Tests of the liver's capacity to transport organic anions and metabolize drugs

INTRODUCTION — A number of blood tests are available that reflect the condition of the liver [1-3]. The most common tests used in clinical practice include the serum aminotransferases, bilirubin, alkaline phosphatase, albumin, and prothrombin time. These tests are often referred to as "liver function tests," although this term is somewhat misleading since most do not accurately reflect how well the liver is functioning, and abnormal values can be caused by diseases unrelated to the liver. In addition, these tests may be normal in patients who have advanced liver disease.

Several specialized tests have also been developed (such as indocyanine green clearance), which, although uncommonly used in clinical practice, can measure specific aspects of hepatic function [4].

Despite their limitations, liver biochemical and function tests have many applications in clinical medicine:

●They provide a noninvasive method to screen for the presence of liver disease. The serum aminotransferases, for example, used to be part of panel of tests used to screen all blood donors in the United States for the presence of transmissible viruses before specific viral tests became available.

●They can be used to measure the efficacy of treatments for liver disease (such as immunosuppressant agents for autoimmune hepatitis). (See "Management of autoimmune hepatitis".)

●They can be used to monitor the progression of a disease such as viral or alcohol-associated hepatitis.

●They can reflect the severity of liver disease, particularly in patients who have cirrhosis. As an example, the Child-Turcotte-Pugh score, which incorporates the prothrombin time and serum bilirubin and albumin concentrations, can predict survival (table 1).

The pattern of abnormalities on these tests is more accurate than any of the individual tests. Elevation of serum aminotransferases indicates hepatocellular injury, while elevation of the serum alkaline phosphatase indicates cholestasis. Recognition of patterns that are consistent with specific diseases can prompt appropriate additional testing.

The liver biochemical and function tests that are used commonly in clinical practice and that are used occasionally for specific circumstances can be categorized as follows:

●Tests that detect injury to hepatocytes – Most of these tests measure the activity of hepatic enzymes, such as the aminotransferases, in the circulation. These enzymes are normally intracellular but are released when hepatocytes are injured. (See "Overview of liver biochemical tests".)

●Tests of the liver's capacity to transport organic anions and metabolize drugs – These tests measure the liver's ability to clear endogenous or exogenous substances from the circulation. The best studied include serum measurements of bilirubin and bile acids and a variety of dynamic tests, including caffeine and lidocaine metabolites, a variety of breath tests, and clearance tests such as bromsulphthalein (BSP) and indocyanine green (ICG).

●Tests of the liver's biosynthetic capacity – The most commonly performed tests to assess the biosynthetic capacity of the liver are the serum albumin and the prothrombin time (which requires the presence of clotting factors produced in the liver). Other tests that have been used are the serum concentrations of lipoproteins, ceruloplasmin, ferritin, and alpha 1-antitrypsin.

●Tests that detect altered immunoregulation, or viral hepatitis – These tests include the immunoglobulins, hepatitis serologies, and specific autoantibodies. Most of these substances are proteins made by B lymphocytes, not by hepatocytes. However, some are quite specific for certain liver diseases, such as antimitochondrial antibodies in primary biliary cholangitis. (See "Clinical manifestations, diagnosis, and prognosis of primary biliary cholangitis".)

This topic review will discuss tests of the liver's capacity to transport organic anions and to metabolize drugs, but will not discuss the serum bilirubin. Most of these tests are used rarely in clinical practice. The other categories of liver biochemical and function tests are discussed separately. (See "Approach to the patient with abnormal liver biochemical and function tests" and "Overview of liver biochemical tests" and "Enzymatic measures of hepatic cholestasis (alkaline phosphatase, 5'-nucleotidase, gamma-glutamyl transpeptidase)".)

DYE TESTS — Because the serum bilirubin is an insensitive and nonspecific measure of hepatobiliary disease, several other tests of hepatic excretory capacity have been developed. These include dye tests, measurement of the serum bile acid concentration, breath tests, and caffeine clearance. Although these tests are more sensitive than the serum bilirubin, they are nonspecific, more difficult to perform, and rarely used.

