Primary biliary cirrhosis # †

This guideline has been approved by the AASLD and represents the position of the association. These recommendations provide a data-supported approach to the management of primary biliary cirrhosis (PBC). They are based on the following: (1) formal review and analysis of the recently published world literature on the topic (Medline search); (2) American College of Physicians Manual for Assessing Health Practices and Designing Practice Guidelines1; (3) guideline policies, including the AASLD Policy on the Development and Use of Practice Guidelines and the American Gastroenterological Association Policy Statement on Guidelines2; and (4) the experience of the authors in the specified topic. Intended for use by physicians, these recommendations suggest preferred approaches to the diagnostic, therapeutic, and preventive aspects of care. They are intended to be flexible, in contrast to standards of care, which are inflexible policies to be followed in every case. Specific recommendations are based on relevant published information. To more fully characterize the quality of evidence supporting recommendations, the Practice Guideline Committee of the AASLD requires a Class (reflecting benefit versus risk) and Level (assessing strength or certainty) of Evidence to be assigned and reported with each recommendation (Table 1, adapted from the American College of Cardiology and the American Heart Association Practice Guidelines3). AASLD, American Association for the Study of Liver Diseases; AIH, autoimmune hepatitis; ALT, alanine aminotransferase; AST, aspartate aminotransferase; IAIH-G, International Autoimmune Hepatitis Group score; Ig, immunoglobulin; PBC, primary biliary cirrhosis; PDC-E2, pyruvate dehydrogenase complex; UDCA, ursodeoxycholic acid. PBC is often considered a model autoimmune disease because of its hallmark serologic signature, the antimitochondrial antibody (AMA) and specific bile duct pathology.4, 5 The etiology of PBC is thought to be due to a combination of genetic predisposition and environmental triggers.6 Although the genetic predisposition is clear, major histocompatibility complex associations are varied.7 Several large epidemiologic studies have been performed and have suggested an association with urinary tract infections, reproductive hormone replacement, nail polish, past cigarette smoking, and toxic waste sites, as well as xenobiotics in an animal model of PBC.8-10 One critical and unique feature of PBC is the high degree of specificity for involvement of the small intrahepatic bile ducts. Staining of small bile ducts with monoclonal antibodies against mitochondrial autoantigens demonstrates an intense staining at the apical surface of biliary epithelial cells.11, 12 The characteristic serologic hallmark of PBC is the AMA, a highly disease-specific autoantibody found in 90%-95% of patients and less than 1% of normal controls.13 The targets of the disease-specific antimitochondrial response are all members of a family of enzymes, the 2-oxo-acid dehydrogenase complexes and include pyruvate dehydrogenase complex (PDC-E2), branched chain 2-oxo-acid dehydrogenase complex, and 2-oxo-glutaric acid dehydrogenase complex. These enzymes catalyze the oxidative decarboxylation of keto acid substrates and are located in the inner mitochondrial membrane.14, 15 Fewer than 5% of patients with PBC are AMA-negative in one study.16 Both immunofluorescence, and now more commonly enzyme-linked immunosorbent assays, are used to test for AMA. There is a 100-fold to 150-fold increase of autoreactive CD4 PDC-E2–specific T cells in liver and regional lymph node compared to blood in patients with PBC, and a 10-fold to 15-fold increase in autoreactive CD8 PDC-E2–specific T cell infiltrates in liver compared to blood. These data strongly suggest that the antimitochondrial response is either directly related to pathology or intimately associated with the etiological insult.17, 18 PBC is a chronic cholestatic disease with a progressive course which may extend over many decades. The rate of progression varies greatly among individual patients. Over the past decades, there have been many changes in the diagnosis and management of PBC. More patients are being recognized with earlier stage disease and many of these patients respond well to medical therapy. In both Europe and North America the number of liver transplants for PBC is falling.19, 20 AMA may be detectable in serum when patients are symptom-free and liver tests are normal. Based on one small study, it is believed that many of these patients may eventually develop abnormal liver tests and symptoms. The median follow-up time from the first positive AMA test to persistently abnormal liver tests in this series was 6 years with a range between 1 and 19 years. However, none of the patients developed cirrhosis during the follow-up.21 It is estimated that 0.5% of the general population is AMA-positive, which means that fewer than 10% of