Technically, this term could refer to any nonnormal artery, but in common usage this term refers to an artery located in the hepatic lobule, instead of their normal location in the portal tract (Fig. 3.1). Aberrant arteries are a common feature of neoplasms, including hepatic adenomas and hepatocellular carcinomas. However, abnormal lobular arteries can also be found in reactive conditions such as focal nodular hyperplasia and in medical, nonmass forming liver disease. Examples of abnormal lobular arteries in medical liver disease include livers with vascular flow abnormalities, such as portal vein thrombi, and steatohepatitis with fibrosis.

A single dead cell is referred to as an acidophil body (Fig. 3.2). In liver pathology, this term is used most commonly for single, scattered dead hepatocytes. Acidophil bodies have dense, eosinophilic cytoplasm and are typically smaller than the adjacent hepatocytes. The nucleus is usually absent or small and shriveled. Acidophil bodies can be seen in a wide variety of hepatic injuries. When prominent lobular acidophil bodies are the only pattern of injury, unaccompanied by inflammation or cholestasis or fatty change, then the differential is primarily drug effect, acute viral infection, and low-grade/transient ischemia.

Accessory lobes affect approximately 10% of livers. They can be mistaken for mass lesions and can sometimes have tumors develop within them.¹, ², ³ They are also at risk for torsion.⁴ The most common accessory lobe is called a Riedel lobe (see separate definition entry). Most accessory lobes are connected to the liver by a thick pedicle and have a normal microscopic organization. Less commonly, accessory lobes are connected to the liver by a thin fibrovascular core containing blood vessels and connective tissue, bile ducts (often), and thin wisps or scattered islands of hepatocytes (occasionally). The accessory lobe itself might lack normal bile ducts and portal veins and can sometimes show nodular regenerative hyperplasia like changes. In contrast to an accessory lobe, ectopic hepatic tissue will have no connection to the liver.

The hepatic acinus is the functional unit of the liver and is usually illustrated as an elliptical or diamond shaped structure with ends in the portal tract and a bulging middle extending from central vein to central vein. Although the acinus itself cannot be readily seen on histology, the acinus forms the basis for microscopic liver zones, which can be seen histologically, with zone 1 referring to hepatocytes near the portal tracts, zone 3 referring to hepatocytes near the central vein, and zone 2 referring to those hepatocytes in between. See also the entry for lobule.

This term is commonly used to refer to any abrupt presentation of liver disease. There are formal clinical definitions for acute hepatitis, but formal definitions are rarely followed strictly outside of research studies. Although the formal definitions vary a bit, one common approach is to define acute hepatitis as an abrupt onset hepatitis that is less than 6 months in duration.

In cases presenting clinically as acute hepatitis, the histological findings will vary considerably depending on the type of injury, but can be predominately hepatitic, cholestatic, biliary, congested, fatty, or bland necrosis. The extent of injury in any of these patterns can range from minimal to marked. Of these injury patterns, a markedly active lobular hepatitis (Fig. 3.3) is almost always an acute hepatitis, as marked lobular inflammation is too injurious to be seen as a chronic pattern. Likewise, substantial hepatocyte necrosis indicates an acute injury for the same reason. Other than in these settings, there are no histological findings that can differentiate an acute hepatitis from a chronic hepatitis, so the term is typically avoided as a pathology diagnosis.

Of note, histological fibrosis indicates a chronic hepatitis, but the converse is not true, as the lack of fibrosis does not indicate an acute hepatitis. If there is a clinical history of an abrupt onset of the clinical disease, but definite fibrosis on the biopsy, then the findings suggest a flare of a chronic liver disease or a superimposed injury on top of an underlying chronic liver disease. In these settings, the term acute on chronic hepatitis is often used.

This term is used in gross pathology. Although it is largely obsolete, occasionally the term is still encountered. In the setting of acute liver failure from extensive necrosis, the liver is often atrophic on gross examination and the surviving hepatocytes can show fatty change and cholestasis, all of which contribute to the liver looking small and yellow. This general pattern of injury can be seen in numerous causes of extensive liver necrosis, ranging from acute viral infection (Fig. 3.4) to drug injury.

Individuals with alcoholic hepatitis can rarely (less than 1%) have a pattern of diffuse microvesicular steatosis (Fig. 3.5).⁵ The explanation for this pattern of liver injury is not clear, but neither the amount nor the duration of alcohol intake appears to be causal factors.⁵

As formally defined, an apoptotic body results when a cell undergoes programmed cell death. In practical pathology usage, an apoptotic body is a single dead cell and can refer to dead hepatocytes, dead bile duct cells, dead lymphocytes, dead tumor cells, or any cell population with scattered single dead cells. Apoptotic bodies can have accompanying inflammation, depending on the cause of the injury.

