Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Glucksberg-Seattle criteria
graft-versus-host disease
hemopoietic cell transplantation
hemopoietic stem cell transplantation
International Bone Marrow Transplant Registry
a Although the prefix CK is widely used in surgical pathology to designate human cytokeratins, consensus nomenclature recommends the replacement of “cytokeratin” with “keratin” and the prefix “CK” with “K.” (Schweizer J, Bowden PE, Coulombe PA, et al. New consensus nomenclature for mammalian keratins. J Cell Biol . 2006;174:169–174)
keratin
loss of intrahepatic bile ducts
periodic acid–Schiff
primary biliary cholangitis
primary sclerosing cholangitis
Several terms have been used interchangeably to indicate a reduction in the number of intrahepatic bile ducts: paucity of intrahepatic bile ducts, vanishing bile duct syndrome , and ductopenia . However, historical usage over the years has led to specific diagnostic connotations for each; thus, the term paucity of intrahepatic bile ducts as initially formulated by Alagille is most often used to describe absence of bile ducts in the eponymous syndrome (discussed in Chapter 5 ). Vanishing bile duct syndrome delineates a clinicopathologic condition that includes disappearance of intrahepatic bile ducts associated with chronic cholestasis of various causes whereas ductopenia has come to imply absence of bile ducts in at least 50% of portal tracts, thus losing its neutral descriptive label and acquiring a quantitative dimension. The absence or loss of bile ducts in portal tracts is a morphologic observation encountered in multiple disease entities; thus, we prefer to apply a descriptive expression, loss of intrahepatic bile ducts (LIBD), to avoid presumptuous associations with specific diseases, pathogenetic backgrounds, or severity of bile duct loss.
LIBD in an adult liver may be due to a variety of diseases, both primary liver disorders, such as primary sclerosing cholangitis (PSC), and hepatic manifestation of systemic illnesses, such as graft-versus-host disease (GVHD) and sarcoidosis. In addition, LIBD may be a consequence of drug-induced liver injury ( Box 28.1 ; Table 28.1 ). There are several other, less frequent or rare causes of LIBD; these have been summarized in a comprehensive review.
Primary biliary cholangitis
Primary sclerosing cholangitis
Chronic liver graft rejection
Graft-versus-host disease
Ischemic-type cholangitis
Drug-induced bile duct injury
Sarcoidosis
Idiopathic adulthood ductopenia
Liver Transplantation | Hematopoietic Cell Transplantation |
---|---|
Acute rejection | Acute graft-versus-host disease |
Chronic rejection | Chronic graft-versus-host disease |
Obstructive cholangiopathy | Obstructive cholangiopathy |
Ischemic-type cholangitis | Sepsis |
Recurrent primary biliary cholangitis | Drug-induced disease |
Recurrent primary sclerosing cholangitis | |
Secondary sclerosing cholangitis (various causes) | |
Sepsis | |
Drug-induced injury |
Although no single mechanism can explain the loss of bile ducts in all cases, two pathways appear to account for the large majority. The first pathway involves immunologic mechanisms, including autoimmune destruction of bile ducts in primary biliary cholangitis (PBC); immune-mediated inflammatory reactions in PSC and sarcoidosis; and alloimmune reactions that occur in donor-host interactions following liver and hemopoietic cell transplantation (HCT). In liver allograft rejection, bile duct injury is the consequence of alloimmune attack by the recipient’s immunocompetent cells on allogeneic cholangiocytes. A reverse situation occurs in GVHD following HCT when the engrafted immunocompetent cells react against allogeneic recipient bile duct antigens. Similar to the native liver, immunologic mechanisms are thought to be responsible for recurrent PBC and PSC, which are important causes of LIBD in the liver allograft. In addition, drug-induced LIBD is often suspected to be a consequence of immunologic damage that occurs either because of unmasking of self-antigens or molecular mimicry. The second major pathogenetic pathway responsible for LIBD appears to be ischemic damage to the biliary tree. This is exemplified by chronic allograft rejection, ischemic cholangiopathy, and some cases of secondary sclerosing cholangitis. Finally, there are a significant number of clinical conditions in which the underlying mechanism is not known, as well as a handful of other conditions such as chronic allograft rejection, in which both immunologic and ischemic components contribute to bile duct damage and loss.
Although classically described as portal “triads” containing a branch each of the hepatic artery and portal vein, and a tributary of the bile duct, slightly more than half of all portal tracts observed in a liver biopsy do not conform to this configuration. A portal tract in a normal adult human liver contains, on average, two interlobular bile ducts, two hepatic arteries, and one portal vein. The bile ducts are of similar caliber to the hepatic arteries and the two structures usually run parallel to each other ( Fig. 28.1 and eSlide 1.1 ). Similarly, portal tracts lacking one of the three structures are not uncommon in the normal liver, and the term portal dyad , instead of portal triad, has been applied to designate this normal variation of the portal tract. Although the absent structure in most portal dyads is the portal vein, in normal peripheral liver tissue sampled in a needle biopsy specimen, approximately 7% of portal tracts may lack a bile duct.
To distinguish normal portal dyads lacking a bile duct from pathologic LIBD necessitates quantification of the bile duct-to-portal tract ratio in a biopsy that contains a sufficient number of portal tracts. Occasionally, identification of bile ducts may have to be facilitated by immunohistochemical stains for the biliary keratins (Ks), K7 or K19, to highlight the bile duct or its remnants ( eSlide 28.1C, eSlide 28.1D ). In general, ductopenia is diagnosed when absence of bile ducts is observed in 50% of portal tracts, provided that an adequate number of portal tracts (between 10 and 20) are available for evaluation in the biopsy. An average liver biopsy of 1.8 cm length contains 21 portal tracts and seven terminal hepatic venules.
Validating the absence of a bile duct in a portal tract is more difficult than confirming its presence because an absent bile duct is easily overlooked, especially when it has totally vanished; the portal tract contains insignificant numbers of inflammatory cells; and there is no ductular reaction. These features are not uncommonly encountered in chronic liver allograft rejection ( Fig. 28.2A–B and eSlide 38.6 ).
When assessing LIBD, the hepatic arteriole serves as the most consistent and reliable profile in a portal tract. In a normal liver, the majority of bile ducts (92%) are accompanied by a hepatic artery (see Fig. 28.1 ) and almost all hepatic arteries (96%) run in parallel with a bile duct. The extent of bile duct loss could therefore be determined by looking for the number of “unpaired arterioles,” which are arterioles that are unaccompanied by (unpaired with) a bile duct. However, care should be exercised when applying this practical rule in those instances when the hepatic artery itself is not guaranteed to be present (eg, loss of hepatic artery branches may occur in small portal tracts in chronic liver graft rejection).
Arterialization of central zones, a phenomenon reported to be common but underrecognized in nonalcoholic steatohepatitis and alcoholic steatohepatitis, presents another pitfall in determining LIBD. In these instances, small arteries without adjacent bile ducts are found in scarred centrizonal areas that may also contain a mild ductular reaction, thus closely mimicking portal tracts. These features may lead to an incorrect identification of a portal tract lacking a bile duct and hence to an erroneous diagnosis of ductopenia. The darker and denser elastic fibers in a portal tract may help differentiate an actual portal tract from an arterialized central zone.
Another mimic of portal tracts are the portal tract–like structures in biopsies derived from focal nodular hyperplasia and hepatic adenomas. These structures contain variable quantities of inflammatory cells, arteries, and bile ductules. However, an actual bile duct paired with an artery of a similar caliber is not present. In addition, the arteries usually show abnormal features such as eccentric intimal thickening ( Fig. 28.3 ). Awareness that the biopsy is acquired from a focal liver lesion is helpful in recognizing the precise nature of these portal tract–like structures.
LIBD presents variable histologic features, depending on the type and stage of the associated diseases and the severity of bile duct damage. The accompanying features such as ductular reaction and inflammatory infiltrate are also dependent on the type and stage of the disease. Many variable disease entities may show overlapping histologic characteristics of bile duct damage preceding LIBD; therefore clinicopathologic correlation is mandatory for accurate interpretation of the histologic findings.
PBC is an autoimmune disease that affects the interlobular and septal bile ducts, leading to LIBD, fibrosis, and eventually biliary-type cirrhosis. The severity of the histologic changes in PBC is categorized in a staging system (stages I to IV), based on the extent and progression of the bile duct injury, portal inflammation, and fibrosis ( eSlide 26.1, eSlide 26.2 ). However, portal tracts and bile ducts are unevenly affected, and features of all stages can be present concurrently in a single liver sample.
In early PBC, portal tracts show dense lymphoplasmacytic portal inflammation with variable numbers of eosinophils and neutrophils. The inflammatory cells tend to be centered on a damaged bile duct showing epithelial distortion with cytonuclear attenuation, an irregular bile duct configuration, and intraepithelial leukocytes. With progression to stage II, there is periportal extension of inflammatory cells, mimicking the interface hepatitis seen in chronic viral hepatitis. At this early stage, the small interlobular bile ducts may start to disappear. The rate of progression and extent of LIBD is variable, not only among individual PBC patients but also among portal tracts within the same patient. The finding of bile duct loss is important in differentiating PBC from diseases such as hepatitis C that show bile duct damage without actual loss of bile ducts ( Fig. 28.4 ). The dense inflammatory infiltrate in early PBC may conceal the bile ducts and augment the difficulty in determining their numbers. In this situation, immunohistochemical stains for biliary antigens, such as K7 or K19, help highlight the bile ducts, including their remnants as well as bile ductules at the periphery of portal tracts ( Fig. 28.5 ). Within the lobule, these markers highlight the intralobular canals of Hering. The K7 stain, and less prominently K19, may highlight groups of small periportal hepatocytes and intermediate hepatobiliary cells sometimes observed in continuity with ductular cells. These intermediate hepatobiliary cells, so named because they show morphologic and immunophenotypical features intermediate between cholangiocytes and hepatocytes, are larger than the small cuboidal ductular cholangiocytes but smaller than hepatocytes. They show mild membranous expression of biliary keratins, in particular K7 staining, giving rise to the designation blushing hepatocytes (N. Theise, oral communication) ( Fig. 28.6 ). Whether they represent hepatic progenitor cell proliferation is still inconclusive. The presence of intermediate hepatobiliary cells is not specific for PBC and is observed regularly in several other cholestatic conditions.
The mechanism of bile duct loss following immune injury is attributed to apoptosis of cholangiocytes and oxidative stress-mediated cellular senescence. The bile duct damage and LIBD, with subsequent impediment of bile flow, elicits a ductular reaction of variable degrees. However, these ductules fail to restore normal bile flow, and although overt cholestasis may still be absent, features of chronic cholestasis such as accumulation of copper and copper-associated protein may be present in periportal hepatocytes. Less pronounced ductular reaction may be due to the fact that the canals of Hering and bile ductules, themselves targets of the autoimmune attack, have been destroyed. As the disease advances, the portal tracts show variable degrees of LIBD and ductular reaction, in a background of decreasing portal inflammation and increasing portal fibrosis; the latter eventually extends beyond the portal tracts to end inevitably in biliary cirrhosis ( eSlide 26.4 ).
PSC is an immune-mediated inflammatory disorder that causes multifocal strictures and segmental ectatic dilatations that produce the pathognomic cholangiographic appearance of a “beaded” biliary tract. However, unlike PBC, which affects the smaller peripheral ducts, the extrahepatic and large intrahepatic ducts are affected in PSC. Thus, the histologic features seen in a liver biopsy reflect proximal changes of biliary tract obstruction, which are not specific for PSC. The diagnostic label small duct PSC is applied to a disease variant in which typical biochemical and histologic features of PSC exist in spite of a normal cholangiogram, thus designating a disease that affects small ducts to the exclusion of large extrahepatic and intrahepatic ducts.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here