Show Summary Details
Page of

Primary biliary cirrhosis 

Primary biliary cirrhosis
Primary biliary cirrhosis

M.F. Bassendine

and D.E.J. Jones


Treatment—discussion of response criteria or UDCA therapy, with need for targeting of second line treatments at non-responders.

A relevant case history from Oxford Case Histories in Gastroenterology and Hepatology has been added to this chapter.

Updated on 27 February 2014. The previous version of this content can be found here.
Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2016. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 22 June 2018

Primary biliary cirrhosis Case History—A 60 yr old woman presenting with abnormal liver blood tests.


Primary biliary cirrhosis is a chronic, cholestatic liver disease in which the biliary epithelial cells lining the small intrahepatic bile ducts are the target for immune-mediated damage leading to progressive ductopenia. The cause is unknown, but presumed to be autoimmune.

The disorder affects women (>90% of cases) and usually has an insidious onset in middle age. Fatigue and pruritus are the most common presenting symptoms. Findings on examination vary widely, ranging from no abnormality to jaundice with hyperpigmentation, scratch marks, and signs of long-standing cholestasis.

Diagnosis of primary biliary cirrhosis is based on three criteria: (1) cholestatic liver function tests, with increases in serum alkaline phosphatase and γ‎-glutamyl transferase, but only modest changes in transaminases; (2) presence of serum antimitochondrial antibodies (AMA), which are found in more than 95% of cases; and (3) compatible liver histology. Many asymptomatic patients are recognized following the incidental discovery of AMA or elevated levels of serum alkaline phosphatase.

Treatment with ursodeoxycholic acid can lead to significant improvement in liver biochemical values and is recommended for some patients. Cholestyramine is used to treat pruritus. No immunosuppressive drug regimen has been proven to be effective. Progression may be slow, but eventually patients can develop cirrhosis, and death may occur from liver failure or complications of cirrhosis such as bleeding oesophageal varices. The disease at one time was a leading indication for liver transplantation, but earlier diagnosis and treatment has now improved the prognosis for many patients.

Historical perspective

In 1851 Addison and Gull described six patients with jaundice and xanthomas, the dominant presenting features of disease for over a century. In 1965 the clear association between antimitochondrial antibodies (AMA) and primary biliary cirrhosis (PBC) was recognized and in 1988 the major mitochondrial autoantigen was identified as an enzyme component of pyruvate dehydrogenase complex. Routine use of laboratory tests for AMA and liver function have changed the clinical presentation of the disease and over half of patients are now asymptomatic at diagnosis. The term ‘primary biliary cirrhosis’ was first used in 1950 but is a misnomer as cirrhosis is not often present. The first randomized controlled trial of therapy was initiated in 1968 and many treatments have since been evaluated but none have been shown to cure the disease; only liver transplantation is curative.



In common with most autoimmune disorders, genetic factors play a role in determining susceptibility to PBC, but the pattern of inheritance is complex. Familial clustering is well documented, and the sibling relative risk is 10.5, similar to values seen in other autoimmune disorders where it is thought that genetic factors may contribute up to 50% of the total risk. The concordance rate of PBC in identical twins is among the highest reported in autoimmunity at 63%. There is no association of the disorder with major histocompatibility complex (MHC) class I antigens. An immunogenetic predisposition has been confirmed by a genomewide association study identifying HLA class II genes (DQB1, DPB1, DRB1, and DRA), IL12A, and IL12RB2 as susceptibility loci. Genetic factors may also impact on the severity of PBC.


Over 95% of patients have antibodies to mitochondria, with the dominant autoantibody response being directed against two components (dihydrolipoamide acetyltransferase (E2) and E3-binding protein) of pyruvate dehydrogenase complex (PDC) (Table The loss of tolerance to these autoantigens is an early event in this progressive disease, with AMA being detectable in serum before abnormalities in liver function and long before the onset of symptoms. One hypothesis is that the development of these AMA marks the exposure of a genetically susceptible individual to an initiating environmental factor. Autoreactive T cells play a central role in the development of various autoimmune diseases and an immunodominant T-cell epitope within PDC-E2 (peptide 163–176) has been identified in patients with PBC. This epitope is within the lipoyl domain of PDC-E2 and in the same region where AMA bind. T-cell clones reactive to this peptide can also be activated by mimicry peptides derived from several xenobiotics and microbial proteins, supporting the hypothesis that autoreactive T-cells present in the peripheral blood and liver can be activated and clonally expanded by antigenic stimulation by mimicry peptides derived from environmental nonself antigen.