Synonym: Acidophil body, spotty necrosis, Councilman body. However, all of these synonyms are reserved for apoptotic hepatocytes.

Ascending cholangitis refers to an infection of the extrahepatic biliary tree that has “ascended” up the biliary tree into the liver, typically manifesting as bile ducts that are dilated, lined by attenuated epithelium, and filled with neutrophils (Fig. 3.6).

Ballooned hepatocytes are enlarged, have rarified cytoplasm, and may contain Mallory bodies. They do not have fatty change, lacking both microvesicular and macrovesicular steatosis. They are best identified using lower power magnification, such as 4× or 10×, where they should truly stand out as being larger and more swollen than the adjacent hepatocytes, typically at least 50% bigger than their neighbors (Fig. 3.7). Ballooned hepatocytes are most commonly seen in fatty liver disease, where they can help make a diagnosis of steatohepatitis. However, balloon cells can be seen in a wide variety of other diseases, in particular with chronic cholestasis. In these other settings, balloon cells in isolation do not indicate an additional component of fatty liver disease.

Also of note, ballooned hepatocytes can be easy to over diagnose if you spend too much time at 40× or 60×, as there are many cells that are a bit bigger than their neighbor and have a hint of cytoplasmic clearing; overall, it’s best to avoid calling these ballooned hepatocytes. You will certainly encounter hepatocytes with findings intermediate between those of classic ballooned hepatocytes and normal hepatocytes, but their diagnostic value is limited (Fig. 3.8). However, finding equivocal balloon cells should prompt continued examination for true balloon cells, as they can co-occur. Classic ballooned hepatocytes, and a subset of equivocal ones, will lose their normal expression of CK8/18.⁶,⁷ However, these immunostains are not routinely employed in current diagnostic pathology to identify ballooned hepatocytes.

Hepatocytes with ballooning degeneration are enlarged, rounded, and have rarified cytoplasm. Many will have Mallory hyaline. The term ballooning degeneration is
used in two settings primarily. First, the term can describe scattered, typically single, injured hepatocytes in the setting of steatohepatitis. Secondly, the term can describe broad swaths of injured hepatocytes in the context of marked hepatitis or in the setting of cholestasis, where ballooning degeneration can be part of the pattern of cholate stasis. Some authors prefer the term balloon cell in the setting of steatohepatitis and ballooning degeneration for other situations, but there is no strong consensus on term usage.

The bile duct is a tubular structure located in the portal tract that is lined by a simple cuboidal to low-columnar epithelium, which is composed of cholangiocytes. The bile duct is usually located in close proximity to the hepatic artery. In a normal portal tract, the bile duct and artery will be about the same size (Fig. 3.9). In severe chronic cholestatic liver disease, this may not be true, as the arteries can become more prominent.

Even in the normal liver, a bile duct may not be visible in up to 20% of the portal tracts. Of note, however, this absence of bile ducts is seen only in the smallest portal tracts and all normal medium-sized and larger portal tracts will have a bile duct. The bile duct should not be confused with bile ductules, which are quite different biologically, despite their similar names.

Bile duct duplication is associated with chronic obstructive biliary tract disease, such as primary sclerosing cholangitis or other stricturing processes. Instead of the normal single bile duct, a portal tract with bile duct duplication will have a small cluster of bile ducts in the center of the portal tract (Fig. 3.10). This change is not the same as proliferating bile ductules, where the ductules are located at the periphery of the portal tract.

Bile ducts can undergo metaplasia, usually in the setting of chronic biliary tract disease.⁸ The metaplasia can take various forms, most commonly intestinal metaplasia (Fig. 3.11) or pyloric gland metaplasia, and very rarely hepatic metaplasia (Fig. 3.12). Intestinal metaplasia is also associated with an increase in the number of neuroendocrine cells within the biliary epithelium. Paneth cells can also be found. Rare cases of clear cell metaplasia have also been reported.⁹

In primary biliary cirrhosis, the bile duct cells can sometimes have increased mitochondrial mass,¹⁰ giving them a densely eosinophilic appearance on light microscopy, a finding called “oncocytic metaplasia.” Oncocytic bile duct metaplasia can occasionally be found in other conditions, in particular with chronic liver disease and advanced fibrosis.

