treatment of biliary atresia, there is no evidence that it improves outcome or decreases complications such as recurrent cholangitis. Further studies are needed to determine whether subgroups of patients with biliary atresia (e.g., those with recurrent cholangitis after the establishment of bile flow) will benefit from UDCA therapy. Clearly, when portoenterostomy is unsuccessful or has not been performed, UDCA is of no benefit in biliary atresia and could be potentially toxic.

Cholestasis Associated with Total Parenteral Nutrition

UDCA may improve liver tests in patients on long-term therapy with TPN, but there is little evidence that it alters the course of the disease. One limitation to the use of UDCA in patients at risk for the development of TPN-associated cholestasis is poor intestinal absorption of UDCA in patients with short gut. However, UDCA may bind bacterial endotoxin in the gut lumen and prevent its absorption, thereby reducing activation of Kupffer cells, inhibiting TNF generation, and reducing liver injury. This mechanism could explain a beneficial effect of UDCA therapy in some children with short gut syndrome and bacterial overgrowth of the small bowel. As an alternative, TUDCA has not been of value in improving the liver disease in infants with short bowel syndrome and cholestasis. Modifications of intravenous lipid emulsions have a more profound effect on reversing cholestasis in this condition.

Hepatic Sinusoidal Obstruction Syndrome and Graft-Versus-Host Disease

UDCA has been used prophylactically in the treatment of hepatic complications related to bone marrow transplantation. Compared with a historical control group, prophylactic treatment with UDCA decreased serum bilirubin levels, reduced the incidence of sinusoidal obstruction syndrome, and improved survival after bone marrow transplantation [15]. While improvement of serum liver tests was observed in patients with graft-versus-host disease of the liver, biochemical abnormalities returned after discontinuation of UDCA. Although further data are desirable, UDCA may be considered for the treatment of liver graft-versus-host disease and in the prevention of sinusoidal obstruction syndrome following bone marrow transplantation, particularly in patients receiving high-risk chemotherapy regimens.

Summary of Ursodeoxycholic Acid Treatment

The use of UDCA has proven long-term benefit in adult PBC. While further studies are needed, it appears prudent to use UDCA in the treatment of CF-associated liver disease, Alagille syndrome, PFIC, graft-versus-host disease, and sinusoidal obstruction syndrome. There has been no proven benefit of its use in PSC, TPN-associated cholestasis, biliary atresia, chronic hepatitis, or orthotopic liver transplantation. It must be pointed out that, at present, no UDCA trials in children with cholestasis have shown that this therapy has altered the ultimate course of the underlying liver disease or survival. However, most experience shows that it is safe to use in infants and children who do not have fixed obstruction to bile flow.

Phenobarbital

Phenobarbital therapy has been used for many years as a choleretic and antipruritic agent for many cholestatic liver diseases. By increasing the bile acid independent fraction of bile flow, enhancing bile acid synthesis, inducing hepatic microsomal enzymes, and increasing hepatic Na⁺/K⁺-ATPase activity, it has been used in cholestasis to decrease serum bilirubin, lower circulating serum bile acids, and, by its hepatic microsomal stimulation and excretory enhancement, possibly help in the elimination of a pruritogenic substance. The usual daily dosage of phenobarbital is 3–10 mg/kg, aiming to achieve a serum level of approximately 10–20 μg/mL. High-dose phenobarbital therapy can be associated with sedation, and alterations in the metabolism of a wide variety of drugs including vitamin D may occur. Chronic phenobarbital therapy in children with seizure disorders has been associated with poor self-esteem, labile moods, neurotic symptoms, frank depression, and an increased risk for suicide and suicidal behavior. Although detailed study has not been conducted on the effects of phenobarbital on cognitive and behavior functions in children with chronic cholestasis, with the availability of other medications that reduce pruritus and stimulate bile flow, phenobarbital treatment is now used rarely for the treatment of cholestasis.

