Guidelines for the management of primary biliary cirrhosis

The Intractable Hepatobiliary Disease Study Group supported by the Ministry of Health, Labour and Welfare of Japan


  • Working Subgroup (English version) for Clinical Practice Guidelines for Primary Biliary Cirrhosis

    Corresponding author
    • *Correspondence: Hiromi Ishibashi, Fukuoka Sanno Hospital, International University of Health and Welfare, 3-6-45 Momochihama, Sawara-ku, Fukuoka, 814-0001, Japan. Email:

    Search for more papers by this author
    • Working Subgroup (English version) for Clinical Practice Guidelines for Primary Biliary Cirrhosis (in alphabetical order): Atsumasa Komori, Clinical Research Center, National Hospital Organization Nagasaki Medical Center; Atsushi Tanaka, Department of Medicine, Teikyo University School of Medicine; Hajime Takikawa, Department of Medicine, Teikyo University School of Medicine; §Hirohito Tsubouchi, Digestive Disease and Life-style Related Disease, Kagoshima University Graduate School of Medical and Dental Sciences and Kagoshima City Hospital; †Hiromi Ishibashi, International University of Health and Welfare/Fukuoka Sanno Hospital and Clinical Research Center, National Hospital Organization Nagasaki Medical Center; Hiroto Egawa, Department of Surgery, Institute of Gastroenterology, Tokyo Women's Medical University; Junko Hirohara, Third Department of Internal Medicine, Kansai Medical University; Ken Shirabe, Department of Surgery and Science, Kyushu University; Kenichi Harada, Department of Human Pathology, Kanazawa University Graduate School of Medicine; Makoto Nakamuta, Department of Gastroenterology, National Hospital Organization Kyushu Medical Center; Mikio Zeniya, Department of Gastroenterology, Jikei University Graduate School of Medicine; Minoru Nakamura, Clinical Research Center, National Hospital Organization Nagasaki Medical Center and Department of Hepatology, Nagasaki University Graduate School of Biomedical Sciences; Nobuyoshi Fukushima, Department of Gastroenterology, National Hospital Organization Kyushu Medical Center; Shinji Shimoda, Medicine and Biosystemic Science, Graduate School of Medical Sciences, Kyushu University; Shotaro Sakisaka, Department of Gastroenterology and Medicine, Fukuoka University Faculty of Medicine; Toshio Morizane, Japan Council for Quality Health Care; Yasuaki Takeyama, Department of Gastroenterology and Medicine, Fukuoka University Faculty of Medicine; ‡Yasuni Nakanuma, Department of Human Pathology, Kanazawa University Graduate School of Medicine; Yoshiyuki Ueno, Department of Gastroenterology, Yamagata University Faculty of Medicine (†Chairperson of the Working Group, ‡Chairperson of the PBC Subcommittee, §Chairperson of the Intractable Hepatobiliary Disease Study Group).

1. Introduction

The Japanese version of the clinical practice guidelines for primary biliary cirrhosis (PBC) was developed in 2012 by the Intractable Hepatobiliary Disease Study Group, with the support of the Ministry of Health, Labour and Welfare of Japan, for the use of general physicians, gastroenterologists and hepatologists who treat patients with PBC.

In preparation for developing the guidelines, the study group reviewed recent studies that provided important evidence or that were published in leading journals with a high impact factor, in addition to considering the formal consensus of experts on PBC or related subjects. Using the core keywords “primary biliary cirrhosis,” a PubMed search was conducted for English-language clinical trials, randomized clinical trials (RCTs) and meta-analyses that were published from January 1998 to December 2009 and that addressed treatment of PBC and its complications, follow-up, indication of and time of consultation for liver transplantation, or time of consultation with specialists. Medical systems and other culture-specific factors in Japan were also taken into account. Members of the task force exchanged ideas frequently during the drafting process to try and establish a consensus. The final draft was made after collecting comments from the public and all the committee members. The level of evidence (LE; Table 1) and the grade of recommendation (GR; Table 2) were based on the Medical Information Network Distribution Service in Japan (MINDS).

Table 1. Level of evidence (LE)
1aSystematic review/meta-analysis of RCTs
1bBased on one or more RCTs
2aBased on non-randomized control study (prospective study)
2bBased on non-randomized control study (historical cohort study and retrospective cohort study)
3Case control study
4Analytical epidemiological study (cross-sectional study)
5Descriptive study (case report or case series)
6Opinion of expert committee, or an expert, not based on patient data
Table 2. Grade of recommendation (GR)
  • Strong recommendation, with high level of evidence
  • Moderate recommendation, with certain level of evidence
  • *Supported by an intermediate level of evidence and considered to be clinically useful
  • *Supported by a high level of evidence but not considered to be clinically very useful
  • *Evidence level is low, but usefulness has already been established in clinical practice
  • C1 Recommendation to be done, without a high level of evidence
  • C2 Recommendation not to be done, without a high level of evidence
  • Recommendation not to be done, as evidence indicates ineffectiveness or harm

After being modified by recent literatures published since 2010, the present English version of the guidelines was developed in order to spread our ideas and exchange opinions with physicians who are involved in the management of PBC patients overseas.

These clinical practice guidelines should be revised at appropriate intervals to incorporate advances in methodology and treatment.

2. Diagnosis of PBC

2.1 Diagnosis of PBC: general principles

PBC is an autoimmune-mediated, chronic cholestatic liver disease that predominantly affects middle-aged women. The initial symptom is most often pruritus, though the disease generally progresses insidiously without symptoms for many years. Jaundice progresses without improvement once it becomes overt, and portal hypertension occurs at a high rate.

Clinically, increased levels of serum biliary enzymes [alkaline phosphatase (ALP) and γ-glutamyl transferase (GGT)] and detection of antimitochondrial antibodies (AMAs) are characteristic. AMAs are found in 95% of patients with PBC when using the most sensitive detection techniques, and have specificity of 98% for the disease (Supporting information Memo 1). Approximately 20–30% of PBC patients are positive for anti-nuclear pore proteins, e.g., anti-gp210, and/or anti-centromere antibodies. Most patients with PBC have an elevated serum IgM concentration, although high serum IgM is not highly specific or sensitive for diagnosis of PBC. The total gamma globulin concentration remains normal until late in the disease when cirrhosis develops.

Histologically, chronic non-suppurative destructive cholangitis (CNSDC) is seen in the intrahepatic small bile ducts at the level of the interlobular and septal bile ducts. Disease progression in PBC results in bile duct loss and liver fibrosis, which develop into biliary cirrhosis and, in some cases, hepatocellular carcinoma.

The differential diagnosis includes autoimmune hepatitis, primary sclerosing cholangitis, drug-induced chronic cholestasis, and paucity of intrahepatic bile ducts, after excluding obstructive jaundice and cholestatic diseases of known etiologies.


  1. Patients with one of the following criteria should be diagnosed with PBC: (1) histologically confirmed CNSDC with laboratory findings compatible with PBC; (2) positivity for AMAs with histological findings compatible with PBC but in the absence of characteristic histological findings of CNSDC; and (3) no histological findings available, but positivity for AMAs as well as clinical findings and a course indicative of typical cholestatic PBC. (GR A)
  2. Diagnosis of PBC should be performed using the criteria endorsed by the Intractable Hepatobiliary Disease Study Group with the support of the Japanese Ministry of Health and Welfare (2010 version, Table 3). (GR A)
  3. Differential diagnosis should be performed for a spectrum of diseases that manifest chronic cholestatic liver dysfunction or immunological disorder with autoantibodies (Table 4). (GR A)
  4. Non-invasive imaging of the liver and biliary trees should be considered mandatory to exclude diseases manifesting as obstructive jaundice. (GR A)
Table 3. Diagnosis of PBC endorsed by the Intractable Hepatobiliary Disease Study Group in Japan (2010 version)

With histological findings

  • 1) Biochemical evidence of cholestasis accompanied by histological evidence of CNSDC
  • 2) Presence of AMA accompanied by histologically compatible features of PBC, even without CNSDC
  • Without histological findings

  • 3) Presence of AMA accompanied by clinical features and course of classical and cholestatic PBC
Table 4. Differential diagnosis of PBC
  • 1) Cholestatic liver diseases
    • Intrahepatic cholestasis: chronic drug-induced cholestasis

    • Primary sclerosing cholangitis

    • IgG4-related sclerosing cholangitis

    • Adult-onset bile duct paucity

    • Obstructive jaundice

  • 2) Diseases with immunological abnormalities
    • Autoimmune hepatitis

    • Drug-induced liver injury

  • 3) Elevation of serum ALP and/or GGT
    • Space occupying lesions of the liver

    • Bone lesions

    • Hyperthyroidism

    • Fatty liver diseases

2.2 Diagnosis of atypical PBC

Pathognomonically-related but atypical PBC cases that do not fulfill the diagnostic criteria should be handled distinctively and appropriately; treatment strategies for these cases are different from those for typical PBC.

Early PBC

AMA may be detectable in the serum of individuals without symptoms of PBC and with normal liver tests. Histopathological changes of PBC with no or mild progression are apparent and this condition is designated early PBC. Follow-up without medication is appropriate.

Autoimmune cholangitis, autoimmune cholangiopathy

Patients whose clinical features are compatible with PBC may be negative for AMA but have a high titer of antinuclear antibody (ANA) in their serum. In 1987, Brunner and Klinge first described this condition as immunocholangitis, while others have used different terminology, such as autoimmune cholangiopathy, primary autoimmune cholangitis, or autoimmune cholangitis. The current understanding is that this condition is atypical PBC.

AMA-negative PBC

Approximately 10% of patients who have biochemical evidence of cholestasis, accompanied by histological features of PBC, are negative for AMA. Autoreactive T cells in these patients react with mitochondrial antigen, despite being negative for AMA. Special consideration for their treatment is not warranted.

PBC–AIH overlap syndrome

Patients with PBC who manifest clinicopathological features of autoimmune hepatitis (AIH) in conjunction with elevated levels of aminotransferases could be recognized. These cases have also been referred to as PBC with features of AIH. Prednisolone may effectively reduce aminotransferase levels in such cases.

2.3 Diagnosis of PBC: symptoms, signs, and complications

Approximately 70–80% of PBC cases are diagnosed in the early and asymptomatic phase. Although this phase is likely to persist for years, the clinical and histological progression precipitates several symptoms (symptomatic PBC). The symptoms and complications of PBC include cholestasis, liver injury, and comorbid autoimmune disease(s) (Table 5).

Table 5. Symptoms and complications of PBC


  • 1) None of the following
  • 2) General fatigue
  • 3) Cholestasis-associated
    • Pruritus (scratching)

    • Jaundice

  • 4) Liver injury and cirrhosis-associated
    • Hematemesis and melena

    • Abdominal fullness

    • Consciousness disturbance

  • 5) Comorbid autoimmune diseases -associated
    • Sicca syndrome, etc.