Bile duct hamartomas are composed of benign duct-like structures, usually with open lumens and bile deposits (Fig. 3.13). The duct-like structures interanastamose and can have a serpiginous appearance. They are found in the smaller branches of the portal tracts. The hamartomas can be small and microscopic, or can be larger grossly visible lesions. Most are less than 5 mm and almost all are less than 15 mm. The larger lesions typically have a background of dense fibrosis. They can be single or multiple and are often subcapsular.

This descriptive term is used to report lymphocytes within the bile duct epithelium, a finding that is often associated with bile duct injury (Fig. 3.14). Bile duct lymphocytosis is a key part of the pattern of injury seen in acute cellular rejection, but is also seen in many other conditions, including drug reactions, chronic viral hepatitis, autoimmune hepatitis, and primary biliary cirrhosis. Its presence in many diverse diseases shouldn’t dissuade anyone from its potential diagnostic usefulness, remembering that the focus should be on determining whether the bile ducts are targeted by the inflammation or whether the finding represents a bystander effect, with focal mild lymphocytosis in the setting of a more generalized moderate or marked portal inflammation.

A bile infarct can be seen with high grade, often acute, obstruction of the biliary tree or with long standing cholestasis. There will be a small circumscribed collection of bile stained and dead hepatocytes (Fig. 3.15). The center of the lesion commonly has extravasated bile. Most often, bile infarcts are located right next
to a portal tract. A giant cell histiocytic reaction to the bile can also be seen.

Bile ductules are different than the centrally located bile duct. Bile ductules are small, epithelial, tubular structures located at the periphery of the portal tracts (Fig. 3.16). Bile ductules are composed entirely of cholangiocytes and most often do not have a visible lumen, except for when the lumens have bile plugs. Bile ductules are a response to liver injury and can be seen in a wide variety of disease conditions, often being most prominent in the setting of obstructive biliary tract disease. The cell of origin for the bile ductules has been a major research interest for many years. At this time, the data suggests bile ductules most commonly originate from cells with stem cell-like features, but less commonly can originate from hepatocytes that undergo metaplasia. Bile ductules connect the canals of Hering to the bile duct.¹¹

Bile ductules will occasionally be distended by bile plugs (Fig. 3.17). This can occur both in the setting of more generalized, profound lobular cholestasis and when the lobules show little or no cholestasis. In the former setting, the cholestasis is a nonspecific finding that reflects the overall severe cholestatic injury. In the latter case, the finding is most often idiopathic. There is older literature, largely based on autopsy studies, that associated bile ductular cholestasis with sepsis. However, in surgical pathology specimens most cases are not sepsis related and remain idiopathic. When a potential cause is identified, more definite confirmation of the cause is often difficult because patients tend to have debilitating illnesses and are on numerous medications, making it almost impossible to isolate a specific etiology. Nonetheless, associations have been made with total parenteral nutrition therapy, congestive heart failure, and alcoholic hepatitis. In the setting of severe alcoholic hepatitis, a bile ductular pattern of cholestasis is associated with an increased risk of subsequently developing clinical sepsis,¹² though the reason(s) are unclear.

Experimental evidence suggests that liver injury can cause a reparative response where mature hepatocytes transform into bile ductules, a process called bile ductular metaplasia.

A bile ductular reaction is defined as increased numbers of ductules at the periphery of the portal tract (Fig. 3.18). In the normal liver, the portal tracts have few or no visible bile ductules. However, in response to liver injury, there can be a ductular proliferation, as the bile ductules are a source of liver progenitor cells and proliferate in response to injury. In many cases of mild liver injury, the bile ductular proliferation is mild and inconspicuous on hematoxylin and eosin stain (H&E), though it can be highlighted with special stains, such as cytokeratin 7. As the lobular injury becomes more severe, the ductular proliferation becomes increasingly visible on H&E. In other cases, the ductular reaction is the major pattern of injury and there is little or no lobular injury. This latter pattern can be accompanied by neutrophils and sometimes by portal tract edema and suggests biliary obstruction.

Many times a bile ductular reaction shows admixed neutrophils (Fig. 3.19). The admixed neutrophils should be in the stroma associated with the proliferating ductules and not in the lumen of the bile duct proper. Neutrophils in the main duct of the portal tract suggest ascending cholangitis, especially if the bile duct is dilated and the lining epithelium is attenuated (Fig. 3.6).

The ductular reaction is one of the key injury patterns in the liver. Historically, Popper and colleagues¹³ were one of the first users of the term ductular reaction. They further subdivided ductular reactions into three types¹⁴: type I, associated with biliary obstruction; type II, associated with significantly active hepatitis; type III, associated with massive liver necrosis. This terminology of types I, II, and III is no longer used, but serves as a useful reminder of the major types of injury that can lead to a ductular reaction.

