Autoimmune Liver Diseases: Primary Biliary Cholangitis

Fig. 16.1
Asymptomatic primary biliary cirrhosis: a study of its natural history and prognosis. Spring J, Cauch-Dudek K, O’Rourke K, et al. Am J Gastro. 1999; 94(1): 47–53


Fig. 16.2
Asymptomatic primary biliary cirrhosis: a study of its natural history and prognosis. Spring J, Cauch-Dudek K, O’Rourke K, et al. Am J Gastro. 1999; 94(1): 47–53

Patients with symptomatic PBC show a more rapid progression to end-stage liver disease and have a worse prognosis than that observed in the asymptomatic patients [16, 1820]. In the late phase of the disease, serum bilirubin levels increase [21] and clinical features of liver failure such as portal hypertension and hepatic encephalopathy develop [20]. The mean survival times in symptomatic patients vary between 6 and 10 years [22, 23]. In a 28-yea r follow-up study of a large cohort that included 770 PBC patients [19], the percentage of patients who developed liver failure was 15.4 % and 26.4 % after 5 years and 10 years of diagnosis, respectively [19]. Of all the variables, serum bilirubin is the best predictor of survival in PBC patients [21, 2428]. Advanced age, male gender, advanced histological stage, elevated serum ALP, low serum albumin, development of esophageal varices, and prolonged prothrombin time have also been associated with poor prognosis in patients with PBC [20].

In some patients, presence of AMA may be the only evidence of PBC. In an early report [29], patients incidentally discovered to have positive AMA (titers ≥ 1:40) but no symptoms of liver disease and normal hepatic biochemistries were followed for over 18 years. Liver biopsies were compatible with or diagnost ic of PBC in 83 % of patients at baseline [29]. During the follow-up period, 76 % of patients developed PBC symptoms, and 83 % had persistently elevated serum alkaline phosphatase (ALP) levels [30]. Repeat liver biopsies in 10 patients showed that two patients progressed histologically by one stage, and two other patients progressed by two stages [30]. None of the patients during the follow-up period died from liver disease [30]. This study suggests that patients who test positive for AMA but have no liver-related symptoms and normal hepatic biochemistries might eventually develop symptomatic but slowly progressive PBC. These findings, however, need to be confirmed in a larger cohort of patients.

Natural History in the UDCA Era

The natural history of PBC has signifi cantly changed since the introduction of UDCA. Responders to UDCA demonstrate survival comparable to age- and sex-matched healthy subjects [31]. Studies have shown that UDCA delays histological progression [32], delays development of esophageal varices [33], and improves the transplant-free survival of PBC patients [3442]. Data compiled from three clinical trials in which PBC patients were randomly assigned to receive either UDCA or placebo for up to 4 years have shown that the survival free of liver transplantation was significantly improved in the UDCA-treated arm compared to the controls (Fig. 16.3) [43]. The effect of UDCA on the development of esophageal varices has been evaluated in a prospective clinical trial of 180 patients who received either UDCA or placebo for 4 years [33]. At baseline, 22.8 % had varices. After 4 years of treatment, the risk of developing varices in the UDCA- versus placebo-treated patients was 16 % versus 58 %, respectively [33].


Fig. 16.3
Combined analysis of randomized controlled trials of ursodeoxycholic acid in primary biliary cirrhosis. Poupon RE, Lindor KD, Cauch-Dudek C, et al. Gastroenterology. 1997; 113: 884–890


A Markov model, using death, liver transplantation, and histological stage progression as the main clinical end points, was used to assess the survival of 262 PBC patients (Fig. 16.4) [44]. In this prospective follow-up study, patients received UDCA at a dose of 13–15 mg/kg daily for a mean of 8 years (range: 1–22 years). The overall 10- and 20-year survival rate was substantially better than that predicted by the model [44]. The predicted survival rate was 92 % at 10 years and 82 % at 20 years [44]. The predicted survival rate without liver transplantation was 84 % at 10 years and 66 % at 20 years [44]. The survival predicted by the updated Mayo model was far poorer than the survival predicted on UDCA therapy [44]. In patients with early histological-stage PBC, the predicted survival rate was 97 % at 10 years and 88 % at 20 years. In the same group, the predicted survival without liver transplantation was 93 % at 10 years and 77 % at 20 years, which was similar to that of a matched control population [44]. However, patients with late histological stage PBC had worse 10- and 20-year survival rate without liver transplantation compared to that of a matched control group (68 % and 48 % versus 82 % and 71 %, respectively) [44].


Fig. 16.4
The effect of urosdeoxycholic acid therapy on the natural course of primary biliary cirrhosis. Corpechot C, Carrat F, Bahr A, et al. Gastroenterology. 2005; 128(2): 297–303

Models using time-fixed Cox proportional hazards have been developed to assess the survival in PBC patients. The Mayo PBC risk score [26] is the most widely used model to assess the survival in PBC patients. This model helps determine treatment success in clinical trials of PBC and also guides timing for liver transplantation. Recently, serum ALP and bilirubin have been shown to be excellent predictors of long-term outcomes (death or liver transplantation) in PBC [45]. These biochemical markers are useful surrogate endpoints when designing clinical trials assessing newer therapies in PBC [45].

Diagnosis of PBC

Liver Biochemical Tests

Elevated serum ALP is the most frequent serum biochemical ab normality detected in patients with PBC [15, 16, 19, 22, 4648]. Other biochemical abnormalities in PBC patients are mildly elevated levels of liver transaminases and elevated immunoglobulin M (IgM) levels [11]. Elevated serum bilirubin is rarely seen in early PBC; levels tend to increase as the clinical and histological stage of the disease progress [21]. In addition to the liver-related serum biochemistries, PBC patients frequently have elevated serum lipids, namely serum cholesterol levels [49].


