If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
This document presents the official position of the American Association for the Study of Liver Diseases (AASLD) on the application of serum alanine aminotransferase (ALT) activity, based upon an analysis of the currently available scientific data. Its authorship was selected by the Public Policy Committee. The document is fully endorsed by the AASLD Governing Board.
Physicians caring for patients with liver disease, predominantly hepatologists and gastroenterologists, have long been aware that measurements of liver enzyme activities (serum aminotransferases, including ALT [alanine aminotransferase] and AST [asparate amniotransferase]) are critical in the diagnosis and assessment of liver disease. These enzymes were formerly referred to as SGPT and SGOT, respectively. The serum ALT activity (hereafter termed ALT) has been regarded as a reliable and sensitive marker of liver disease. ALT may also be a good indicator of overall health, particularly in the context of obesity, the metabolic syndrome, and presence of cardiovascular disease, as many patients affected by these conditions also are at risk of having non-alcoholic fatty liver disease.
Despite all these considerations, abnormal ALT activity is often ignored or minimized by practitioners as most patients are asymptomatic. Minor elevations are often construed to be clinically insignificant, in part because of lack of a longitudinal perspective about the impact of abnormal ALT on long-term outcome such as end-stage liver disease or premature mortality. This document summarizes the position of the American Association for the Study of Liver Disease regarding ALT and includes review of its physiology, its distribution in health and disease, and its role as a screening and diagnostic test and clinical tool. Specifically, the significance of ALT measurements for determining general health, liver health and liver disease is addressed. The purpose of this document is to reinforce that the significance and etiology of a persistently elevated ALT must be evaluated regardless of the degree of elevation and to examine ALT as a population screening tool for early detection of liver disease.
Physiology of ALT
Alanine aminotransferase (ALT) is an enzyme that catalyzes the transfer of amino groups to form the hepatic metabolite oxaloacetate.1 It is composed of 496 amino acids, which are encoded by a gene located in the long arm of chromosome 8.2, 3 ALT is found abundantly in the cytosol of the hepatocyte. ALT activity in the liver is about 3000 times that of serum activity. Thus, in the case of hepatocellular injury or death, release of ALT from damaged liver cells increases measured ALT activity in the serum. Although it is generally thought to be specific to the liver, it is also found in the kidney, and, in much smaller quantities, in heart and skeletal muscle cells.
ALT released in the blood is catabolized in the liver with a resulting plasma half life of 47 ± 10 hours, which is considerably longer than that of AST (17 ± 5 hours).1 ALT activity varies day to day, by 10% to 30%. Within a given day, there is a significant diurnal variation, with ALT activities being up to 45% higher in the afternoon than in the early morning.4, 5
In acute hepatocellular injury, serum AST levels usually rise immediately, reaching a higher level than ALT initially, due to the higher activity of AST in hepatocytes and its release with liver injury. Within 24 to 48 hours, particularly if ongoing damage occurs, ALT will become higher than AST, because of its longer plasma half-life. In chronic hepatocellular injury, ALT is more commonly elevated than AST; however, as fibrosis progresses, ALT activities typically decline, and the ratio of AST to ALT gradually increases, so that by the time cirrhosis is present, AST is often higher than ALT.6, 7 One notable exception to the predominance of serum ALT activity in chronic liver disease is alcoholic liver disease where AST activity is generally higher than ALT levels.
AIH, autoimmune hepatitis; ALD, alcoholic liver disease; ALT, alanine aminotransferase; ANA, antinuclear antibofy; AST, aspartate aminotransferase; BMI, body mass index; HBeAg, hepatitis B e antigen; HBV, hepatitis B virus; HDL, high-density lipoprotein; MAST, Michigan alcoholism screening test; NAFLD, non-alcoholic fatty liver disease; SMA, smooth muscle actin; SMR, standardized mortality ratio; ULN, upper limit of normal.