The term bridging fibrosis is used when fibrous bands extend from either one portal tract to another portal tract (Fig. 3.20) or from a portal tract to a central vein. Synonyms are septal fibrosis and fibrous septa. Of note, however, there is nonuniform usage of these latter terms, especially the term fibrous septa. Although most authors use fibrous septa as a synonym for bridging fibrosis, occasional authors use the term fibrous septa to describe short fibrous extensions
from the portal tracts that do not actually connect two portal tracts, and thus would not be the same as bridging fibrosis.

With bridging necrosis, the lobules show an irregular band of dead hepatocytes that extends from central vein to central vein, central vein to portal tract, or less commonly from portal tract to portal tract. In the band of necrosis, the tissue often shows collapse and variable inflammation. This pattern is not specific for an etiology, but does indicate a severe liver injury. This pattern is most commonly seen in the setting of markedly active hepatitis or toxin exposure.

The canal of Hering connects the lobules to the biliary tree. Canals of Hering are formed partly by cholangiocytes and partly by hepatocytes. They are not visible on H&E stains, but with keratin immunostains can be seen as thin, linear structures that are formed by cuboidal cells and extend from the portal tracts into zone 1 of the hepatic lobules. The canals of Hering often go in and out of the plane of section, so they may appear discontinuous on a single slide, but serial sections show a continuous line of cells extending from lobules into the portal tracts.

Cholangiocytes are the epithelial cells that line the bile ducts and bile ductules. The canals of Hering are lined partly by hepatocytes and partly by cholangiocytes.

Cholate stasis is defined as swollen, pale hepatocytes located in zone 1 (periportal), resulting from injury because of chronic exposure to bile acids (Fig. 3.21). Cholate stasis results from chronic cholestasis. The hepatocytes in cholate stasis are sometimes confused for ballooned hepatocytes. Although there are some general similarities resulting from cell swelling, they are quite different. In cholate stasis, larger groups of often contiguous periportal hepatocytes are pale and swollen, whereas ballooned hepatocytes tend to be found as single, enlarged cells. As noted previously, cholate stasis affects the zone 1 hepatocytes, in contrast to the balloon cells in steatohepatitis, which are found mostly as scattered single cells, often in zone 3. The background changes are also distinct, as cholate stasis is seen in the setting of chronic cholestasis, whereas ballooned hepatocytes primarily occur in the setting of steatohepatitis. Mallory hyaline can be found in both cholate stasis and ballooned hepatocytes. Copper stains typically show scattered, though often sparse, granules of copper in cholate stasis. A CK7 is also positive for intermediate hepatocytes in cholate stasis, but does not stain balloon cells in fatty liver disease.

Histological cholestasis is defined by the presence of bile in the liver, visible on H&E stains. The location of the bile varies in different cases and in most cases can be seen in multiple different compartments, with hepatocellular cholestasis the most common, followed in descending order of frequency by cholestasis involving the bile canaliculi, the proliferating ductules, and the bile duct proper. In general, the nonhepatocellular forms of cholestasis take longer to develop and are associated with more severe cholestasis. Outside of this general correlation with the severity and length of cholestasis, there is little or no additional diagnostic information contained in identifying bile in the hepatocytes versus nonhepatocyte location. The ductular cholestasis pattern has been linked to sepsis,¹⁵,¹⁶ but is actually seen in many other severely cholestatic conditions and has little or no diagnostic specificity for sepsis. Furthermore, the major pattern of injury in sepsis is actually a nonspecific hepatitis and fatty change.¹⁶

Chronic hepatitis is often used somewhat informally by pathologists to describe any lymphocytic inflammation in the liver, but such usage should be discouraged, as the term chronic hepatitis does have a distinct clinical implication, where it is commonly defined as elevated serum enzyme levels for greater than 6 months. The best histological finding that can support a diagnosis of chronic hepatitis is convincing fibrosis, though the opposite is not true, as many cases of chronic hepatitis can take years to decades to start fibrosing.

This term is no longer used, but is of interest for historical reasons.¹⁷ A major clinical question, one that persists to this day, is why some individuals with chronic hepatitis (viral, autoimmune, etc.) progress to cirrhosis whereas others do not. One of the earliest attempts to better understand this question focused on histology, dividing cases into chronic active hepatitis versus chronic persistent hepatitis. Chronic active hepatitis was thought to have the greatest risk for fibrosis progression and was defined by the presence of piecemeal necrosis (now called interface activity) and either periportal or septal fibrosis. In contrast, chronic persistent hepatitis (also see entry below) had a good prognosis and was defined by having absent or slight piecemeal necrosis and either no fibrosis or portal fibrosis. This approach deeply influenced the thinking of the pathology community but eventually had to be abandoned when data showed it did not predict disease progression. However, the notion of piecemeal necrosis (interface activity) still has clinical relevance and is part of essentially all modern grading systems for chronic hepatitis. The duality of portal fibrosis versus periportal fibrosis has also persisted in some staging systems, but perhaps more because of inertia than biological relevance, as mostly they are about the same thing.

