Laboratory Tests in Liver Disease

Transaminases

AST and ALT, formerly referred to as serum glutamic-oxaloacetic transaminase (SGOT) and serum glutamic-pyruvic transaminase (SGPT), respectively, have been regarded as reliable and sensitive markers of liver injury. They are soluble enzymes present in the mitochondria and cytosol of hepatocytes; at times of hepatic injury, when cells are damaged or cell membranes become leaky, these enzymes are released into the circulation. However, the source of normally circulating aminotransferases is unclear. Interestingly, elevation of ALT in at least one animal model, the methionine-choline-deficient diet–nonalcoholic steatohepatitis (MCD-NASH) animal, resulted from increased messenger RNA (mRNA) transcription. Although aminotransferases are considered sensitive markers of hepatocyte injury, transaminemia also occurs in nonhepatic diseases such as hemolysis, rhabdomyolysis (with concurrent elevation in creatine phosphokinase), various muscular dystrophies, muscle exertion, anorexia nervosa, acute myocardial infarction, infarcted bowel, and celiac disease.

The range of normal values is calculated by each laboratory as the mean (±2 standard deviations [SD]) in a group of healthy people from the local geographic area. Frequently, these “healthy volunteers” may have unrecognized nonalcoholic fatty liver disease (NAFLD), resulting in elevated range of normal values. It has therefore been suggested that an ALT greater than 19 U/L for women and greater than 29 U/L for men should be considered abnormal. However, fluctuating levels are not uncommon in an individual. AST levels normally measure at approximately 0.8 of ALT levels; they vary slightly with age and gender, with men having higher levels than women.

AST is also present in cardiac muscle, skeletal muscle, kidneys, brain, pancreas, lungs, leucocytes, and erythrocytes. Its half-life is 17 ± 5 hours, and serum levels show day-to-day variation of 5% to 10%. When stored, the activity of AST is stable at room temperature for 3 days, in the refrigerator for 3 weeks (<10% loss), and for years when frozen (<15% loss). The organ-to-serum activity ratios suggest that AST activity in the liver is 9000 times higher than in serum. African Americans appear to have 15% higher levels, and strenuous exercise can lead to a threefold increase in serum levels. Rarely, an isolated elevation in AST without concurrent elevation in other liver enzymes is noted, and this is a result of the presence of the macro-AST enzyme. In this condition, AST forms complexes with other proteins, primarily immunoglobulin G (IgG), resulting in a chronically elevated AST level, which is of no clinical consequence.

ALT, although believed to be liver specific, is also present in muscle and kidney. Its half-life is 47 ± 10 hours, longer than that of AST. When stored, the activity of ALT in serum is stable at room temperature for 3 days and in the refrigerator for 3 weeks with less than 10% loss, but it is markedly decreased by freezing and thawing. The organ-to-serum activity ratio of ALT is 7600:1 in liver and 750:1 in muscle. Although day-to-day variation of serum ALT (10% to 30%) is comparable to that of serum AST, ALT also shows variability of approximately 45% during the day, with highest levels observed in the afternoon. Unlike AST, strenuous exercise results in 20% lower values of serum ALT.

In almost all liver diseases, ALT is higher than AST, except in alcoholic liver disease and advanced fibrosis. In alcoholic hepatitis, AST is greater than ALT, typically in a 2 to 1 ratio because alcohol increases the mitochondrial release of AST while concurrently decreasing the cytoplasmic production of ALT due to pyridoxine deficiency. In acute hepatocellular injury, the lower organ-to-serum activity and longer half-life of ALT in comparison with AST leads to a higher and immediate rise of AST followed by subsequent higher levels of ALT. Serum ALT has been shown to be a good indicator of overall health and mortality risk; in two recent studies, ALT levels correlated with mortality from all causes as well as death from cardiovascular disease.

“Biliary” Enzymes

ALP is a family of isoenzymes found in liver, bile duct epithelium, placenta, and intestine, and it occurs to a lesser extent in the kidneys and leukocytes. When stored in the laboratory, it is stable for up to 7 days in the refrigerator and several months in a freezer.

There is a day-to-day variation of 5% to 10% in the levels of ALP. Similar to AST, levels in African-American individuals appear to be 15% higher. ALP levels are higher in children and pregnant women; separate reference limits are therefore used for these two population groups. Although exercise does not appear to have an effect on ALP values, they are elevated after a fatty meal in individuals with blood type O or B.

ALP is elevated in the presence of biliary obstruction; chronic cholestatic diseases such as primary biliary cholangitis (PBC), primary sclerosing cholangitis (PSC), and drug-induced liver injury (DILI); infiltrative diseases such as sarcoidosis, granulomatous diseases, and diffuse metastatic disease; and space-occupying lesions such as polycystic liver disease. In situations of isolated elevation in ALP, either isoenzyme analysis of ALP by electrophoresis or concurrent elevation in gamma-glutamyl transpeptidase (GGT) and 5′-nucleotidase may be needed to confirm hepatobiliary origin. Elevated levels of GGT are found in a broad range of liver diseases; thus, although advocated as a marker for alcoholism, its utility is limited by lack of specificity for this condition. Elevated GGT levels can therefore be used only to support a suspicion of alcohol abuse rather than to make a definitive diagnosis.