Table Reactivity of disease-specific autoantibodies in PBC

Molecular mass (×103)

Occurrence (%)

Mitochondrial antigens

Pyruvate dehydrogenase complex (PDC): lipoyl domain of E2 acetyltransferase



PDC: lipoyl domain of E3-binding protein



PDC E1α‎ decarboxylase



PDC E1β‎ decarboxylase



2-oxoglutarate dehydrogenase complex: lipoyl domain of E2 succinyl transferase



Branched chain 2-oxo-acid dehydrogenase complex: lipoyl domain of E2 acyltransferase



Nuclear antigens

Glycoprotein of the nuclear-pore membrane



Nucleoporin p62






Antinuclear antibodies are found in a minority of patients with PBC (Table and display unique immunofluorescence patterns such as nuclear dots or a nuclear ring-like pattern. Disease-specific nuclear antigens include a 210-kDa glycoprotein of the nuclear-pore membrane (gp 210), nucleoporin p62, and Sp100, an interferon-inducible 100-kDa nucleoprotein.

Despite progress in characterizing the reactivity of the disease-specific autoantibodies and autoreactive T-cell responses in PBC, the elucidation of early events in the pathogenesis of bile duct damage remains elusive. Progress has been hampered by the practical limitations in accessing human liver tissue and the lack of a suitable animal model. There is a controversial report that PBC is associated with a retroviral infection; however, the balance of evidence in PBC remains strongly in favour of an autoimmune process in which autoreactive effector mechanisms are directed at epitopes within self-PDC-E2 expressed normally or aberrantly by biliary epithelial cells.


The characteristic early lesion of PBC is inflammatory duct destruction. Later there is fibrosis, often patchy, and eventually a frank cirrhotic picture. Histologically this disease appears to evolve from a florid duct lesion to cirrhosis. This has led to a morphological classification into four stages. It must be recognized, however, that overlaps between stages is common in different parts of the liver. In stage 1, the duct lesion is florid (Fig. with the epithelium irregular, hyperplastic, or ulcerated. There is a heavy infiltrate of lymphocytes, plasma cells, and neutrophils, with occasional eosinophils. Aggregates of histocytes with granulomas ranging from foci of epithelioid cells to rounded lesions with multinucleated giant cells are present. In stage 2 there is established duct destruction and the bile ducts may be replaced by lymphoid aggregates with fibrosis. In stage 3 there is relatively little inflammation, though lymphoid aggregates may be present and fibrous septa extend from the portal tract. In stage 4 there is an established cirrhosis, paucity of bile ducts, and lymphoid infiltration (Fig. Mallory bodies similar to those seen in alcoholic liver disease may be present adjacent to the areas of inflammation and there is excess stainable copper-binding protein, a reflection of the cholestasis.

Fig. Bile duct lesion in PBC. There is granulomatous destruction of a medium-sized bile duct radicle in which the epithelium appears hyperplastic. Epithelioid macrophages are surrounded by a chronic inflammatory cell infiltrate. Haematoxylin and eosin.

Bile duct lesion in PBC. There is granulomatous destruction of a medium-sized bile duct radicle in which the epithelium appears hyperplastic. Epithelioid macrophages are surrounded by a chronic inflammatory cell infiltrate. Haematoxylin and eosin.

(Courtesy of A D Burt.)

Fig. Stage 4 PBC: an established micronodular cirrhosis; the halo effect seen around the nodules is a characteristic feature of biliary cirrhosis. Haematoxylin and eosin.

Stage 4 PBC: an established micronodular cirrhosis; the halo effect seen around the nodules is a characteristic feature of biliary cirrhosis. Haematoxylin and eosin.

(Courtesy of A D Burt.)


There is a marked geographical variation in the prevalence of the disease; it is commonest in northern Europe but rare in the Indian subcontinent and Africa. It was previously considered to be rare and account for fewer than 5% of patients dying of cirrhosis in Western communities, but appears to be becoming commoner. In the north-east of England the prevalence rose from 202 per million adults and 541 per million women over 40 in 1987 to 335 per million adults and 940 per million women over 40 in 1994. In Finland a similar increase in prevalence has been documented, from 161 per million women in 1988 to 292 per million in 1999. It remains unclear whether this represents better diagnosis or a true increase in prevalence. In Victoria, Australia the prevalence in British-born immigrant women is 344 cases per million, compared to 160 per million in Australian-born women, suggesting that environmental factors may play a role in the aetiology of PBC. A significant role for environmental factors in the triggering of PBC is also suggested by the demonstration, using formal cluster analysis, of disease ‘hot spots’ in the north-east of England.

