This procedure has been greatly simplified, and its morbidity and mortality markedly reduced, by the introduction of small-caliber biopsy needles such as the Menghini. Nevertheless, there is a small but definite risk. Relative contraindications to biopsy include a PT in the anticoagulant range or a platelet count less than 50,000/mm3. Liver biopsy is especially useful in the following circumstances:

1. To differentiate among the many etiologies of liver function test abnormality when the clinical picture and laboratory test pattern are not diagnostic. This most often happens when the AST level is less than 10 or 20 times the upper reference limit and the ALP level is less than 3 times the upper limit. In cases of possible obstructive jaundice, extrahepatic obstruction should be ruled out first by some modality such as ultrasound.
2. To prove the diagnosis of metastatic or primary hepatic carcinoma in a patient who would otherwise be operable or who does not have a known primary lesion (in a patient with an inoperable known primary lesion, such a procedure would be academic).
3. In hepatomegaly of unknown origin whose etiology cannot be determined otherwise.
4. In a relatively few selected patients who have systemic diseases affecting the liver, such as miliary tuberculosis, in whom the diagnosis cannot be established by other means.

A discussion of liver biopsy should be concluded with a few words of caution. Two disadvantages are soon recognized by anyone who deals with a large number of liver specimens. First, the procedure is a needle biopsy, which means that a very small fragment of tissue, often partially destroyed, is taken in a random sample manner from a large organ. Localized disease is easily missed. Detection rate of liver metastases is about 50%-70% with blind biopsy and about 85% (range, 67%-96%) using ultrasound guidance. Second, many diseases produce nonspecific changes that may be spotty, may be healing, or may be minimal. Even with an autopsy specimen it may be difficult to make a definite diagnosis in many situations, including the etiology of many cases of cirrhosis. The pathologist should be supplied with the pertinent history, physical findings, and laboratory data; sometimes these have as much value for interpretation of the microscopic findings as the histologic changes themselves.

In summary, liver biopsy is often indicated in difficult cases but do not expect it to be infallible or even invariably helpful. The best time for biopsy is as early as possible after onset of symptoms. The longer that biopsy is delayed, the more chance that diagnostic features of the acute phase have disappeared or are obscured by transition to healing.