|
|
||||||||
Enzymes and Protein Markers |
-glutamyltransferase, and mean corpuscular volume in teetotalers
1
Department of Clinical Neuroscience, Karolinska Institute, Center for Dependency Disorders at St. Görans & Karolinska Hospital, S-10229 Stockholm, Sweden.
2
Nova Medical Calab, St. Görans Hospital, S-11281
Stockholm, Sweden.
a Address correspondence to this author at: Alcohol & Drug Dependence Unit, St. Görans Hospital, S-11281 Stockholm, Sweden. Fax 46-8-6721994; e-mail anders.helander{at}bekl.csso.sll.se.
| Abstract |
|---|
|
|
|---|
-glutamyltransferase (GGT) in serum, and erythrocyte mean
corpuscular volume (MCV) were collected once every 12 weeks over ~5
months from 10 female and 4 male teetotalers. Mean values for serum CDT
(using the CDTectTM assay) ranged from 9.9 to 29.4 units/L
(median, 14.2 units/L), and the highest results were obtained in the
women. The mean values for serum GGT ranged from 0.15 to 0.49 µkat/L
(median, 0.30 µkat/L, or 18 U/L) except for one woman with a very
high mean of 3.07 µkat/L. For MCV, the mean values ranged from 79.5
to 91.5 fL. Two women showed several CDT results above the upper
reference limit (mean values, 27.6 and 29.4 units/L, respectively);
however, their GGT and MCV values fell within the reference intervals.
One of these women exhibited an increased total transferrin
concentration (mean value, 5.38 g/L), which was possibly related to the
use of oral contraceptives and/or a low serum iron concentration. When
the CDTect value was expressed relative to total transferrin, a ratio
within the reference interval was observed for this woman but not for
the other woman with increased CDTect values. The present study
demonstrates a considerable variation between individuals in CDT, GGT,
and MCV without drinking any alcohol. The results also show that these
baseline values are fairly constant over time within the same
individual. | Introduction |
|---|
|
|
|---|
-glutamyltransferase (GGT, EC 2.3.2.2), the major advantage is that
CDT produces considerably less false-positive test results (i.e., a
superior diagnostic specificity). Apart from being increased by chronic
drinking, there are many non-alcohol-related causes of a raised GGT,
such as severe liver disease, certain medications, obesity, and
smoking, thereby limiting its diagnostic specificity (3)(4)(5). The cutoff limit used to differentiate between health-related and abnormal values for biochemical markers is traditionally calculated as the mean plus or minus two standard deviations of the values in an apparently healthy control population (light/moderate social drinkers). It should be noted that this practice will yield a specificity of <100% because a small fraction of the control values will always lie outside the reference interval. Additionally, although the response in CDT to a given dose of alcohol is known to vary considerably between individuals, people with low baseline values of CDT may be able to drink much more alcohol on average than those with high baseline values before exceeding the upper reference limit. However, during long-term monitoring with repeated sampling in outpatient treatment programs, the sensitivity of biochemical alcohol markers can be enhanced considerably by introducing individualized reference limits instead of the conventional, population-based a priori cutoffs (6)(7). This strategy, which uses each individual's baselineor abstinencevalue as the starting point, not only facilitates early detection of relapse into heavy drinking but also improves the reliability of the test results.
When the stability of CDT over time was compared by sequential measurements in control subjects (light/moderate social drinkers and one teetotaler) and alcohol-dependent individuals undergoing treatment in an outpatient program, the single lowest coefficient of variation was observed for the teetotaler (7). This observation suggested that occasional intake of small to moderate amounts of alcohol can cause increases in the CDT concentration, leading to a greater intraindividual fluctuation (8). To determine the intra- and interindividual variability of CDT at baseline, i.e., without drinking any alcohol, serial monitoring was performed in teetotalers. For comparison, measurements of two other routinely available clinical tests of excessive drinking, GGT and the mean corpuscular volume of erythrocytes (MCV), were also evaluated.
| Materials and Methods |
|---|
|
|
|---|
Venous blood samples for determination of MCV and serum CDT and GGT were drawn from individuals once every 12 weeks over a period of ~5 months (between January and June). All samples were collected on the same time of day and by the same nurse; however, the sampling procedure was not further standardized. One man dropped out after only a few weeks, and the data for this subject were excluded from the study. The blood samples for MCV determination were kept at room temperature and analyzed within 4 h after they were drawn. Serum samples for CDT and GGT determination were frozen within 4 h after collection and stored at -20 °C until use. Samples were then thawed overnight at 4 °C, and analysis was performed in single analytical runs. The total number of blood samples obtained from each subject ranged from 15 to 22 (mean, 18 samples). Altogether, 252 blood samples were collected.
measurements
Measurement of CDT in serum was carried out in duplicate using a
commercial test kit (CDTectTM, Pharmacia & Upjohn
Diagnostics) based on the initial separation of transferrin isoforms by
microanion exchange chromatography on disposable microcolumns.