Bilirubin , the principal pigment in bile, is derived from the degradation of hemoglobin. The daily formation of 250 to 350 mg of bilirubin is an intracellular process occurring within reticuloendothelial cells involved in the clearance of senescent erythrocytes. Bilirubin derived from this process passes from the reticuloendothelial cells into the blood, where it is bound to albumin to form unconjugated bilirubin. It is rapidly cleared from the blood in less than 5 minutes by hepatocytes and then conjugated with glucuronic acid by uridine diphosphate (UDP)-glucuronyl transferase to form a more water-soluble compound that can be excreted in bile. Under high-performance liquid chromatography (HPLC), bilirubin separates into four peaks ( Table 3.1 ): alpha, which is unconjugated bilirubin; beta, which is singly conjugated bilirubin (usually to glucuronic acid); gamma, which is doubly conjugated bilirubin; and delta, which is bilirubin covalently bound to albumin during times of liver injury or cholestasis. Accordingly, delta bilirubin (δ-bilirubin) has a half-life similar to serum albumin (17 to 20 days).

Conjugated and unconjugated bilirubin are often interchangeably used with direct and indirect bilirubin, respectively. Most clinical laboratories today measure total and direct bilirubin based on their reaction with diazo dyes and calculate indirect bilirubin. However, because direct bilirubin detected by this method also measures the delta-fraction, it is not an accurate measurement of conjugated bilirubin (see Table 3.1 ). Accordingly, assays have been developed to measure conjugated and unconjugated bilirubin, and it has been suggested that conjugated bilirubin is better than direct bilirubin at assessing recovery from liver injury. Determination of direct and total bilirubin is used in differentiating the cause of jaundice and hyperbilirubinemia. On the basis of the extent of contribution by either direct (>50%) or indirect (>80%) fraction to elevated total bilirubin, hyperbilirubinemia is termed conjugated and unconjugated , respectively. Hemolysis (elevated lactate dehydrogenase and low haptoglobin) and Gilbert syndrome commonly cause unconjugated hyperbilirubinemia and are not associated with any liver injury. Conjugated bilirubinemia is usually seen in liver diseases, both obstructive and hepatocellular.

Measures of Coagulation

PT and INR are clinically used to assess hepatic synthetic function because blood clotting factors II, V, VII, IX, and X are produced in the liver. PT is the time taken by a plasma sample to clot after adding tissue factor, a biologically obtained product. Clot formation is usually measured optically but may sometimes have to be measured mechanically to eliminate interferences in lipemic and icteric samples. Because of differences between batches and manufacturers of tissue factor, the INR was devised to standardize results. INR is the ratio of a patient’s PT to a normal (control) sample, raised to the power of the international sensitivity index (ISI) value for the control sample used. Because the tissue factors were previously calibrated in plasma from patients on vitamin K antagonists to assign the ISI value for conversion of PT to INR, investigators have proposed using the term INR (liver) where calibration is made by using plasma from patients with cirrhosis to calculate ISI (liver). Hemostasis is a fine balance between procoagulants and anticoagulants and despite a lowering of actual levels, the thrombin generation potential may remain intact in cirrhotic patients. Studies using whole blood clotting assays such as thromboelastography that reflect overall hemostatic balance may be superior for clinical use.

PT and INR are prolonged in chronic liver disease but not until more than 80% of the liver’s synthetic function is compromised, making them relatively insensitive markers in the evaluation of chronic liver disease. However, because factor VII has a short half-life of only about 6 hours, it is a sensitive and significant indicator of acute liver injury. Therefore it is used in the evaluation of fulminant liver injury as an indicator of rapid changes in liver synthetic function. Factors II, VII, IX, and X are vitamin K–dependent clotting factors; therefore it is important to remember that PT and INR may be prolonged not only by conditions that compromise liver synthetic function but also by those that compromise vitamin K absorption.

Other Tests

Albumin, blood ammonia, and platelet count are frequently used in the evaluation of liver disease. Although serum albumin is used as an index of liver synthetic function, it is affected by several nonhepatic factors, making interpretation a bit difficult. Moreover, albumin has a plasma half-life of 3 weeks, resulting in a slow change in serum concentration in response to acute alterations in hepatic function. Isolated hypoalbuminemia with no other liver test abnormality should raise suspicion of a nonhepatic cause. Measurement of blood ammonia is often performed in patients with cirrhosis and known or suspected hepatic encephalopathy presenting with altered mental status. However, blood ammonia levels, either arterial or venous, do not accurately correlate with mental status of patients with liver disease, and increased levels are not required to make the diagnosis of hepatic encephalopathy. Thus there is the variable use of clinical utility of blood ammonia level for monitoring therapy of cirrhosis patients with hepatic encephalopathy. The accuracy of the blood ammonia value is affected by factors such as fist clenching, tourniquet use, and placement of the sample on ice. Platelet count is lower in patients with cirrhosis and portal hypertension because of splenomegaly and depressed bone marrow production. Several noninvasive tests such as the AST-to-platelet ratio index (APRI), Forns test, FibroIndex, and FibroMeter use platelet count in their scoring algorithm to predict the degree of hepatic fibrosis.