Glucocorticoids

While steroids have not been used as long-term choleretic agents, high-dose “bursts” of intravenous methylprednisolone have been shown to be possibly effective in stimulating bile flow during episodes of refractory cholangitis after hepatic portoenterostomy treatment for extrahepatic biliary atresia. Short-term intravenous corticosteroid therapy is frequently used in Asia following portoenterostomy for biliary atresia. Several US and European uncontrolled studies have showed improvement in conjugated bilirubin levels and transplant-free survival in a group of patients with biliary atresia following portoenterostomy when treated with high-dose steroids. However, a large multicenter, double-blind randomized trial of high-dose steroid therapy following hepatic portoenterostomy failed to show statistically significant treatment differences in bile drainage at six months and in transplant-free survival at 24 months of age [16]. Despite a Cochrane review concluding that there is insufficient evidence to support this treatment, the use of steroids in biliary atresia remains controversial.

Cholecystokinin

Cholecystokinin, a peptide hormone secreted by the intestine in response to a meal, stimulates gallbladder contraction and relaxation of the sphincter of Oddi, and increases intestinal motility. A synthetic cholecystokinin-octapeptide (sincalide) has been developed and was believed to be of potential benefit in treating cholestasis associated with abnormal gallbladder function, such as TPN-cholestasis where lack of enteral nutrition is associated with a decrease in endogenous cholecystokinin secretion. While administration of sincalide may cause a decline in serum conjugated bilirubin levels, no significant improvement in serum aminotransferases or in the course of liver disease was observed. A large multicentered randomized controlled trial of sincalide in infants at risk for TPN cholestasis found that it did not affect conjugated bilirubin levels, sepsis incidence, time to achieve 50% and 100% energy intake enterally, mortality rate, incidence of cholelithiasis, and number of days in intensive care and in hospital. Based on these results, sincalide should not be recommended for prevention of TPN cholestasis in infants at risk.

Nuclear Receptor Agonists

The identification of the proteins involved in hepatic bile acid uptake, transport, and excretion has provided a more mechanistic model of liver transport functions in both health and disease. Identification of these transport proteins has provided insight into the pathogenesis of many cholestatic liver disorders, while also advancing our basic knowledge of normal liver transport functions. One exciting area has been the identification of the regulatory pathways involved in the transcription and expression of these transport proteins, which involve the binding of ligands to specific nuclear receptors that regulate transcription [2, 17]. Several nuclear receptors have been identified that regulate bile acid transport proteins and enzymes involved in bile acid synthesis, including the farnesoid X receptor (FXR), the constitutive androstane receptor (CAR), the pregnane X receptor (PXR), and PPARα. Identification of these nuclear receptors suggests a new category of agents to treat cholestasis, namely specific receptor activators that will alter the expression of bile acid transporters directly.

Farnesoid X receptor is highly expressed in the liver and activated by bile acids, such as the hydrophobic bile acid chenodeoxycholic acid. Binding of bile acids to FXR upregulates transcription of genes coding for BSEP, MDR3 and multidrug resistance-associated protein 2, proteins responsible for the export of bile acids from the hepatocyte. Activation by FXR also inhibits the transcription of CYP7A and CYP8B, both involved in bile acid synthesis, and hence provides feedback inhibition of bile acid synthesis. Another membrane receptor for bile acids, TGR5, has expression and function that is distinct from FXR [17]. These two bile acid receptors complement each other in maintaining bile acid homeostasis and mediating bile acid signaling. Presence of TGR5 may be important for preserving the bile acid pool and for preventing bile acid-induced toxicity.

There has been considerable interest in developing agonists for these receptors in order to harness their potential to influence bile acid metabolism and transport as well as metabolism of lipids and carbohydrates. Obeticholic acid (6α-ethyl-chenodeoxycholic acid, OCA) is a synthetically modified bile acid that is 100 times more potent than chenodeoxycholic acid as an FXR agonist [18]. It has been approved as treatment in adults for primary biliary cholangitis, and has shown promising results in non-alcoholic steatohepatitis. Some patients have developed pruritus and reduced HDL-cholesterol levels in these studies. Severe hepatotoxicity has been observed in patients with advanced liver disease treated with higher doses of the drug. OCA is not approved for use in children, however it is in clinical trials.