Quantification of CDT was carried out by a double antibody
radioimmunoassay. In this method, the CDT content is expressed as an
absolute amount (in units/liter; according to the manufacturer, 1 unit
of CDT in the CDTect assay refers to ~1 mg transferrin) of the
transferrin isoforms with pI values >5.7 (asialo-, monosialo- and part
of disialotransferrin). The reference limits between health-related and
abnormal CDT values were 20 units/L for men and 27 units/L for women.
The analytical precision (intraassay CV) of the method was 6.1%
(n = 15). To assess the possible influence on CDT of variations in
the serum transferrin concentration, total transferrin was quantified
in five samples from each subject (the samples were evenly distributed
throughout the 5-month collection period).
GGT, MCV, and serum transferrin (traceable to CRM 470) were determined at the local clinical chemistry laboratory, using accredited routine methods. The upper reference limit for GGT was 0.8 µkat/L for healthy women and 1.3 µkat/L for healthy men (1 µkat = 60 U); the intraassay CV of the measurements was 0.95% (n = 14) at 1.9 µkat/L. The corresponding limit for MCV was 96 fL for both men and women, and the interassay CV of the measurements was 0.75% (n = 12) at 87 fL. The standard reference interval for serum total transferrin was 1.943.26 g/L, and the intraassay CV of the method was 2.5% (n = 5).
| Results and Discussion |
|---|
|
|
|---|
|
|
In two of the women (subjects M and N), several CDT values exceeded the
upper reference limit used to indicate prolonged heavy drinking, the
respective mean values being 27.6 and 29.4 units/L. Both women showed
GGT and MCV values within the reference intervals. Although transferrin
synthesis and glycosylation are two distinct processes, an association
between serum CDT and total transferrin concentrations has been
reported (10)(13)(14). The results
of the present study (Fig. 2
a and Table 1
) confirm that the serum transferrin concentration
is fairly stable over time within the same individual (15).
However, two women (subjects L and N) exhibited transferrin
concentrations above the upper reference limit of 3.26 g/L for healthy
adult individuals (the mean values ± SD were 3.74 ± 0.24
g/L and 5.38 ± 0.18 g/L, respectively). These were the only
subjects taking oral contraceptives (estrogens), which are known to be
associated with a ~1216% increase on average in serum transferrin
synthesis (the reference interval for women using estrogens is
2.253.85 g/L) (16)(17)(18) and possibly also with the amount
of CDT (19), although this was not confirmed by others
(11)(20). If these two subjects were excluded
from the calculations, the interindividual CV value for serum
transferrin concentration was markedly reduced, thereby approaching
that reported by others (Table 1
) (21). When the ratio of
CDT (CDTect) to total transferrin was calculated in units/gram (Fig. 2b
), only subject M showed a relative value above the reference range
for healthy adults for this ratio (6.4 ± 1.7 units/g) reported by
Sorvajärvi et al. (14). This suggests that the
abnormal test result according to the CDTect assay for subject N can be
explained by the increased total transferrin concentration, which may
in turn be related to intake of oral contraceptives and/or a low serum
iron concentration (9.9 ± 3.2 µmol/L). However, the cause of
the high CDT (CDTect) values in subject M remains unknown but may
possibly involve partial coelution of the trisialotransferrin isoform
in the initial microcolumn fractionation step (22).
|
Although CDT is considered a very specific marker for recent excessive alcohol consumption, values outside the reference interval are occasionally found even without prior heavy drinking. Known or likely causes for false-positive CDT results include severe hepatic failure (usually primary biliary cirrhosis and chronic viral hepatitis), the uncommon transferrin D variant, a rare inherited defect in glycoprotein metabolism, and transplantation (1)(10)(23)(24)(25)(26)(27)(28). During pregnancy and, as indicated above, oral estrogen administration and also in iron deficiency anemia, there is a general increase in total transferrin concentration, which could yield false-positive results if absolute CDT values are used (29)(30). On the other hand, using relative values (%CDT) might, at least theoretically, yield false-negative results in women who abuse alcohol during pregnancy because most of the increase in transferrin synthesis is accounted for by the highly sialylated isoforms (15), leading to a gradual decline in relative CDT content over time (31). Whether the CDT result should preferably be expressed as an absolute amount (units or milligrams CDT/liter serum) or the amount relative to total transferrin (%CDT) has been a matter of some debate. In general, correlation studies with the two different ways of expressing the CDT result have shown good agreement, and the small variations in diagnostic efficiency observed are mostly not important in clinical practice (14)(32)(33). However, in the present study, expressing CDT as the relative amount helped to identify one individual who consistently showed false-positive results with the CDTect assay. In cases where the CDT result and the clinical experience are not congruent, an independent CDT method based on HPLC or isoelectric focusing should be recommended because the visible measurements of the latter procedures reduce the risk of obtaining false-positive results caused by genetic variants or other chromatographic interferences.