  • 1) Cholestasis-associated
    • Osteoporosis

    • Hyperlipidemia

  • 2) Liver injury and cirrhosis-associated
    • Portal hypertension

    • Hepatocellular carcinoma

    • Ascites

    • Hepatic encephalopathy

  • 3) Comorbid autoimmune diseases -associated
    • Sjögren's syndrome

    • Rheumatoid arthritis

    • Hashimoto thyroiditis, etc.

Pruritus accompanied by cholestasis is characteristic of PBC. It may occur initially before overt jaundice. Prolonged cholestasis results in jaundice, xanthoma coupled with lipid abnormalities, and osteoporosis-related bone lesions/fractures.

Persistent fatigue is another common symptom, occurring in 20–70% of Caucasian patients, although less frequently in Japanese patients. No correlation has been found between fatigue and age, sex, jaundice, liver function parameters, or histological stage of the disease. PBC patients can experience profound distress associated with fatigue.

Cirrhosis-associated symptoms include esophagogastric varices. Portal hypertension is more likely to occur in PBC than in liver diseases with other etiologies, and can develop even in the non-cirrhotic stage of PBC; some patients are diagnosed by the presence of esophagogastric variceal bleeding as an initial symptom.

Prevalent comorbid autoimmune diseases include Sjögren's syndrome, Hashimoto thyroiditis, and rheumatoid arthritis. aPBC may be masked by the symptoms of comorbid autoimmune diseases. Appropriate diagnosis of comorbid autoimmune diseases is important because they may influence the outcome of PBC.

2.4 Diagnosis of PBC: pathological examinations

PBC is histologically characterized by autoimmune-mediated progressive destruction of the intrahepatic small bile ducts, as well as by chronic intrahepatic cholestasis, hepatocellular damage, fibrosis, and septal formation.

Bile duct damage

PBC is histologically characterized by CNSDC and progressive bile duct loss, which preferably affects the intrahepatic small bile ducts, especially the interlobular bile ducts. Non-caseating epithelioid granuloma formation is often seen in the portal tracts. Granulomatous cholangitis consisting of CNSDC and periductal granuloma formation is valuable for pathological diagnosis. CNSDC is characterized by marked lymphoplasmacytic accumulation around the damaged bile ducts, and lymphoid cell infiltration is found in the biliary epithelial layer of CNSDC. Some biliary epithelial cells in CNSDC show eosinophilic apoptotic changes and swelling. Moreover, chronic cholangitis, which does not fulfill the criteria of CNSDC, is also found. Bile duct loss is seen during the progression of PBC, and the interlobular bile ducts are mostly lost in the terminal cirrhotic stage. The presence of arteries in the absence of bile ducts is useful for identification of bile duct loss or ductopenia.

Parenchymal change

In the early stage of PBC, non-specific necroinflammatory changes are found in the parenchyma. Interface hepatitis and chronic cholestatic changes are also found. During the progression of irreversible bile duct damage and loss, there are several characteristic findings that reflect cholestasis, including ductular reaction (proliferating bile ductules), copper deposition (orcein-positive granules), bile plaques, hepatocellular ballooning (cholate stasis), Mallory–Denk bodies, and feathery degeneration. These features are associated with the progression of biliary fibrosis and biliary cirrhosis. Changes similar to small cell dysplasia are also often found in zone 1 (periportal area), which is useful for the diagnosis of PBC. In addition to these cholestatic changes reflecting bile duct loss, chronic hepatitic changes resembling autoimmune hepatitis, such as interface and lobular hepatitis, are also found in most PBC cases, and are involved in the progression of hepatic fibrosis and cirrhosis.

Histological staging

The characteristic histological findings of PBC are heterogeneously distributed throughout the liver. Thus, in small specimens such as those taken from needle liver biopsy, sampling errors are likely to be recognized when using the classification systems of Scheuer and Ludwig, because these two systems define each stage by a sole histological feature (Supporting information Memo 2). Therefore the novel staging system of Nakanuma (2009) (Tables 6-8) is recommended for histological staging of PBC, as this system could avoid the sampling errors caused by the heterogeneous distribution of histological features.

Table 6. Histological staging of PBC
  • Stage 1 no progression
  • Stage 2 mild progression
  • Stage 3 moderate progression
  • Stage 4 advanced progression
Table 7. Histological staging system of PBC (Nakanuma et al.)
I. Histological findings and scoring for the determination of stage of PBC
A. Scoring of fibrosis 
Score 0No portal fibrosis, or fibrosis limited to portal tracts
Score 1Portal fibrosis with periportal fibrosis or incomplete septal fibrosis
Score 2Bridging fibrosis with variable lobular disarray
Score 3Liver cirrhosis with regenerative nodules and extensive fibrosis
B. Scoring of bile duct loss 
Score 0No bile duct loss
Score 1Bile duct loss in < 1/3 of portal tracts
Score 2Bile duct loss in 1/3 to 2/3 of portal tracts
Score 3Bile duct loss in > 2/3 of portal tracts
C. Scoring of deposition of orcein-positive granules 
Score 0No deposition of granules
Score 1Deposition of granules in a few periportal hepatocytes in < 1/3 of portal tracts
Score 2Deposition of granules in several periportal hepatocytes in 1/3 to 2/3 of portal tracts
Score 3Deposition of granules in many hepatocytes in > 2/3 of portal tracts
II. Staging by sum total of two (A and B) or three (A, B, and C) criteria
StageSum of scores
Two criteriaaThree criteriab
  1. aTwo criteria: fibrosis and bile duct loss.
  2. bThree criteria: fibrosis, bile duct loss and deposition of orcein-positive granules.
Stage 1 (no progression)00
Stage 2 (mild progression)1–21–3
Stage 3 (moderate progression)3–44–6
Stage 4 (advanced progression)5–67–9
Table 8. Grading of chronic cholangitis and hepatitis activity in PBC
I. Cholangitis activity (CA) 
CA0 (no activity)No cholangitis, but mild duct damage may be present
CA1 (mild activity)One bile duct with evident chronic cholangitis
CA2 (moderate activity)≥2 bile ducts with evident chronic cholangitis
CA3 (marked activity)≥1 bile duct with CNSDC
 (CNSDC: chronic nonsuppurative destructive cholangitis.)
II. Hepatitis activity (HA) 
HA0 (no activity)No interface hepatitis, and no or minimum lobular hepatitis
HA1 (mild activity)Interface hepatitis affecting > 10 continuous hepatocytes in 1 portal tract or fibrous septa, and mild to moderate lobular hepatitis
HA2 (moderate activity)Interface hepatitis affecting > 10 continuous hepatocytes in ≥2 portal tracts or fibrous septa, and mild to moderate lobular hepatitis
HA3 (marked activity)Interface hepatitis affecting > 20 continuous hepatocytes in ≥ 1/2 of portal tracts, and moderate lobular hepatitis or bridging or zonal necrosis


