Liver biopsy is essential in the evaluation of patients with liver disease. This is not without risk and complications; however, they are uncommon and, fortunately, usually respond to conservative management (Rockey DC, Caldwell SH, Goodman ZD et al. Liver biopsy. AASLD Practice Guidelines. Hepatology, in press). Coagulopathy due to hepatocellular dysfunction and thrombocytopenia due to portal hypertension and hypersplenism are major concerns for an increased bleeding risk in patients with more clinically overt liver disease.395 Routine hematologic evaluation before liver biopsy includes reviewing the results of a recent international normalized ratio (INR) and platelet count. Drugs with antiplatelet activity, such as ticlodipine, aspirin, and nonsteroidal anti-inflammatories, need to be discontinued for at least 7 days before biopsy. Warfarin therapy should be discontinued for 3-5 days with documented normalization of INR. Typically, a platelet count o50 000 and an INR 41.5 are regarded as contraindications to blind percutaneous liver biopsy (Rockey DC, Caldwell SH, Goodman ZD et al. Liver biopsy. AASLD Practice Guidelines. Hepatology, in press). However, there is a controversy in recent medical literature about whether any platelet count level or INR derangement truly separates out those patients with liver disease most likely to bleed after liver biopsy.396
A study performed in the early 1980s in 200 patients undergoing laparoscopic liver biopsy with direct visualization of the site failed to establish any relationship between duration and extent of bleeding and prothrombin times, platelet count, or whole clot time.397 Recently, a systematic review of bleeding, including that associated with liver biopsy, also failed to establish a relationship between risk and conventional tests of coagulation.396 Although attempts at correction of coagulopathy with plasma replacement are common, there is also a lack of evidence that they reduce the risk of bleeding. This uncertainty has prompted a multicenter NIH-funded trial in the United States of plasma replacement in patients with an INR of 1.3-1.9 undergoing invasive hepatic procedures, and this should help determine whether plasma replacement is indicated to reduce the risk of postliver biopsy bleeding.398 Even with a normal INR and platelet count, there remains a concern about performing liver biopsy because of platelet dysfunction associated with uremia (Rockey DC, Caldwell SH, Goodman ZD et al. Liver biopsy. AASLD Practice Guidelines. Hepatology, in press). As with most of the other literature regarding liver biopsy, it is impossible to quantify the increased risk, if any, related to uremic platelet dysfunction. Although not supported by data on efficacy, there has been increasing use of desmopressin acetate (also known as DDAVP) (0.3 mg per kg body weight) infused immediately before liver biopsy in patients with CKD, although no specific serum level of creatinine or degree of reduction in GFR is currently determined for the use of DDAVP.399 The presence of a prolonged INR and/or a platelet count below 100 000 in a patient with liver disease are generally reliable indicators of underlying cirrhosis and may obviate the need for liver biopsy to determine histologic severity in a patient with HCV.
The majority of liver biopsies are currently obtained by the percutaneous transthoracic route. Ultrasound is being increasingly performed to identify the optimal biopsy site, although it remains controversial whether this maneuver increases the safety of the procedure.400,401 Typically, transvenous liver biopsy, via the transjugular or transfemoral route, is used in the presence of ascites, coagulopathy, or thrombocytopenia of such severity that a percutaneous approach is considered to be contraindicated, or when additional diagnostic information is required-notably, free and hepatic wedge pressures-to confirm the presence of portal hypertension.
Liver biopsy in patients with chronic HCV infection is indicated not only to assess disease severity with particular attention to the amount of fibrosis and necroinflammatory activity but also to exclude other concomitant causes of hepatic dysfunction, such as nonalcoholic fatty liver disease. A sufficiently large core of tissue is crucial for adequate interpretation to reduce sampling error. Gauge 16 or larger biopsy needles are recommended, ideally with a minimum length of 2.0-2.5 cm, to reduce sampling error. One study evaluated the accuracy of quantification of fibrosis based on biopsy core size, correlating an automated image analysis technique with Metavir score on hepatic resection specimens. 402 A core length of 2.5 cm allowed a more accurate assessment of fibrosis compared to smaller specimens. Smaller specimens may fail to identify serious liver disease in 20% of patients with liver disease based on a core length of 1.5 cm (Rockey DC, Caldwell SH, Goodman ZD et al. Liver biopsy. AASLD Practice Guidelines. Hepatology, in press).284 Sampling error is an important concern. A study performed laparoscopic liver biopsies from the right and left lobes of 124 patients with chronic HCV.284 Liver biopsy stage varied by at least 1 on a scale of 0-4 in one-third of patients between the wo lobes. Furthermore, in 18 patients (14%), cirrhosis was diagnosed in one lobe, but only stage 3 was observed in the other lobe. Reassuringly, a difference of two stages was observed only in a small minority of patients (3 (2.4%)), suggesting that although there may be intrahepatic variation in histologic severity, it does not usually exceed one stage. To accurately assess the severity of HCV infection, a minimum of 11 portal tracts are necessary.403 Guidelines on liver biopsy developed by the AASLD recommend a biopsy core that is ideally 3 cm long obtained by a 16-gauge needle (Rockey DC, Caldwell SH, Goodman ZD et al. Liver biopsy. AASLD Practice Guidelines. Hepatology, in press). Irrespective of who is performing the biopsy, it is imperative that an adequate core of tissue be obtained.
To ensure reproducibility of liver biopsy interpretation, a number of scoring systems have been devised in an attempt to quantify inflammation and fibrosis. An early attempt was a complex scoring system,404 and it was the basis for the subsequent Ishak system.405 The two most commonly used scoring systems at present are Metavir406 and Ishak. The Metavir system assigns a score of 0-4 for fibrosis, whereas the Ishak system scores fibrosis from 0 to 6 ranging from no fibrosis to established cirrhosis (Table 29).59 The simpler I-IV scoring system of Metavir has found favor with many pathologists for routine diagnostic use, whereas the more complex Ishak system has found application in large clinical trials of antiviral agents. In case of diagnostic uncertainty, consultation should be sought from a pathologist with expertise in liver biopsy, as up to 25% of cases reviewed by a hepatopathologist at a referral center has led to a substantial change in interpretation.407 Although there is currently considerable interest in noninvasive markers of hepatic fibrosis, they generally are most accurate in patients with either no fibrosis or advanced fibrosis.408 There is no information about their use in patients with HCV and CKD, and for now the prognostic information afforded by liver biopsy remains unsurpassed.409