For many years, the activity of GGT in serum has been the most widely
used laboratory test for detection of alcohol abuse. The major
disadvantage of GGT is that it can be increased by a variety of other
conditions besides heavy drinking, thereby reducing its diagnostic
specificity. Nonetheless, many of the confounding factors are well
known today and can often be controlled for in clinical situations. In
the present study, the mean values for serum GGT ranged from 0.15 to
0.49 µkat/L (median value, 0.30 µkat/L, or 18 U/L), except for one
women (subject E) showing an abnormally high mean of 3.07 µkat/L
(Fig. 1b
). These results are in good agreement with earlier findings in
teetotalers (34)(35), and both intra- and
interindividual CVs were concordant with reference values from the
literature (Table 1
) (36). The woman with increased GGT
showed an initial value of 3.4 µkat/L, and the GGT value peaked in
the third sample (7.2 µkat/L, or 432 U/L), after which there was a
gradual decline until the last three specimens, which were 1.0, 0.8,
and 1.3 µkat/L, respectively. The reason for her abnormally high and
fluctuating GGT values is unknown; however, it may at least partly be
related to her weight (a body mass index of 30 kg/m,
compared with a reference value of
25 kg/m)
(37). However, her corresponding concentrations of CDT
(range, 812 units/L) and MCV (8892 fL) were stable and fell within
reference limits during the entire 5-month observation period (Fig. 1
).
An increased erythrocyte MCV is often observed in alcoholic patients,
and this hematological marker has been widely used as an indicator of
alcohol abuse (38). However, the sensitivity of MCV is much
too low to motivate its use as the sole evidence of heavy drinking, and
there are also several other explanations for increased values besides
excessive alcohol consumption. The mean values for MCV in the present
study ranged from 79.5 to 91.5 fL (Fig. 1c
). All test results fell
within the reference interval except for patient F, who showed three
consecutive values of 97 fL, i.e., being just above the upper reference
limit. The intra- and interindividual CVs were small (Table 1
), which
agree with previous results (36). The smaller variability in
MCV over time compared with the CDT, GGT, and transferrin values may be
related to the long life span of erythrocytes in the circulation
(~120 days).
The present study performed on healthy teetotalers demonstrates a considerable variation between individuals in their baseline values of the alcohol markers CDT, GGT, and MCV. The results also show that without drinking any alcohol the changes over time within the same individual are generally not very large.
| Conclusions |
|---|
|
|
|---|
| Acknowledgments |
|---|
| Footnotes |
|---|
-glutamyltransferase; and MCV, mean corpuscular volume of erythrocytes. | References |
|---|
|
|
|---|
-glutamyltransferase as markers of heavy alcohol consumption: gender differences. Alcohol Clin Exp Res 1994;18:747-754.
[Web of Science][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
![]() |
A. Helander, G. Eriksson, H. Stibler, and J.-O. Jeppsson Interference of Transferrin Isoform Types with Carbohydrate-deficient Transferrin Quantification in the Identification of Alcohol Abuse Clin. Chem., July 1, 2001; 47(7): 1225 - 1233. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Arndt Carbohydrate-deficient Transferrin as a Marker of Chronic Alcohol Abuse: A Critical Review of Preanalysis, Analysis, and Interpretation Clin. Chem., January 1, 2001; 47(1): 13 - 27. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Brathen, K. S Bjerve, E. Brodtkorb, and G. Bovim Validity of carbohydrate deficient transferrin and other markers as diagnostic aids in the detection of alcohol related seizures J. Neurol. Neurosurg. Psychiatry, March 1, 2000; 68(3): 342 - 348. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Helander Absolute or Relative Measurement of Carbohydrate-deficient Transferrin in Serum? Experiences with Three Immunological Assays Clin. Chem., January 1, 1999; 45(1): 131 - 135. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |