Is a biopsy really necessary?

Read this article from the American Journal of Gastroenterlogy: It seems to indicate that there are alternatives to liver biopsies.

--------------------------------------------------------------------------------

Liver Biopsy in Chronic Hepatitis C: Routine or Selective

Gabriel Garcia, M.D.a and Emmet B. Keeffe, M.D.a

--------------------------------------------------------------------------------

A percutaneous liver biopsy can be useful in patients with chronic hepatitis C virus (HCV) infection by providing information regarding the stage of fibrosis and grade of inflammation. Expert consensus groups from the United States and Europe have recommended the routine performance of liver biopsy before initiation of antiviral therapy for chronic hepatitis C (1, 2). The purpose of a pretreatment liver biopsy is to distinguish patients most likely to benefit from therapy (i.e., those with moderate inflammation and advanced stages of fibrosis) from patients who may be less likely to benefit (i.e., those with mild inflammation and minimal or no portal fibrosis). A liver biopsy has been shown to be useful in determining the likelihood of progression of liver injury in chronic hepatitis C; moderate or severe inflammation and fibrosis have been associated with a more rapid progression to cirrhosis (3, 4).

Most clinicians use liver biopsy to determine the urgency to treat their patients, and the current consensus guidelines recommend that only patients with significant fibrosis (Metavir stages F2 [a few septa], F3 [many septa], and F4 [cirrhosis]) should be considered for antiviral therapy (1, 2). In addition, a biopsy occasionally may yield an unexpected diagnosis or document a coexisting condition such as nonalcoholic fatty liver disease, alcoholic liver disease, autoimmune hepatitis, or significant iron loading. Finally, if current investigational methods of treating fibrosis come of age and become clinically useful, then an acceptable method of quantifying liver fibrosis becomes an essential requirement to monitor and determine endpoints of therapy.

However, patients with chronic hepatitis C are not always eager to have a liver biopsy. They frequently have anticipatory anxiety, which would be expected of a procedure that is associated with pain in 30% of patients, severe complications in 0.3%, and death in 0.03% (5, 6). In a study conducted in our center, the duration of pain after biopsy extended beyond the day of the biopsy in 40% of patients, and extended for over 1 wk in a small number (I. Barnard and G. Garcia, unpublished data). In our study, 15% of patients who had a liver biopsy would not have agreed to have the procedure done if they had known how they would feel during and after the procedure. Liver biopsy also adds significant direct costs (equipment, observation time, and time of a skilled clinician and pathologist) and indirect costs (time away from work and home) to the management of patients with chronic hepatitis C. Finally, a cost-effectiveness analysis suggested that the best strategy in the management of chronic HCV infection is to offer therapy to all patients and not perform liver biopsies (7). Thus, selective rather than routine liver biopsy might be used to assist in decision making regarding therapy (e.g., the patient who prefers to defer therapy rather than undergo current therapy if no or minimal fibrosis is present). Other patients prefer an attempt to achieve a sustained virological response to antiviral therapy irrespective of whatever a liver biopsy might show, and yet other patients not only opt for therapy but also want to know their degree of hepatic fibrosis.

An approach to the determination of the need for therapy because of the presence of advanced fibrosis without the performance of liver biopsy is to find the right combination of biochemical markers, and/or other clinical and imaging findings, with high positive and negative predictive values of advanced fibrosis. In this issue Pohl and colleagues ( Cool

propose a simple formula based on readily available tests to determine the likelihood of Metavir stage F3 or F4 fibrosis in patients with chronic hepatitis C. In their patient population, an AST/ALT ratio of 1 in combination with a platelet count of <150,000/mm3 detected patients with severe fibrosis with an excellent specificity (99.1%) and positive predictive value (93.1%) but with a relatively low sensitivity (41.2%). This formula could not be used in patients with significant alcohol consumption. The combination of the two laboratory values, AST/ALT ratio 1 and platelet count < 150,000/mm3, would result in the elimination of a liver biopsy in a relatively small number of patients (15/211 [7.1%]). The investigators were unable to distinguish patients with no or minimal fibrosis using an AST/ALT ratio of <1 and various combinations of platelet counts. Although a good model will hold true for the population in which it was developed, one cannot predict with certainty how well that model will perform in all patients with chronic hepatitis C until it is tested and shown to have general applicability.

The authors suggest a decision tree for staging fibrosis and determining the need for liver biopsy. The decision regarding liver biopsy, however, is likely more complex than outlined in their algorithm. Examples of other important issues regarding whether or not to perform liver biopsy are patient preferences, cost-effectiveness strategies, presence of relative contraindications, and the finding of other laboratory abnormalities such as a high-titer antinuclear antibody with hypergammaglobulinemia or abnormal iron studies.

Other investigators have studied the AST/ALT ratio in patients with chronic hepatitis C with results that are consistent with the findings of Pohl et al. (9, 10, 11). Imbert-Bismut et al. (12) used a more complex formula and have achieved good performance of their model in identifying patients with or without significant fibrosis. Use of their formula could correctly classify 46% of their patients with chronic HCV infection into these two categories. However, their formula required three assays not generally performed or routinely available in patients with chronic viral hepatitis C (2-microglobulin, haptoglobin, and apolipoprotein A1) in addition to three routine tests (-globulin, -glutamyltranspeptidase, and total bilirubin).

One of the important endpoints and justifications for antiviral therapy in patients with chronic hepatitis C is the prevention of progression to cirrhosis and thus elimination of the risk of decompensated liver disease and hepatocellular carcinoma. As long as we continue to treat our patients with this endpoint in mind, it will be important to have dependable estimates of the degree of liver fibrosis. It is possible that specific markers of fibrosis, such as hyaluronic acid or procollagen-III peptide, will be shown to reliably reflect the stage of hepatic fibrosis, but it is more likely that the role of these tests will be in following disease progression and the impact of antiviral therapy (13, 14). An estimate of hepatic fibrosis will be of even greater importance as the promise of generic antifibrotic therapies becomes a reality.

The work of Pohl and others reassures both clinicians and patients that not everyone with chronic hepatitis C who is a candidate for antiviral therapy needs to undergo a procedure for which there is a finite complication rate, personal and societal costs, and much apprehension. Thus, we may be moving to the selective rather than the routine use of liver biopsy in patients with chronic HCV infection. Further work is needed to establish a model that accurately predicts the degree of fibrosis, including the early stages; can be generalized to all patients with chronic hepatitis C; and performs well regardless of alcohol intake.