| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
Department of Clinical Biochemistry, Odense University Hospital, DK-5000 Odense C, Denmark.
2
The Danish Center for Demographic Research and
Epidemiology, Institute of Public Health, University of Southern
Denmark, Main Campus: Odense University, DK-5000 Odense, Denmark.
3
Terry Sanford Institute, Duke University, Durham, NC
27708-0245.
4
Max Planck Institute for Demographic Research, Rostock
D-18057, Germany.
a Address correspondence to this author at: Department of Clinical Biochemistry, Odense University Hospital, DK-5000 Odense C, Denmark. Fax 45-6541-1911;
Lise.Bathum{at}ouh.fyns-amt.dk.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We conducted a survey among Danish twins, 73–102 years
of age, identified in the population-based Danish Twin Registry. Among
the 2749 individuals in the study population, an interview was
conducted with 79%. From these, a blood sample was collected from 290
same-sex twin pairs, total of 580 subjects, within a 6-month period and
analyzed for alanine aminotransferase (ALT), lactate dehydrogenase
(LDH), 
-glutamyltransferase (GGT), bilirubin, and albumin. The
interview included questions about alcohol consumption and body mass
index (BMI; self-calculated height and weight). Heritability
(proportion of the population variance attributable to genetic
variation) was estimated using structural-equation analyses before and
after correction for alcohol consumption and BMI.
Results: Structural-equation analyses revealed a substantial heritability (35–61%) for the four biochemical liver function tests: ALT, GGT, LDH, and bilirubin. The remaining variation could be attributed to individuals’ nonfamilial environments. Adjustment for alcohol consumption and BMI had no influence on the heritability for ALT, GGT, LDH, and bilirubin. For albumin, two models fit equally well before adjustment for alcohol and BMI: a model including additive genetic and nonshared environmental factors (AE), and a model including shared and nonshared environmental factors (CE). After adjustment, the model including shared and nonshared environment was clearly the best fitting model.
Conclusions: For both males and females, the effect of genetic factors on the biochemical liver function tests ALT, GGT, LDH, and bilirubin is substantial and accounts for one-third to two-thirds of the variation among individuals 73–102 years of age. The heritability is equal for males and females and does not change notably after controlling for alcohol consumption and BMI. For albumin, no major impact of genetic factors was found independent of BMI and alcohol consumption. An understanding of the genetic mechanisms underlying biochemical liver function tests among the very old may be of value in the interpretation of these tests in this age group.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-glutamyltransferase (GGT)
indicate the degree of cellular injury that has occurred during the
past few hours (1). GGT is believed to be the most suitable
of these liver function tests for examining long-standing excessive
alcohol consumption, because alcohol induces formation of the enzyme,
and for the recognition of cholestasis (2)(3)(4). The albumin
concentration indicates the protein synthetic function of the liver,
but albumin synthesis decreases very rapidly (in a few days) in persons
on a protein-deficient diet (5). Hypoalbuminemia in liver
disease indicates that a chronic liver disease is present
(1). Bilirubin is one of the most commonly used liver
function tests. Bilirubin is a metabolic breakdown product of heme
derived from senescent red blood cells. The liver conjugates and
excretes bilirubin into the bile (1). Increased
concentrations of unconjugated bilirubin have a high predictive value
in the diagnosis of many hepatobiliary disorders. Liver function tests are used for identifying patients with liver disease; in the differential diagnosis of jaundice; in monitoring the severity, course, and response to treatment of disease; and in detecting hepatotoxicity caused by drugs. Findings of transient, asymptomatic increases in liver function test results, particularly transaminase concentrations are common (6)(7). In particular, treatment with many common drugs [e.g., nonsteroidal anti-inflammatory drugs among others (8)] may cause increases in liver enzyme activity in the absence of clinical liver disease. Although these liver function tests have a low sensitivity for detecting liver diseases and the values frequently are temporarily increased, they are among the most frequently used biochemical tests in screening for diseases in older age groups. Abnormal values usually lead the physician to request a repeated evaluation and new testing, which is accompanied by concern on the patient’s part and additional expense. One study showed that 25–30% of asymptomatic workers screened by five liver tests (bilirubin, ALT, LDH, alkaline phosphatase, and aspartate aminotransferase) had serum concentrations exceeding the reference intervals (6).
Alcohol abuse usually is regarded as the most likely cause of increased liver enzyme activity (2). However, increased body mass index (BMI) has also been shown to be positively correlated with increased liver enzyme activity (9)(10). Furthermore, when changes in biochemical liver tests are studied over time, a significant increase has been shown in subjects who gain weight (11).
