Cirrhosis

CHAPTER 10 Cirrhosis




Diagnosis of cirrhosis by liver biopsy


The ease with which the pathologist can diagnose cirrhosis from a biopsy specimen depends on the sample as well as on the criteria used. The sample may be sufficiently big, and the nodules sufficiently small, to make the diagnosis obvious. On the other hand a slender core from within a large cirrhotic nodule can be difficult to identify as such (Fig. 1.4). There are occasions when the pathologist can do no more than hint at the possible diagnosis.


The type of biopsy needle used also influences the ease of diagnosis. Very narrow needles may be adequate to obtain tumour samples, but may be inadequate for the accurate diagnosis of medical conditions. For example, in staging chronic hepatitis thin-needle biopsies obtained under computed tomographic guidance may underdiagnose cirrhosis for advanced bridging fibrosis.3 Some clinicians prefer to use the Trucut type of needle when cirrhosis is suspected in order to lessen the risk of fragmentation,4,5 but suitable samples can usually be obtained with needles of the aspiration type.6,7 Transjugular biopsy is used when there is a risk of haemorrhage by other routes. The combination of biopsy with laparoscopy has been advocated.8,9 Operative wedge biopsies of cirrhotic liver give an accurate idea of the relative proportions of parenchyma and stroma in the liver as a whole.10


The histological criteria for a diagnosis of cirrhosis are outlined in Table 10.2. The two fundamental criteria, nodularity and fibrosis, reflect the definition of cirrhosis. When there are well-defined, rounded nodules surrounded by fibrous septa the diagnosis is easily established. Underestimating the stage of fibrosis because of specimen fragmentation (see below) is a concern, particularly when scoring biopsies in chronic hepatitis.10a Correlation with clinical and laboratory data help surmount this problem. Occasionally, a nodular appearance just deep to the liver capsule is not representative of the whole liver but has resulted from transection of a tongue or peninsula extending from the main bulk of the parenchyma.


Table 10.2 Cirrhosis: diagnostic criteria











A. Fundamental

B. Relative


In many patients the relative criteria listed in Table 10.2 are equally important. They allow a tentative diagnosis of cirrhosis to be reached, readily converted to a firm diagnosis when correlated with other data. A diagnosis of cirrhosis therefore requires communication between pathologist and clinician, and cannot be exactly equated with a histological stage.11



Fragmentation


Fragmentation of the specimen, either at the time of biopsy or during processing in the laboratory, should itself suggest the possibility of cirrhosis (Fig. 10.1). The specimen is more likely to break into fragments when needles of the aspiration type (e.g. Menghini) are used. Other biopsy specimens that are likely to fragment are metastatic tumours surrounded by reactive fibrous tissue, and hepatocellular carcinoma.




Abnormal structure


Structural changes should be assessed by means of a reticulin preparation, preferably not counterstained. This may show two features not readily seen with other stains. First, although nodules are readily cored out of the dense fibrous stroma of a cirrhotic liver during aspiration biopsy, a thin layer of connective tissue tends to adhere to the nodules over much of their surface (Fig. 10.2). This layer may be difficult to see even with the help of collagen stains, and is easily missed in haematoxylin and eosin (H&E)-stained sections (Fig. 10.3). Second, minor alterations of structure become apparent even in those nodules which closely mimic normal liver. Such alterations include abnormal orientation of reticulin fibres resulting from different patterns and rates of growth in different areas (Fig. 10.4) and approximation of portal tracts and terminal venules. The number of venules may be abnormally large in relation to the number of portal tracts (Fig. 10.5), and the latter are sometimes abnormally small and poorly formed (see Fig. 1.4). A more obvious structural abnormality in cirrhosis is the presence of septa linking central veins (terminal hepatic venules) to portal tracts. These septa must be distinguished from recently formed necrotic bridges.



image

Figure 10.3 Cirrhosis: selective sampling.


Same field as in Fig. 10.2. In a haematoxylin and eosin preparation the thin layer of connective tissue is not easily seen. (Needle biopsy, H&E.)




In wedge biopsies, excess fibrous tissue in and near the capsule and crowding of vessels must be distinguished from the changes of cirrhosis. The latter extend through the specimen, whereas the former is confined to the capsular and immediately subcapsular area.12 Very occasionally a wedge biopsy of part of a large, well-differentiated regeneration nodule fails to show the histological features of cirrhosis.



Hepatocellular changes


In some biopsies from cirrhotic livers the hepatocytes are normal in appearance and arrangement, so that diagnosis rests on the structural changes discussed above. In others there are more or less obvious abnormalities of growth.


Regeneration is suggested by thickening of the liver-cell plates (Fig. 10.6). In any liver an oblique plane of sectioning will cause a few plates to appear more than one cell thick, but widespread double-cell plates are seen when there is active growth. Hepatocytes in hyperplastic areas contain little or no lipofuscin pigment, even near terminal venules. Regeneration is not always evident in cirrhosis because it is not a continuous process. Its absence does not therefore exclude the diagnosis. Conversely, its presence does not prove cirrhosis, because it is found also in other circumstances, for example after an acute hepatitis and in the precirrhotic stages of chronic biliary diseases.



A very characteristic feature of cirrhosis is the presence of adjacent populations of hepatocytes growing at different rates and having different cell and nuclear characteristics (Fig. 10.7). This pleomorphism gives rise to the abnormalities of reticulin pattern already mentioned, notably a tendency for reticulin fibres in the different growth areas to lie in different directions.



In a minority of cirrhotic livers the hepatocytes show structural atypia of a degree sufficient to warrant a label of dysplasia, an appearance further discussed in Chapter 11. Two types have been described, large-cell dysplasia13 and small-cell dysplasia.14 Because of the controversial status of either type as a precursor of malignant change,15 some authors prefer to call them large-cell change and small-cell change.15,16 In the large-cell form, the cells are enlarged and their nuclei are hyperchromatic and irregular in shape, with prominent nucleoli (Fig. 10.8). Nuclear–cytoplasmic ratio is normal or only moderately increased.17 This type of dysplasia was first described in an African population with a high incidence of hepatocellular carcinoma (HCC) and hepatitis B virus (HBV) infection.13 It is most often seen in patients with HBV and HCV infection but may also be evident in other chronic liver diseases.18 There is evidence of an association of large-cell dysplasia with an increased risk of development of HCC independently of other risk factors.19,20 Decreased expression of cell cycle checkpoint markers, presence of cytoplasmic DNA micronuclei and shortened telomeres in large-cell change are evidence favouring a disposition to HCC.21 Demonstration of an increased hepatocyte proliferation rate as a risk for carcinoma is also important.22 Care should be taken not to interpret the nuclear atypia which may be associated with cholestasis as large-cell dysplasia.16



In small-cell dysplasia the nuclear–cytoplasmic ratio is increased but the overall size of the affected cells is less than normal (Fig. 10.9). Zones of dysplastic hepatocytes of either type support a diagnosis of cirrhosis, and are regarded by some clinicians as an indication for increased monitoring for HCC. A finding of dysplasia of either type should therefore be specifically mentioned in liver biopsy reports.



Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jul 25, 2017 | Posted by in GASTROENTEROLOGY | Comments Off on Cirrhosis

Full access? Get Clinical Tree

Get Clinical Tree app for offline access