In addition to FXR agonists, PPARα agonists, which are in clinical trials for PBC, increase MDR3 expression and its subcellular redistribution to the canalicular membrane and, in turn, stimulate biliary phospholipid secretion, reduce bile acid synthesis (via suppression of CYP7A1 and CYP27A1), induce bile acid detoxification (via induction of CYP3A4), and have anti-inflammatory, anti-fibrotic and antipruritic properties [2].

Emerging Pharmacologic Approaches

A number of new pharmacologic agents under development, based on the physiology of bile formation and secretion (Figure 9.2), are currently being tested in phase 1 and 2 studies in adults with PBC and PSC, many with promising early results [2]. Nor-UDCA, a side chain shortened UDCA derivative that lacks a methylene group and has resistance to amidation, enhances cholehepatic shunting of bile acids secreted by hepatocytes from within the bile duct lumen back to the hepatocyte. This induces a bicarbonate-rich hypercholeresis that counteracts intrinsic bile acid toxicity to biliary epithelia, the so-called “bicarbonate umbrella.” Nor-UDCA also has anti-inflammatory, anti-lipotoxic, antiproliferative, and anti-fibrotic qualities. Nor-UDCA might be of benefit in CF-induced cholestasis among other cholestatic disorders.

Figure 9.2 Emerging therapies under development for the treatment of cholestasis (blue bubbles represent pharmacologic agents). Multiple sites within hepatocytes, ileal enterocytes and cholangiocytes are targets for new pharmacological agents that could potentially treat cholestasis. In hepatocytes, inhibition of basolateral membrane transporters (e.g., NTCP) could reduce the hepatocyte bile acid burden and the toxicity of retained bile acids to the hepatocyte. FXR agonists can upregulate canalicular and hepatocyte bile acid efflux pumps (BSEP, MRP2, MRP3, MRP4, OSTα/β) and similarly reduce hepatocyte toxicity while increasing bile flow. UCDA, molecular chaperones (e.g., 4-phenylbutyrate for BSEP) or other agonists may improve bile flow and fat absorption. PPARα agonists may increase phospholipid secretion into bile and protect cholangiocytes against bile acid toxicity through activation of PPARα. Other strategies that inhibit bile acid synthesis by suppressing CYP7A1 (FGF19 analogues, FXR agonists, short interfering RNAs) may also reduce bile acid toxicity to the hepatocyte. In the ileum, inhibiting ileal enterocyte bile acid re-uptake by ASBT inhibitors may lower the bile acid pool size, change bile acid composition and alter enterocyte FXR signaling. FXR agonists may also activate ileal enterocyte FXR and increase FGF19 secretion with subsequent inhibition of CYP7A1 in the hepatocyte. In the cholangiocyte, nor-UCDA may increase bile pH to protect the cholangiocyte from bile acid injury. 4-PB: 4-phenylbutyrate; BA: bile acid; PC: phosphatidylcholine; Bili: conjugated bilirubin; Cl: chloride; PS: phosphatidylserine; UDCA: ursodeoxycholic acid.

Adapted from Feldman AG, Sokol RJ. Neonatal cholestasis: emerging molecular diagnostics and potential novel therapeutics. Nat Rev Gastroenterol Hepatol 2019;16:346–60, with permission.

Circulating conjugated bile acids are taken up by the hepatocyte through the basolateral bile acid uptake protein (NTCP). Inhibitors of NTCP, such as myrcludex B, could potentially reduce the bile acid burden in the liver and ameliorate bile acid toxicity to the cholestatic hepatocyte.

FGF19, secreted by the ileum in response to FXR activation by ileal enterocyte uptake of intestinal bile acids through the apical ASBT transporter, binds to its hepatocyte receptor FGFR4/βKL and downregulates hepatocyte bile acid synthesis by suppression of cholesterol 7-apha hydroxylase. FGF19 analogues have been developed to down regulate bile acid synthesis and are in trials in cholestatic diseases in adults.

For cholestatic disorders in which genetic variants reduce synthesis or trafficking of canalicular bile transport proteins, small-molecule chemical chaperones are being explored. Chaperones, such as 4-phenyl butyrate, are proposed to bind reversibly to active sites of a missense mutant enzyme, correct protein misfolding and enhance trafficking of the protein to the correct target. Potential chaperones are being tested for PFIC types 1 and 2 and other diseases that result in cholestasis. This opens the possibility for personalized therapy based on the molecular characterization in individual patients of their genetic cholestatic disorder.