The serological hallmark of PBC is the presence of AMA; a highly disease-specific autoantibody found in nearly 95 % of cases of PBC [8, 9]. An AMA titer of ≥1:40 is indicative of PBC. This autoantibody targets members of a family of mitochondrial enzymes, the 2-oxo-acid dehydrogenase complexes, and include pyruvate dehydrogenase complex (PDC-E2), branched chain 2-oxo-acid dehydrogenase complex (BCOADC-E2), and 2-oxo-glutaric acid dehydrogenase complex (OADC-E2) [57]. More specifically, the lipoylated domains of the E2 and E3 binding protein (E3BP) components of the PDC-E2 and the E2 components of the OADC-E2 and BCOADC-E2 are the epitopes recognized by the AMAs [5, 6]. In HEp-2 cell monolayers, AMAs typically exhibit cytoplasmic “string of pearls” fluorescence staining with a coarse, filamentous, granular, and speckled pattern [50]. The presence of AMA in the sera of patients with PBC was first described in 1965 [4], and in 1987, the AMA antigens were cloned and identified [7, 51, 52]. Enzyme-linked immunosorbent assay (ELISA) is the most widely used method of detection of AMA in commercial laboratories [12]. The magnitude of antibody level correlates poorly with the stage of PBC [11], and may persist after liver transplantation [53]. AMAs are rarely found in healthy individuals. In an Italian study involving 1530 individuals, 0.5 % tested positive for AMA [10].

In addition to AMA, antinuclear antibodies (ANAs) may also be detected in patients with PBC [54]. Some ANAs have been found to be of diagnostic and prognostic value. Anti-Sp100 antibodies (present in 17–41 % of PBC cases) [5561], anti-Sp140 antibodies (present in 11–15 % of PBC cases) [62], and antinucleoporin p62 antibodies (present in 13–32 % of PBC cases) [63, 64] are thought to be specific for PBC, therefore, they can be useful when the diagnosis of PBC is uncertain. In a Japanese study that included 276 PBC patients [65], the presence of anti-gp210 antibodies wa s identified as an important risk factor for progression to liver failure and need for transplantation [65], whereas the presence of anti-centromere antibodies was identified as an important risk factor for development of esophageal varices and hepatocellular carcinoma [65]. Other ANAs found in the sera of PBC patients are the anti-promyelocytic leukemia proteins antibodies [66], anti-SUMO antibodies [67], and anti-lamin B receptor antibodies [6870]. The clinical significance of the latter ANAs is yet to be determined.


PBC is characterized by chronic, nonsuppurative cholangitis that mainly affects the interlobular and septal bile ducts [71]. The term “florid duct lesion ” is used to describe the intense inflammatory lesions around the bile ducts [7274]. The intense inflammatory infiltrate consists of lymphocytes, plasma cells, macrophages, polymorphonuclear cells, and in some cases, epithelioid granulomas [7579]. Histological staging systems in PBC have been developed by Rubin, Popper, and Schaffner [80], Scheuer [81], and Ludwig [71]. Of all the staging systems available, Ludwig’s staging system is the most widely used, in which stage I is characterized by inflammation limited to the portal space, stage II is characterized by inflammation involving the periportal areas as well, stage III is characterized by septal fibrosis or inflammatory bridging, and stage IV represents cirrhosis. Nodular regenerative hyperplasia (NRH) is a known complication in PBC [8286] and should be differentiated from cirrhosis. Liver biopsy is not routinely used in clinical practice to diagnose PBC, as ~95 % of cases of PBC are AMA-positive. Biopsy may be indicated when the suspicion for PBC is high in the absence of AMA [11]. Liver biopsy may also be indicated in patients in whom the suspicion for PBC-Autoimmune hepatitis (AIH) overlap syndrome is high [11]. In these circumstances, patients may h ave histological features of AIH such as periportal or periseptal lymphocytic piecemeal necrosis [87].

Role of Imaging

Imaging is not necessary to establish the diagno sis of PBC, but may be performed at the time of presentation to exclude biliary obstruction. Ultrasound or magnetic resonance cholangiography (MRC) are typically performed. In a study of 117 PBC patients, all extra-hepatic ducts were normal on cholangiograms [47]. The intra-hepatic ducts, however, were abnormal in 9 % of patients, revealing mild tapering, narrowing, and irregularity [47]. Transient elastography (TE), a simple and noninvasive procedure, has been shown to be useful for assessing liver fibrosis in PBC when compared with other surrogate markers of liver fibrosis [88]. Larger and longer-term studies are needed to validate these findings.

Diagnostic Approach

The diagnosis of PBC can be made when two of the three following criteria are met (the American Association for the Study of Liver Disease (AASLD) Guidelines) [11]:

  1. (a)

    Biochemical evidence of cholestasis, mainly elevated serum ALP


  2. (b)

    Presence of AMA


  3. (c)

    Histopathological evidence of PBC, when liver biopsy is performed.


Clinical Manifestations of PBC


Fat igue

Fatigue is the most common symptom in PBC, affecting nearly 80 % of individuals [89]. Severe fatigue can have a severe negative impact and has been associated with an increased mortality, depression, and poor quality of life [9097]. Fatigue in PBC does not correlate with disease activity and seems not to respond to therapies, including UDCA [96]. The etiology of fatigue in PBC is poorly understood. It is, however, thought that chronic cholestasis that occurs in PBC causes degenerative changes in areas in the brain that regulate autonomic functions, ultimately resulting in impaired delivery of oxygen to the peripheral tissue which in turn leads to expression of fatigue and its associated cognitive impairment through secondary dysfunction of peripheral muscles [96, 98]. Evidence in favor of an organic central nervous system process in PBC comes from neurophysiological studies which suggest organic brain changes in PBC. Newton et al. [99] found that 53 % of PBC patients had concentration and memory problems, and that they repetitively failed neuropsychiatric testing. These findings progressed over a 2-year follow-up period [99]. Patients with symptomatic PBC have worse outcomes when compared to patients with asymptomatic PBC [100]. Current therapies seem to be ineffective in the treatment of fatigue in PBC, including liver transplantation, as a significant proportion of patients with PBC continue to suffer severe fatigue even a fter liver transplantation [101]. This highlights the need to understand the underlying mechanism(s) of fatigue in PBC, as it will help develop therapies for this debilitating symptom.