Laboratory Investigation of Acute Liver Injury

Acute hepatic injury/acute liver injury is defined as the presence of abnormal liver tests for less than 6 months in a patient without preexisting liver disease. When the acute hepatic injury is associated with coagulopathy (INR ≥ 1.5) and encephalopathy, the terms acute hepatic failure and fulminant hepatic failure are used to indicate the severity of liver disease. The acute hepatic injury may be associated with jaundice or other constitutional symptoms of acute illness, and elevated liver tests accompany it. Liver test patterns are usually hepatocellular (R > 5) or mixed (2 < R < 5). The more common liver diseases that present as acute hepatic injury are viral hepatitis, autoimmune hepatitis (AIH), Wilson disease, alcoholic hepatitis, ischemic hepatitis, and drug-induced or herbal and dietary supplement–induced hepatitis.

The pattern of liver injury in viral hepatitis is mostly hepatocellular and characterized by an increase in AST/ALT greater than 5 × ULN, with ALP elevation less than 3 × ULN. Rarely, patients with hepatitis A virus (HAV) infection can develop a cholestatic pattern of liver injury characterized by an increase in bilirubin and ALP greater than 3 to 5 × ULN with only a mild increase in transaminases. The presence of anti-HAV IgM antibody confirms the diagnosis; this immunoglobulin usually disappears after 4 to 6 months and is replaced by anti-HAV IgG antibodies, which persists for life. Acute hepatitis B virus (HBV) infection is recognized by the presence of anti-core IgM antibodies and hepatitis B surface antigen (HBsAg). Anti-core IgG persists for years and helps differentiate immunity acquired through a previous infection (anti-HBs positive, anti-HBc IgG positive) from immunity acquired through vaccination (anti-HBs positive, anti-HBc IgG negative). Testing for antibodies against hepatitis delta virus (HDV) is indicated in patients with prior chronic hepatitis B who have acute hepatitis because of high-risk behavior. Acute hepatitis C virus (HCV) infection is recognized by detecting HCV RNA in serum because anti-HCV antibodies do not typically develop for 6 to 8 weeks after exposure and may take as long as 6 months (range, 2 to 6 months) to appear in serum. Serum aminotransferases become elevated approximately 6 to 12 weeks after exposure (range, 1 to 26 weeks); however, serum ALT levels are variable. Other types of viral hepatitis include those caused by hepatitis E virus (HEV) detected by anti-HEV antibody, cytomegalovirus (CMV) detected by CMV antigenemia or histologic presence of typical nuclear inclusions, and Epstein-Barr virus (EBV) detected by the Monospot test or IgM antibodies against EBV viral capsid antigen or EBV viral levels.

Drug-induced liver injury (DILI) typically has variable patterns of injury; however, certain drugs create signature patterns that help identify them as the causative agent. Acetaminophen overdose, either intentional or unintentional, results in very high transaminases (100 × ULN) and is associated with renal failure in approximately 10% to 15% of patients. The Hy law, or Hy rule, is often used by the United States Food and Drug Administration (FDA) as a way of assessing a drug’s risk of causing serious hepatotoxicity; an increase in bilirubin (to at least 2 × ULN) and a concomitant transaminase elevation (to at least 3 × ULN), with no competing etiology, is associated with a high rate of bad outcomes, including 10% to 50% mortality or transplantation. Alcoholic hepatitis typically shows marked elevation in bilirubin with only slight elevation in transaminases (<10 × ULN) and AST/ALT greater than 2; a history of active alcohol use or recent abstinence following heavy use is usually obtained. Hemodynamic instability in critically ill hospitalized patients results in ischemic hepatitis with a sudden rise in aminotransaminases (>100 × ULN). Dramatically improving aminotransferases in the recovery period are equally typical of ischemic hepatitis and help substantiate the diagnosis. On occasion, the recovery period after the hemodynamic insult may be associated with a cholestatic pattern, which is often (and inaccurately) termed “cholestasis of sepsis.” In situations of severe congestive heart failure, surreptitious acetaminophen intake, and arrhythmias, liver tests fluctuating at high levels (10 to 100 × ULN) may be observed. AIH may present acutely and is characterized by a hepatocellular pattern of liver injury with associated increase in the serum globulin fraction (total protein-albumin >3.5), increase in total IgG, and presence of autoantibodies. Rarely, Wilson disease may present as acute hepatitis associated with acute renal failure and Coombs-negative hemolytic anemia; the ALP-to-total bilirubin ratio is characteristically low (<2) in these patients. Extrahepatic biliary obstruction caused by either gallstones or malignancy often presents as acute liver injury with a cholestatic or mixed pattern of liver injury. An algorithmic approach to the diagnosis of acute liver injury is shown in Fig. 3.1 . The etiology of new-onset jaundice in a metropolitan area is shown in Table 3.4 .