Fibrosis

Fibrosis is the common outcome of chronic liver injury that may evolve to cirrhosis, portal hypertension, liver failure, and hepatocellular carcinoma. The process may result from infectious, obstructive and metabolic causes. Hepatocyte damage results in inflammation and activation of hepatic stellate cells, which are thought to be the major source of myofibroblasts in the liver [19]. Altered intestinal barrier function and dysbiosis of the small intestine and colon in cholestasis lead to absorption of pathogen-associated molecular patterns that may activate hepatic innate immune pathways and contribute to the induction of fibrosis. Extracellular matrix, including type I collagen, which constitutes the fibrous tissue in the liver, is largely produced by myofibroblasts which proliferate in response to cytokines, chemokines, and growth factors. The therapeutic intervention points for treating fibrosis seem conceptually straightforward, including eliminating the underlying noxious stimulus, suppressing hepatic inflammation, interfering with the activation of stellate cells, and promoting degradation of extracellular matrix [20]. Despite increased understanding of the pathogenesis of fibrosis and numerous experimental strategies under investigation, effective treatment of liver disease is currently the only approach to preventing or reversing fibrosis. This is particularly important in children in whom regression of fibrosis or even cirrhosis is more commonly observed than in adults.

Pathogenesis

The pathogenesis of the pruritus of cholestasis is complex, multifactorial and remains incompletely understood [22]. Putative pruritogens include histamine, substance P, bile acids, lysophosphatidic acid, endogenous opioids and a yet unidentified factor, but to date no single substance has been conclusively shown to be the causative pruritogen in cholestasis [23]. Penicillate intraepidermal nerve endings, which arise from unmyelinated subepidermal free nerve endings, have been implicated as the sensor that mediates general pruritus; however, the mediators that stimulate these nerve endings during cholestasis are still unknown. Earlier studies had suggested that elevated serum and skin concentration of bile acids were responsible, but a direct causal relationship between itching and bile acid levels in skin and/or serum has not been confirmed. Other evidence refuting the role of bile acids in pruritus is the reduction of pruritus in patients with uremia and polycythemia vera by cholestyramine treatment, disease states not associated with bile acid retention. However, the absence of pruritus in children with bile acid synthesis defects and those with low serum concentrations of bile acids, despite significant cholestasis, argues for a role of circulating bile acids.

There is a significant component of the pruritus that may be of central neurogenic origin, possibly involving the opiate receptor system. This is based on the observation that pruritus is a recognized side effect of morphine and other opiate receptor agonists. Indeed, physicians who have used meperidine for sedation prior to procedures are familiar with the “itching of the nose” behavior associated with the administration of this medication. This opioid-associated pruritus is reversed by opiate receptor antagonists (naloxone, naltrexone) but not by antihistamines. The central effect of opiates is mediated via opiate receptors in the brain. Bergasa et al. [21] injected serum from patients with PBC into the medullary dorsal horn of monkeys and induced itching, which was blocked by the opioid-receptor antagonist, naloxone. In a rat model of cholestasis, binding of a selective μ-opioid receptor ligand to μ-opioid receptors is altered in cholestasis. These μ-opioid receptors are downregulated, suggesting that cholestasis may be associated with chronically elevated levels of endogenous opioids. In chronic cholestatic liver disease, nalmefene, a specific oral opiate receptor antagonist, produces symptoms strikingly similar to the “withdrawal reaction” of opiate addiction. This observation suggests that patients with cirrhosis and impaired hepatocellular function are chronically exposed to increased levels of endogenous opiate receptor agonists, and it is further supported by the finding of elevated levels of the endogenous opiate ligands metenkephalins and leuenkephalins in these patients. Furthermore, other evidence suggests that these elevated plasma levels of pentapeptide enkephalins allow them to cross the blood–brain barrier. Reports on the beneficial effects of opiate receptor antagonists (naloxone, naltrexone and nalmefene) in the pruritus of cholestasis likewise support the concept that increased availability of endogenous opiate ligands at central opiate receptors may stimulate the pruritus of cholestasis.