Pruritus is a less frequent, but more specific, sympt om than fatigue in patients with PBC [102, 103]. It affects 20–70 % of PBC patients [102106]. The pruritus of cholestasis tends to be generalized. It leads to scratching, sometimes violent, resulting in excoriations and prurigo nodularis [107]. This type of pruritus can lead to sleep deprivation, depression, and in some patients, to suicidal ideations [107]. A survey was conducted in 239 PBC patients to understand how patients with cholestatic pruritus perceive pruritus [108]. Of these, 69 % reported itching. Seventeen percent reported that itch was “relentless” or so severe that it lead to wanting to “tear the skin off”, and 3.6 % of patients reported that they itched until they bled [108]. Seventy-four percent of the 162 respondents who addressed the qu estion reported that itch affected sleep, 65 % that the itch was worst at night, and 11 % reported that nothing provided relief [108]. The etiology of pruritus in cholestasis is unknown. Accumulation of bile acids in tissues [109, 110], excess of histamine in patients with liver disease and pruritus [111], and excess of substance P [112, 113] (an excitatory neurotransmitter) have been proposed as mechanisms by which pruritus is triggered in patients with liver disease. More recently, lysophosphatidic acid (LPA) and autotaxin, the serum enzyme converting lysophosphatidylcholine into LPA, have been found in higher concentrations in the sera of patients with cholestatic disorders, including PBC, compared to control subjects, suggesting that LPA and autotaxin play a role in the pathogenesis of cholestatic pruritus [114]. The natural history of pruritus in PBC has been inadequately studied and most data is derived from clinical trials of therapies in PBC. Talwalkar et al. [103] examined the natural course of pruritus in PBC patients enrolled in a multicenter, randomized, placebo-controlled clinical trial of UDCA in PBC. They reported that the overall prevalence of pruritus in the placebo group did not differ between study entry and follow-up at 36 months (56 % versus 49 %) [103]. In addition, 30 % of patients in the UDCA group reported symptom improvement compared to 24 % of the placebo-treated patients after 1 year of therapy [103]. Conversely, only 7.9 % of patients in the UDCA group reported development of pruritus compared to 14.5 % patients in the placebo group after 1 year of therapy [103]. Similar to fatigue, pruritus can have a negative impact on the patients’ quality of lives [107].

Other Conditions and Symptoms Associated with PBC

A number of conditions a re associated with PBC [115]. These include Hashimoto’s thyroiditis (12.5 %), Grave’s disease (1.9 %), Raynaud’s disease (18 %), Sjogren’s syndrome (34.3 %), systemic lupus erythematosus (2.2 %), scleroderma/CREST (6.1 %), rheumatoid arthritis (6.1 %), cutaneous autoimmune diseases (5 %), celiac disease (1.4 %), and vasculitis (2.2 %) [115]. Female patients are more likely to have PBC in association with these conditions than male patients [115]. The presence of these conditions does not reduce the survival in PBC patients [115].

Physical Examination

The physical examination in patie nts with PBC is usually normal. Signs of hyperlipidemia such as xanthomas and xanthelasmas can be found. Ascites, splenomegaly, hepatomegaly, and spider angiomata are frequently found when PBC is complicated by portal hypertension [11, 116]. Jaundice and hepatic encephalopathy are signs of advanced disease [20].

Portal Hypertension

Portal hypertension is a feared complication of PBC. The d evelopment and burden of esophageal varices in PBC has been prospectively examined in 265 patients with PBC (69 % had stage III and IV PBC at baseline) enrolled in a clinical trial [117]. Patients were followed for a median of 5.6 years. Esophageal varices developed in 31 % of patients, 48 % of whom experienced ≥1 episodes of variceal bleeding. After the development of varices, the 3-year survival was 59 %, and after the initial variceal bleeding episode, the 3-year survival was 46 % [117]. Unlike other liver diseases, patients with PBC can develop portal hypertension and gastroesophageal varices in the pre-cirrhotic stages of PBC [84, 86]. In a study of 325 patients with PBC enrolled in two clinical trials, 127 patients were identified as early-stage (stage I and II) PBC at baseline; 6 % of those with early-stage PBC had gastroesophageal varices at baseline [118]. A number of noninvasive tools using simple biochemical markers have been developed to assist clinicians in identifying PBC patients whom might benefit from screening upper endoscopy for gastroesophageal varices. Patanwala et al. [119] developed the Newcastle Varices in PBC (NVP) score that uses the following parameters: serum ALP, serum albumin, and platelet count. This noninvasive tool has been developed in a large well-characterized cohort of PBC and has been validated internally and exter nally with excellent performance (93 % sensitivity, 93 % negative predictive value, and a discriminating value “AUROC” of 0.86).

Bone Disease

Osteoporosis is a frequent complication of PBC [120]. Numerous studies have reported a strong association between low bone mass and PBC [121126]. The finding of lower levels of osteocalcin (a marker of bone formation) and higher levels of urinary hydroxyproline (a marker of bone resorption) among PBC patients than in controls lends support to this phenomenon [127]. Studies have reported a prevalence of 20–35 % of osteoporosis among the PBC population [121, 127, 128]. Patients with PBC have a 30-fold increased risk of osteoporosis when compared to the normal population, and patients with advanced-stage PBC have a 5.4-fold increased risk of developing osteoporosis compared to their counterparts with early-stage PBC [128]. Reports evaluating the risk factors for developing osteoporosis in PBC have identified increasing age, low body mass index, previous fractures, increasing serum bilirubin, severity of cholestasis, and advanced histological stage of PBC as independent risk factors for osteoporosis in PBC [127, 128]. The rate of bone loss during the early histological course of PBC is slower than that in patients with advanced histological-stage PBC [128]. As the histological course in patients with early PBC progresses, the rate of bone loss equals that in patients with advanced histological-stage PBC [128]. Typically, PBC patients suffer from osteoporosis of the lumbar spine and hip area, and the rate of bone loss in the lumbar spine correlates with that in the hip bone [128].