Serum ALT as a Blood Test
ALT measurement affords a readily available, low-cost blood test that is utilized throughout the United States as a tool for detection of liver disease. ALT is a valuable screening test to detect otherwise inapparent liver disease, such as asymptomatic viral hepatitis and non-alcoholic fatty liver disease, both of which represent an epidemic that remains largely undiagnosed in the United States. Apart from liver disease, however, serum ALT activity may be affected by a number of factors not associated with hepatic necrosis. ALT levels differ with gender, with higher values in men than in women.8 Additional factors that affect serum ALT levels include body mass index (BMI) and triglyceride levels, regardless of gender.9–11 Total cholesterol levels and alcohol consumption among men have a positive correlation, whereas smoking, physical activity and age have a negative correlation with ALT levels.11–13 Glucose levels, in women, have a positive correlation with ALT activities, whereas use of oral contraceptives tends to lower ALT values.
Some of these correlations (such as BMI) may be explained by ALT being higher in people with fatty liver disease. Similarly, patients with hyperlipidemia or hyperglycemia may also have fatty liver disease, as a part of the metabolic syndrome. In light of the increasing prevalence of obesity in Americans, the distribution of ALT in the apparently healthy population has changed such that some patients with non-alcoholic fatty liver disease may have ALTs in the normal range as determined by the mean ± 2 standard deviations.9 In response, some physicians have advocated lowering the normal range.8
To the extent that there is a rough correlation between ALT and degree of hepatic inflammation in general, patients with high ALT levels tend to have more severe inflammation in the liver than those with normal ALT values. In contrast, the correlation between ALT and degree of hepatic fibrosis, the parameter that is most relevant to the prognosis of the patient, is not very strong, as exemplified by the common observation that a cirrhotic patient may have normal or only mildly elevated ALT.
ALT as an Indicator of Liver Disease
Since serum ALT levels rise in disease states that cause hepatocellular injury, serum ALT levels can effectively identify an ongoing liver disease process. The probability of clinically significant liver disease increases, particularly if the elevated ALT is associated with symptoms such as fatigue, anorexia or pruritus.
The utility of additional evaluation of a patient with asymptomatic elevation of ALT depends upon the findings of history and physical examination, the length of time that ALT has been elevated and the level of ALT elevation. Whereas one study suggested that the majority of asymptomatic people with elevated ALT do not have significant liver disease, a Scandinavian study of 151 consecutive patients with mild to moderate elevations of serum aminotransferase levels for at least 6 months revealed that liver disease was common.14, 15 Diagnoses included non-alcoholic steatohepatitis and hepatic steatosis (noted in 42%), chronic HCV (15%), alcoholic liver disease (8%) and autoimmune hepatitis, primary biliary cirrhosis and alpha1 antitrypsin deficiency in smaller numbers.
The level of ALT also guides the urgency and extent of further investigation. A serum ALT level less than 5 times the upper limit of the normal range should be rechecked before an extensive work-up is undertaken. If elevated ALT levels are confirmed and if they remain persistently elevated, additional work-up is indicated. ALT levels greater than 5 times the upper limit of the normal range suggest a potentially serious, active liver disease process and work-up should be initiated without waiting to confirm the persistence of abnormal ALT.
ALT levels greater than 15 times the normal range indicate severe acute liver cell injury and evaluation should be initiated immediately. The differential diagnosis for patients with severe acute liver injury (ALT levels >15 times the normal range) is relatively limited. Acute viral hepatitis (A-E), ischemic hepatitis or other vascular disorders such as acute venous outflow occlusion (Budd-Chiari), or toxin-mediated hepatitis should be considered. Acute autoimmune hepatitis, hepatic lymphoma or acute biliary occlusion may also present with highly elevated ALT activity. The diagnosis may be made upon historical grounds [ischemic episode, risk factors of acquisition of viral hepatitis, medication or hepatotoxin exposure (e.g., isoniazid) or overdose (e.g., acetaminophen)]. Blood testing (hepatitis and autoimmune serologies) may be helpful where applicable, whereas abdominal imaging may be helpful in other settings (e.g., venous outflow obstruction, biliary obstruction or abnormal lymphadenopathy).
Non-alcoholic Fatty Liver Disease (NAFLD).
NAFLD is probably the most common cause of abnormal ALT values among US adults and may affect up to 3% of the US population.16, 17 Risk factors for NAFLD include obesity, diabetes and hyperlipidemia.18 Elevated ALT may be a component of the metabolic syndrome, the hallmark of which is insulin resistance, manifested by hyperglycemia, hyperlipidemia, abdominal obesity and hypertension. The role of NAFLD as an increasing threat to public health is highlighted by the well-publicized trend in the proportion of overweight or obese Americans.19 Similarly, the prevalence of the metabolic syndrome is also increasing rapidly.20 In these patients, testing for ALT will facilitate timely diagnosis of NAFLD before irreversible fibrosis of the liver is established. Elevated ALT activities may be the only clue to this entity since there are no definitive blood tests to confirm the diagnosis. Furthermore, patients with high ALT among those with the metabolic syndrome may represent a subgroup with a propensity for systemic inflammation that may, in turn, increase the risk of atherosclerosis, leading to coronary artery or cerebrovascular disease.21
Elevated ALT levels may correlate with the severity of NAFLD. In a study in which 233 morbidly obese women were examined, 60% had some degree of hepatic fibrosis, and the majority of these patients had an elevated ALT value. Twenty-eight percent of patients with mild fibrosis and 68% of patients with advanced fibrosis had raised ALT activity. ALT levels were elevated in only 17% of patients without fibrosis.22 These observations are helpful in correlating elevated ALT with severity of liver damage. Therefore, ALT represents an excellent screening test to detect significant NAFLD.23
Alcoholic Liver Disease (ALD).
ALD remains the most common cause of liver-related morbidity and mortality in the United States.24 In alcoholic liver injury, AST activity is characteristically elevated in comparison to ALT activity, although mild elevation of ALT level is common.25 This is thought to be due to the longer half-life of mitochondrial AST released in response to alcohol and the coexistence of deficiency of pyridoxal-6-phosphate in alcoholics, which is a cofactor for the enzymatic activity of ALT.26 History of alcohol use should be ascertained by accurate questioning such as with the CAGE questionnaire27 or the MAST (Michigan alcoholism screening test)28 in all patients with serum aminotransferase elevations. Random blood alcohol level is sometimes useful in distinguishing ALD from NAFLD. The histology of ALD may be indistinguishable from that of NAFLD.29
Hepatitis C Virus (HCV) Infection.
Chronic HCV infection is the most common chronic blood-borne infection in the United States, affecting approximately 2% of the population.30, 31 However, ALT levels fluctuate in HCV and values may occasionally fall into the normal range.32 Since HCV infection is frequently asymptomatic, ALT elevations noted upon routine blood testing often stimulate the work-up whereby HCV infection is diagnosed. Sixty-nine percent of 248 asymptomatic blood donors who tested positive for HCV antibody had elevated ALT activity.33 Sixty-eight percent of patients positive for HCV RNA had elevated ALT levels, compared with 17% of those without detectable RNA. Patients with severe liver damage on liver biopsy in this cohort had at least 1 elevated ALT determination. Twenty-nine percent of HCV-infected patients with initially normal ALT values, when followed, will develop persistently elevated ALT levels, and 57% will develop transient elevation in ALT activities within 5 years.34 HCV patients with persistently normal ALT levels (at least 2 normal ALT values within 6 months) are more likely to be females35 and tend to have lower necroinflammatory and fibrosis scores on liver biopsy when compared to similar patients with elevated ALT activities.36, 37 Significant fibrosis was found in 8% to 20% of patients with normal ALT levels compared to 60% of patients with elevated ALT activities. While ALT analysis alone may fail to detect a minority of persons infected with HCV, it is most effective in detecting those persons whose liver disease is more severe. Such a characteristic enhances the value of ALT as a screening tool for detection of clinically important liver disease. Moreover, the sensitivity of ALT analysis can be improved with serial measurements and long-term follow-up.
Hepatitis B Virus (HBV) Infection.
Chronic HBV infection, a common etiology of elevated ALT values worldwide, afflicts at least 1.3 million individuals in the United States.38 Certain risk groups, such as individuals born in endemic countries, with a history of injection drug use, or on hemodialysis, may be identified in whom prevalence of HBV infection is particularly high in the United States.39 Chronic HBV infection is also frequently asymptomatic and is sometimes discovered because of an elevated ALT level identified upon routine blood testing. Among HBV patients, the level of ALT is associated with progression of liver disease and development of morbidity. The cumulative risk of development of complications is highest in patients with ALT values at least 1 to 2 times above the upper limits of normal (ULN).40 Among patients who are hepatitis B e antigen (HBeAg)-positive, ALT is also predictive of the likelihood of HBeAg seroconversion.41 Thus, in HBV patients, ALT is useful not only in determining the presence of significant liver disease and need for treatment but also in gauging the future course in the natural history of the infection.
The use of many medications has been associated with elevated ALT levels.42 Over-the-counter medications and herbal preparations are also implicated. If elevated ALT levels are confirmed, unnecessary medications should be discontinued, and ALT levels should be monitored. If ALT activity remains elevated, other etiologies should be sought. If the medication must be maintained for clinical benefit, ALT activity should be monitored. If ALT values continue to increase or are associated with development of symptoms or alteration of hepatic synthetic function, the offending medication must be discontinued.
Autoimmune and Cholestatic Liver Diseases.
Autoimmune hepatitis (AIH) may also be identified by recognition of mild to moderate elevations of ALT activity.43 Patients may be asymptomatic or have nonspecific symptoms such as fatigue and arthralgias. Once the diagnosis is confirmed with serologic testing such as antinuclear antibody (ANA) and smooth muscle antibody (SMA) and a liver biopsy, immunosuppressive therapy may be considered. Aminotransferase activity plays an important role in determining treatment candidacy and also treatment response in those who undergo immunosuppressive treatment. Although AST activity has been traditionally used in these criteria, ALT activity is important in these management decisions. ALT levels may also be variably elevated in cholestatic hepatic processes such as primary biliary cirrhosis or primary sclerosing cholangitis.44, 45
Metabolic Liver Diseases.
Mild elevations in ALT level may be noted in hereditary hemochromatosis, a relatively common genetic disorder of iron overload in people of Northern European descent.46 Elevated iron saturation and serum ferritin levels are usually present. Homozygosity of the HFE gene mutation (C282Y/C282Y) confirms the diagnosis, although a liver biopsy with iron quantification remains a useful diagnostic procedure to define the extent of liver injury and amount of iron deposition. Liver biopsy is informative in patients with elevated ALT levels, elevated serum iron studies and unremarkable HFE gene testing.47 Symptoms of the disease are not usually noted until the fourth or fifth decade in men and the fifth or sixth decade in women. It is particularly important to identify patients with hemochromatosis early in life, because liver injury can be prevented with periodic therapeutic phlebotomy. Clinical manifestations of disease can be avoided if treatment commences before complications occur. Mild elevations in ALT activity may also identify other less common genetic disorders such as Wilson disease and alpha-1-antitrypsin deficiency.48 Furthermore, mild elevations of ALT levels are also observed in the setting of celiac disease.49
ALT as a Measure of Overall Health and Mortality Risk
While ALT is useful as an initial test in detecting liver disease, emerging data highlight its potential value as a measure of overall health and survival. There is a strong relationship between ALT activity and mortality, even when the life-threatening process does not originate from the liver.
The strongest population-based data to address the association between elevated ALT values and subsequent mortality risk was based on a cohort of participants of a large health insurance program in Korea.50 In this study, there were 142,055 individuals of ages between 35 and 59 years in whom baseline demographic and laboratory data obtained between 1990 and 1992 were available. This cohort was followed up to 2000, when death certificates were used to determine survival and causes of death.
Figure 1 summarizes the impact of different levels of ALT on mortality. In men, 9% of the subjects had ALT ≥ 40 U/L, whereas only 5% of women had ALT ≥ 30 U/L. ALT activity, in men, correlated with higher mortality from all causes and liver disease. As expected, the effect of ALT was much larger on liver-specific mortality. For example, compared to those with ALT < 20 U/L, men with ALT ≥ 100 U/L had 59 times the risk of death from liver disease. In women, a similar trend was seen, but the number of subjects and events in the highest ALT category was small, making the risk estimation in this group imprecise.
In the same study, ALT activity correlated with the risk of cardiovascular mortality as well. Compared with those with ALT < 20 U/L, men with ALT ≥ 100 U/L had nearly 3 times the risk of death from cardiovascular causes. A similar trend was suggested in women, but the incidence of cardiovascular events was low, making extrapolation of the data in women more difficult.
A similar analysis has recently been undertaken in Olmsted County, MN.51 Based on a community-wide database, all county residents who had their ALT determined in the calendar year 1995 were identified and followed forward. Of 47,182 county residents who had healthcare encounters in 1995, 6,823 (14.5%) had their ALT measured. Of those, 5,912 had results within normal limits and 911 (13.4%) abnormal. The standardized mortality ratio (SMR) associated with ALT between 1 and 2 times the ULN was 1.21 (P = 0.23), whereas ALT greater than 2 times the ULN was 1.51 (P = 0.02). On the other hand, ALT less than the ULN was associated with lower risk of death than expected (SMR = 0.61, P < 0.01).
The question of ALT being a marker of cardiovascular health has recently been evaluated by Ioannou and coauthors, who used the third National Health and Nutrition Examination Survey to correlate ALT activity and risk of coronary artery disease in the general US population.52 Of 19,620 adult participants in the survey, 8,381 met the eligibility criteria for the study, which consisted of (1) lack of previous myocardial infarction or congestive heart failure and (2) complete laboratory data including ALT activity drawn after at least 8 hours of fasting. Of those, 7,526 did not have hepatitis B or C or history of excessive alcohol use. These included 7,259 subjects whose ALT was within normal limits (≤43 U/L) and 267 who had elevated ALT activities. When these 2 groups were compared to each other, those with elevated ALT activity had higher total cholesterol level, lower high-density lipoprotein (HDL) level, and higher blood pressure and were more likely to be diabetic. These and other risk factors for coronary artery disease were used in a formula (the Framingham Risk Score) to estimate the risk of developing coronary artery disease. Men with elevated ALT levels were estimated to have 1.3-fold increase in the risk of coronary artery disease within 10 years. In women, there was a 2.1-fold increase in risk.
These data highlight that ALT activity is predictive of future mortality in the general population. While mortality may be due to unrecognized liver disease, it may also be related to other risk factors for ALT elevation including obesity, serum cholesterol, and plasma glucose concentration, in addition to alcohol consumption, which are linked to non-liver health risks. The cardiovascular mortality risk associated with ALT activity described may in part be explained by the metabolic syndrome commonly present in patients with non-alcoholic fatty liver disease. Further, ALT may serve as a marker of a proinflammatory state that is associated with higher cardiovascular risk even among individuals with the metabolic syndrome.21, 53
ALT as a Population Screening Test
The biochemical, clinical and epidemiological information presented so far suggest that ALT may be useful as a screening test for early detection of asymptomatic liver disease and possibly for other causes of premature mortality. Screening is defined as the presumptive identification of unrecognized disease by tests, examinations, or other procedures which can be applied easily and conveniently.54 A screening test is not intended to be diagnostic; rather, it is designed to classify individuals with a high probability of disease from those with a low probability. In evaluating a screening test or program, the most widely recognized gauge is the criteria proposed by Wilson and Jungner.55 The following discussion applies the 10 items of the Wilson-Jungner criteria to ALT as a screening test for early detection of liver disease in the population.
1. The Condition Being Screened for Should Be an Important Health Problem.
The primary condition for which ALT is used to screen is chronic liver disease, which may ultimately lead to liver cirrhosis, end-stage liver disease and/or hepatocellular carcinoma. Chronic liver disease is the 10th leading cause of death in the United States.56 In addition, hepatocellular carcinoma, which almost exclusively occurs in patients with chronic liver disease, is one of the most common malignancies around the globe.57, 58 It is now well established that the incidence of and mortality from HCC in the United States has been increasing in the recent past, further highlighting the importance of chronic liver disease as a public health problem.59–61
2. The Natural History of the Condition Should Be Well Understood.
Progression of chronic liver disease is correlated with accumulation of hepatic fibrosis, eventually leading to cirrhosis, although the rate at which this progression occurs varies by the specific liver disease and by individual patients.62 Clinicians use diagnostic tests to evaluate the degree of fibrosis, which has most commonly been a liver biopsy, although noninvasive methods for this assessment are increasingly being developed. These techniques help clinicians gauge the progression of disease in individual patients.
3. There Should Be a Detectable Early Stage.
In most chronic liver disease, the disease span between the onset of disease and end-stage liver disease is measured in years and decades. This provides ample opportunities for screening with ALT to detect liver disease in a pre-cirrhotic stage.
4. Treatment at an Early Stage Should Be of More Benefit Than at a Later Stage.
Cirrhosis is known as an irreversible condition. Chronic liver disease amenable to effective therapy may be treated at an early stage to prevent progression to cirrhosis. In certain diseases, such as chronic hepatitis C, patients with advanced fibrosis have poorer response to therapy than in those with earlier stage disease.
5. A Suitable Test Should Be Devised for the Early Stage.
ALT is a suitable test to identify subjects with chronic liver disease in an asymptomatic phase. Most, if not all, chronic liver disease entails a component of hepatic parenchymal inflammation and hepatocellular degeneration, which ALT is thought to represent. Although high quality data to demonstrate the diagnostic accuracy of ALT in the detection of liver disease in the general population are lacking, aforementioned data by Prati or by Kim strongly suggest suitability of ALT as a test to detect chronic liver disease in its early stage.
6. The Test Should Be Acceptable.
As a simple blood test, ALT is as acceptable as many other established tests, such as mammography, colon cancer screening modalities, and serum cholesterol measurement.
7. Intervals for Repeating the Test Should Be Determined.
Patients with chronic liver disease are commonly diagnosed in their middle age or later and ALT testing sometime before the fifth decade of life may identify most asymptomatic patients with liver disease. However, further data are needed to delineate at what age(s) ALT screening must be performed or repeated to optimize the performance of a program of screening using ALT.
8. Adequate Health Service Provision Should Be Made for the Extra Clinical Workload Resulting From Screening.
Presently, elevated serum aminotransferase activities constitute 1 of the most common indications for hepatology or gastroenterology consult. No formal study is available to assess whether sufficient manpower exists to address increased demands that may arise if population-wide screening of ALT were to be instituted. A carefully constructed diagnostic algorithm to identify patients with chronic liver disease that may benefit most from hepatology consultation may alleviate extra clinical workload generated from such a screening program.
9 and 10. The Risks, Both Physical and Psychological, Should Be Less Than the Benefits. The Costs Should Be Balanced Against the Benefits.
These last 2 indicators address the risk–benefit and cost-effectiveness ratios of ALT as a screening test. Given its low cost and absence in general of immediate prospect for morbidity and mortality associated with abnormal ALT, the physical and psychological risks and economic costs of an ALT-based screening program likely compare favorably against the benefits of early diagnosis of chronic liver disease. However, formal studies assessing the risk–benefit and cost-effectiveness of population screening ALT have not been conducted.
ALT is an integral part of the evaluation of patients with liver disease. Its importance as a screening test for liver disease is highlighted by the fact that most patients with common liver diseases such as viral hepatitis B and C and non-alcoholic fatty liver disease have elevated ALT, even though they remain without symptoms to prompt a medical evaluation. Thus, although the interpretation and practical use of ALT analysis may differ across specific liver disease categories, ALT is a sensitive test to detect individuals with liver disease. The importance of ALT activity as an indicator of liver disease has recently been demonstrated in population-based studies which documented a strong association between ALT and subsequent mortality from liver disease.
Furthermore, emerging data suggest that ALT has a role as a predictor of mortality independent of liver disease. This association is generally construed to signify NAFLD as a component of the metabolic consequences of insulin resistance, which facilitates the development of atherosclerotic cardiovascular disease. ALT activity may be important not only as a marker of liver diseases but also as an indicator of general health.
Overall, although measurement of ALT is commonly performed as a part of the hepatic panel, the significance of this test may have been underestimated. In examining ALT as a screening tool for the population, we found that ALT meets most of the accepted criteria for a screening test. However, additional data will strengthen the rationale and inform optimal implementation of ALT screening. These include determination of the optimal schedule for ALT screening and assessment of the practical impact of its implementation as well as its cost-effectiveness. While we wait for these data, we highlight that ALT is an excellent screening test in individuals at risk of liver disease. Subsequently, an abnormal ALT result, as determined by a properly defined normal range, must trigger an appropriate clinical evaluation.
The authors thank the Public Policy Committee of AASLD for the opportunity to contribute to creating this document on its behalf. The committee consisted of Adrian M. Di Bisceglie, MD (Chair), Henry C. Bodenheimer, Jr, MD, Karen L. Lindsay, MD, Hal Yee, MD, John Goss, MD, Lee M. Kaplan, MD, Robert G. Gish, MD, W. Ray Kim, MD, Arun Sanyal, MD (Board Liaison).