Clinical features

Primary biliary cirrhosisA 60 yr old woman presenting with abnormal liver blood tests.

Patients with early disease may be asymptomatic or complain of fatigue or symptoms of coexisting autoimmune disease. Profound fatigue affects 50% of patients and can be a significant cause of disability. It is not related to the severity of underlying liver disease or itch. Pruritus can be local or diffuse, is worse at night, and usually precedes the onset of jaundice by months to years. Those with more advanced disease have evidence of cholestasis, with jaundice, light stools, easy bruising, and weight loss; itch may become the most distressing symptom. Rarely patients present with ascites, gastrointestinal bleeding from oesophageal varices, or associated peptic ulcer.

Findings on examination vary widely. At one extreme, there may be no abnormality, whereas at the other the patient is jaundiced, with hyperpigmentation, scratch marks, and signs of long-standing cholestasis. The planus form of xanthoma occurs characteristically as xanthelasmas around the eyes and in the palmar creases.

The liver is often enlarged and firm, and splenomegaly may be present, with or without portal hypertension. Spider naevi and palmar erythema are less frequent than in patients with alcoholic cirrhosis. Fluid retention with ascites and oedema is usually a late complication, as is bleeding from oesophageal varices. Steatorrhoea occurs primarily in patients who have advanced cholestasis, leading to malabsorption of fat-soluble vitamins, especially vitamin D. Deficiency of vitamin K sometimes results in easy bruising or other haemorrhagic phenomena. Bone pain due to osteomalacia can occur, as can liver failure with encephalopathy. However, such late manifestations of disease are now rarely seen in Western countries as liver transplantation is performed in most patients before their development. Osteoporosis in PBC is related to advancing age and disease severity; there is a twofold increase in both the absolute and relative fracture risk in people with PBC compared with the general population.

Hepatocellular carcinoma (HCC) is a recognized complication of cirrhosis from any cause. Men are afflicted at least twice as often as women, and PBC is no exception to this rule. HCC is a relatively common cause of death in male PBC patients with cirrhosis and surveillance with regular liver ultrasound is recommended.

PBC is associated with past smoking and a number of other autoimmune diseases. These include Sjögren’s syndrome, seropositive and seronegative arthropathy, thyroiditis, scleroderma, and renal tubular acidosis. The CRST syndrome (calcinosis, Raynaud’s phenomenon, sclerodactyly, and telangiectasia), pulmonary fibrosis, psoriasis, and coeliac disease have also been reported.

Differential diagnosis

The main differential diagnosis is from other causes of cholestasis. Good ultrasound examination of the liver and biliary tree is mandatory to exclude extrahepatic biliary obstruction or gallstones. CT, magnetic resonance cholangiopancreatography (MRCP), or endoscopic retrograde cholangiopancreatography (ERCP) may be necessary for patients without detectable antimitochondrial antibody, many of whom have a positive antinuclear antibody and may be thought to have ‘autoimmune cholangitis’. There is an overlap with autoimmune hepatitis, which can be diagnosed on liver histology, whilst primary sclerosing cholangitis will be evident on MRCP or ERCP.

Clinical investigation

The serological hallmark of PBC is a positive AMA, which may antedate all other abnormalities. The presence of AMA is not sufficient by itself to allow the diagnosis of PBC but indicates a substantial risk of onset of PBC over the next decade. Liver function tests reflect cholestasis, with increases in serum alkaline phosphatase and γ‎-glutamyl transferase, but only modest changes in transaminases. At presentation total serum bilirubin is usually normal or only modestly increased. The serum globulins are usually raised, particularly the IgM, but the serum albumin is usually maintained until late in the disease. Other tests such as erythrocyte sedimentation rate and autoantibodies other than antimitochondrial antibodies are less specific. Serum lipid levels may be strikingly elevated in PBC but it is not clear if this is associated with atherosclerotic risk. The increased cholesterol levels found in PBC are primarily due to LP-X, an abnormal low-density lipoprotein (LDL) particle with antiatherogenic properties.

The use of liver biopsy has diminished in recent years in the setting of ‘typical’ PBC, i.e. a patient with symptoms of fatigue or itch, cholestatic LFTs, and a strongly positive AMA. However the histological features are very specific, and although several different histological stages may be found in one biopsy, the presence of fibrosis or cirrhosis indicates a worse prognosis. Liver stiffness measurement (transient elastography) is a promising new noninvasive test for assessment of liver fibrosis and monitoring its progression over time. In patients with more advanced disease, surveillance for osteopenia and hepatocellular cancer should be undertaken.