patients with AMA will develop PBC.22 The proportion of asymptomatic patients (which has been variably defined) who will subsequently develop PBC-related symptoms has been investigated in several series from the United Kingdom, North America, and Sweden.23-28 All of these studies provide evidence of progressive disease in a substantial proportion of patients, with between 36% and 89% becoming symptomatic during average follow-up periods ranging from 4.5-17.8 years. In the two most recent studies,27, 28 the median time from diagnosis to the appearance of symptoms was found to be 2 and 4.2 years. Patients with early disease in the absence of ursodeoxycholic acid (UDCA) therapy have a shortened survival comparable to a healthy population regardless of symptoms.27, 28 The 10-year survival of asymptomatic patients in three contemporary series ranged from 50%-70%; whereas the median duration of survival for symptomatic patients ranged from 5-8 years from the onset of symptoms.27, 28 In an older study of 279 patients from the United States,24 the median survival of symptomatic patients was 7.5 years, much shorter than the median survival of 16 years for asymptomatic patients. This marked difference in survival has not been found in the study from Northeast England, a finding possibly explained by an excess of deaths unrelated to liver disease in asymptomatic patients who were on average a decade older.29 Histologic stages have been found to predict survival.30, 31 The rate of histologic progression has been assessed in three large groups of patients in the absence of a therapeutically effective agent.30, 32, 33 The median time to develop extensive fibrosis was 2 years. After 4 years, the probability of remaining in the early stage was 29% (confidence interval: 15%-52%), while cirrhosis was diagnosed in 50% of patients who initially had only interface hepatitis without fibrosis. Only a minority (20%) of patients who were precirrhotic showed histologic stability. Overall, the histologic stage progressed by one stage every 1.5 years. The development of liver failure (ascites, bleeding, hepatic encephalopathy, or hyperbilirubinemia [>6 mg/dL]) during a follow-up of 5 years has been estimated to be 15% in the large community-based study of 770 patients in Northeast England27 and 25% of the 236 patients enrolled in the European azathioprine trial.30 The rate of development of esophageal varices and its impact on survival were evaluated in a prospective study of 256 patients (28% of whom had cirrhosis) who were observed for a median time of 5.6 years.34 A total of 31% of patients developed esophageal varices. After the development of varices, the 3-year survival was 59%, whereas after a first bleeding episode, it was 46%. UDCA is currently the only drug approved for the treatment of patients with PBC. Several randomized trials, combined analyses, and long-term observational studies have shown that this agent not only improves biochemical indices but also delays histologic progression and improves survival without transplantation.32, 35-46 Accordingly, most patients are now treated with UDCA. In an early study, the rate of histologic progression to cirrhosis was significantly less in the UDCA group than in the control group (13% versus 49%).35 In a trial involving 192 patients, UDCA therapy significantly delayed histologic stage progression after a median follow-up of 3.4 years.39 In the French trial of UDCA, the risk of progression from stages I-II to stages III-IV was 7% ± 2% with UDCA and 34% ± 9% with placebo.32 Predictive factors for cirrhosis developing included serum bilirubin higher than 1 mg/dL, and moderate to severe lymphocytic piecemeal necrosis on the liver biopsy.47 The effect of UDCA therapy on the development of esophageal varices was addressed in a prospective study of 180 patients who received UDCA versus placebo and were observed for up to 4 years.48 A total of 139 patients had no varices and 41 had varices at baseline. After 4 years, the risk of developing varices was 16% for the UDCA-treated patients and 58% for those receiving the placebo. However, UDCA did not reduce the low rate of bleeding. To overcome the lack of power of clinical trials in assessing the long-term effectiveness of therapy, a Markov model has been used to study the effect of UDCA on the natural history of PBC.46 The study included 262 patients who had received 13-15 mg/kg UDCA daily for a mean of 8 years (range 1-22 years), and their survival was substantially better than that predicted by the model. The overall survival rates without liver transplantation were 84% and 66% at 10 years and 20 years, respectively. The survival rate was better than the spontaneous survival rate as predicted by the updated Mayo model (relative risk: 0.5, P < 0.01). In early-stage patients, 6% were predicted to progress to liver transplantation or death after 10 years and 22% by 20 years. The survival rate of these patients was similar to that in the control