Chapter 51. Primary Biliary Cirrhosis

• Most patients with primary biliary cirrhosis (PBC) are asymptomatic at the time of diagnosis but eventually develop symptoms, including fatigue and pruritus.
• Consider PBC in any patient in whom liver enzymes are elevated in a cholestatic pattern; an isolated elevated alkaline phosphatase level always requires further evaluation.
• Antimitochondrial antibodies have 95% sensitivity for PBC and are central to the diagnosis.

Primary biliary cirrhosis (PBC) is a classical autoimmune disease. Although it is predominantly a disease of middle-aged woman, with a median age of diagnosis of approximately 50 years, 5–10% of patients are men, and the reported age range is 22–93 years. It is reported throughout the world, but with varying geographic incidence and prevalence; the highest reported incidence and prevalence is in northern Europe. Studies have suggested the incidence and prevalence are increasing worldwide. Antimitochondrial autoantibodies (AMAs) are found in 95% of patients with PBC. The targets of the autoantibodies are members of the family of the 2-oxo-acid dehydrogenase complexes, most particularly the E2 subunits of the pyruvate dehydrogenase complex (PDC-E2). The immunologic injury is marked by a T-cell–mediated destruction of the intrahepatic bile ducts. The only proven therapy for PBC is ursodeoxycholic acid (UDCA). Despite early diagnosis and treatment, most patients have an inexorable disease progression leading ultimately to cirrhosis and end-stage liver disease.

Figure 51–1

Genetic Factors

There have been several studies indicating the role of genetics in determining the susceptibility to PBC. Particularly important observations include the following:

1. First-degree relatives of patients with PBC have a 4–6% prevalence of disease.

2. PBC has the highest reported concordance rate (62.5%) in monozygotic twins of any autoimmune disease.

3. The prevalence of AMAs in first-degree relatives of patients with PBC was 13.1% compared with 1% of controls.

A number of candidate genes have been implicated in PBC including CTLA4 and IL12A.

Environmental Factors

As with other autoimmune conditions, molecular mimicry is felt to be the mechanism for the initiation of autoimmunity. Several candidate environmental triggers have been suggested, including infectious agents (viruses and bacteria) and chemicals. These potential triggers have significant homology to human mitochondrial proteins, particularly PDC-E2. The role of transient environmental agents in the pathogenesis of PBC was supported by a recent study that demonstrated highly statisitically significant space-time clustering in 1015 cases of PBC in northeast England.

Immunologic Factors

The T cells infiltrating the portal triads are specific for the PDC-E2 autoepitope. However, it is not clear why the immune attack has such specificity for the biliary epithelium when the target mitochondrial proteins are present in all nucleated cells. Studies have suggested that the mechanism of the metabolic processing of PDC-E2 in biliary epithelium results in the autoantigen remaining immunologically intact and available for presentation to antigen-presenting cells. In all other cell types the autoepitope is destroyed during apoptosis.

Symptoms and Signs

The common practice of checking serum liver enzyme tests as part of annual physical examinations has increased the percentage of patients with PBC who are asymptomatic at the time of diagnosis, raising it to more than 50%. Most patients will go on to develop symptoms over time. For those who are symptomatic, fatigue and pruritus are the most common complaints, each found in approximately 20% of patients. Both of these symptoms may be found even in early-stage disease. The fatigue can be severe enough to impair quality of life. The pruritus can be either local or diffuse, worse at night, exacerbated by heat, and profoundly distressing. Other symptoms seen less commonly are dry eyes and dry mouth (sicca symptoms), and unexplained right upper quadrant discomfort. Jaundice is a late finding and a poor prognostic sign as are other symptoms of hepatic decompensation, including fluid retention (ascites and peripheral edema), gastrointestinal bleeding, and encephalopathy.

The asymptomatic patient almost always lacks physical findings. Early-stage findings include skin hyperpigmentation, excoriations from itching, xanthelasmas, xanthomas, and hepatomegaly. Later stages are marked by findings of portal hypertension and hepatic decompensation: splenomegaly, caput medusa, ascites, spider nevi, palmar erythema, jaundice, proximal and temporal muscle wasting, and asterixis.

Laboratory Findings

The liver enzymes in patients with PBC are typically in a cholestatic pattern; that is, the alkaline phosphatase (ALP) is elevated out of proportion to the aminotransferases. An isolated elevated ALP level should first be evaluated to verify its biliary origin by obtaining values for γ-glutamyl transferase (GGT), 5′ nucleotidase, or fractionation of the ALP. The degree of ALP elevation does not correlate with disease severity; rare cases have been reported of patients with normal ALP diagnosed with early-stage PBC on the basis of positive AMAs and consistent histology.

The presence of autoantibodies, particularly antimitochondrial antibodies, is the hallmark of PBC. The AMA titer does not correlate with disease activity in patients with PBC. Also, the absence of AMAs does not rule out the diagnosis of PBC. There are well-described patients with "AMA-negative PBC" who have the typical clinical, biochemical, and histologic features of PBC. Other autoantibodies—most commonly antinuclear antibodies (ANAs)—are reported in PBC. Nuclear rim and nuclear-dot patterns of ANAs are highly specific for PBC. Serum immunoglobulin M levels are often elevated.

Histologic Findings

There are four histologic stages of PBC. Stage 1 is marked by the presence of the florid bile duct lesion (a bile duct at the center of a dense lymphocytic infiltrate). PBC is initially a ductocentric process; the inflammation does not extend beyond the portal tracts. Stage 2 is marked by loss of normal bile ducts, development of bile duct reduplication, and extension of the inflammation into the hepatic parenchyma. Stage 3 is marked by bridging fibrosis of the portal triads and progressive loss of bile ducts. Stage 4 is frank cirrhosis and end-stage liver disease. The presence of copper deposition is supportive of a cholestatic process.

The liver is not involved uniformly in PBC, a fact that increases the likelihood of sampling variability. All four stages of PBC may be seen in a single large biopsy. The convention is to report either the highest stage of disease or the range of stages. Although some have questioned the need for histologic staging in the diagnosis of PBC, it can both aid in determining prognosis and guiding treatment by providing baseline histologic findings against which treatment response can be assessed.

Primary Therapy

UDCA is the only therapy approved by the U.S. Food and Drug Administration for the treatment of PBC. The mechanism of action of UDCA remains unknown. It is a choleretic agent. It has been shown to slow progression of disease in patients with early-stage disease and is highly effective as monotherapy in 25–30% of patients. As is seemingly true with all therapies used for PBC, UDCA is not as effective in advanced disease. There is clearly a point in the natural history of PBC where sufficient numbers of bile ducts are destroyed and the condition has passed the point of reversibility. The appropriate dosage of UDCA is 13–15 mg/kg/day split into two doses. It is well tolerated with the most common side effect being diarrhea.

Although many hepatologists do not recommend consideration of any therapy beyond UDCA (Figure 51–2), there is a growing consensus that it is important for the clinician to differentiate between patients who have responded to UDCA and those who have not; studies have shown that patients with a biochemical and histological response to UDCA have a better prognosis than those who do not respond. These data support a more aggressive approach to management in these selected patients who are at a higher risk of disease progression and in young patients with early-stage disease with the ultimate goal of improving outcomes (Figure 51–3). If patients have an incomplete biochemical or histologic response to UDCA, the use of alternative therapies including colchicine, methotrexate, and budesonide should be considered.

In contrast to some studies that failed to show a benefit of colchicine, several prospective double-blind trials found that colchicine improved biochemistry profiles, symptoms (pruritus), and histologic findings. A meta-analysis reported that colchicine delayed the need for transplantation and reduced the incidence of major complications.

Methotrexate has been shown in some studies to improve biochemical and histologic findings in patients with an incomplete response to UDCA monotherapy. Other studies have failed to show any benefit of methotrexate, either alone or in combination with UDCA. One 10-year study suggested that patients with early-stage disease were most likely to benefit from methotrexate.

Short duration, randomized, placebo-controlled trials of budesonide have shown that it improves biochemical and histologic findings in PBC. Longer trials will be needed to assess the impact on survival and ensure osteopenia is not an issue. Budesonide is contraindicated in patients with cirrhosis.

Figure 51–3 highlights the stepwise approach for patients with stage 1–3 disease. This approach is based on the observation that clinical response to medical therapy varies greatly from patient to patient and that a significant percentage of patients will progress on UDCA monotherapy. In this approach, histology is a critical component in decision making, both in determining whether to add additional agents to UDCA and whether to continue those agents. The severity of the portal inflammation is typically the major deciding factor in determining whether to add additional therapies.

Liver transplantation is the only effective therapy for patients with end-stage PBC. These patients have post-transplant outcomes that exceed most other indications. Disease recurrence has been reported—30% at year 10 in one series—but it is exceedingly rare to lose a graft to recurrent disease.

Secondary Therapy

Pruritus

As with primary therapy, management of pruritus is most effective when one proceeds in a logical and stepwise fashion. Cholestyramine is effective in more than 90% of patients and should be the first-line agent. The starting dosage is typically 4 g, two to three times a day, and some improvement of symptoms is usually seen within 2–4 days. Tolerance can be an issue, with gastrointestinal side effects (unpleasant taste and constipation) being prominent. It is critical to educate patients regarding the need to separate the dosing of cholestyramine from other medications. The second-line agent for pruritus is rifampin; a dosage of 150 mg twice daily is well tolerated and has little risk of toxicity. Oral opioid receptor antagonists are a reasonable third-line agent. Sertraline, 75–100 mg a day was shown of benefit in one study of pruritus in patients with cholestatic hepatitis. Antihistamines have little role aside from use as a sleep aid. Plasmapheresis has been effective in truly refractory cases.

Fatigue

This symptom can be incapacitating, even in patients with early-stage disease. Modafinil has been shown to be effective in two small, uncontrolled pilot trials and in personal experience. Controlled trials are awaited. The dosage is 100–200 mg/day.

Osteoporosis

Approximately one third of patients with PBC develop osteoporosis, which can manifest before the onset of jaundice. Bone density should be checked every 2 years in patients with PBC and osteopenia managed aggressively. While benefit has not been proven, these patients should be on calcium (1500 mg a day) and vitamin D (1000 IU/day) supplementation. Alendronate has been shown to be beneficial in two trials and estrogen replacement in two studies of postmenopausal women; however, this benefit needs to be weighed against the increased risk of vascular disease and cancer.

Hyperlipidemia

Although serum lipids are often elevated in PBC, patients seemingly do not have an increased risk of atherosclerosis. This finding is explained partly by the presence of high levels of high-density lipoprotein and partly by the presence of a subfraction of low-density lipoproteins, called lipoprotein X, that has anti-atherogenic properties. Thus, it is rare that PBC patients require lipid-lowering agents. However, statin drugs can be used safely, if indicated, so long as standard monitoring for hepatotoxicity is performed.

Fat Soluble Vitamin Deficiency

Deficiency of fat-soluble vitamins occurs in patients with PBC, particularly those with advanced stage disease. Vitamin A deficiency can occur in up to one third of patients and cases of night blindness have been reported. Fat-soluble vitamin levels should be checked every 2 years and supplemented as needed.