Experimental evidence has implicated the lysophospholipase autotaxin and its product, lysophosphatidic acid, as potential mediators of cholestatic pruritus. In a recent study, increased serum autotaxin was specific for pruritus of cholestasis but not for other pruritus-associated disorders such as uremia or Hodgkin disease [24]. Rifampin treatment significantly reduced itch intensity and autotaxin activity in cholestatic patients with pruritus. In vitro studies showed that this effect required the expression of PXR. Other effective treatments for severe, refractory pruritus using the molecular adsorbents recirculation system or nasobiliary drainage improved itch intensity and was correlated with the reduction of autotaxin levels.

A recent study has also demonstrated that pathophysiologic levels of bilirubin in a mouse model excite peripheral itch sensory neurons and elicit pruritus through Mas-related G-protein receptor, a family of G-protein coupled receptors expressed in primary sensory neurons [25].

Overall, the current understanding is that cholestatic pruritus results from a complex interplay of direct or indirect actions of bile acids and/or their metabolites, increased opioidergic neurotransmission, upregulation of autotaxin and increased levels of LPA resulting in triggering of pruritoceptive nerve fibers. Findings advancing roles for lysophosphatidic acid, opioids and other mediators in the pathogenesis of pruritus in cholestasis may not be mutually exclusive. Many mediators may be involved.

Treatment

The therapeutic agents most commonly used for pruritus in cholestasis are oral bile acid-binding resins (cholestyramine or colestipol), phenobarbital, rifampicin, UDCA, carbamazepine and sertraline [26]. ASBT inhibitors are under study in Alagille syndrome and PFIC for pruritus. Cool baths, moisturizers, topical steroid creams, topical anesthetics, antihistamines, and sedatives have offered little long-term relief, although they may be of temporary benefit in individual patients. In small children, fingernails should be trimmed, long-sleeve nightshirts worn, and occasionally the hands covered securely with stockings at night to minimize the effects of scratching. Plasmapharesis and ultra-violet light treatment have improved pruritus in adults with PBC. The possible use of opioid antagonists is being explored. Finally, partial biliary diversion, ileal exclusion, and liver transplantation are considered when all other therapeutic options have been exhausted.

Non-Absorbable Ion Exchange Resins

Cholestyramine, colestipol, and colesevelam hydrochloride are non-absorbable anion exchange resins that bind bile acids, cholesterol, many drugs, and presumably other toxic agents in the intestinal lumen, thereby increasing fecal excretion of these substances [22]. These bile acid-binding agents interrupt the enterohepatic circulation of bile acids, decreasing the negative feedback to the liver, enhancing conversion of cholesterol to bile acids, and possibly stimulating a choleresis. Because cholestyramine relieves pruritus without causing a change in serum bile acids, it is possible that it also removes other anionic molecules that may be contributing to pruritus.

Cholestyramine and colestipol are usually administered mixed with juice or water at a daily dose of 0.25–0.50 g/kg, given both before and after breakfast when bile flow is maximal or, less commonly, divided among two or three daily meals. Colestipol appears to be better tolerated than cholestyramine; however, cholestyramine bars are now available and are more palatable. No other medications or vitamins should be given orally for the two hours preceding or following administration of these resins because of the risk of binding to the resin and impaired absorption. Several other factors limit the use of cholestyramine and colestipol: the unpalatable nature of the compounds (which may lead to poor compliance); increased steatorrhea and fat-soluble vitamin deficiency because of further reduction in the already low concentrations of free bile acids in the intestinal lumen; constipation; intestinal obstruction from inspissation of the drug; and hyperchloremic metabolic acidosis. Owing to these problems, these agents should be regarded as second-line therapies. These compounds are generally contraindicated in the infant with biliary atresia and a Roux-en-Y portoenterostomy because of the risk that the compound may accumulate in and obstruct the reconstructed biliary intestinal conduit, leading to ascending cholangitis.