This term is obsolete but of historical interest. Early efforts to understand risk factors for fibrosis progression divided cases of chronic into one of two broad histological patterns: chronic active hepatitis versus chronic persistent hepatitis. In this duality, chronic persistent hepatitis had a better prognosis and was defined as having absent or “slight” piecemeal necrosis and either no fibrosis or mild portal fibrosis. See also the entry for “chronic active hepatitis.”

Cirrhosis is a diffuse fibrosis of the liver leading to parenchymal nodularity. Rene Laennec (1781 to 1826), a distinguished French physician who invented the stethoscope, is often given credit for the first description of a cirrhotic liver, which he named after “kirrhos,” the Greek word for “tawny.” As is often the case in history, Laennec thought he was the first, but he was not, as cirrhosis was described more accurately and earlier by two British pathologists, John Browne (1642 to 1700) and Matthew Baillie (1761 to 1823). Nonetheless, Sir William Osler (1849 to 1919), a prominent and dominant North American physician, one of the four founders of Johns Hopkins Hospital, gave Laennec the credit for describing cirrhosis in his very influential English language medical text books, and this seems to have stuck.¹⁸

If the necrosis pattern in a liver specimen is seen as scattered, single cells, then “spotty necrosis” or similar terms are used. However, if the apoptosis/necrosis involves multiple adjacent hepatocytes, then the term confluent necrosis is used. Confluent necrosis indicates a more severe pattern of hepatic injury. Confluent necrosis most commonly affects zone 3 hepatocytes, but with severe necrosis can extend to other zones of hepatocytes. The distinction between single cell necrosis and confluent necrosis in clinical practice should be made using common sense and with the goal in mind of conveying the overall findings in the biopsy. For example, if a biopsy specimen shows predominately spotty necrosis, but by carefully searching you also find a single area with 3 adjacent dead hepatocytes, the best descriptor for the overall pattern of injury is still spotty necrosis. In cases with severe confluent necrosis, bridging necrosis is also commonly found.

Councilman body is no longer a commonly used term, being replaced in the current literature by the terms apoptotic body or acidophil body. However, the term is of historic interest and is occasionally encountered in modern literature. Councilman bodies are named after the American pathologist William Councilman (1854 to 1933), who described apoptotic hepatocytes as a dominant finding in the pathology of yellow fever.

Ductal plate malformations can be diffuse in the setting of inherited polycystic liver and kidney disease or in the setting of congenital hepatic fibrosis, being found in most of the portal tracts. However, similar lesions can also be acquired, being found as single or small numbers of lesions, seen more commonly in cirrhotic than in noncirrhotic livers.

Ductal plate malformations are composed of elongated, interanastomosing bile ducts. They can encircle the portal tract in syndromic settings such as congenital hepatic fibrosis (Fig. 3.22). In other cases, particularly the sporadic lesions, they can form small nodules at the edges of portal tracts. Larger lesions can be grossly visible. The ducts typically have open lumens and bile plugs and are growing in a dense fibrotic background. Sporadic lesions are also commonly called von Meyenburg complexes.

Ductopenia is defined as a reduction in the number of bile ducts. When ductopenia is well established, it’s not that hard to recognize, if you think to look for it. However, early ductopenia can be very challenging to diagnose because in normal liver tissue some of the smaller portal tracts will not have bile ducts, potentially leaving you to wonder if your case is within normal limits or shows very early duct loss.

As a general rule, a biopsy is considered to be within normal limits if up to 20% of the smaller-sized portal tracts do not have bile ducts evident on H&E. With a keratin stain, rare portal tracts may still not have evident bile ducts, but fewer than on the H&E. Chronic inflammation in the portal tracts can obscure bile ducts and an immunostain should be used to confirm bile duct loss. Also of note, ductopenia can be accompanied by patchy bile ductular proliferation, so make sure you assess the bile ducts proper and not the bile ductules. One useful clue can be to look for portal tracts with hepatic arteries that are “unpaired” because they do not have a nearby bile duct.¹⁹ When using this approach, be aware of this potential pitfall: scarred central veins can recruit hepatic arteries²⁰ and by doing so can mimic a portal tract with an unpaired hepatic artery.