| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters |
1
Department of Medical Sciences, Clinical Chemistry, University Hospital, SE-751 85 Uppsala, Sweden
2
Box 575, SE-296 21 Åhus, Sweden
aAuthor for correspondence. Fax 46-18-552562; e-mail anders.larsson{at}clm.uas.lul.se.
To the Editor:
We read with great interest the articles by Dufour et al., "Diagnosis and monitoring of hepatic injury. I. Performance characteristics of laboratory tests" (1) and "Diagnosis and monitoring of hepatic injury. II. Recommendations for use of laboratory tests in screening, diagnosis, and monitoring" (2). These articles address an important but somewhat controversial subject. Broughton and Worthington (3) reported that an order for liver function tests produced 17 different test combinations in 19 different laboratories. This shows that there is no real consensus on the use of liver function tests. We thus want to add some remarks to the articles regarding aspartate aminotransferase (AST) and alanine aminotransferase (ALT). In our opinion, the use of AST can be reduced, and we think that it is not necessary to order AST with ALT in routine practice.
The information presented by Dufour et al. (1) on factors affecting the transaminases is important, but drug effects must be added. Several hundred drugs have been reported to affect AST or ALT (4).
Many laboratories still offer "standardized liver test panels". In our opinion, laboratory investigations should always be ordered selectively, depending on the clinical situation. Several reports have shown that a high AST value in relation to the ALT value is associated with alcohol-dependent liver damage. It is known that alcoholics often suffer from minor trauma and that alcohol affects other organs in addition to the liver. Part of the AST activity may thus be from other organs. Such an increase in AST is probably not attributable to muscle injury because creatine kinase typically is only slightly increased in these patients (5). Patients with suspected liver damage also have damage to other organs (e.g., trauma patients and patients with autoimmune diseases) and thus have high AST/ALT ratios that may be interpreted as alcohol dependent.
Several of the studies showing a high AST/ALT ratio were performed many years ago. In the older studies, ALT may have been analyzed without addition of pyridoxal-5'-phosphate. Alcoholics have low pyridoxal-5'-phosphate, which in animal models is associated with decreased ALT activities (6). The high AST/ALT ratio seen in alcoholics was at least partly attributable to low pyridoxal-5'-phosphate (7). At present, all AST and ALT tests in Sweden are preincubated with pyridoxal-5'-phosphate. This should cause a reduction in the AST/ALT ratio. Were the ratios presented in Table 1 of Ref. (2) analyzed in the presence of pyridoxal-5'-phosphate? What are the specificity and sensitivity of the AST/ALT ratio in an emergency ward setting with modern AST and ALT assays?
Serum AST decreases more rapidly than ALT because of a shorter half-life. This means that the AST/ALT ratio changes over time, which further reduces the usefulness. To be able to evaluate the ratio, we thus need to know when the damage occurred. If the time of damage is known, it is likely that the cause is also known. Patients with viral hepatitis have a low AST/ALT ratio in comparison with alcoholics (2). Could this be at least partly attributable to differences in disease duration in previous studies?
In several intervention studies in Sweden, we have tried to optimize the use of clinical chemistry investigations in primary care. In a primary care setting we have recommended that ALT is sufficient to screen for liver damage. The recommendations produced a reduction in AST ordering. Initially, AST and ALT tests were ordered with equal frequency, and this decreased to 0.23 AST tests per ALT ordered (8). In the rheumatology outpatient unit in Uppsala, the ordering ratio decreased to 0.2 (9), and in the University Hospital in Odense to 0.006. We believe that the use of AST should be questioned in a dialog with our clinical colleagues (10). Our experience is that discussion of the value and use of AST leads to a significant reduction in the ordering of AST tests.
One of our most important tasks as clinical chemists is to have good communication with clinicians for the benefit of the patients. The articles by Dufour et al. (1)(2) are very good examples of such work.
References
Veterans’ Affairs, 50 Irving Street NW, Washington, DC 20422
aAuthor for correspondence. Fax 46-18-552562; e-mail anders.larsson{at}clm.uas.lul.se.
To the Editor:
I appreciate the comments of Larsson and Tryding on our guidelines for use of laboratory tests in hepatic injury. It is of interest to see the perspective and experiences of other communities in approaching appropriate test utilization.
We did not include a discussion of drug effects in the tables on preanalytic variation of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) because drugs do not increase plasma activities of these enzymes except by causing hepatic injury. We did, however, include drug-induced injury in the differential diagnosis of both acute and chronic liver injury and emphasized the importance of taking a drug history. We agree with the authors that this is an important consideration.
The issue of whether to include both AST and ALT in "standardized liver tests panels" was of considerable interest to our group. Our position, as stated in the guidelines, is that both enzymes, along with several other tests, should be included in a "hepatic panel" (1)(2). We agree, however, that there are cases where a complete hepatic panel is not needed. For example, we did not recommend use of AST or ALT for monitoring patients with acute hepatitis once they have begun to decrease. We also stated that ALT is the most important test in screening for chronic hepatic injury.
There is considerable evidence, however, that AST and ALT together are of use in the differential diagnosis of acute hepatic injury and in monitoring patients with chronic hepatic injury. In acute hepatic injury, AST typically is increased to a greater degree than ALT in the very acute stages of injury, as with acetaminophen or ischemic liver injury. Recognition of this ratio often provides a clue to the presence of one of these two etiologies. As pointed out by Larsson and Tryding and in our guidelines, a high ratio is typical of alcoholic hepatitis. Although they suggest that modern assays with pyridoxal-5'-phosphate may not show such a ratio, the article by Matloff et al. (3) showed that correction of plasma pyridoxine deficiency did not abolish the high AST/ALT ratio in alcoholic hepatitis. In unpublished studies of patients with hepatitis C, we have consistently seen a much higher AST/ALT ratio in persons with both alcohol abuse and hepatitis C than in those with hepatitis C alone, and in 7% of samples, only AST is increased. Furthermore, as indicated in our guidelines, an increasing AST/ALT ratio is relatively specific for development of cirrhosis in patients with chronic hepatitis C infection, and may be present long before clinical symptoms of decompensation develop. Thus, we believe that use of both AST and ALT is indicated in monitoring persons with chronic hepatic injury.
We do agree with the authors that "routine" use of both AST and ALT in screening tests is excessive. We applaud their intervention efforts in reducing AST ordering after consultation with clinical colleagues. For screening purposes, as our guidelines indicate, ALT is a superior test and can usually be used alone. In patients with known liver disease, we believe that use of both tests provides additional clinical information that is worth the small additional cost.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