Rifampicin

Rifampicin, a potent CYP3A4 inducer, should be regarded as a first-line treatment of cholestatic pruritus [22, 23]. Studies indicate that rifampicin, an antibiotic for tuberculosis and other infections, is an effective treatment for severe pruritus in PBC and in children with chronic cholestatic liver disease [23]. Yerushalmi et al. [27] treated 24 children with cholestasis with rifampicin (10 mg/kg daily, in two divided doses). After an average of 18 months of therapy, 10 patients had a complete response and 12 patients had a partial response as assessed by a clinical scoring system. Treatment was associated with a reduction in GGT and no clinical or biochemical toxicity of rifampin was observed. Complete response was more common in children with extrahepatic cholestasis (e.g., biliary atresia) than intrahepatic cholestasis (64% vs. 10%).

Rifampicin is a ligand for PXR, which activates many pathways for biotransformation [22, 26]. Proposed mechanisms for the effect of rifampicin on pruritus include enhancement of multidrug-resistance protein 2 production, activation of enzymes (UDP-glucuronosyltransferase-1A and cytochrome P4503A4), and stimulation of 6α-hydroxylation of bile acids, thereby promoting urinary excretion of dihydroxy and monohydroxy bile acids. The capacity to reduce lysophosphatidic acid levels may be more important than its effects on bile acid metabolism. Despite the apparent amelioration of pruritus with rifampicin, its propensity for toxic hepatitis requires careful monitoring. The other potential adverse effects associated with its use are drug interactions, hemolytic anemia, and renal failure. We have been very pleased with the response of children with a variety of cholestatic disorders to rifampicin treatment.

Opioid Antagonists

Given the theory that cholestasis-associated pruritus may be caused by centrally mediated increased opioid tone, the use of several opioid antagonists have been investigated, including naloxone, nalmefene, and naltrexone [22]. A small cross-over study in eight patients with PBC and a larger double-blind, controlled, cross-over trial in 29 patients with liver diseases of various causes demonstrated marked improvement in pruritus during intravenous, 24-hour naloxone infusions compared with placebo [21]. Mild neuropsychiatric disturbances, described as “ill-defined anxiety,” were reported in four patients in the larger study and no patients in the pilot study. This complication may be explained by a mild opiate withdrawal effect in the presence of the chronic increased opioid tone postulated to exist in patients with cholestasis. Because of the opioid receptor specificity of naloxone, these findings support the hypothesis that a mechanism underlying the pruritus of cholestasis is modulated by endogenous opioids. Although effective, naloxone has several limitations for long-term use, including a short half-life and large first-pass metabolism, which necessitate intravenous administration.

Nalmefene, another opioid antagonist, has a longer duration of action compared with naloxone and can be given orally, however, at the present time it is only available in the USA as a parenteral product. In an initial report of 11 patients with cirrhosis, nalmefene therapy (starting at a dose of 5 mg/day and gradually increasing to a maximum of 20–40 mg three times daily) resulted in a significant reduction in patient’s pruritus scores and sense of fatigue. Distressingly, all 11 patients experienced withdrawal reactions consisting of nausea, abdominal pain, diaphoresis, tremor, and occasional hallucinations. Larger doses of nalmefene (up to 300 mg) given to healthy subjects have not produced withdrawal reactions, once again supporting the theory of increased opioid tone in patients with cholestasis-associated pruritus. A recent open-label trial of oral nalmefene also demonstrated a beneficial effect in relieving pruritus, but with fewer adverse side effects reported [21]. In this study of 14 adults with cholestasis, the initial starting dose was 2 mg twice a day and this was gradually increased, over two to four weeks, until a satisfactory clinical response was achieved (average maintenance dose was 60 mg/day, with a range of 20–240 mg/day) and continued for 2 to 26 months. Only five patients experienced withdrawal-like reactions, which did not preclude continuation of therapy. A significant decrease in visual analogue scores was noted in 13 patients and a decrease in scratching activity was noted in 12 patients. Possible tolerance occurred in three patients and three patients experienced a marked exacerbation of pruritus after therapy was suddenly discontinued. This uncontrolled study suggests that orally administered nalmefene is of benefit to patients with cholestasis-associated pruritus and is associated with fewer withdrawal reactions with the lower starting dose.

The opioid antagonists investigated to date have severe limitations [21]. Naloxone has a short half-life and can only be administered parenterally, while nalmefene treatment is associated with a severe opiate withdrawal reaction and is not currently licensed for clinical use. These limitations have prompted investigation of other opioid antagonists. Naltrexone is an opiate receptor antagonist with a bioavailability and half-life that lies between that of naloxone and nalmefene, and it can be administered orally. It is a structural analogue of naloxone and nalmefene that undergoes extensive first pass metabolism; however, the main metabolite, 6β-naltrexol, reaches higher plasma levels than the parent drug and exerts long-lasting opiate antagonist activity. A large, double-blind, placebo-controlled study demonstrated significant decreases in both daytime and night-time itching (as recorded by the patient using the visual analogue scale) [28]. In this study, 16 adults with cholestasis were randomized, with eight receiving oral naltrexone (50 mg/day for four weeks) and eight receiving placebo. Compared with the placebo group, the naltrexone-treated group had significantly decreased pruritus scores at the end of treatment, with associated improvement in sleep satisfaction and less fatigue compared with baseline scores prior to treatment. Withdrawal reactions were noted in four patients in the treatment group but were generally transient, with the exception of one patient who required discontinuation of treatment. Naltrexone may be an effective alternative therapy for patients with cholestasis-associated pruritus unresponsive to other anti-pruritics; however, larger, long-term studies are needed. The initial concerns over possible hepatotoxicity of naltrexone in studies of alcoholism were not validated in a review of adults. Nausea (9%) and headaches (6%) were the most common side effects of naltrexone. Preliminary reports of successfully using 1–2 mg/kg/day of naltrexone in children with cholestasis-induced pruritus were promising although the safety of this approach in young children has not been fully evaluated.

The use of opioid antagonists may provide an effective alternative treatment for patients with severe pruritus unresponsive to other therapies; however, the significant side effects and withdrawal reactions may severely limit the general use of these medications. Further placebo-controlled trials are needed to determine safety, proper dosage, and long-term efficacy in children with cholestatic liver disease. Concerns regarding the effects of chronic opioid antagonism in the developing brain will also need to be addressed.

Phenobarbital

In addition to its choleretic effects, phenobarbital therapy has been beneficial in improving cholestasis-associated pruritus [22]. The mechanism of action in ameliorating pruritus is not entirely clear. The antipruritic action of phenobarbital has been demonstrated without corresponding decreases in circulating levels of bile acids, suggesting that the effect of phenobarbital may not be entirely explained by a decrease in bile acid levels. However, through microsomal enzyme stimulation and excretory enhancement, phenobarbital may eliminate another, as of yet unidentified, pruritogenic substance. The beneficial effect of phenobarbital in relieving cholestasis-associated pruritus has been demonstrated in a number of cholestatic disorders including adults with PBC and children with intrahepatic cholestasis. However, studies comparing the efficacy of phenobarbital with other anti-pruritics have not been as favorable. Rifampicin appeared to improve cholestasis-associated pruritus to a greater degree and with fewer side effects than phenobarbital. As mentioned above, the sedative effects, irritability, and altered performance associated with phenobarbital therapy are undesirable and limit its chronic use in children.

Ursodeoxycholic Acid

As discussed above, UDCA is a potent choleretic and has been shown to improve biochemical parameters associated with several cholestatic disorders [13]. Reports in children with chronic intrahepatic cholestasis suggest that UDCA administration may similarly result in significant improvement in refractory pruritus. However, in infants with biliary atresia and poor bile drainage following portoenterostomy, administration of UDCA may worsen pruritus and possibly lead to a significant worsening of liver dysfunction. The other side effect reported in children has been occasional diarrhea.

Intestinal Bile Salt Transport Inhibitors

Potent inhibitors of the ileal bile acid transporter/apical sodium‐dependent bile acid transporter (ASBT) (SLC10A2) are under development to treat the pruritus associated with cholestasis. A randomized, double‐blind, placebo‐controlled, parallel group, multicenter trial of ileal bile acid transporter, maralixibat, has been conducted in children with Alagille syndrome and pruritus [29]. The primary outcome was the change from baseline to week 13 in a novel itch score (ItchRO). Statistically significant decreases were observed in the itch score with doses of 70 and 140 µg/kg/day but not 280 µg/kg/day relative to placebo or all doses combined. The data in total suggest that maralixibat is generally safe and well tolerated and may reduce pruritus in some patients with Alagille syndrome (ALGS). However, no ASBT inhibitor has been approved for use in children.

In another study of adult patients with primary biliary cholangitis and pruritus, 14 days of ileal bile acid transporter inhibition by GSK2330672 was generally well tolerated without serious adverse events, and demonstrated efficacy in reducing pruritus severity [30].

Further studies are required to determine the role of ASBT inhibitors in the treatment of cholestatic pruritus in children, including whether certain genotypes are more likely to respond to these agents.

Partial Biliary Diversion and Ileal Exclusion

Partial external diversion of bile is being used as a treatment for refractory pruritus in children with severe intrahepatic cholestasis. This surgical procedure consists of the construction of a 10–15 cm jejunal conduit from the dome of the gallbladder to the abdominal wall. The gallbladder is anastomosed end to side to the blind proximal portion of a jejunal conduit with the distal end of the conduit brought out to the skin as a permanent cutaneous stoma. Others have described using the appendix as a conduit between the gallbladder and skin. The bile collected in the stoma appliance is discarded. Still others have performed partial internal biliary diversion in which the gall bladder is anastomosed to the ascending colon via a conduit of jejunum with apparently good results.

Another variant of this approach is ileal exclusion in which a surgical internal ileal–colonic bypass of the distal ileum is created. This results in an interruption of the enterohepatic circulation of bile acids by the failure of bile acids to be actively transported in the terminal ileum. In patients who have undergone cholecystectomy, ileal exclusion has been suggested as an alternative for increasing fecal excretion of bile acids and reducing pruritus. Others have now used this surgery as primary therapy. Initial studies suggest that ileal exclusion may be as effective as partial biliary diversion (and cosmetically more acceptable to the child and family); however, its benefits appear to diminish over time. There have been several small studies suggesting that interruption of the enterohepatic circulation of bile acids could improve pruritus and liver test abnormalities in children with PFIC and Alagille syndrome.

The outcomes of surgical management of genetically defined PFIC patients resulting from mutations in the ATP8B1 gene (encoding familial intrahepatic cholestasis 1 (FIC1) deficiency) or the ABCB11 gene (BSEP deficiency) have recently been reported [31]. This retrospective multicenter study involved 42 patients with FIC1 deficiency (FIC1 patients) and 60 patients with BSEP deficiency (BSEP patients) who had undergone one or more surgical procedures (57 diversions, six exclusions). BSEP patients were divided into two groups, BSEP-common (bearing common missense mutations D482 G or E297 G, with likely residual function) and BSEP-other (with more severe mutations). Overall, diversion improved liver tests, pruritus, and growth, with substantial variation in individual response. BSEP-common or FIC1 patients survived longer after diversion without developing cirrhosis, being listed for or undergoing liver transplantation, or dying, compared to BSEP-other patients [31]. Diversion can improve clinical and biochemical status in FIC1 and BSEP deficiencies, but outcomes differ depending on genetic etiology. It is a concern that three patients with the E297 G mutation of ABCB11 developed hepatocellular carcinoma, so these patients require careful follow-up even with a good clinical response. For many patients, particularly BSEP-other, diversion is not a permanent solution and transplantation is required.

The mechanism by which this procedure produces these results is poorly understood. It is not known if a toxin in bile is discarded, if hepatic bile acid metabolism is altered, or if another process is taking place. Recently, it has been shown that the ileal bile acid transporter may be upregulated in FIC1 disease, providing an explanation for the significant cholestasis and elevated bile acids in this disorder as well as providing a rationale for the use of biliary diversion. Partial external biliary diversion or ileal exclusion is an option that should be considered in planning the management of children with progressive intrahepatic cholestasis, particularly in those with intractable pruritus uncontrolled by medical therapy.

Recently, nasobiliary drainage has been employed to effectively reduce pruritus in patients with exacerbations of benign recurrent intrahepatic cholestasis, a milder form of FIC1 and BSEP diseases [32]. Ultraviolet light and the molecular adsorbent recirculating system have also been used with variable success.