Hence, several environmental factors that influence biochemical liver
tests are known. Numerous genetic defects that influence biochemical
liver tests are also known (e.g., Gilbert syndrome, hemochromatosis,
and 
1-antitrypsin deficiency). However, the
overall impact of genetic factors on the most widely used
biochemical liver tests is unknown. Twin studies are among the best
methods to identify the effects of genes and environment. We evaluated
the extent to which different biochemical liver tests are influenced by
genetic and environmental factors, using elderly twins from the
Longitudinal Study of Aging Danish Twins (LSADT-97).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
BMI was calculated as weight (kg)/height (m2) based on self-reported data. BMI was categorized in three groups: <22, 22–28, and >28 kg/m2. The interview included questions concerning how many beers, glasses of wine, and drinks of strong alcohol the study subjects drank per week. These were summarized in one variable and categorized in three groups: 0, 1–14, and 15+ drinks per week. In the analysis, alcohol consumption and BMI were treated as categorical variables. Of the 290 same-sex twin pairs who donated blood samples, 257 pairs (88.6%) also provided information on both alcohol consumption and BMI for both twins in the pair. The age range for this group was 73–94 years.
laboratory assays
Serum samples for all study subjects were kept frozen at
-80 °C. The samples were thawed at room temperature, mixed, and
analyzed within the same day. All analytes were analyzed on the
Technicon AXON® System (methods nos.
SM4-2157C96, SM4-2172A94, SM4-2179A94, SM4-2142F90, and SM4-2131F90).
ALT was measured as described by Wroblewski and LaDue
(14) in accordance with the 1978 recommendations from the
IFCC (15). LDH was measured by a direct colorimetric
measurement of the change in oxidation state of NADH occurring by the
conversion of pyruvate to lactate (14)(16).
Total bilirubin was analyzed by measuring azobilirubin after the
addition of accelerators and diazotized sulfanilic acid
(17). GGT was measured by the optimized method described by
Shaw et al. (18). Albumin was measured by a direct
colorimetric procedure using the specific dye bromcresol green
(19). As quality control, a multirule Shewhart chart was
used (20). Within-assay CVs for the analyses during
the period were as follows: 4.7% for ALT, 6.6% for LDH, 3.7% for
bilirubin, 4.4% for GGT, and 3.1% for albumin. The reference
intervals for the various analytes were as follows: ALT, 10–50
U/L for males and 10–35 U/L for females; LDH, 200–700 U/L for
subjects >70 years; bilirubin, <20 µmol/L; GGT, 5–80 U/L for males
and 5–50 U/L for females; albumin, 36–46 g/L.
statistical analysis
In humans, two types of twinning occur: monozygotic (MZ) twins
share all genetic material; whereas dizygotic (DZ) twins, like ordinary
siblings, share, on average, 50% of their genes. In the classical twin
study, MZ and DZ intraclass correlations for a trait are compared. A
significantly higher correlation in MZ twins indicates that genetic
factors contribute to the variation. To estimate the heritability of
the biochemical liver function test (i.e., the proportion of the
population variance attributable to genetic variation), the twin data
were analyzed using structural-equation biometric models
(21). The usual assumptions of a classic twin study [no
gene-environment interaction or correlation, no assortative mating or
epistasis, and that the degree of environmental similarity is equal for
MZ and DZ twins (the equal environment assumption)] were made. For a
more detailed description of twin methodologies, see Neale and Cardon
(22), Plomin et al. (23), and McClearn et al.
(24).
It was assumed that the total phenotype variance (V) could
be decomposed as:
 |
where A refers to the variance contribution of additive
genetic effects, D refers to the variance contribution of
genetic effects attributable to dominance (intralocus interaction),
C refers to the variance contribution of shared
environmental effects (i.e., environmental factors that are shared by
twins reared together and are thus a source of similarity), and
E refers to the variance contribution of nonshared
environmental effects (i.e., environmental factors that are not shared
by twins reared together and are thus a source of random variation).
This component also includes measurement error, which in this case is
the analytical and intraindividual variation in the biochemical liver
function tests. Assuming that shared environmental effects contribute
equally to the traits of MZ and DZ twins, the expected twin covariances
are given by:
 |
 |
Variance components were estimated from the observed twin variances and covariances by the maximum likelihood method using the Mx software (21). To correct for unequal variances between twin 1 and twin 2, data were entered twice and the degrees of freedom adjusted accordingly (25).
Because the distribution of the data was positively skewed, all results from the liver function tests were transformed into symmetric distributions by taking the loge before analysis. Only intact pairs, i.e., pairs where results from the biochemical analysis and information regarding alcohol consumption and BMI were present for both subjects in the pair, were used in the statistical analysis.
The following models were fitted: ACE, ADE, AE, DE, CE, and E. The ACE
model contained an additive genetic component, a shared environmental
component, and a nonshared environmental component. The ADE model left
out the shared environment, assuming that it was too small to
contribute to the overall variation, but contained a dominant genetic
component. The AE model contained an additive genetic component and a
nonshared environmental component. The CE model contained only the
shared and nonshared environmental components assuming that genetic
factors do not contribute to the variation, and the E model tested
whether the differences in variation between the MZ and DZ twin pairs
could be explained by purely nonshared environmental factors. The fit
of each model was assessed by a goodness-of-fit
2 test that had between 6 (ACE) and 10 (E)
degrees of freedom.
The goal in model fitting is to explain the observed data as well as
possible with as few parameters as possible. The Akaike Information
Criterion (AIC) (26), which equals the 2
value minus twice the degrees of freedom, was used to find the best
fitting model. The model with the lowest AIC value reflected a balance
between goodness of fit and parsimony.
On the basis of the best fitting model, the final step in our twin analysis was the estimation of the proportion of variance in the biochemical liver function tests attributable to A, D, C, and E.
To test whether there was a sex difference in genetic and environmental factors, models in which parameter estimates were constrained to be equal across sex groups were compared with models in which estimates were allowed to differ among sexes.
All model-fitting analyses were done twice, before and after adjustment for alcohol consumption and BMI. Adjustment for alcohol consumption and BMI was assessed using residuals from sex-wise regression models.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. For technical reasons, it was not possible to test one
individual (MZ male) for GGT and one individual (DZ female) for LDH.
Both members of the pairs of these individuals were omitted from
statistical analysis. There were no major differences in the results
from the biochemical liver function tests between the different sex and
zygosity groups. However, GGT and bilirubin concentrations were
slightly higher in males than in females, and LDH concentrations were
slightly higher in females than in males. The MZ males had lower
LDH values than DZ males, but there was no difference in LDH
concentrations between MZ and DZ females.
|
In Table 2
, the data from single individuals, not pairs, are stratified by
alcohol consumption and BMI. Subjects without data for alcohol
consumption or BMI were categorized as missing. Only pairs for whom
information on BMI and alcohol consumption was provided for both twins
were used in the model fitting. As shown in Table 2
, this did not
produce any bias because the biochemical liver function test results
from subjects who did not provide this information did not differ from
subjects who provided this information. As can also be seen in Table 2
,
there was a difference in alcohol consumption and BMI between males and
females. The female twins had a tendency toward a lower BMI and
lower alcohol consumption compared with the male twins.
|
The intraclass correlations are shown in Table 3
. The correlations for BMI were much higher for MZ twins than
for DZ twins, indicating a high genetic component in determination of
BMI. For females, all MZ intraclass correlations were significantly
larger than 0 and consistently exceeded the corresponding DZ
correlations, except for albumin. The female twin correlations were
higher compared with the male twin correlations except for bilirubin
and albumin before adjustment for alcohol/BMI and for bilirubin after
adjustment. The intraclass correlations were almost unchanged after the
adjustment except for albumin for MZ males.
|
Biometrical analyses revealed that for ALT and GGT, a model including
dominant genetic factors and nonshared environment was the best fitting
model (lowest AIC; Table 4
). The fit of the full model estimating separate parameters in
the two sexes was compared with the fit of a constrained model
specifying equality of the genetic and environmental parameters across
the two sexes. Thus additive genetic (A) and shared
environmental (C) factors were not needed to account for the
observed data. For bilirubin and LDH, a model including additive
genetic factors and nonshared environmental factors was the
best-fitting model. Thus, the dominant genetic (D) and the
shared environment (C) factors were not needed to account
for the observed data. For albumin, two models fitted equally well
before adjustment for alcohol and BMI, the AE and the CE model. After
adjustment, the CE model was clearly the best-fitting model. This
change from the fit of two models (AE and CE) to the fit of one model
(CE) for albumin was the only effect of adjusting for alcohol
consumption and BMI. The models indicated that the heritability is the
same for the two sexes.
|
The analyses revealed a substantial heritability (35–61%) for the four biochemical markers ALT, GGT, bilirubin, and LDH, but a negligible heritability for albumin. This general pattern was unaltered when adjusted for alcohol consumption and BMI.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The adjustment for BMI and alcohol consumption was based entirely on
self-reported data. It could be suspected that subjects with a high BMI
or alcohol consumption are reluctant to answer questions regarding
these issues and that data would therefore be missing. As a
consequence, it is possible that these subjects and their co-twins
would be left out of the statistical analysis. This nonrandom omission
could introduce a bias. However, as seen in Table 2
, this did not seem
to be the case. The results from individuals were stratified in Table 2
as a function of alcohol consumption and BMI, and the results from
individuals with missing information on BMI and alcohol consumption did
not deviate from the other groups. Therefore, it seems likely that
subjects who did not provide information regarding BMI and alcohol
consumption did not differ from the rest of the group.
As seen in Table 2
, there are large differences in the distribution of
BMI and alcohol consumption between males and females. This was
accounted for in the statistical analyses by sex-wise regression
analysis. The correlation for BMI was considerably higher in MZ twins
than in DZ twins for both males and females, indicating a substantial
genetic contribution to the variation in BMI. This large genetic
component in BMI has been shown in several other studies
(27)(28)(29). The correlation for alcohol consumption in MZ
females was twice as high as the correlation in DZ females, whereas
there was no difference between the correlation in MZ and DZ males.
Previous twin studies have shown quite high heritability estimates for
alcohol consumption in both sexes (30)(31), and
this has been shown for both frequency of consumption and average
quantity consumed (31). The heritability estimates in these
studies were higher for females than for males, but it is still
surprising that there were no major difference in the correlations for
MZ and DZ males in our study.
Twin studies are based on the assumption that the degree of intrapair environmental similarity is equal in MZ and DZ pairs. If, in fact, the environmental similarity is greater among MZ twins than among DZ twins, an overestimation of the genetic influence will occur. However, it seems unlikely that shared environmental differences of this type should be of major importance for these twins, of whom the majority separated more than 50 years ago.
The estimate that one-third to two-thirds of the variation in biochemical liver function test results among our study population is attributable to genetic factors could be an underestimation for several reasons. We did not take the interviewed persons’ disease records into consideration, and the blood samples were collected 3–6 months after the interview. In addition, we did not use information regarding medical history in our models. This means that temporary deviations attributable to, e.g., current or recent acute diseases or different medications will introduce additional variability that will tend to decrease twin similarity. Furthermore, any measurement error in the biochemical analysis or interviewer effect on the information regarding BMI or alcohol consumption will also lead to decreased twin similarity.
In conclusion, the effect of genetic factors on the biochemical liver function tests ALT, GGT, LDH, and bilirubin is substantial and accounts for one-third to two-thirds of the variation among individuals 73–94 years of age, and the heritability is equal among males and females. The adjustment for alcohol consumption and BMI does not alter the correlations in this group of healthy, older twins. For albumin, there is no major impact of genetic factors. The biochemical liver function tests are widely used in the screening for diseases in older age groups, and an understanding of the genetic mechanisms underlying these tests may be of value in their interpretation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
-glutamyltransferase; BMI, body mass index; MZ, monozygotic; DZ, dizygotic; AIC, Akaike Information Criterion. 
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-glutamyltransferase [(-glutamyl)-peptide:amino acid -glutamyltransferase, EC 2.3.2.2]. IFCC methods for measurement of enzymes. Part 4. International Federation of Clinical Chemistry, Scientific Committee, Analytical Section, Expert Panel on Enzymes. J Clin Chem Clin Biochem 1983;21:633-646.[Web of Science][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  B. Kollerits, S. Coassin, S. Kiechl, S. C Hunt, B. Paulweber, J. Willeit, A. Brandstatter, C. Lamina, T. D Adams, and F. Kronenberg A common variant in the adiponutrin gene influences liver enzyme values J. Med. Genet., February 1, 2010; 47(2): 116 - 119. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Bose, V. S. Voruganti, M. E. Tejero, J. M. Proffitt, L. A. Cox, J. L. VandeBerg, M. C. Mahaney, J. Rogers, J. H. Freeland-Graves, S. A. Cole, et al. Quantitative Loci Regulating Plasma Levels of {gamma} Glutamyl Transferase and Albumin and Their Genetic Correlations with Cardiovascular Risk Factors Exp Biol Med, December 1, 2009; 234(12): 1519 - 1524. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kimura, J. v. B. Hjelmborg, J. P. Gardner, L. Bathum, M. Brimacombe, X. Lu, L. Christiansen, J. W. Vaupel, A. Aviv, and K. Christensen Telomere Length and Mortality: A Study of Leukocytes in Elderly Danish Twins Am. J. Epidemiol., April 1, 2008; 167(7): 799 - 806. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Ichihara, Y. Itoh, C. W. K. Lam, P. M. K. Poon, J.-H. Kim, H. Kyono, N. Chandrawening, D. Muliaty, and Science Committee for the Asian-Pacific Federation Sources of Variation of Commonly Measured Serum Analytes in 6 Asian Cities and Consideration of Common Reference Intervals Clin. Chem., February 1, 2008; 54(2): 356 - 365. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. B. Whitfield Serum {gamma}-Glutamyltransferase and Risk of Disease Clin. Chem., January 1, 2007; 53(1): 1 - 2. [Full Text] [PDF]
|
|
|
|
|
|
J. B. Whitfield, G. Zhu, J. E. Nestler, A. C. Heath, and N. G. Martin Genetic Covariation between Serum {gamma}-Glutamyltransferase Activity and Cardiovascular Risk Factors Clin. Chem., September 1, 2002; 48(9): 1426 - 1431. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