population. In contrast, the probability of death or liver transplantation was significantly increased in patients treated in late stages of the disease (relative risk: 2.2, P < 0.05). Several clinical, biochemical, and histologic features have prognostic significance in PBC although bilirubin level is the best predictor of survival and is the most important component in all mathematical models of prognosis in PBC.49, 50 Some of these models have been useful in predicting survival in UDCA-treated patients as well (http://www.mayoclinic.org/gi-rst/mayomodel1.html). The diagnosis of PBC should be suspected in the setting of chronic cholestasis after exclusion of other causes of liver disease. The diagnosis is suspected based on cholestatic serum liver tests and largely confirmed with tests for AMA. A liver biopsy can be used to further substantiate the diagnosis if needed. Most patients with PBC have abnormal liver tests including elevations of alkaline phosphatase, mild elevations of aminotransferases (alanine aminotransferase [ALT] or aspartate aminotransferase [AST]) activity, and increased levels of immunoglobulins (mainly immunoglobulin M [IgM]). Some patients with PBC may have high ALT or AST activities associated with hyperglobulinemia (increase in IgG). The changes in biochemical tests are related in part to the stage of the disease and to severity of histologic lesions.30, 51, 52 In patients without cirrhosis, the degree of elevation in alkaline phosphatase is strongly related to the severity of ductopenia and inflammation; the increase in aminotransferase activity and IgG levels reflects mainly the degree of periportal and lobular necrosis and inflammation; hyperbilirubinemia reflects the severity of ductopenia and biliary piecemeal necrosis. A rise in serum bilirubin, gamma globulins, and hyaluronic acid together with a fall in serum albumin and platelet count are the early indicators of the development of cirrhosis and portal hypertension.51, 52 As in other cholestatic diseases, serum cholesterol levels are often elevated.53 Individual serum bile acid levels can be elevated but are not routinely determined. AMA is found in nearly 95% of patients with PBC.5 Antinuclear antibody and anti–smooth muscle antibody are found in nearly half of patients with PBC.5 In approximately 5%-10% of the patients, AMA antibodies are absent or present only in low titer (≤ 1/80), when immunofluorescent techniques are used. The presence or absence of antibody, rather than the magnitude of antibody level, is most important. In some patients, antinuclear antibodies, particularly anti-GP210 and/or anti-SP100 are present and may correlate with prognosis54; in some other AMA-negative patients, antibodies against the major M2 components (PDC-E2, 2-oxo-glutaric acid dehydrogenase complex) are present using enzyme-linked immunosorbent assay or western blotting techniques. PBC is characterized by chronic, nonsuppurative cholangitis that mainly affects interlobular and septal bile ducts. When focal lesions show intense inflammatory changes and necrosis around bile ducts, the term "florid duct lesion" is often used. The inflammatory infiltrate consists essentially of lymphocytes and mononuclear cells in close contact with the basal membrane of cholangiocytes undergoing necrosis. The infiltrate consists of plasma cells, macrophages, polymorphonuclear cells (especially eosinophils), and in some cases epithelioid granulomas which are present more often in the early stage of disease.5 There are few (if any) arterial lesions. In contrast, portal venules are often compressed and occluded by the inflammatory reaction. Terminal hepatic venules are often retained in their central location with progression of fibrosis and sometimes even in cirrhosis. Bile duct paucity or ductopenia is usually defined in less than 50% of portal tracts containing ducts. The size of the liver biopsy specimen is important. The probability of observing cholangitis and bile duct destruction increases with the number of portal tracts because of the typical patchy distribution of the lesions. At least 10-15 portal tracts should be present, and multiple sections should be reviewed to adequately appreciate or rule out cholangitis and ductopenia. These would include periportal/periseptal copper deposition, periportal/periseptal feathery degeneration with or without Mallory-Denk bodies, and cholestatic rosettes. Actual bile stasis is not appreciated until decompensated liver disease has occurred. Histologic lesions are classically divided into four stages. Stage I is characterized by portal inflammation with or without florid bile duct lesions. In this stage, inflammation remains confined to the portal triads. Disease progression is characterized by the gradual increase of periportal lesions extending into the hepatic parenchyma referred to as interface hepatitis (stage II). Periportal regions become focally irregular, and the lesion is characterized by cellular