  1. The novel system for histological grading and staging of PBC proposed by Nakanuma et al. is recommended (LE6, GRC1).

2.4 Diagnosis of PBC: clinical staging and disease severity

Clinical staging

PBC is classified into two groups depending on the absence or presence of symptoms caused by liver damage: aPBC and sPBC. aPBC is considered the non-advanced stage (stage I), while sPBC is considered the advanced stage. sPBC is further classified as s1PBC, with serum bilirubin level <2.0 mg/dL, and s2PBC, with serum level ≥2.0 mg/dL (Table 9). s1PBC is considered a non-icteric advanced stage (stage II), and s2PBC is considered an icteric advanced stage (stage III).

Table 9. Clinical staging of PBC
  1. aSymptoms caused by liver damage: jaundice, pruritis, esophageal varices, ascites, encephalopathy, and hepatocellular carcinoma.
  • 1) Asymptomatic PBC (aPBC): Condition absent from symptoms caused by liver damagea
  • 2) Symptomatic PBC (sPBC): Condition with symptoms caused by liver damagea.
    • 2-1) s1PBC: serum bilirubin level below 2.0 mg/dL
    • 2-1) s2PBC: serum bilirubin level equal or over 2.0 mg/dL

Disease severity

PBC progresses insidiously on a chronic course without acute exacerbation, and a good hepatic reserve is maintained for a long period. Therefore, severity is evaluated at the advanced stage (sPBC) and the modified Child–Pugh grading system with a modified total bilirubin level is applied (Table 10).

Table 10. Modified Child-Pugh score for evaluating the severity of PBC
  1. Grade A: 5–6 points; Grade B: 7–9 points; Grade C: 10–15 points.
T.Bilirubin (mg/dL)1–44–10>10
Albumin (g/dL)3.5<2.8–3.5<2.8
PT (%)70%<40–70%<40%
EncephalopathyNoneGrade1–2Grade 3–4

2.5 Diagnosis of PBC: clinical type and prognosis

Clinical type

The progression of PBC varies among individuals, and more than 70% of those with aPBC do not progress over 10 years. PBC is largely classified into three clinical types (Fig. 1) . Many patients progress gradually and remain in the asymptomatic stage for longer than a decade (gradual progressive type). However, some patients progress to portal hypertension presenting without jaundice (portal hypertension type), and others progress rapidly to jaundice and ultimately hepatic failure (jaundice/hepatic failure type). The jaundice/hepatic failure type tends to affect relatively younger patients compared to the other two types. Patients with the jaundice/hepatic failure-type PBC are often positive for anti-gp210 antibody, while those with the portal hypertension-type PBC have anti-centromere antibodies (Supporting information Memo 3).

Figure 1.

Three types of natural course of PBC.


Several models for predicting the prognosis of PBC have been proposed. In the updated Mayo Clinic Natural History Model for PBC, the key factors are age, serum total bilirubin, albumin, prothrombin time (PT), edema/ascites, and use of diuretics. This model is used worldwide to predict the prognosis of PBC patients. The updated version is better than the original one for prediction of shorter prognosis (Supporting information Memo 4).

In the logistic model developed by the Japanese Liver Transplantation Study Group (Ref.VII-1) (Supporting information Memo 5), serum total bilirubin and aspartate aminotransferase (AST)/alanine aminotransferase (ALT) ratio are necessary. The probability of death after 6 months is calculated by means of a logistic regression formula, and transplantation is recommended if the value exceeds 50%.

Finally, for the MELD (Model for End-Stage Liver Disease) score, the serum creatinine level, total bilirubin, and prothrombin time (PT) are the key factors. The MELD score is used for the evaluation of end-stage liver failure. The score is high if hepatorenal syndrome is present, and the pre-transplantation value correlates well with the likelihood and magnitude of complication after liver transplantation. Therefore, it is recommended that transplantation should be performed before complication by hepatorenal syndrome (Supporting information Memo 6).

The common factor among these different schemes is serum total bilirubin. A life expectancy of 10 years is predicted for patients with a serum bilirubin level <2.0 mg/dL, 5 years for 2.0–3.0 mg/dL, and 1 year for >6.0 mg/dL.


  1. Total bilirubin, prothrombin (INR), albumin, and the serum creatinine level, which are essential to calculate the MELD score, should be measured when considering liver transplantation. (LE 2b (2a in part), GR A)
  2. Patients with PBC should be referred to transplant hepatologists when serum total bilirubin level is >5 mg/dL. To encourage the patients to prepare for liver transplantation, an earlier and appropriate explanation of liver transplantation is desirable. (LE 4, GR B)

3. Treatment and Management of PBC

3.1 Treatment and management of PBC: general principles

Although there is no completely curative treatment for PBC, ursodeoxycholic acid (UDCA) is currently considered the first-line treatment for the disease. UDCA delays the progression of PBC, although it does not have a significant benefit for PBC at the advanced stage.

The clinical usefulness of UDCA is evaluated according to the following factors: (i) improvement of serum biochemical markers, such as ALP, GGT, AST, ALT and total bilirubin; (ii) histological improvement of cholangitis, liver inflammation and liver fibrosis; and (iii) delay in the disease progression until end-stage liver disease, death, or liver transplantation. The following Paris and Barcelona criteria are useful for evaluating the clinical outcome of UDCA treatment. (i) Paris criteria: total bilirubin ≤1.0 mg/dL, ALP ≤3 × the upper normal limit (UNL), and AST ≤ 2 × UNL at 1 year after introduction of UDCA. (ii) Barcelona criteria: decrease of ALP ≥40% at 1 year after introduction of UDCA.

Liver transplantation is the only therapeutic approach for patients in the advanced stage when medical treatment shows little improvement. Prevention and treatment strategies for comorbid autoimmune diseases, cholestasis, and cirrhosis-related symptoms and complications are required.

Although the term cirrhosis is included in the name PBC, most patients (70–80%) with PBC have little clinical and histological evidence of liver cirrhosis. Patients should be informed accordingly to prevent misunderstanding of their prognoses. Currently, patients are likely to be diagnosed at earlier stages and disease progression is likely to be delayed by UDCA. Therefore, the prognosis of patients with aPBC, as long as they remain asymptomatic, is equivalent to that in the general population. No restrictions are necessary in daily life for patients with aPBC. By contrast, some restrictions in daily life and nutritional education are required for patients with sPBC, depending on symptoms, expected future complications, and disease severity.

3.2 Prescription

Ursodeoxycholic acid (UDCA)

Extensive clinical trials including randomized clinical trials (RCT) and meta-analyses were carried out for UDCA after the first report by Poupon et al. After lively debates, it was concluded that UDCA not only improves the serum biochemical values of PBC patients but also prolongs the period to death or liver transplantation.

The clinical guidelines for PBC by the European Association for the Study of the Liver (EASL) and the American Association for the Study of Liver Diseases (AASD) recommend that UDCA be given at a dose of 13–15 mg/kg/day, whereas in Japan, it is usually given at 600 mg/day. In clinical trials performed with Japanese PBC patients, 600 mg/day UDCA was given to PBC patients for 48–132 weeks and then the results of liver tests were analyzed. Improvement was demonstrated in 81.8% (27/33) of cases. Therefore, 600 mg/day is considered as a standard dose, irrespective of body weight. The dose can be increased up to 900 mg/day or decreased depending on weight and adverse events. Co-administration with bezafibrate is then considered if 900 mg/day UDCA has little effect. UDCA results in biochemical improvement, but is not likely to act against the “core” pathogenesis of PBC; administration is usually maintained throughout life.


  1. UDCA should be used to improve liver biochemical tests and histological findings, and to prolong the time until death or liver transplantation, though it does not provide significant benefit for those at the advanced stage. (LE 1a, GR A)
  2. In general, UDCA should be administered at 600 mg/day, and increased to 900 mg/day if the response is suboptimal. (LE 2a, GR B)
  3. UDCA is usually given TID, but the effects have been shown to be similar even if it is given as a single daily dose or BID. (LE 2a, GR B)
  4. The following definitions are proposed by the Intractable Hepatobiliary Disease Study Group of Japan for evaluation of the effects of UDCA after starting therapy. Good response: serum ALP, ALT and IgM become normal within 2 years; Fair response: serum ALP, ALT and IgM become <1.5 × UNL at 2 years; Poor response: serum ALP, ALT and IgM remain >1.5 × UNL at 2 years. (LE 6, GR C1)
  5. UCDA is the only drug shown to have long-term efficacy. (LE 2a, 2b, C, GR C1)


Bezafibrate, a peroxisome proliferator-activated receptor α (PPAR α) agonist, has been reported to show biochemical improvements and effectiveness in patients with PBC, mainly by Japanese researchers. However, the long-term effects of bezafibrate have not yet been evaluated, and the use of the drug for PBC is not recommended in the clinical guidelines by EASL and AASLD.

When possible, bezafibrate should be administered in combination with UDCA, because the drugs have different pharmacological mechanisms of action and demonstrate additive effects. Bezafibrate is given at 400 mg/day in patients who exhibit a suboptimal response to UDCA. However, in Japan, prescription of bezafibrate is only approved for patients with hypertriglyceridemia; PBC patients are still subject to off-label use.

Some reports indicate that fenofibrate, the other PPARαagonist, is also effective against PBC. Both bezafibrate and fenofibrate are known to increase the risk of rhabdomyolysis, and elevation of ALT is occasionally observed as an adverse effect of fenofibrate.


  1. Administration of bezafibrate (Bezatol®, 400 mg/day) may be considered in patients who exhibit a suboptimal response to UDCA. (LE 2a, GR B)

Prednisolone (PSL)

PSL has been considered to be contraindicated for PBC, because it brings about little improvement of PBC and may even cause deterioration of osteoporosis in postmenopausal women. Co-administration of PSL with UDCA is indicated for patients with PBC–AIH overlap syndrome, especially those whose symptoms of hepatitis are clinically and histologically relevant. The recommended initial corticosteroid dose is <0.5 mg/kg/day. It is advised to switch to UDCA monotherapy after hepatitis subsides.

3.3 Treatment by clinical stage


UDCA improves liver tests as well as histological findings, and as a consequence, prolongs the time until death or liver transplantation. However, the therapeutic effects of UDCA are negligible in patients with advanced PBC and marked jaundice. Liver transplantation is indicated for these patients of advanced stage.


There is no evidence to assist in the decision as to whether PBC with mildly elevated ALP should be treated, and thus no consensus has been reached. Although it has been proposed that all PBC patients should be treated as soon as the diagnosis is established, some physicians have argued that patients with mild elevation of ALP could be followed up without UDCA treatment until disease progression is apparent, in consideration of the costs and adverse effects of UDCA.


  1. UDCA should be initiated immediately when the serum ALP level is increased to up 1.5 × UNL. In patients with ALP <1.5 × UNL, liver enzymes should be measured every 3–4 months and UDCA treatment should be started when an increase in serum ALP to 1.5 × UNL is detected. (LE 6, GR C1)
  2. As elevation of AST and/or ALT suggests hepatitis features of PBC and likely disease progression, UDCA should be administered in these cases. (LE 6, GR C1)

Early PBC

Early PBC is defined as PBC without any elevated liver tests. Patients in this category require no treatment and are followed up every 1–2 years. Development of overt PBC may be preventable in these patients if etiology-oriented medical treatment becomes available in the future.

Autoimmune cholangitis

When the response to UDCA is not optimal, PSL administration should be considered. It is advisable to switch to UDCA monotherapy after hepatitis subsides, as in cases of PBC–AIH overlap syndrome.

AMA-negative PBC

The diagnosis of PBC should be confirmed by liver histology. The treatment strategy is identical to that for AMA-positive PBC.

PBC–AIH overlap syndrome

PSL is recommended in addition to UDCA for cases that are considered to be PBC–AIH overlap syndrome, because superimposed AIH could deteriorate the clinical course of PBC toward cirrhosis.


  1. When patients with PBC are diagnosed with PBC–AIH overlap syndrome due to clinical and histological features of AIH, and meet the criteria for corticosteroid use for PBC–AIH overlap syndrome (Table 11), PSL administration is strongly recommended. (LE 2b, GR B)
  2. It is advised that treatment should be switched to UDCA monotherapy when hepatitis features subside. (LE 3, GR C1)
Table 11. Diagnostic criteria for corticosteroid use in PBC–AIH overlap syndrome (Intractable Hepatobiliary Disease Study Group in Japan, 2011)
PSL is recommended in addition to UDCA for cases that are considered to be PBC–AIH overlap syndrome and meet the two following criteria simultaneously:
(1) diagnosed with PBC using the criteria of the Intractable Hepatobiliary Disease Study Group in Japan (2010)
(2) diagnosed as probable/definite AIH using International Autoimmune Hepatitis Group (IAIHG) simplified criteria (2008). (Supporting information Memo 7)
As for liver histology, HA scores in the PBC grading/staging systems in Table 8 should be used as follows: 0 for HA score 0 or 1, 1 point for HA score 2, and 2 points for HA score 3.

3.4 Liver transplantation

General principles

Liver transplantation is considered in cases with continuous elevation of total bilirubin, intractable pleural effusion and/or ascites, hepatic encephalopathy, repeated rupture of esophageal and/or gastric varices, and markedly reduced quality of life (QOL) due to severe pruritus. On the other hand, liver transplantation is generally contraindicated for patients with severe complications, such as lung and kidney disease, other organ disease, infection, and malignancy.

It should be borne in mind, however, that not every patient for whom liver transplantation is indicated succeeds in finding a donated liver. Living donor liver transplantation (LDLT) is more common in Japan because deceased donor livers are scarcely offered for transplantation. In order to plan for LDLT, a 1-month period is desirable for the living donor. This period is required for medical examination, preparation for early rehabilitation and approval by the appropriate ethical committee. Earlier registration for deceased donor liver transplantation (DDLT) is recommended. Given this situation, there is no difference in timing between cases in which LDLT is indicated and those in which DDLT is indicated. Moreover, there is no difference in the outcome of PBC patients who undergo LDLT and DDLT.


  1. When PBC progresses to cholestatic cirrhosis, medical treatment has little effect on further disease progression and liver transplantation is the only therapeutic approach for survival. (LE 1, GR B) Appropriate timing of liver transplantation is the most important consideration. (LE 2b, GR B)
  2. The following criteria (Table 12) should be consulted to determine whether liver transplantation is indicated. (LE 6, GR A)
Table 12. Criterion for the indication of liver transplantation
  • 1) Both of the following items (I) and (II) should be met.
    1. Sum of Child–Pugh score ≥8.

    2. Serum levels of total bilirubin ≥5.0 mg/dL, with at least one complication depicted below (a–g).

      • a) Hepatic coma
      • b) Gastrointestinal bleeding with portal hypertension
      • c) Refractory ascites and/or pleural effusion
      • d) Spontaneous bacterial peritonitis, hepatorenal syndrome, hepatopulmonary syndrome
      • e) Hepatocellular carcinoma
      • f) Itch sensation, causing insomnia
      • g) Severe general malaise, and deterioration of QOL, by severe osteomalacia

Indication for liver transplantation

As described in the Prognosis portion of section 2.5, three scoring systems have been widely implemented for predicting prognosis in PBC. The most popular system is the updated Natural History Model for PBC from the Mayo Clinic. Once the Mayo risk score is >7.8, the outcome after liver transplantation is poor. Furthermore, this score was a significant predictor for liver-related death before liver transplantation, but not for post-transplantation prognosis. Thus, liver transplantation should be performed before the Mayo risk score reaches 7.8.

Secondly, the indication model of the Japanese Liver Transplantation Indication Study Group recommends liver transplantation when the mortality rate after 6 months is >50%, as estimated by a logistic model. In this model, the severity of disease is estimated as a score of 1, 3, 6, 8 or 10 points. At present, patients with scores >6 points, which means the expected mortality rate after 6 months is >70%, are candidates for DDLT. In contrast, patients in whom the expected mortality rate after 6 months is >50% are candidates for LDLT.

Finally, the cumulative survival rate at 1 year after liver transplantation is about 50% in patients whose MELD score is >20. The higher the MELD score, the poorer is the outcome after liver transplantation. In a previous report from a single center in Japan, the outcome became poorer when the MELD score was >25. The average MELD score is −15 in patients who have undergone liver transplantation, and thus the timing of consultation may be adequate when the MELD score reaches 12.


  1. The use of scores for the evaluation of liver failure is mandatory.
    1. Updated Natural History Model for PBC from the Mayo Clinic: risk score >7.8 (LE 2b, GR B)
    2. Mortality rate after 6 months: ≥ 50%, as estimated by the Japan liver transplantation indication society model (LE 2b, GR, B)
    3. MELD score ≥15 (LE 2b, GR, B)

Management of patients after liver transplantation

Liver-transplanted patients should be administered immunosuppressive agents and closely monitored. Postoperative complications, acute/chronic rejection, recurrence of PBC, and infections should all be carefully monitored. Postoperative recurrence of PBC is an important cause of graft dysfunction. The five-year recurrence rate after liver transplantation is reported as 0–33% in representative facilities in Japan. The ten-year survival rate of patients with PBC after liver transplantation is equal to the survival in those with other diseases.

3.5 Management of symptoms and complications

Dermal pruritus

Pruritus is the most specific symptom in PBC, and may appear even before development of jaundice. Although it has been debated whether increased concentrations of bile salts, histamine, progesterone metabolites or endogenous opioids are potential pruritogens in cholestasis, recent experimental evidence has implicated the lysophospholipase, autotaxin (ATX), and its product, lysophosphatidic acid (LPA), as potential mediators of cholestatic pruritis. Pruritus is more often exacerbated at night more than in the daytime, and may decrease along with progression of liver damage.

Cholestyramine is a non-absorbable basic anion-exchange resin, and is a drug of first choice for pruritus in PBC. It improves pruritus by inducing adsorption of bile acids in the intestinal tract. In patients treated with UDCA, an interval of a few hours is necessary in order to avoid the attenuating effect caused by binding of cholestyramine and UDCA. Cholestimide, which is also a basic anion-exchange resin, is used empirically in Japan.

Antihistamines are also frequently prescribed in Japan due to their ease of use. They can be effective for insomnia due to their sedative action.

The efficacy of rifampicin, which is an anti-tuberculosis agent, for pruritus has been validated in two meta-analyses. As there is a possibility of various side effects, including liver damage, close and regular follow-up are necessary. A dose of 150–300 mg twice daily is used for pruritis.


  1. Cholestyramine is effective against dermal pruritus in PBC patients, and should be considered the first choice agent. (LE 2a, A)
  2. Antihistamines might be effective against severe dermal pruritus which may cause insomnia. (LE 5, GR C1)
  3. Rifampicin is effective against dermal pruritus in PBC patients. (LE 1a, GR B)


Osteoporosis is frequently observed in patients with PBC because intestinal absorption of fat-soluble vitamins is disturbed due to reduced secretion of bile acids, and PBC is common in middle-aged and postmenopausal women. For prevention of osteoporosis, abundant oral intake of calcium (1 to 1.2 g/day) and vitamin D (plentiful in fish and mushrooms) and weight-bearing exercise are recommended, and medical treatment should be given if necessary. Bisphosphonates, bioactive vitamin D3 agents, and vitamin K2 are prescribed.

Among bisphosphonates, alendronate improves bone density more than etidronate. Nevertheless, there is no evidence that alendronate suppresses bone fracture. Administration once weekly is preferable to daily administration. Alendronate is contraindicated for cases with esophageal stenosis due to sclerotherapy for esophageal varices.

Vitamin D3 and vitamin K2 formulations have frequently been prescribed for PBC in Japan. Both drugs have been proven to be effective for osteoporosis itself, and are regarded as Grade B in guidelines for the prevention and treatment of osteoporosis.


  1. It is desirable to start treatment for the prevention of fractures in cases with a T score below −1.5. (LE 4, GR C1)
  2. Alendronate improves bone density in PBC patients. (LE 1b, GR A)
  3. Although there is scarce evidence in PBC patients, vitamin D3 and vitamin K2 formulations can be effective for osteoporosis. (LE 1b, GR C1)


Hypercholesterolemia is likely to develop in PBC due to cholestasis. Xanthoma is seen around the eyelids. No specific treatment for hypercholesterolemia in PBC is required in most cases, while bezafibrate is expected to be effective for both PBC and hypercholesterolemia.

Sicca syndrome

Sicca syndrome, a major symptom of Sjögren's syndrome, is frequently complicated with PBC. The diagnosis of Sjögren's syndrome should be made by detection of serum anti-SS-A/SS-B antibodies, presence of corneal erosion, and lip biopsy if necessary. Artificial lachrymal fluids are indicated for eye symptoms. If the response is not favorable, pilocarpine hydrochloride and cevimeline hydrochloride hydrate are used under the guidance of ophthalmologists. As for oral symptoms, artificial saliva should be used first, and pilocarpine hydrochloride and cevimeline hydrochloride hydrate can also be prescribed.


  1. Cevimeline hydrochloride and pilocarpine hydrochloride may be effective for xerostomia in PBC, although there are no studies evaluating their potential to alleviate the symptoms occurring in PBC patients with concurrent Sjögren's syndrome. (LE 6, GR B)
  2. Patients with PBC frequently experience cholestasis, comorbid autoimmune diseases, and symptoms associated with liver injury and cirrhosis. Prevention and management of these symptoms are required. (GR A)

3.6 Follow-up

Prognosis of asymptomatic PBC is excellent with little progression, but 25% of patients with aPBC develop some symptoms within 10 years. Serum total bilirubin and cholestatic enzymes (ALP, GGT) are important for assessing the activity and progression of PBC. Liver biochemical tests should be done every 3–6 months. In addition, thyroid hormone (every year) and bone mineral density (every 2–4 years) tests are recommended because PBC is likely to be complicated with other autoimmune diseases, such as Sjögren's syndrome, chronic thyroiditis, and rheumatoid arthritis.

Regular upper gastrointestinal endoscopy, depending on stage (1 or 2 times per year), is required because esophageal/gastric varices may develop even in patients without jaundice. Abdominal ultrasound (US) and serum AFP testing every 6–12 months are necessary in patients with definite or suspected liver cirrhosis. Liver cirrhosis, older age, and male sex are high risk factors for developing hepatocellular carcinoma (HCC). Therefore, testing for tumor markers and imaging studies [US and computed tomography (CT)] are required for early detection of HCC in patients with advanced PBC. Management for other complicating autoimmune diseases should be done depending on each symptom.

Finally, special attention should be paid to pregnancy in PBC and patients who have a desire to bear children. The chance for pregnancy could be the same in the early stage of aPBC as in the normal population; there is no evidence to recommend avoidance of pregnancy in patients with aPBC. In sPBC, however, if worsening of icterus or varices is reported, then avoidance of pregnancy could be justified.

The impact of pregnancy on PBC is unclear because both exacerbation and improvement of cholestasis have been reported. Estrogen could potentially worsen cholestasis; pruritus may become severe in pregnancy and could be prolonged even after delivery. Conversely, it should be noted that cholestasis could be symptomatic after pregnancy. After a patient has become pregnant, monitoring for varices is necessary as in other cirrhotic patients, especially after the second trimester, due to increase in circulating blood volume. The use of β blockers is considered to be safe. It is also advisable to shorten the second trimester of pregnancy, if possible.


  1. The blood and other clinical tests should be undertaken regularly to investigate complicating comorbidities, prevent complications, and detect portal hypertension and liver cancer as early as possible. (Table 13) (LE 3, GR B)
  2. It is advisable to consult with hepatologists when the diagnosis of PBC is made, or when patients with PBC become symptomatic. In patients with atypical forms of PBC such as PBC–AIH overlapping syndrome, earlier referral is recommended. (Table 14) (LE 6, GR A-B)
  3. For patients in the symptomatic stage, there is a likelihood of worsening of pruritus or icterus in the pregnancy, as well as an increased possibility of variceal rupture. It is advisable that these patients undergo upper gastrointestinal endoscopy by the second trimester of pregnancy. (LE 5, GR C1)
  4. Administration of UDCA or bezafibrate should be withheld, if the patient with PBC is possibly pregnant or in the early stage of pregnancy. In the third trimester of pregnancy, administration of UDCA is possible for cholestasis if necessary. (LE 5, GR C1)
Table 13. Clinical tests for follow-up of patients with PBC
  1. aIntervals of tests vary depending on the stage of the patient.
1) Assessment of activity and progression of PBC
       image Liver tests (albumin, total bilirubin, AST, ALT, ALP, GGT, PT) every 3–6 monthsa 
2) Assessment of complications
       image Thyroid function (TSH, free T4)Every 1 yeara
       image Bone mineral density testingEvery 2–4 yearsa
       image Upper GI endoscopyEvery 1–2 yearsa
       image Abdominal ultrasound and serum AFPEvery 1 yeara
 (every 3–6 months in liver cirrhosis)a
Table 14. Recommendation for consultation to specialists
  • 1) Decision of initial treatment approach, in particular, diagnosis and evaluation of atypical cases
  • 2) Decision of treatment strategy
  • 3) Suboptimal response of UDCA
  • 4) Apparent progression to symptomatic PBC
  • 5) ≥5 mg/dL of total bilirubin (consult to liver transplant surgeons; earlier explanation is necessary for patients)


This study was supported by Grants-in-Aid from the Research Program of lntractable Disease provided by the Ministry of Health, Labor and Welfare of Japan.

Conflicts of Interest

Shotaro Sakisaka is given research funds from MSD K.K., Mikio Zeniya is given research funds from Daiichi Sankyo Co. Ltd. and Chugai Pharmaceutical Co., Ltd., Hirohito Tsubouchi is given research funds from Chugai Pharmaceutical Co., Ltd., MSD K.K. and KAN Research Institute, Inc. All other authors have no conflicts of interest to declare.