Hyperlipidemia is commonly associated with PBC, occurring in 75–95 % of cases [129]. Typically, patients with PBC have markedly elevated total cholesterol (up to 1775 mg/dL has been reported [130]), elevated high-density lipoprotein (HDL), and elevated low-density lipoprotein (LDL) [130]. Patients with advanced PBC tend to have higher LDL levels when compared to patients with early PBC [130]. The mechanisms of hyperlipidemia in PBC are different than those in other conditions. In vitro studies suggest that biliary cholestasis, lipid reflux from the biliary ducts into the bloodstream, and an increased cholesterol synthesis lead to the hyperlipidemia seen in cholestatic disorders [131136]. Hyperlipidemia associated with PBC does not place PBC patients at increased risk for atherosclerotic-related deaths. In one study, the reported percentage of atherosclerosis-related deaths in a cohort of 312 patients with PBC whom were followed for 7.4 years was only 2.2 % [130]. Importantly, the incidence of atherosclerosis death among the PBC patients was not different when compared with a matched U.S. control population [130].

Vitamin Deficiency

PBC patients may have decreased bile acid secret ion into the intestines, leading to an increased risk for lipid malabsorption. Clinically important deficiencies of fat-soluble vitamins A, D, E, and K are uncommon in PBC patients [137141]. Fat-soluble vitamins may be decreased in patients with advanced PBC, leading to night blindness, neuropathy, and prolonged prothrombin time [11, 141].

Etiology of PBC

The etiology of PBC is poorly understo od. Findings from several studies suggest a role for genetics and environmental factors in the pathogenesis of PBC. Family studies revealed that the prevalence of PBC is ~0.72 % and 1.2 % in first-degree relatives and offspring of affected individuals respectively [142, 143]. A large study of first-degree relatives of PBC patients found that 20 % of sisters, 15 % of mothers, and 10 % of daughters of PBC patients were positive for AMA [144]. Several genome-wide association studies (GWAS) have identified loci (such as HLA, IL12A, and IL12RB2, SPIB, IRF5TNPO3 and 17q12-21, STAT4, DENND1B, CD80, IL7R, CXCR5, TNFRSF1A, CLEC16A, and NFKB1) strongly associated with PBC [145149]. Specific mutations of the X chromosome, particularly X monosomy, have been linked to the development of PBC [150].

Data from several studies suggest that environmental factors may play a role in the development of PBC. Infections, particularly urinary tract infections caused by E. coli, and the xenobiotics have been linked to the development of PBC [143]. An early association between PBC and UTI has been reported, as in one study, 59 % of 1032 PBC patients reported a history of UTI [151]. Interestingly, PBC patients’ sera react with both E. coli and PDC-E2, and there is cross-reactivity between antibodies in PBC patients and enzymes secreted by E. coli [152, 153]. The xenobiotic 2-octynoic acid, used as a food additive and in manufacturing nail polish, reacts to AMA, and when injected into mice, it results in hig h titers of AMA and development of PBC-like histological lesions [154, 155]. The clustering of cases of PBC around areas of superfund toxic waste sites has been recently reported [156], suggesting that toxin exposure may play a role in the development of PBC. Smoking cigarettes and use of hormone replacement therapies have also been associated with PBC [12, 143].

Epidemiology of PBC

Studies report a prevalenc e of PBC ranging from 19 to 365 cases per million in the United States, Canada, Australia, and Europe [157159]. In Olmsted County, Minnesota, the reported age-adjusted incidence of PBC per million persons was 45 for women and 7 for men, with an overall incidence of 27 per million persons [158]. In a recent Canadian epidemiological study [160], the reported overall age- and sex-adjusted annual incidence of PBC in the Calgary Health Region was 30.3 cases per million (48.4 per million in women and 10.4 per million in men) [160]. European epidemiological studies have estimated PBC incidence rates of 4–58 per million persons-years [161167]. Recent reports suggest that the incidence and prevalence of PBC might be increasing. In Sheffield, United Kingdom, the incidence of PBC has increased from 5.8 to 20.5 cases per million between the years 1980 and 1999 [46, 168]. In Finland, the incidence and prevalence of PBC increased from 12 cases to 17 cases and from 103 cases per million to 180 cases per million, respectively, in the time period between 1988 and 1999 [159]. In the Calgary Health Region, Canada, the prevalence increased from 100 cases per million in 1996 to 227 cases per million in 2002 [160]. It is still unknown whether the trends in the PBC epidemiology reflect true increase in the frequency of PBC cases or an increase in awareness of the disease by physicians and, perhaps, prolonged survival of PBC patients after UDCA has been introduced. Indeed, recent reports suggest that the absolute number of PBC patients undergoing transplantation for PBC has been decreasing [169], reflecting a change in the natural history of PBC follo wing the introduction of UDCA.

Therapy for PBC

Food and Herbals

No clinical evidence exists to support the use or avoidance of specific foods or herbal supplements in PBC patients.

Herbal and alternative medicines have seldom been e xamined in patients with PBC. Silymarin has tested in combination with UDCA in PBC but offered little benefit [170]. Currently, no clinical evidence exists regarding safety or efficacy of other herbal products.


In addition to being safe, several randomized controlled clinical studies reported that the use of UDCA in patients with PBC not on ly improves liver biochemistries, but also delays histological progression, delays the development of esophageal varices, improves the liver-transplantation-free survival, and in a selected group of PBC patients, it improves the survival [3234, 3641, 43, 44, 171]. At least four mechanisms have been proposed by which UDCA exerts its therapeutic effects: (1) UDCA inhibits the intestinal absorption of toxic bile acids [172176], (2) UDCA stimulates biliary secretion of bile acids and organic anions, thereby preventing cholestasis induced by toxic bile acids [177180], (3) UDCA exerts cytoprotective effects against the hepatotoxic effects of the toxic bile acids [181185], and (4) UDCA may have anti-inflammatory and immunomodulatory properties, based on results from animal experiments [186, 187]. UDCA is the only medical therapy approved by the FDA for treatment of PBC. The recommended dose is 13–15 mg/kg/day, and it should be started in all patients with PBC regardless of the stage of the disease, as long as liver biochemistries are abnormal [11]. Lifelong treatment with UDCA is recommended. Biochemical response to UDCA at 1 year of treatment is a strong predictor of long-term prognosis [38, 188, 189]. Biochemical response has been defined by numerous criteria: the Mayo Clinic criteria (decrease in serum ALP < 2 times the upper limit of normal (ULN)) [190], the Spanish criteria (decrease in ALP < 40 % from baseline or to normal value) [38], the French criteria (decrease in ALP < 3 times ULN, decrease in aspartate aminotransferase <2 times ULN, and decrease in bilirubin <1.0 mg/dL) [188], and the Dutch criteria (normalization of bilirubin and/or albumin after treatment if one or both were abnormal at baseline) [189] have been commonly used. Approximately 40 % of PBC patients have incomplete response to UDCA. This group of patients is at high risk for serious outcomes [191].

UDCA is generally safe; no serious adverse events have been reported. Weight gain, loose stools, and hair thinning have been infrequently reported and have not raised issues of patients’ noncompliance to therapy [11, 192].

Management of Symptoms of PBC


A supportive positive approach to the management of sy mptoms in PBC, in particular fatigue, is vital and in itself can lead to improvements in quality of life. The two most important features associated with fatigue in PBC are excessive daytime sleepiness and autonomic dysfunction [96]. Therefore, it is crucial to rule out other causes of excessive daytime sleepiness and autonomic dysfunction such as obstructive sleep apnea, anemia, malabsorption, cardiac failure, hypothyroidism, adrenal insufficiency, diabetes mellitus, and use of sleep aid medicines, sedatives, narcotics, and antihypertensive medicines.

UDCA, Ondansetron (a serotonin receptor 3 antagonist) and Fluoxetine (a selective serotonin reuptake inhibitor) have not improved PBC-associated fatigue [193, 194]. Modafinil, a stimulant, has been investigated as a treatment option in PBC patients suffering fatigue [195]. In an open label study using modafinil [196], 14 PBC patients achieved objective short-term benefits in terms of daytime excessive sleepiness and fatigue. At 14 months follow-up, 66 % of patients failed to tolerate modafinil long-term [196]. Larger and longer-term placebo-controlled studies are needed to investigate the role of modafinil in PBC-associated fatigue. A small placebo controlled clinical trial did not show benefit [197].

Liver transplantation is reserved for PBC patients with fatigue who failed conservative therapies. Although it has been reported that liver transplantation improves fatigue in patients with PBC [101], a considerable proportion of patients with PBC continue to suffer from severe fatigue after liver transplantation [101].


UDCA does not relieve pruritus in patients with PBC. The effects of several antipruritic agents have been investigated in patients with PBC. Liver transplantation is reserved for patients with intractable pruritus after they fail all existing therapies.

Procedures That Remove the Pruritogens from the Body

The most commonly used drug in this class is Cholestyramine [198]. It is a resin that is not absorbed from the intestines and binds anions in the small intestines increasing their fecal excretion, including bile acids and cholesterol [199]. The use of cholestyramine has been associated with improvement of pruritus in patients with PBC [198, 200]. The recommended dose in PBC-related pruritus is 4 g per dose and not exceed 16 g/day, given 2–4 h before or after UDCA [11]. The side effects of this resin tend to be minor (mainly bloating and diarrhea) [199]. It is recommended to take cholestyramine immediately before and after breakfast, as the rationale for its use is to bind the pruritogens that accumulate in the gallbladder during the overnight fast and that are secreted into the small intestine after breaking the fast. Coleavalam, also a resin, was tested in a placebo-controlled clinical trial. The effect of this resin was not better than that of placebo [201].


The use of Rifampicin , an antibiotic, has been associated with relief of pruritus in PBC patients [202205]. The mechanism of action of this antibiotic as an antipruritic agent is poorly understood. The recommended dose ranges between 300 mg daily and 600 mg (in 2 divided doses) daily [11]. It was concluded in a meta-analysis study of several clinical trials that rifampicin is safe [206]; however, there is a risk of hepatotoxicity [207], severe hemolytic anemia and nephrotoxicity [204] with the use of rifampicin, but these are rare events. Serial liver chemistries and renal function tests are recommended in patients using this drug.

Opiate Antagonists

Naloxone and Naltrexone are opioid antagonists used for the treatment of the pruritus of cholestasis [208217]. They act by decreasing the opioidergic tone [112]. To decrease the probability of an opiate withdrawal-like reaction (characterized by tachycardia, abdominal pain, high blood pressure, goose bumps, nightmares, and depersonalization) that some patients experience, it is recommended by some experts to initiate treatment in a controlled environment (surgery suite, specialty clinic, etc.) using intravenous infusions before instituting the oral forms [199]. The dose should be increased gradually to avoid opiate withdrawal-like reactions, until relief of pruritus is achieved. The metabolism of naltrexone is slow in patients with cirrhosis [218], therefore caution should be exercised when using this drug in this population. A rare but serious adverse event associated with using naltrexone is hepatic failure [219], therefore, routine liver chemistries are recommended in patients using this medication.

Other Agents

Serotonin Antagonists

The serotonin system participates in the neurotransmission of nociceptive stimuli. Odansetron, a serotonin receptor 3 antagonist, has been examined in PBC patients with pruritus but provided only minimal relief of pruritus [220223].


Selective serotonin reuptake inhibitors have been shown to have antipruritic effects. Sertraline (75–100 mg daily) has been associated with relief of pruritus in PBC patients [224].


Phenobarbital has been shown to have antipruritic effects [225, 226]. The sedative effect of phenobarbital may also be associated with its ameliorating effects on pruritus.


Antihistamines have been associated with relief of pruritus in PBC patients [227]. The sedative effect of antihistamines may also help patients sleep, as deprivation of sleep is a significant problem in PBC patients suffering pruritus.

Other Options

A transient relief of pruritus has been reported in association with anion adsorption and plasma separation, and the extracorporeal liver support systems [228230]. This option should be reserved only for patients with intractable pruritus who failed other therapies.

Plasmapheresis seems to be effective in relieving pruritus in PBC patients who have intractable pruritus and failed medical therapy. Old reports have shown that plasmapheresis results in prompt relief of pruritus, decrease in serum bilirubin, and a decrease in the bile acid pool [231, 232]. The duration of clinical response may vary; up to 5 months of pruritus relief following plasmapheresis has been reported by a few patients [232]. Unfortunately, patients reported return of symptom to the pre-plasmapheresis degree 2–3 weeks after the last session of plasmapharesis. The mechanism by which plasmapheresis results in pruritus relief is removal of pruritgens, immune complexes, and bile salts, but this is not clear yet [231]. Most patients with PBC tolerated plasmapheresis quite well, and no adverse events related to plasmapheresis in PBC patients have been reported [233].

Management of Sicca Syndrome

Patients with Sicca syndrome generally suffer from dry eyes and dry mouths, with their consequent complications. Therefore, measures should be taken to prevent complications of these symptoms. The use of artificial tears and humidification of the house environment are recommended [11]. Cyclosporine ophthalmic solution can be used under the supervision of an ophthalmologist when conservative measures fail [234]. Measures to improve oral health include regular visits to the dentist, use of fluoride-containing toothpastes, daily flossing, and avoidance of sugar-containing snacks between meals [235, 236]. Chewing sugar-free gum can improve saliva production as well as the use of cholinergic agents such as pilocarpine and cevimeline [11]. Oral candidiasis can occur as a complication of dry mouth and requires specific intervention [11]. In mild cases, nystatin solution might help reduce the symptoms and duration of infections. Oral and systemic antifungals such as fluconazole are indicated in severe cases of oral candidiasis.

Management of Sjogren’s Syndrome and CREST Syndrome

The management of CREST (C-calcinosis, R-Raynaud’s, E-esophageal dysfunction, S-sclerdactyly, T-telangiectasia) requires a specialized team effort and these patients should be referred to rheumatologists and other appropriate subspecialties.

Complications Related to Cirrhosis

Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is a recognizable complication of PBC. The exact frequency of HCC in the PBC population is unknown but is estimated to be between 0.7 and 5.9 % [237242]. HCC development significantly affects the survival of patients with PBC. The reported 5- and 10-year survival times for patients with PBC who did not develop HCC versus those who did develop HCC was 95 % vs. 75 % and 85 % vs. 45 %, respectively [242]. Older age, male sex, previous blood transfusion, presence of portal hypertension, and advanced histological stage of PBC have been identified as independent risk factors for development of HCC in patients with PBC [240242]. Recently, it has been reported that patients with PBC who developed HCC and underwent liver transplantation had better survival than those who did not undergo liver transplantation [238]. Regular screening for HCC with cross-sectional imaging with or without alpha fetoprotein at 6- to 12-month intervals is recommended for all patients with liver cirrhosis [243].

Portal Hypertension

Portal hypertension is a frequent complication in cirrhotic patients. There is still a debate on timing of screening PBC patients for esophageal varices. A number of noninvasive tools have been proposed to be used as indicators for the presence of esophageal varices in PBC patients: (a) a platelet count of <200,000/mm3, an albumin level <4.0 g/dL, and a bilirubin level >1.2 mg/dL [244], (b) a Mayo risk score of ≥4.0 [190], (c) a platelet count <140,000/mm3 and/or a Mayo risk score of ≥4.5 [245], (d) the MABPT model (M-male sex, A-albumin <3.5 g/dL, B-bilirubin ≥1.2 mg/dL, PT-prothrombin time ≥12.9 s) [118], (e) the Newcastle Varices PBC score (uses platelet count, albumin, and ALP level) [119]. Only two scores have been cross-validated in independent sets of PBC patients [119, 245].

Management of Portal Hypertension

The management of esophageal varices in patients with PBC follows the guidelines published by the AASLD [246]. Screening upper endoscopy is indicated all PBC patients when the diagnosis of cirrhosis is made. Nonselective beta blocker therapy (propranolol or nadolol) or endoscopic variceal ligation is recommended in patients who have medium to large varices that have not bled and have a high risk of hemorrhage (Child score B/C or variceal red wale markings on endoscopy) [246].

Unlike other liver diseases, varices can develop in the early histological stages of PBC [118, 247]. In patients with pre-cirrhotic PBC and varices who fail traditional therapies (i.e. nonselective beta blockers, ligation, or both), a distal spleno-renal shunt (DSRS) might be an alternative to prevent recurrent bleeding [248]. This approach does not deprive the liver of its blood supply and helps preserve the hepatic function in patients with PBC [248]. DSRS is rarely performed nowadays for this indication.

Complications Related to Chronic Cholestasis

Osteopenia and Osteoporosis

Patients with PBC are at significantly higher risk for osteopenia and osteoporosis and their consequent complications when compared to a matched population [120, 127, 128, 249]. Dual-energy X-ray absorptiometry (DXA) is used to measure the bone mineral density (BMD) and is the gold standard for diagnosis of osteopenia and osteoporosis. A baseline screening DXA is recommended in all patients when a diagnosis of PBC is confirmed, with follow-up DXA every 2–3 years [11]. Calcium (1500 mg daily) and vitamin D (1000 IU daily) supplements may be used if there is no history of renal stones [11]. Weight-bearing and muscle-strengthening exercise, smoking cessation, and avoidance of excessive alcohol intake are generally recommended [250]. Bisphosphonate therapy (namely alendronate 70 mg orally weekly) is recommended in PBC patients with osteoporosis. In a randomized controlled clinical trial, alendronate significantly improved bone mineral density when compared to etidronate and placebo [251253]. Alendronate should not be used in patients with acid reflux or known varices [11].


PBC is frequently complicated by hypercholesterolemia, which poses no additional risk for atherosclerotic-related death [49, 130]. When classic risk factors for cardiovascular and cerebrovascular diseases are present, such as family history of myocardial infarction, diabetes mellitus, and hypertension, the use of statins is appropriate provided no contraindications to their use exists. The use of statins in PBC is safe [129], even in the presence of abnormal liver biochemistries. UDCA might be helpful in reducing serum cholesterol levels in PBC patients. In a randomized placebo-controlled clinical trial of 177 PBC patients with hypercholesterolemia, UDCA significantly reduced serum total cholesterol levels compared to placebo [254].

Liver Transplantation

In the 1980s, PBC was the leading indication for liver transplantation across North America and Europe [11]. Following the introduction of UDCA, the natural history of PBC has changed significantly [31]. With the use of UDCA in PBC patients, the transplant-free survival rates have considerably improved, even among patients with advanced stages of PBC when they demonstrate biochemical response to UDCA, as defined by various international groups. Patients with early-stage PBC on UDCA live just as long as the healthy population [31, 44]. In the United States the burden of PBC on liver transplantation has reduced over a 12-year period. Lee et al. [169] reported that the absolute number of liver transplants for PBC has decreased an average of 5.4 transplants per year and the absolute number of cases of PBC added to the liver transplant waiting list has also significantly decreased between the years 1995 and 2006 [169]. Despite these facts, 40 % of PBC patients have inadequate response to UDCA and are at high risk for adverse outcomes and many progress to end-stage liver disease [191].

Liver transplantation remains the only curative option for PBC patients with end-stage liver disease and is the sixth leading indication for liver transplantation (Fig. 16.5) [11, 255]. The reported 5-year survival in PBC patients who underwent liver transplantation in North America and Europe in the late 1990s ranged between 78 and 87 % [255257]. Acute rejection of the transplanted graft occurs in 46–56 % of PBC patients but is rarely of clinical significance, as it responds very well to increasing the immunosuppression [258260]. Chronic rejection of the transplanted graft is a more serious but rare problem, occurring in 2–9.3 % [261].


Fig. 16.5
Actual (Kaplan–Meier) survival after transplantation in 161 patients with primary biliary cirrhosis and estimated survival without transplantation as predicted by the Mayo Model (simulated control). Markus BH et al. N Engl J Med 1989; 320: 1709–1713

About 20–25 % of PBC patients who undergo liver transplantation develop recurrent PBC [11]. Liver biopsy remains the gold standard for diagnosis of recurrent PBC in the transplanted graft, as AMAs may persist in the sera of PBC patients even after liver transplantation [53]. Tacrolimus-based immunosuppression post-transplantation, male sex and advanced recipient’s age have been identified as risk factors for recurrent PBC in the transplanted graft [258, 260]. It has been reported that liver transplantation improves fatigue and pruritus in PBC patients [101]. However, a significant proportion of PBC patients continue to suffer from severe debilitating fatigue after liver transplantation [101].

General Advice

Pregnancy, Hormone Replacement, and PBC

Because up to 25 % of cases of PBC present younger than the age of 50, some women with PBC will be of childbearing age at the time of diagnosis [262]. Pregnancy in women with PBC is frequently symptomatic, with up to 53 % of pregnant women with PBC developing de novo pruritus, and up to 71 % requiring symptom-specific therapy [263]. Liver biochemistries remain stable through the pregnancy in 70 % of cases, however, 72 % of PBC cases develop biochemical flare up in the post-partum period [263]. Pregnancy in PBC women is mostly uneventful, with 91 % of women reporting at least one successful live birth [263]. Moreover, UDCA appears to be safe during pregnancy and lactation. In one study, 6 women with PBC took UDCA at various time points during pregnancy without adverse fetal consequences [263].

As with all other women with cirrhosis who become pregnant, it is advisable to screen for varices in the second trimester [11]. Nonselective beta blockers are safe during pregnancy [11].

Estrogens promote cholestasis, so oral contraceptives and pregnancy can induce or worsen pruritus in patients with PBC [11].

Screening Family Members

First-degree family members of PBC patients are at i ncreased risk for developing PBC [144]. Screening for PBC is done by measuring serum ALP and AMA. The clinical value of screening family members of patients with PBC is unclear.

Long-Term Follow-Up

UDCA should be continued indefinitely [11]. Liver biochemistries should be assessed at 3- to 6-month intervals. Thyroid function tests should be performed annually or when suspicion for hypothyroidism or hyperthyroidism is high [11]. Patients with a new diagnosis of cirrhosis should undergo screening endoscopy for esophageal varices [246]. No consensus exists on screening pre-cirrhotic PBC patients for esophageal varices; using any of the noninvasive markers as a guiding tool for screening for esophageal varices is reasonable. DXA at baseline and every 2–3 years is recommended in all patients with PBC to screen for osteopenia and osteoporosis and to monitor bisphosphonate therapy [11]. Cross-sectional imaging with or without measuring alpha fetoprotein level every 6–12 months in PBC patients with cirrhosis is recommended [243].

Special Cases of PBC

AMA-Negative PBC

AMAs are present in 95 % of PBC patients. Five percent of PBC cases are AMA-negative [264]. In these circumstances, a liver biopsy is indicated to establish the diagnosis. PBC-specific ANAs mig ht also be helpful in this setting. The percentage of AMA-negative patients might decrease in the future with the development of newer, more sensitive ELISA screening techniques [265]. The natural course of patients with AMA-negative PBC is similar to that observed in patients with classic AMA-positive PBC [266]. All patients with AMA-negative PBC should receive UDCA, and treatment should continue indefinitely [11]. The same screening procedures and long-term care in patients with classic PBC apply to patients with AMA-negative PBC.

PBC/AIH Overlap Syndrome

The diagnosis of PBC/AIH overlap is suspecte d when patients demonstrate features of both diseases. The true prevalence of this condition is unknown; however, the reported estimated prevalence ranges between 2 and 20 % [87]. Hispanics may be more likely to have PBC/AIH overlap syndrome than non-Hispanics [267].

Diagnosis of PBC/AIH

Diagnosis of PBC/AIH is challenging, largely due to the lack of consensus on the diagnostic criteria for this syndrome. The two most widely used criteria for the diagnosis of PBC/AIH overlap syndrome are the Paris Study Group Criteria [268] and the International Autoimmune Hepatits Group (IAIHG) [269]. The diagnosis of the PBC/AIH overlap syndrome is based on the presence of at least two of the three diagnostic criteria for each disease [87]. For PBC, the diagnostic criteria are (1) ALP levels >2 times ULN, (2) positive AMA, and (3) liver biopsy showing bile duct lesions consistent with PBC. For AIH, the diagnostic criteria are (1) ALT levels >5 times ULN, (2) serum immunoglobulin G >2 times ULN, and (3) liver biopsy showing periportal and/or periseptal lymphocytic piecemeal necrosis. In addition to the mentioned diagnostic criteria, some of the serological markers have been found to be of diagnostic value. The anti-dsDNA antibodies are found more frequently in patients with PBC/AIH overlap syndrome (60 %) than in patients with PBC alone (3 %) or AIH alone (26 %) [270], and positivity for both AMA and anti-dsDNA antibodies have been found in 47 % of cases of overlap syndrome as opposed to 1 % in AIH and 3 % in PBC [270].

Clinical Course of PBC/AIH

The natural history of PBC/AIH is poorly understood. It has been reported that patients with PBC/AIH overlap syndrome had worse outcomes in terms of complications of portal hypertension and need for liver transplantation when compared to patients with PBC alone [271]. More recently, Levy et al. [267] reported that Hispanic subjects with PBC/AIH had worse outcomes in terms of development of portal hypertension, variceal hemorrhage, need for liver transplantation, or death, when compared to a non-Hispanic group of patients with PBC/AIH.

At this time, there is no consensus on the management of patients with PBC/AIH. A combination of UDCA and immunosuppression is a reasonable approach. Given the rarity of this combination, randomized clinical trials are unlikely to occur.

Consecutive PBC/AIH

Rarely, PBC may morph into AIH over time. A review of 282 cases of PBC followed long term suggest that 4.3 % (12/282) develop features of AIH [272]. Of those who developed AIH following PBC, 16.6 % died from hepatic failure secondary to HCC [272].

AMA-Positive AIH

There are case reports of patients with AIH who test positive for AMA, but on long-term follow-up, these patients did not develop PBC [273].

Future Therapies

Obeticholic acid (OCA, INT-747) is a farnesoid X receptor (FXR) agonist [274] that has shown promising results in early clinical trials in patients with PBC who had inadequate response to UDCA [275, 276]. Preliminary reports from the ongoing phase III randomized placebo-controlled clinical trial of OCA in PBC patients showed that 47 % of patients in the 5 g—OCA arm and 46 % of patients in the 5 mg—followed by 10 mg-OCA arm reached the composite primary endpoint of a reduction of ALP to <1.67 times ULN, a total bilirubin within normal limits, and at least a 15 % decrease in serum ALP, compared to only 10 % of patients who received placebo (data presented in the European Association for the Study of Liver 2014 meeting).

NGM282, a novel specific inhibitor of the cholesterol 7α hydroxylase enzyme (the rate-limiting enzyme in bile acid synthesis), is currently evaluated in a phase II clinical trial in PBC patients.

Expert Commentary

Primary biliary cholangitis (PBC), a progressive disease of the biliary tree characterized by cholestasis and damage of the small bile ducts, can lead to cirrhosis and liver failure. The introduction of ursodeoxycholic acid (UDCA) has favorably changed the natural course of PBC. Once a leading indication for liver transplantation, PBC is now the sixth leading indication for liver transplantation. UDCA has been shown to prolong the survival of PBC patients without liver transplantation. In patients with early stages of PBC, institution of UDCA early in the course of the disease can improve the survival. Despite the documented efficacy of UDCA, approximately 40 % of PBC patients show inadequate response to UDCA. These patients have are at high risk for serious complications. Liver transplantation, the only curative option in patients with end-stage liver disease due to PBC, is an invasive procedure and expensive procedure. PBC recurs in the transplanted graft in up to 25 % of cases. These facts underscore the need for newer, more effective therapies for PBC. Obeticholic acid (OCA), a first-in-class farnesoid X receptor (FXR) agonist, has shown promising results during the first year of an ongoing phase III clinical trial in patients with PBC. The use of the genome-wide association studies (GWAS) might help us identify loci of therapeutic importance, which would help us identify future potential therapeutic agents.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Nov 20, 2017 | Posted by in GASTROENTEROLOGY | Comments Off on Autoimmune Liver Diseases: Primary Biliary Cholangitis

Full access? Get Clinical Tree

Get Clinical Tree app for offline access