necrosis or apoptosis, separation of hepatocytes by inflammatory cells, and macrophages. There are two main types of interface hepatitis. The first is lymphocytic piecemeal necrosis, the association of hepatocellular necrosis or apoptosis with lymphohistiocytic cells. This is similar to the lesion found in autoimmune hepatitis (AIH). Second is biliary piecemeal necrosis, which is marked by a striking ductular reaction, sometimes referred to as ductular proliferation, and accompanied by edema, neutrophil infiltration, periductular fibrosis, and necrotic hepatocytes, the latter associated with cholestasis. The French have shown that severity of interface hepatitis is highly predictive of development of extensive fibrosis.47, 55 Stage III is characterized by a distortion of the hepatic architecture with numerous fibrous septa. Cirrhosis with the existence of regenerative nodules defines stage IV. Nodular regenerative hyperplasia is a known complication of PBC and should be differentiated from cirrhosis. With the high disease specificity of a positive AMA test, the role of liver biopsy to diagnose PBC is questionable with alkaline phosphatase activity ≥ 1.5 times normal and AST values < 5 times normal.56 Liver biopsy may be recommended in AMA-negative patients and to exclude other concomitant diseases such as AIH and nonalcoholic steatohepatitis.46, 47, 55 Expert noninvasive imaging of the liver and biliary tree is mandatory in all patients with biochemical evidence of cholestasis. If the diagnosis is uncertain, then cholangiography may be necessary preferentially with noninvasive magnetic resonance imaging or endoscopically to exclude primary sclerosing cholangitis or other biliary tract diseases. Transient elastography (Fibroscan; Echosens, Paris, France) is a new noninvasive tool to evaluate the degree of liver fibrosis, which has been studied in patients with PBC,57 but it is not yet approved by the U.S. Food and Drug Administration. The diagnosis of PBC is generally based on the following criteria: (1) biochemical evidence of cholestasis with elevation of alkaline phosphatase activity; (2) presence of AMA; and (3) histopathologic evidence of nonsuppurative cholangitis and destruction of small or medium-sized bile ducts if a biopsy is performed. The differential diagnosis includes a cholestatic drug reaction, biliary obstruction, sarcoidosis, AIH and primary sclerosing cholangitis. The diagnosis of PBC can be established when two of the following three criteria are met: Biochemical evidence of cholestasis based mainly on alkaline phosphatase elevation. Presence of AMA. Histologic evidence of nonsuppurative destructive cholangitis and destruction of interlobular bile ducts (Class I, Level B). Fatigue is the most common symptom in PBC and has been found in up to 78% of patients.56-62 Fatigue is nonspecific and occurs in many conditions other than PBC. Fatigue does not correlate with the severity, histologic stage, or duration of PBC. Severe fatigue may impair the quality of life in patients with PBC61, 62 and may be associated with decreased overall survival.63 Its etiology is unknown. Recently, an autonomic neuropathy has been described in association with fatigue in patients with PBC.64 Fatigue does not improve with treatment of depression, is often constant over time,65 is frequently associated with and may be a of which occurs in of patients with is a more specific symptom of PBC than fatigue and in of patients with PBC. It is now less common because patients with PBC are often asymptomatic at It can be or usually at while in and is often by contact with other or occurs in PBC, its severity may over However, it is to without treatment until a cirrhosis and liver The of in PBC is It is that the of including that to PBC, is at least in part by increased while other studies a role for components of and/or is and are associated symptoms. The is usually normal. and are and are found in the setting of portal is a late finding in patients with liver disease. to other liver diseases, portal most often late in the course of PBC, when patients have cirrhosis. However, in contrast to other liver diseases, portal may also develop in patients with precirrhotic PBC. These patients may from esophageal varices, varices, or portal normal or normal liver Nodular regenerative hyperplasia is associated with of the portal venules and may to portal in some of these Patients can for many years after without liver and hepatic may develop in patients with histologic PBC and cirrhosis. is the most often in and occurs in up to of The risk for in PBC compared to an and healthy population is It is usually not associated with specific and by The disease that was and often by multiple is now The of in PBC is Patients with PBC to have in which is and is low or is normal in patients with PBC for those with and may be elevated in The of in PBC is from that in other of cholesterol are elevated and such as may cholesterol levels were and in two studies of patients with PBC and levels ranged from in individual cholesterol is elevated compared to and patients with PBC are not at increased risk of death from Although patients with PBC may have decreased bile acid in increased risk of important of the and are is and serum levels of and are usually normal in most patients, including those with The occurs in patients who are liver transplantation who may also have and levels may be in and decreased activity, Patients with AMA-negative PBC to those who lack AMA but clinical liver and natural history are nearly to patients with typical PBC. all of these patients have antinuclear and/or anti–smooth muscle in and between the and AMA-negative is on the apical of biliary epithelial cells from with AMA-negative as well as PBC, that their is The diagnosis of AMA-negative PBC requires a liver biopsy that demonstrates the typical features of bile duct destruction in PBC. The diagnosis is more if granulomas are A recent large study has shown AMA-negative cases of PBC have less and more autoimmune diseases and levels are in AMA-negative than patients with A recent has published of patients treated for AMA-negative PBC, which only 52 patients. The authors no difference in biochemical response to UDCA was observed when patients with and AMA-negative PBC were There is no formal of the between PBC and features of PBC usually to AIH in patients who have a diagnosis of PBC and not to patients with AIH who have AMA. reported to are of size to with degree of a diagnosis of PBC with AIH is from PBC. observational data that response to therapy with UDCA for is no from that observed in patients with PBC A may also to patients with PBC followed by AIH as described recently in a less AIH followed by PBC has been There are two that have been used to evaluate patients with PBC for evidence of Both of these are are by without the of long-term follow-up The first is the International Autoimmune Hepatitis Group the of which was in two diagnosed with This was subsequently and it is this that has been used in several recent studies to this was for AIH and positive for AIH are when there is an absence of factors unrelated to a diagnosis of PBC, hepatitis and In for AMA and/or features of biliary disease would be assigned by A has been used to the presence of a by for the presence of two of the three following (1) ALT activity 5 times of (2) IgG ≥ 2 times of normal and/or positive anti–smooth muscle and (3) liver biopsy with moderate or severe periportal or There have been several with PBC who have been a diagnosis of PBC and have then been evaluated for of using one of these two However, it is if the biochemical, and data were at the time as the liver factors such as drug or concomitant diseases may and possibly all of these studies have reported in patients with patients with who were followed for a mean of years were compared with patients with This study a in of of portal or for liver in patients with PBC and a or However, an estimated 50% of patients in either group had received treatment with UDCA and some in both groups had received a of other UDCA with or without therapy has been but no in therapy for these patients There are no data which best to patients thought to have There are of patients with PBC who respond to UDCA therapy yet subsequently present with clinical features of These patients may no have AMA and liver more typical of AIH which to therapy. Patients with PBC may have florid duct lesions and all have evidence of bile duct usually with cholestatic A review of cases of PBC followed for the term that have features of PBC and AIH and develop an AIH on their These authors to cases of AIH who then developed PBC. More patients from more than patients with PBC were described who developed AIH after years of There are few data on the of detectable serum AMA in patients who have typical features of These data may be from histologic review of patients with AIH, in whom small bile duct pathology was on a of In this none of the patients who positive for AMA had bile duct changes on of liver There are of patients with AIH who but on long-term these patients not develop there is a for better long-term analysis the natural history of both PBC and The effect of therapy on the and its components will to be and only then will the clinical significance of these features become UDCA in a of 13-15 is the only therapy for PBC approved by the U.S. Food and Drug Administration. The drug is and generally in two divided A number of studies have shown the benefit of UDCA in this Individual studies have evidence of liver Some studies with follow-up have also shown from data to increase which has of the of Some have these these include studies of duration and those that have used is now known to be an of UDCA is used and has the to a in for liver transplantation for this The drug is used for patients with stage of PBC as as their liver are A liver biopsy in not for the diagnosis for PBC in many