Bromsulphthalein — The use of bromsulphthalein (BSP, sulfobromophthalein sodium) for the assessment of liver function was first introduced in 1924 and continued to be used as an indicator of liver function until the 1970s [5]. BSP is soluble in water, binds rapidly to albumin and alpha-1-lipoprotein after intravenous injection, and is then avidly taken up by hepatocytes [6]. Infusions of BSP have been used to quantify the liver's excretory capacity by measuring the hepatic uptake, storage, and maximal biliary excretion rate of BSP [7]. Similar information can be obtained by determining clearance after a standard intravenous injection [8].

BSP is seldom used now, and has been withdrawn from commercial markets due in part to a decline in sales from fear about potentially fatal anaphylactic reactions after intravenous administration [9]. The test is still used occasionally in clinical investigations.

Indocyanine green — Indocyanine green (ICG) is another dye that is cleared exclusively by the liver and has been used to evaluate liver function [10]. The principle of ICG clearance is similar to that of BSP except that ICG has a higher hepatic extraction ratio than BSP and is bound to albumin and alpha-1-lipoprotein more avidly [6,11]. ICG has a few other advantages since it is not affected by fever and its clearance is normal in patients with the Dubin-Johnson syndrome and in neonates [10,12].

ICG can be measured directly by spectrophotometry. It can also be measured by dichromatic earlobe densitometry [10], but this method is not always reliable and the earlobe readings do not correlate closely with simultaneously obtained plasma values. An ICG meter with a fingertip optical sensor permits continuous measurement of serum ICG concentration [13]. Pulse dye densitometry technology has further improved ICG monitoring [14].

ICG clearance can be measured by giving a single intravenous injection (around 0.5 mg/kg) and determining a blood level 15 minutes later [15]. Although these low doses are safe, they are well below the threshold needed to saturate the liver's excretory capacity for ICG, which is more than 72 mmol/kg. As a result, they provide a less accurate measure of hepatic excretory function than BSP clearance. The test is used primarily in Asian rather than Western countries.

Use for estimating hepatic blood flow — ICG provides a good estimate of hepatic blood flow [16]. Infusion of ICG at a rate below the liver's capacity to clear it leads to a steady state within one hour, at which time the clearance rate is equal to the infusion rate. Hepatic blood flow can be estimated by use of the Fick equation:

 Estimated hepatic blood flow = R ÷ ([ICGa] - [ICGhv])

where R equals the hepatic removal rate of ICG, which equals the infusion rate at steady state, and ICGa and ICGhv equal the concentration of ICG in the hepatic artery and portal vein blood and in hepatic vein blood, respectively. ICGa is the same as the ICG concentration in peripheral venous blood since other organs do not clear ICG. It is impractical to obtain a direct measurement of ICGhv in most circumstances since it requires catheterization of the hepatic vein. Instead, the hepatic clearance rate is usually assumed to be 100 percent (because first-pass extraction of ICG by the liver is high), and therefore ICGhv is considered to be zero.

Although this method permits noninvasive testing, it underestimates hepatic blood flow because ICGhv is always more than zero when actually measured [16]. Despite this limitation, ICG infusion provides a good approximation of blood flow in healthy adults, but not in patients with cirrhosis who have a marked decrease in extraction of ICG [17,18].

Use in liver transplantation — A potential role for ICG elimination for optimizing use (estimating the optimal dose) of tacrolimus after liver transplantation has been suggested [19]. In another report, the addition of ICG measurement to the Model for End-stage Liver Disease (MELD) score improved the prediction of survival in patients with advanced cirrhosis compared with the standard MELD score or the MELD score combined with plasma sodium concentration (MELD-Na) [20]. With the availability of improved ICG monitoring, ICG clearance continues to be studied, particularly after liver transplantation and hepatic resection [14,21-23].

SERUM BILE ACID CONCENTRATION — Although sensitive tests to measure the serum bile acid concentration have been available for many years, their role in the evaluation of patients with suspected liver disease is not well established. Measurement of serum bile acids (cholic and chenodeoxycholic acids and their conjugates) may be helpful in patients with primary biliary cholangitis and primary sclerosing cholangitis and in women who are suspected of having intrahepatic cholestasis of pregnancy (ICP). (See "Intrahepatic cholestasis of pregnancy".)

Physiology — Bile acids are synthesized from cholesterol in hepatocytes, conjugated to glycine or taurine, and then secreted into bile. Approximately 80 to 90 percent of the secreted pool of bile acids is stored in the gallbladder between meals. The remaining 10 to 20 percent is secreted continuously into the duodenum, which accounts for the bile acids normally present in serum after a long fast (concentration of approximately 5 to 10 micromoles/L) [24].

The gallbladder contracts and discharges its pool of bile acids into the duodenum during a meal. Bile acids move rapidly down the intestinal tract, where they are actively absorbed by carrier-mediated transport in the terminal ileum and carried back to the liver by way of the portal vein [25]. A small proportion of bile acids are absorbed in the more proximal small intestine and colon by nonionic passive diffusion, and some pass into the feces.

The net effect of this enterohepatic recirculation is that a larger quantity of bile acids reaches the liver after a meal, and since the proportion extracted by the liver is constant, there is a two to fivefold elevation in the serum bile acid concentration compared with fasting values [26]. In healthy individuals, serum bile acids are derived entirely from intestinal input; none comes directly from the liver [27].

As a result of these processes, the serum bile acid concentration is affected by a number of factors including hepatic blood flow, hepatic uptake, secretion of bile acids, intestinal motility, and the intestinal microbiota [25,28]. Ethnicity and diseases that influence any of these functions can alter the serum bile acid concentration [29].

Measurement — There are several accurate methods to measure the serum bile acid concentration. These include enzymatic assays in which the bacterial enzyme 3-alpha-hydroxysteroid dehydrogenase is coupled to either fluorometric [30] or bioluminescence assays [31], gas-liquid chromatography [32], radioimmunoassay [33,34], and a highly specific assay combining gas-liquid chromatography and mass spectrometry [35].

Only the enzyme assays and radioimmunoassays can be performed easily by the clinical chemistry laboratory. Although the radioimmunoassays can be automated, a variety of individual radioimmunoassays that detect cholic and chenodeoxycholic acids and their conjugates would have to be used to measure the total concentration of bile acids in serum. This is usually not necessary if the purpose of the assay is to detect liver disease.

Clinical significance — The serum bile acid concentration may be disproportionately elevated in certain cholestatic liver diseases.

●Serum total bile acid concentrations can increase by 10- to 100-fold in intrahepatic cholestasis of pregnancy, and may be the first or only laboratory abnormality [36]. Serum cholic acid rises more than chenodeoxycholic acid, resulting in a marked elevation of the cholic/chenodeoxycholic acid ratio compared with pregnant women without intrahepatic cholestasis [37]. The ratio of glycine to taurine conjugates is decreased. However, the diagnostic accuracy of serum bile acid measurements in intrahepatic cholestasis of pregnancy may be overestimated [38]. (See "Intrahepatic cholestasis of pregnancy".)

●Serum bile acid levels are increased 25-fold in some patients with primary biliary cholangitis and primary sclerosing cholangitis [39]. Furthermore, the response to treatment with ursodeoxycholic acid can be monitored by measuring the concentrations of the different serum bile acids. With an adequate dose, ursodeoxycholic acid should eventually constitute approximately 50 percent of serum bile acids [39,40]. (See "Overview of the management of primary biliary cholangitis".)

As might be expected, the serum bile acid concentration is more sensitive than the serum bilirubin for all types of hepatobiliary disorders [41,42] and is as sensitive as serum aminotransferases in detecting acute viral hepatitis [43]. However, the serum bile acid concentration is less sensitive than serum aminotransferases as a screening test for detecting subclinical liver disease [44,45]. Serum bile acid levels are normal in patients with Gilbert syndrome [46,47] and Dubin-Johnson syndrome [48], which may help to confirm these diagnoses.

The serum bile acid profile is invariably abnormal in cirrhosis of any cause [42,49], a change that is as or more sensitive than the serum albumin or prothrombin time. This is due to the decrease in functioning liver cell mass, decrease in bile excretion, and portosystemic shunting usually present in chronic liver diseases, all of which affect serum bile acid levels.

It was once suggested that serum bile acid testing could predict histologic severity and might replace percutaneous liver biopsy in chronic hepatitis [50]. However, other investigators have found a poor correlation between serum bile acids and histologic severity in chronic hepatitis and alcohol-associated liver disease [41,51,52].

BREATH TESTS AND CLEARANCE TESTS — Breath tests were introduced as a practical means for evaluating hepatic functional reserve. Many breath tests are available, including 14C-aminopyrine [53], 14C- or 13C-galactose [54], 14C-phenacetin [55], 14C-diazepam [56] and 13C-methacetin [57,58] breath tests. Other tests, such as the clearance of galactose or caffeine or the formation of lidocaine metabolites, can provide similar information.

Although breath tests have been available for many years in the United States, they are still used infrequently since they are less convenient to perform than a simple blood test [59]. The equipment needed to perform the tests is not widely available [60]. Furthermore, diagnostic tests using 14C cannot be performed in children.

The principles and methodology underlying all of the breath tests are similar. A substance known to be metabolized primarily in the liver is labeled with 14C or 13C and is given orally or parenterally. As long as absorption is complete or reproducible, the oral route is more convenient. The labeled CO₂ exhaled in the breath is collected at various intervals in an alkaline medium that serves as a CO₂ trap. Metabolism of the labeled agent by the liver can be determined semiquantitatively by multiplication of the specific activity of exhaled 14CO₂ over a given time interval by a value for endogenous CO₂ output of 9 mmol/kg per hour.

An ideal drug for use in a breath test of hepatic drug-metabolizing capacity should have the following characteristics:

●The test drug must be metabolized primarily by the liver, a property which is often difficult to validate in humans.

●When the drug is given orally, its absorption must be rapid and complete or at least predictable.

●The drug should have a low hepatic extraction ratio (about 20 to 30 percent) so that changes in hepatic blood flow have little influence on its clearance from blood.

●The generated 14CO₂ should be distributed evenly in the body and not sequestered in an unavailable compartment.

●For ease of collection of 14CO₂, the drug should have a short elimination half-life.

●The drug must be safe.

Aminopyrine breath test — The greatest experience has been obtained with the 14C-aminopyrine breath test, which measures microsomal hepatic function [61]. The radioactive methyl groups of aminopyrine undergo demethylation and are eventually converted to formaldehyde, formate, and bicarbonate with exhalation of 14CO₂ in breath.

Studies in rats and humans have shown that the rate of 14CO₂ appearance in breath correlates with the hepatic mixed-function oxidase system mass [53,62,63]. The rate of 14CO₂ appearance can be influenced by drugs that affect this system, increasing after pretreatment with phenobarbital, which induces microsomal enzymes, and decreasing after the administration of disulfiram, which inhibits microsomal enzymes [53].

The presumption that the breath test is a measure of functioning hepatic microsomal mass is further supported by observations that patients with hepatocellular disease have low rates of 14CO₂ excretion, which correlate well with other tests that reflect functional mass (such as the prothrombin time, albumin, and serum bile acid concentration) [50,53,63-65]. Reduced aminopyrine breath test results also correlate well with the degree of necrosis and inflammation in liver biopsy specimens from patients with alcohol-associated cirrhosis [64] and with the presence of inflammation in patients with nonalcoholic fatty liver disease [66]. A similar test that uses 13C is also available and avoids exposure to radioactivity [67,68]. However, its use requires sophisticated and expensive mass spectrometers, which are not widely available.

Method — The aminopyrine breath test is usually performed after an overnight fast. A known dose of 14 or 13 C-aminopyrine is administered orally and breath samples are collected at 30-minute intervals for up to four hours. The expired CO₂ is trapped in alkali and counted. There is an excellent correlation between the percentage of administered 14C expired in two hours, the plasma clearance rate of aminopyrine, and the fractional disappearance rate of 14CO₂ [69]. As a result, it is usually satisfactory to rely upon a single sample collected at two hours. However, a single measurement at 60 minutes is more convenient for patients, and yields similar information [70].

Normal subjects excrete 6.6 ± 1.3 percent of the administered dose in the breath in two hours. The value is significantly lower in patients with hepatocellular injury. The results can be influenced by diet, folate deficiency, and use of other drugs.

The 13C-aminopyrine test is performed with serial breath samples obtained every 30 minutes for three hours [67]. Results are expressed as the percentage of the administered dose recovered over time.

Clinical significance — A single breath result cannot distinguish among different types of liver diseases. The degree of depression of breath test results overlaps considerably in all types of liver disease, including cirrhosis, hepatitis, hepatic cancer, and various histologic forms of alcohol-associated liver disease [63,70,71]. The test has been reported to have clinical value in quantifying residual functioning liver cell mass and in establishing prognosis in diseases such as alcohol-associated hepatitis [70], but it does not correlate with changes in the Child-Pugh class after transarterial chemoembolization in patients with hepatocellular carcinoma [72].

The aminopyrine breath test appears to be as sensitive for the detection of hepatocellular disease as the serum aminotransferases and more sensitive than the prothrombin time and serum bilirubin and albumin concentrations [60,71]. It may be more specific for detecting histologic severity in chronic hepatitis and alcohol-associated liver disease than conventional liver biochemical and function tests [64,73]. Sensitivity and specificity were 86 and 68 percent for detecting cirrhosis in a study of 61 patients with various causes of liver disease (of whom 21 had compensated cirrhosis) undergoing the 13C breath test [67].

The results may also correlate with survival in patients with alcohol-associated hepatitis [74] and progression of fibrosis in patients with chronic hepatitis C [75]. In one study, for example, the aminopyrine breath test predicted short-term survival and clinical improvement of patients with alcohol-associated hepatitis more reliably than conventional liver biochemical and function tests [74]. However, studies evaluating the utility of the aminopyrine breath test for predicting survival in patients with cirrhosis have produced conflicting results [76-79].

Galactose clearance and breath test — Galactose clearance is another method to estimate functional hepatic mass and cytosolic hepatic function [80]. After intravenous administration of galactose (0.5 mg/kg), serial blood samples are collected and assayed. The rate of galactose clearance is markedly reduced in patients with cirrhosis and chronic hepatitis but is rarely abnormal in patients with biliary obstruction [81]. Galactose clearance has been proposed as a prognostic marker in patients with severe acetaminophen-induced hepatotoxicity [82].

A breath test has also been developed using 14C-galactose [54], producing results similar to the intravenous galactose clearance test. However, the utility of these tests is uncertain. Neither is better than the serum albumin in distinguishing healthy subjects from patients with cirrhosis [54,83], and galactose clearance appears to add little to standard laboratory studies in predicting outcome in patients with acute liver failure [84] or cirrhosis [85,86].

A 13C-galactose breath test is also available. The sensitivity and specificity for detecting cirrhosis were 71 and 85 percent, respectively, in a study of 61 patients with various causes of liver disease (21 of whom had compensated cirrhosis) [67]. These values were 100 and 93 percent, respectively, when combined with the results of a 13C-aminopyrine breath test.

Caffeine clearance — Measurement of the rate of caffeine clearance is another method to quantify functional hepatic capacity and microsomal function [87]. The test is usually performed by the ingestion of oral caffeine (280 to 366 mg) followed by measurement of the plasma caffeine concentration at 1, 2, 3, 6, 12 and 24 hours [88,89]. In a variation of the method, caffeine is taken at noon and the concentration of caffeine is measured from samples of saliva collected at bedtime and the next morning upon awakening [89]. Caffeine can also be measured in scalp hair [90] and (using 13C-caffeine) in breath [91].

The results of caffeine clearance tests are similar to those obtained with the 14C-aminopyrine test without the need for radioisotopes or breath collections. The test can distinguish decompensated cirrhosis from noncirrhotic liver disease but does not correlate with the Child-Pugh class or MELD score [92]. Smoking is a confounding factor, since it increases caffeine clearance, while increasing age and certain drugs may reduce caffeine clearance [88].

Lidocaine metabolite formation — Tests of lidocaine metabolism take advantage of the conversion of lidocaine to monoethylglycinexylidide (MEGX) within the hepatic cytochrome P450 system. Fifteen minutes after the intravenous administration of lidocaine (1 mg/kg), a serum sample is taken for determination of the MEGX concentration, which is measured using a fluorescence polarization immunoassay [93].

Numerous studies have assessed the prognostic value of this test. In one, the survival of patients with cirrhosis who had MEGX concentrations greater than 30 ng/mL serum was significantly greater than that of patients with lower values [94]. Another study found that a decline in the MEGX concentration correlated with histologic worsening of chronic hepatitis and cirrhosis and may therefore provide a noninvasive means of following the course of chronic liver disease [95]. A limitation of the test is inhibition by co-administered drugs that are also metabolized by cytochrome P450 3A4 [96].

The MEGX test may also have a role in identifying suitable organs for liver transplantation. Good early graft function has been associated with higher donor MEGX concentrations 15 minutes after the infusion of lidocaine [97-99]. Unfortunately, overlapping values [99] and the shortage of organs currently make this approach impractical.

SUMMARY AND RECOMMENDATIONS

●Serum liver biochemical tests – A number of blood tests are available that reflect the condition of the liver. The most common tests used in clinical practice include the serum aminotransferases, bilirubin, alkaline phosphatase, albumin, and prothrombin time. These tests are often referred to as "liver function tests," although this term is somewhat misleading since most do not accurately reflect how well the liver is functioning and abnormal values can be caused by diseases unrelated to the liver. In addition, these tests may be normal in patients who have advanced liver disease. (See "Approach to the patient with abnormal liver biochemical and function tests".)

●Dye tests – Because the serum bilirubin is an insensitive and nonspecific measure of hepatobiliary disease, several other tests of hepatic excretory capacity have been developed. These include dye tests, measurement of the serum bile acid concentration, breath tests, and caffeine clearance. Although these tests are more sensitive than the serum bilirubin, they are nonspecific, more difficult to perform, and rarely used. (See 'Dye tests' above.)

●Serum bile acid concentration – Although sensitive tests to measure the serum bile acid concentration have been available for many years, their role in the evaluation of patients with suspected liver disease is not well established. Measurement of serum bile acids (cholic and chenodeoxycholic acids and their conjugates) may be helpful for early diagnosis in patients with primary biliary cholangitis and primary sclerosing cholangitis and in women who are suspected of having intrahepatic cholestasis of pregnancy. (See "Intrahepatic cholestasis of pregnancy" and 'Serum bile acid concentration' above.)

●Breath tests and clearance tests – Breath tests were introduced as a practical means for evaluating hepatic functional reserve. Many breath tests are available, including 14C-aminopyrine, 4C- or 13C-galactose, 14C-phenacetin, 14C-diazepam, and 13C methacetin breath tests. Other tests, such as the clearance of galactose or caffeine or the formation of lidocaine metabolites, can provide similar information. Although breath tests have been available for a number of years in the United States, they are used infrequently since they are less convenient to perform than a simple blood test. The equipment needed to perform these tests is not widely available. (See 'Breath tests and clearance tests' above.)