Criteria for diagnosis

As many patients are asymptomatic at presentation, diagnosis is currently based on three criteria: cholestatic liver function tests, presence of serum AMA, and liver histology that is compatible with PBC. A definite diagnosis requires the presence of all three criteria and a probable diagnosis requires two of these three.


This consists of therapy aimed at modifying the disease process and progression to cirrhosis, and treatment of symptoms and late complications.

Numerous trials of specific therapy have been undertaken in the last 40 years; the main agents that have been assessed are shown in Table Over recent years the naturally occurring bile acid, ursodeoxycholic acid (UDCA), which is safe and well tolerated, has become an established treatment of PBC. The mechanism of action of UDCA is uncertain and is probably multifactorial. Many randomized controlled trials comparing ursodeoxycholic acid with placebo have been published. All the trials reported an improvement in biochemical liver indices. However, a recent updated systematic review to evaluate the benefits and harms of UDCA in patients with PBC did not demonstrate any benefit of UDCA on mortality or liver transplantation in patients with PBC. Another meta-analysis that was confined to trials using an appropriate dose of ursodeoxycholic acid (>10 mg/kg body weight per day) and with sufficient follow-up (at least 2 years) found that treatment with UDCA resulted in significant improvement in liver biochemical values. In addition subjects without evidence of fibrosis (histological stages I and II) who were treated with UDCA had slower disease progression than did subjects in the control group. UDCA therapy does not seem to benefit the symptom of fatigue and has a variable effect on pruritus. The practice guidelines of the American Association for the Study of Liver Diseases state that appropriately selected patients with PBC and abnormal liver biochemical values should be advised to take UDCA at a dose of 13 to 15 mg/kg daily either in divided doses or as a single daily dose. If cholestyramine is used to treat pruritus, 4 h should elapse between the administration of cholestyramine and that of UDCA.

Table Therapeutic agents evaluated in PBC





13–15 mg/kg per day

Improvement in biochemistry. May delay progression to cirrhosis and normalize survival rate when given in early stages of disease. Approved by FDA for use in PBC. Well tolerated

UDCA + budesonide

6 mg/day

Improvement in liver histology and biochemistry. May worsen osteopenia

UDCA + fibrate

Not determined

Small studies show improvement in biochemistry compared to UDCA monotherapy


2.5–4mg/kg per day

Limited efficacy; renal toxicity and hypertension


15 mg/week

Some benefit but toxicity possible. Use not recommended outside randomized trials


30 mg/day reducing to 10 mg/day

Improved hepatic function in one small study


1–2 mg/kg per day

Limited efficacy

Mycophenolate mofetil

1 g/dy

Pilot study showed limited efficacy


0.5–4 mg/day

Potentially toxic


0.6–1.2 mg/day

Minor benefits but insufficient evidence to support use


250–100 mg/day

No convincing benefit. Excessive toxicity

UDCA, ursodeoxycholic acid.

Primary biliary cirrhosisConflicting evidence regarding the benefit of UDCA therapy in PBC has been explained by the identification of subgroups of patients showing complete and incomplete/absent biochemical response. Response criteria have been defined by several groups (eg “Barcelona” criteria: responders defined by reduction in alkaline phosphatase by >40% or normalization, both after 1 year of UDCA at 13–15mg/kg. “Toronto” criteria: responders defined by reduction in alkaline phosphatase to less than or equal to 1.67 times the upper normal limit by 2 years of treatment). Patients failing to meet response criteria, who are typically younger at disease presentation, are at substantially increased risk of death from PBC or need for transplantation. Second line therapies currently under evaluation are targeted at UDCA non-responders. Non-responding patients should be considered for trials and undergo enhanced follow-up. There is a continued need for new therapeutic options in patients with a suboptimal biochemical response to UDCA, and treatments for disease-related symptoms that do not typically respond to UDCA (e.g., fatigue and pruritus) are also needed.

Immunosuppressive agents

As PBC appears to be a classic autoimmune disease, a number of immunosuppressive agents have been assessed. There is currently insufficient evidence to support or reject the use of glucocorticosteroids. Budesonide combined with UDCA improves liver histology and the results of biochemical tests of liver function, but may worsen osteopenia in patients with more advanced disease; monitoring of bone mass density is recommended. Other immunosuppressive drugs that have been evaluated but found to be either ineffective or toxic include azathiaprine, chlorambucil, cyclosporin, methotrexate, and mycophenolate mofetil.

Other agents

d-Penicillamine was used for patients with PBC because of its ability to decrease hepatic copper and modulate the immune response, but did not reduce the risk of mortality or morbidity, and led to more adverse events. Colchicine has been evaluated because of its immunomodulatory and antifibrotic potential, but there is insufficient evidence to support its use outside clinical trials. Statins effectively reduce cholesterol levels but do not improve cholestasis in PBC. Fibrates have been used in nonicteric patients who have not shown a complete response to UDCA; combination of bezafibrate or fenofibrate with UDCA appears promising in preliminary studies, with significant improvement in serum alkaline phosphatase and IgM, compared with UDCA monotherapy.

Treatment of symptoms

Fatigue in PBC may be associated with the presence of daytime somnolence, and modafinil has been used with good effect in a pilot study. Itching can be an intolerable symptom and the first line of treatment is with cholestyramine. Improvement in itching has also been reported with rifampicin and opioid antagonists (naloxone and naltrexone). A prolonged prothrombin time is treated with intramuscular vitamin K, 10 mg monthly. Injections of vitamin A (100 000 IU) and vitamin D (100 000 IU) are usually given every 2 to 3 months in jaundiced patients and vitamin E supplements may also be required. Osteomalacia is now rare, given such treatment. The principles developed for monitoring and treating postmenopausal osteoporosis can be followed for patients with PBC. The complications of portal hypertension and of liver failure are treated appropriately.

Liver transplantation

This has greatly improved survival in patients with PBC, and it is the only effective treatment for those with liver failure. The requirement for liver transplantation for endstage PBC is falling in both Europe and North America. Referral to a transplant centre should be considered as the bilirubin approaches 100 μ‎mol/litre, though patients with particular problems such as intractable itching may need to be considered individually. Current data suggest that the survival benefit from transplantation occurs when the MELD score exceeds 15. Survival rates at 5 and 10 years are in excess of 80% and 50% respectively. Recurrence of PBC may occur but graft loss from PBC recurrence is rare. The main risk factor for recurrence is the use of tacrolimus compared with ciclosporin for immunosuppression.


People with PBC have a threefold mortality increase when compared with the general population, which is somewhat reduced by regular treatment with UDCA. The clinical and biochemical response to UDCA can be used to predict outcome. Patients with a decrease in serum alkaline phosphatase of at least 40% or a decrease to the normal range at 1 year have a prognosis similar to that of an age-matched healthy population. The most reliable determinant of prognosis is the level of the serum bilirubin; other factors associated with poor prognosis include weight loss, hepatomegaly, splenomegaly, histological stage, patient age, and impaired liver synthetic function. Several similar prognostic models have been validated in clinical studies; the most widely used is the Mayo Risk Score (an online tool for calculating this is available at

Future developments

Progress in understanding the aetiopathogenesis of PBC has been hampered by the absence of a suitable model, but three spontaneous autoimmune biliary disease mouse models have recently been reported. These will help in the study of genetics and immunoregulation in early PBC and may suggest new therapeutic approaches to reverse the disease process. The role of combination regimens of UDCA with other agents should become clearer in the next 5 to 10 years after assessment in appropriately designed trials.

Further reading

Corpechot C, et al. (2005). The effect of ursodeoxycholic acid therapy on the natural course of primary biliary cirrhosis. Gastroenterology, 128, 297–303. [Study using multistate modelling approach to assess the effect of UDCA on the natural history of PBC, showing that UDCA alone normalizes the survival rate of patients when given at early stages.]Find this resource:

    EASL and collaborators (2009). EASL Clinical Practice Guidelines: management of cholestatic liver diseases. J Hepatol, 51, 237–67. [Guidelines developed under the auspices and approved by the European Association for the Study of the Liver.]Find this resource:

      Gershwin ME, et al. (2005). USA PBC Epidemiology Group. Risk factors and comorbidities in primary biliary cirrhosis: a controlled interview-based study of 1032 patients. Hepatology, 42, 1194–202. [Multicentre study evaluating factors associated with increased risk of PBC.]Find this resource:

        Gong Y, Christensen E, Gluud C (2007). Azathioprine for primary biliary cirrhosis. Cochrane Database Syst Rev 3, CD006000. [Review; 63 references.]Find this resource:

          Gong Y, Christensen E, Gluud C (2007) Cyclosporin A for primary biliary cirrhosis. Cochrane Database Syst Rev 3, CD005526 [Review; 55 references.]Find this resource:

            Gong Y, Gluud C (2005). Colchicine for primary biliary cirrhosis: a Cochrane Hepato-Biliary Group systematic review of randomized clinical trials. Am J Gastroenterol, 100, 1876–85. [Systematic review of 10 trials involving 631 PBC patients, 4 of which were high-quality trials, concluding there is insufficient evidence to support the use of colchicine.]Find this resource:

              Gong Y, Klingenberg SL, Gluud C (2006). Systematic review and meta-analysis: d-penicillamine vs. placebo/no intervention in patients with primary biliary cirrhosis—Cochrane Hepato-Biliary Group. Aliment Pharmacol Therapeut, 24, 1535–44. [Systematic review with meta-analysis of seven trials including 706 PBC patients, showing d-penicillamine did not reduce the risk of mortality or morbidity and led to more adverse events.]Find this resource:

                Gong Y, et al. (2007). Ursodeoxycholic acid for patients with primary biliary cirrhosis: an updated systematic review and meta-analysis of randomized clinical trials using Bayesian approach as sensitivity analyses. Am J Gastroenterol, 102, 1799–807. [Meta-analysis of all placebo-controlled trials of UDCA in PBC.]Find this resource:

                  Hirschfield GM, et al. (2009). Primary biliary cirrhosis associated with HLA, IL12A, and IL12RB2 variants. N Eng J Med, 360, 2544–55. [A genomewide association study identifying susceptibility loci in primary biliary cirrhosis.]Find this resource:

                    Jones DE (2007). Pathogenesis of primary biliary cirrhosis. Gut, 56, 1615–24. [Review of evidence supporting autoimmune aetiology of PBC.]Find this resource:

                      Kaplan MM, Gershwin ME (2005). Medical progress: primary biliary cirrhosis. N Eng Med, 353, 1261–73. [Comprehensive review of the condition.]Find this resource:

                        Lindor K (2007). Ursodeoxycholic acid for the treatment of primary biliary cirrhosis. N Engl J Med, 357, 1524–9. [Review of major clinical studies using UDCA based around a ‘typical’ case vignette and explaining areas of uncertainty.]Find this resource:

                          Lindor KD, et al. (2009). Primary biliary cirrhosis. Hepatology, 50, 291–308.Find this resource:

                            McNally RJ, et al. (2009). Are transient environmental agents involved in the cause of primary biliary cirrhosis? Evidence from space-time clustering analysis. Hepatology, 50, 1169–74. [Analysis of >1000 PBC cases indicates that transient environmental agents may play a role in the aetiology of PBC.]Find this resource:

                              Prince MI, et al. (2004). Asymptomatic primary biliary cirrhosis: clinical features, prognosis, and symptom progression in a large population based cohort. Gut, 53, 865–70. [Follow-up study of 770 patients with PBC, 61% asymptomatic at diagnosis.]Find this resource:

                                Prince M, Christensen E, Gluud C (2005). Glucocorticosteroids for primary biliary cirrhosis. Cochrane Database Syst Rev 2, CD003778. [Review with 59 references.]Find this resource:

                                  Prince MI, James OFW (2003). The epidemiology of primary biliary cirrhosis. Clin Liver Dis, 7, 795–819. [Comprehensive review of epidemiology.]Find this resource:

                                    Rautiainen H, et al. (2005). Budesonide combined with UDCA to improve liver histology in primary biliary cirrhosis: a three-year randomized trial. Hepatology, 41, 747–52. [Trial in 77 patients with precirrhotic PBC showing budesonide combined with UDCA improved liver histology.]Find this resource:

                                      Rieger R, Gershwin ME (2007). The X and why of xenobiotics in primary biliary cirrhosis. J Autoimmunity, 28, 76–84. [Review of data supporting an environmental trigger for PBC.]Find this resource:

                                        Shi J, et al. (2006). Long-term effects of mid-dose ursodeoxycholic acid in primary biliary cirrhosis: a meta-analysis of randomized controlled trials. Am J Gastroenterol, 101, 1529–38. [Meta-analysis confined to placebo-controlled trials using an appropriate dose of UDCA and with sufficient follow-up.]Find this resource:

                                          Yeaman SY, et al. (1988). Primary biliary cirrhosis: Identification of two major M2 mitochondrial autoantigens. Lancet i, 1067–70. [Landmark paper characterizing reactivity of AMA.]Find this resource: