The incidence of autoimmune hepatitis is somewhat unclear given that many early studies did not exclude the possibility of viral hepatitis C in patients with idiopathic chronic active hepatitis. Furthermore, standard diagnostic criteria for autoimmune hepatitis have only been recently developed. The likely incidence of autoimmune hepatitis ranges from 0.7 to 2.0/100,000; however, the prevalence and incidence of autoimmune hepatitis varies widely among different populations.¹ In particular, autoimmune hepatitis is more common in Caucasian and North American populations, compared with those from Asia. In the United States, the highest prevalence is seen in Alaska Natives (42.9/100,000).² The incidence of autoimmune hepatitis is roughly similar to primary biliary cirrhosis, but is twice as common as primary sclerosing cholangitis.

The current hypothesis is that hepatocyte injury due to a variety of factors (viral hepatitis, toxin, medication injury, etc.) triggers an aberrant and dysregulated immune response resulting in an autoimmune hepatitis.³ The factors that increase the risk of autoimmune hepatitis have not been fully characterized. The strongest associations are with major histocompatibility complex class I and II alleles.

As with other autoimmune diseases, autoimmune hepatitis is more common in women than in men, with a ratio of roughly 4:1. Autoimmune hepatitis can affect any age, but may be somewhat more
aggressive in younger-age populations. Although more common in younger individuals, one-fifth of autoimmune hepatitis patients are diagnosed after the age of 60 years. Older individuals are more likely to have concomitant autoimmune thyroid or collagen-vascular disease.

The clinical presentation of autoimmune hepatitis is quite varied, ranging from mild elevations in serum liver enzymes to liver failure. The majority of patients present with fatigue, lethargy, anorexia, and abdominal pain. Concurrent autoimmune diseases are common, including type 1 diabetes, celiac disease, autoimmune thyroiditis, and rheumatoid arthritis. The serum liver enzymes may fluctuate quite dramatically because of the waxing and waning nature of autoimmune hepatitis (Fig. 12.1). Up to 33% of patients, mainly those with advanced age, present with sequelae of cirrhosis including ascites, splenomegaly, and esophageal varices. Furthermore, many patients will have some degree of fibrosis on liver biopsy at first presentation, indicating that autoimmune hepatitis has a prolonged subclinical course in many patients.⁴

Up to 20% of patients may be entirely asymptomatic and only have elevations in liver enzymes that are detected during routine medical examination. Autoimmune hepatitis can uncommonly present clinically as an acute hepatitis with scleral icterus. Autoimmune hepatitis is thought to only rarely present as acute liver failure; however, many patients in this setting do not undergo a liver biopsy because of the risk of hemorrhage and it is likely that some patients with unexplained acute liver failure have an autoimmune etiology.

Often the trigger for gastroenterologists and hepatologists to consider a diagnosis of autoimmune hepatitis is the presence of increased serum transaminases and autoantibodies. Serum transaminase levels vary widely in patients with autoimmune hepatitis and do not accurately reflect disease severity. Bilirubin is often elevated, but alkaline phosphatase levels are usually normal or only mildly elevated in most patients.

Autoantibodies commonly seen in autoimmune hepatitis include antinuclear antibody (ANA), anti-smooth muscle antibodies (anti-SMAs), and anti-liver/kidney microsomal (anti-LKM) antibodies (Table 12.1).⁵ Anti-SMAs are the most common antibodies seen in autoimmune hepatitis and have been shown to be directed against filamentous-actin (F-actin). Measurement of F-actin antibodies may be more sensitive than measurement of anti-SMA.⁶ Anti-LKM antibodies are uncommon, but occur more frequently in children. Other rare autoantibodies include antibodies against soluble liver antigen/liver-pancreas antigen, liver cytosol Type 1, and asialoglycoprotein receptor, among others. Serum immunoglobulins are often elevated, usually 1.1 to 1.3 times normal levels.

Autoimmune hepatitis has been classified into two different types based on clinical and serologic features (Table 12.1).¹,³,⁵,⁷ Type 1 autoimmune hepatitis is characterized by elevations in ANA and/or anti-SMA, often with marked elevations in serum IgG levels. Type 1 autoimmune hepatitis can occur at any age and is by far the most common subtype. Type 2 autoimmune hepatitis is characterized by elevations in anti-LKM antibody and/or anti-liver cytosol antibodies
with only mild elevations in serum IgG. This subtype is more common in the pediatric population and may progress to cirrhosis more often than type 1 autoimmune hepatitis. Histologically, there are no differences between these subtypes. A third type of autoimmune hepatitis has been proposed based on elevations in soluble liver antigen/liver-pancreas antibodies; however, this type has not been proven to be a distinct subtype.⁸

Autoantibodies are useful in the diagnosis of autoimmune hepatitis, but the presence of autoantibodies in the setting of elevated liver enzyme levels does not necessarily indicate autoimmune hepatitis. Autoantibodies, particular ANA, can be elevated in a variety of diseases of the liver and in patients with other autoimmune diseases.⁹ Importantly, autoantibody negative autoimmune hepatitis also occurs and accounts for 5% to 10% of cases.¹⁰, ¹¹, ¹² Although the majority of autoantibody negative patients will have elevations in serum immunoglobulins, IgG levels may also be normal. Patients with autoantibody negative autoimmune hepatitis may have a more aggressive course. Although these patients respond to immunosuppressive therapy, they are less likely to achieve remission.¹⁰

Given the importance of making the diagnosis of autoimmune hepatitis, clinical scoring systems have been developed. The International Autoimmune Hepatitis Group in 1999 modified a scoring system
that evaluates clinical, serologic, and histologic data.¹³ More recently a simplified autoimmune hepatitis scoring system has been developed that again relies predominantly on serologic and histologic features (Table 12.2).¹⁴ Both have similar sensitivity and specificity for the diagnosis of autoimmune hepatitis.¹⁵ In both of these scoring systems, a high-titer ANA or anti-SMA receives additional points; however, many laboratories in the United States do not report titer levels, which decreases the utility of these scores. Moreover, the titers may not reflect the severity of the liver disease.¹⁶

The liver biopsy plays an important role when clinicians are considering a diagnosis of autoimmune hepatitis. Classically, autoimmune hepatitis is characterized by dense lymphoplasmacytic portal inflammation with brisk interface activity, wherein portal inflammatory cells extend from the portal tract into the adjacent hepatic parenchyma, resulting in hepatocyte injury (Fig. 12.2).¹⁷,¹⁸ The hepatic parenchyma also demonstrates marked lobular inflammation with clusters of lymphocytes, macrophages, and plasma cells (Fig. 12.3). The hepatocytes are often swollen and occasional multinucleated hepatocytes are seen. The hepatic plate architecture is disrupted by the inflammatory infiltrates and hepatocyte injury (Figs. 12.4 and 12.5). Often the necroinflammatory activity is more severe than seen in chronic viral hepatitis, particularly viral hepatitis C; however, in some cases, autoimmune hepatitis may be quite mild with only minimal to
mild interface and lobular activity. These cases can be quite difficult to arrive at the correct diagnosis.

There have been numerous studies looking at histologic features that are helpful in differentiating autoimmune hepatitis from other forms of hepatitis, including viral hepatitis as well as drug-induced liver injury. Compared with chronic viral hepatitis, autoimmune hepatitis on average demonstrates more interface hepatitis, heavier plasma cell infiltrates, rosettes, and emperipolesis (lymphocytes within hepatocytes).¹⁹ Emperipolesis (Fig. 12.6) and rosette formation (Fig. 12.7) were found to be superior to plasma cell infiltration and marked interface hepatitis in some studies.¹⁹, ²⁰, ²¹ Indeed, these features were also identified as features characteristic of autoimmune hepatitis in the simplified autoimmune hepatitis scoring system published in 2008.¹⁴ However, while emperipolesis and rosettes appear useful in studies, at the practical level emperipolesis is not reliably identified and the rosettes are a nonspecific finding, so most liver pathologists still rely on dense lymphoplasmacytic portal inflammation with brisk interface and lobular activity in the proper clinical setting to arrive at the correct diagnosis.

More recently, the presence of Kupffer cells with prominent, glassy, intracytoplasmic inclusions has been described in patients with autoimmune hepatitis, both adults and children.²² Immunohistochemically, this material is shown to be immunoglobulin. Although
this feature may be seen more commonly in patients with autoimmune hepatitis, it is unlikely to be used as a major criterion for diagnosis.

Immunohistochemical stains for various T cell and B cell markers have limited utility in the diagnosis of autoimmune hepatitis; however, the lymphocytes within autoimmune hepatitis tend to be CD4+ T cells. The IgG:IgM ratio is distinct in autoimmune hepatitis compared with primary biliary cirrhosis (IgM positive plasma cells predominate in primary biliary cirrhosis; IgG in autoimmune hepatitis), and immunostains can be helpful when the differential is autoimmune hepatitis versus primary biliary cirrhosis. However, IgG-positive plasma cells also predominate in other conditions such as infections and drug effects, so are not specific for autoimmune hepatitis. Finally, IgG4-positive plasma cells are scattered and rare in liver biopsies with autoimmune hepatitis. Marked IgG4-positive plasma cells should raise the possibility of IgG4-mediated disease.²³

Fibrosis in a liver biopsy indicates chronic injury and, as autoimmune hepatitis is a chronic liver disease, liver biopsies in this setting often demonstrate some degree of fibrosis.²⁴,²⁵ Determining the degree of fibrosis, however, can be quite challenging when there is marked necroinflammatory activity. In cases of confluent or bridging necrosis, one should carefully evaluate the trichrome and reticulin stains to not overestimate the degree of fibrosis. Bridging necrosis should appear more pale blue on a trichrome stain and demonstrate a collapse of reticulin fibers (Figs. 12.8 and 12.9). If there is any uncertainty regarding the degree of fibrosis, this should be clearly stated in the pathology report because bridging necrosis can heal without marked collagen deposition. Given that elastic fibers are deposited only in areas of long-standing fibrosis, evaluation for the presence of elastic fibers using an orcein stain may be helpful to separate reticulin collapse from true fibrosis (Fig. 12.10). Often in the setting of marked inflammatory activity and necrosis, it may be adequate just to indicate that there is or is not advanced fibrosis (bridging or cirrhosis). If the clinicians are concerned about advanced fibrosis, treatment with immune suppression followed by a
repeat biopsy may be the best way to document the stage of disease. In addition, acute autoimmune hepatitis without evidence of fibrosis does occur.²⁴ This finding should be recognized so as not to automatically assume the presence of fibrosis when considering a diagnosis of autoimmune hepatitis.