Industrial Health 1998, 36, 218-222

Liver Transferase Activity in Healthy Japanese

Employees Aged 18-39 Years

Kazutoshi NAKAMURA'*f, Yutaka MOTOHASHI2, Shoko KIKUCHI',

Masatoshi TANAKA' and Shinichi NAKANO'

' Department of Hygiene and Preventive Medicine

, Fukushima Medical College, l Hikarigaoka, Fukushima 960-
1295, Japan
2 Department of Public Health
, Akita University School of Medicine, 1-1-1 Hondo, Akita 010-8543, Japan
Present address: Department of Hygiene and Preventive Medicine, Niigata University School of Medicine, 1-757

Asahimachi-dori, Niigata City 951-8510, Japan

Received November 17, 1997 and accepted February 9, 1998

Abstract: The aims of the present study were to identify correlates of alanine aminotransferase
(ALT or GPT) and aspartate aminotransferase (AST or GOT) activities among a healthy working
population aged 18-39, and to discuss liver transferase abnormalities. Subjects included 1,009
employees of a company in Fukushima, Japan. Pregnant women, employees exposed to organic
solvents, and employees with a history of liver diseases were excluded. Serum ALT and AST levels
were measured by an enzymatic method. Other information including BMI, job type and lifestyles
was recorded. Mean ALT and AST levels were significantly higher for males than females (P<0.001).
Multiple linear regression analysis revealed that sex and BMI explained 45% and 31% of the
variability in ALT and AST, respectively. The prevalence of abnormal ALT levels (>40 IU) was
16.3% for males and 0.4% for females. Sex, BMI, and shift work were independently associated
with abnormal ALT levels by logistic regression analysis. It is concluded that ALT and AST levels
are well-correlated with sex and BMI, and that abnormal liver transferase activity is prevalent in
male employees but rare in females, suggesting that liver function tests should be introduced for
male employees under 40 years of age.

Key words: Aaanine aminotransferase (ALT), Aspartate aminotransferase (AST), BMI, Liver enzyme,
Occupational epidemiology, Sex difference, Shift work

Introduction aged 35 years and 40 years or older5~. However, they have

not been required to test younger employees. Therefore,
Liver enzyme tests have been commonly used in there remains little known about liver transferase activity
occupational settings to screen for liver diseases such as and its correlates, including abnormalities, among those aged
fatty liver and hepatitis1_4~. Liver enzyme tests including 39 years or younger. Those information on liver transferase
alanine aminotransferase (ALT or GPT) and aspartate activity must be useful for health professionals in
aminotransferase (AST or GPT) are used in health check- occupational medicine to appropriately evaluate values of
up programs nationwide in Japan. liver transferase activity in young adults.
In Japan, employers have been legally obliged to perform The present study aimed to identify correlates of ALT
blood tests, including liver enzyme tests, for their employees and AST levels among a healthy working population aged
18-39 years, and to discuss the importance of liver transferase
*To whom correspondence should be addressed . abnormality in the working population.
LIVER TRANSFERASE ACTIVITY OF EMPLOYEES 219

Subjects and Methods

Subjects
Subjects included 1,186 (646 males, 540 females)
employeesaged 39 years or youngerof a computer and printer
assembly company in Fukushima, Japan. Fukshima has never
had a recorded viral hepatitis epidemic. Among them, 1,067
employeesparticipated in the 1995annual health examination
carried out in September. Fourteen pregnant women,.40
employees who had used and possibly been exposed to
organic solvents, and 4 employees who had had a history of
any liver disease were excluded from the analysis. As a result,
1,009 (523 males, 486 females) employees were studied.

Blood test Fig. 1. Percent frequency distribution of alanine aminotransferase

An overnight-fasting blood specimen was drawn from (ALT) for employees aged 18-39 years

998 subjects between 8:00-9:00 a.m. Serum ALT and AST

activities were determined by the enzymatic method6~with
Table 1. Demographic characteristics of the subjects aged 18-
RA-AST and RA-ALT reagents (Bayer Corporation, New 39 years by sex
York, USA). Details of the blood test procedures have been
described elsewhere'.

Other information
We measured body weight (kg) and body height (m). Body
mass index (BMI) was calculated as kg/m2. Subjects' age
and job type (white-collar and blue-collar) were transcribed
from their health record. Thirty-four male employees were
engaged in 3 shift work (morning, day, and night shift).
Alcohol consumption, smoking, and exercise habits were
determined via a self administered questionnaire. To estimate
usual alcohol consumption, subjects who drank any alcoholic
drink almost every day were asked about their alcohol
consumption. A pure ethanol volume of their averagevolume
of any kinds of alcohol taken were calculated by the volume
of an alcoholic drink multiplied by the ethanol content. The
variable `alcohol consumption' was then classified into 4
categories: 0, never; 1, chance drinker; 2, ethanol <60 ml/
day; 3, ethanol >60 mllday, where 60 ml of ethanol
approximately corresponds to the amount of ethanol in two
large bottles (633 ml) of beer. Smoking was coded as: 0,
never; 1, sometimes or ex-smoker; 2, every day; and the
exercise habit as: 0, never; 1, <3 per week; 2, >_3per week.
Statistical analysis
Normality was assessed for all continuous variables.
Distributions of ALT (Fig. 1) and AST were skewed to higher
values, and a natural logarithmic transformation was made
(termed log ALT and log AST, respectively). Student's t-
test was used to test the difference in mean values between
two variables. Multiple linear regression analysis with the
stepwise method was used to explore correlates of liver
transferase activity. Logistic regression analysis with the
stepwise method was used to find factors associated with
the presence of `liver transferase abnormality', which was
defined as liver transferase activity greater than the reference
value of 40U/L for both ALT and AST. For stepwise methods
in the multivariate analyses, independent variables were;
sex, age, BMI, shift work, job type, alcohol consumption,
smoking, and exercise habits.
Results
Demographic characteristics of the subjects by sex are
shown in Table 1. The mean age of male employees was 3
years older than female. The frequencies of drinking,
smoking, and exercise habits by sex are presented in Table
2. Male employeeswere more likelyto drink (df=3,, =207.4,
P<0.001), smoke (df=2, , =263.5, P<0.001), and to exercise
220 K NAKAMURA et al.

Table 2. Frequency of drinking, smoking, and exercise (df=2, %2=47.1,P<0.001) than female.
habits in the subjects by sex Table 3 illustrates ALT and AST levels stratified by age
and sex. In any age group, male employees showed
significantlyhigher values in both ALT and AST than female.
Results of multiple linear regression analysis are shown in
Table4. Log ALT was significantlyassociatedwith sex, BMI,
and exercise. Likewise, log AST was significantlyassociated
with sex, BMI, and job type. Combined partial R2of sex and
BMI was 46% and 32% for log ALT and 1ogAST,respectively.
Table 5 shows the proportion of subjects whose liver
transferase levels were greater than the laboratory reference
value of 40 U/L. In any age group except for teenagers,
male employees showed a significantly higher proportion
of liver transferase abnormality in both ALT and AST than
female employees. Only 2 of 486 women showed liver
transferase abnormality. Correlates of liver transferase
abnormality were demonstrated with logistic regression
analyses (Table 6). Liver transferase abnormality was

Table 3. Mean ±SD of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) by sex and age group

Table 4. Summary of multiple linear regression analysis (stepwise methods) with liver
transferase activities as dependent variables

Industrial Health 1998, 36, 218-222

LIVER TRANSFERASE ACTIVITY OF EMPLOYEES 221

Table 5. Proportion of subjects (%) whose liver transferase activity was greater
than the reference value (40 U/L)

Table 6. Summary of the logistic regression analysis with stepwise methods

significantly associated with sex and BMI in both ALT and associated with ALT activity. Despite the lack of any
AST. Adjusted odds ratios of male for abnormal ALT and independent associations, a cluster of such addictive
AST were 47.5 (95% Confidence Interval: 6.6, 343.6) and behaviors for males may enhance this sex difference. BMI,
28.8 (95% Confidence Interval: 3.9, 212.6), respectively. an index of obesity, explained an additional 11% of the
In addition, shift work was associated with AST abnormality. variability in ALT levels in this study. Consequently,
combined sex and BMI explained 45% and 31% of the
Discussion variability in ALT and AST, respectively, and thus these two
factors should be taken into account in the evaluation of
ALT and AST are distributed in many human tissues, with liver transferase activity in a subject.
the highest concentrations of ALT in the liver, and of AST There was a surprisingly remarkable sex difference; the
in cardiac tissue, the liver, and skeletal muscles8. Both prevalences of abnormal ALT and AST levels were 16.3%
transferases are used for the evaluation of hepatocellular and 9.1 %, respectively, for male employees, whereas they
disorders, and ALT is a more liver-specific marker8. were only 0.4% for both ALT and AST for females. From
Previous epidemiological studies" 3,4' 9-11>have reported Japan, Saito et al. 12)reported that the incidence of abnormal
that liver transferase activity is correlated with sex, age, and liver function was 14.7% for males and 3.8% for females
BMI. The results of this study were generally in line with by investigating subjects who visited a health examination
previous work with the exception of age. In our population center. However, such a setting may have a sampling bias,
aged <40, age was not a predictor for liver transferase activity. since subjects are already likely to have health problems.
Sex explained 35% and 29% of the variability in ALT and We targeted all subjects in a company and thus our study
AST, respectively. Kahn et al.9~ suggested that the sex was free from such a sampling bias.
difference may be explained by male-specific behavior as Logistic regression analyses revealed that sex, BMI, and
well as physiological sex differences. Our data did not show shift work were related to AST abnormality. They also
that alcohol drinking or smoking were independently demonstratedthat male employees were at least 29-fold more
222
likely to have liver transferase abnormalities than females.
Two possible explanations may account for this marked sex
difference. Male employees may indeed be a high risk group
for liver transferase abnormality, and/or the cut-off point of
the transferases (40 IU/L for both sexes) may be
inappropriate. Separate cut-off points for each sex may be
needed. In addition, the adjusted odds ratio of shift work
for liver transferase abnormality was greater than three.
What is the source of these liver transferase abnormalities?
Saito et a1.12~ reported approximately 40% of adults (20s
and 30s) had fatty liver diagnosed by the ultrasonography.
Should less serious fatty liver, which cannot be found by
ultrasonography,be taken into account, fatty liver is probably
the leading cause of liver transferase abnormality in subjects
aged 20-40 years. The influence of chronic hepatitis virus
infection is expected to be much smaller than that of fatty
liver. It has been reported, from areas of relatively high
prevalence of hepatitis virus in Japan, that the prevalence
of hepatitis B surface antigen (HBsAg) positivity in subjects
aged 20-40 years was less than 4%13),and the prevalence
of anti-hepatitis C virus antibody (anti-HCVAb) positivity
was less than 2% in this groupl3,14)
The present study among employees aged 18-39 found
that: 1) ALT and AST levels are well-correlated with sex
and BMI, and therefore health professionals in occupational
settings should carefully assess the liver transferase activity
of employees; 2) abnormal liver transferase activity may
occur in up to 15.9% of males whereas it is very rare in
females, suggesting liver function tests should be introduced
for male employees less than 40 years of age.
Acknowledgments
The authors wish to thank Ms Tomita for her helpful
assistance in data collection and Ms Sampei, Ms Honda,
and Ms Yusa for biochemical analysis.
References
1) NomuraF, Ohnishi K, SatomuraY, Ohtsuki T, Fukunaga
K, Honda M, Ema M, Tohyama T, Sugita S, Saito M,
Iida S, Okuda K (1986) Liver function in moderate
obesity: study in 534 moderately obese subjects among
4613 male company employees. Int J Obes 10, 349-54.
2) Wright C, Rivera JC, Baetz JH (1988) Liver function
testing in a working population:three strategies to reduce
false-positive results. J Occup Med 30, 693-7.
3) Salvaggio A, Periti M, Miano L, Tavanelli M, Marzorati
K NAKAMURA et al.
D (1991) Body mass index and liver enzyme activity
in serum. Clin Chem 37, 720-3.
4) Bums CJ, Boswell JM, Olsen GW (1996) Liver enzyme
activity and body mass index. J Occup Environ Med
38,1248-52.
5) Miura T, et al. editors (1994) Handbook of occupational
health. 2nd ed. Suppl. The institute for Science of
Labour, Tokyo.
6) International Federation of Clinical Chemistry,
Committee on Standards (1978) Provisional
recommendationsof IFCC methods for the measurement
of catalytic concentrations of enzymes. Clin Chem 24,
720-1.
7) Nakamura K, Shimai S, Kikuchi S, Tominaga K,
Takahashi H, Tanaka M, Nakano S, Motohashi Y,
Nakadaira H, Yamamoto M (1997) Shift work and risk
factors for coronary heart disease in Japanese blue-collar
workers: serum lipids and anthropometric
characteristics. Occup Med 47,142-6.
8) Hurtuk BL, Krefetz RG (1996) Enzymes. In: Clinical
chemistry: Principles, procedures, correlations. 3rd ed.
eds. by Bishop ML, Duben-EngelkirkJL, Fody EP, 207-
236, Lippincott, Philadelphia.
9) Kahn RA, Johnson G, Aach RD, Hines A, Ellis FR,
Miller WV (1992) The distribution of serum alanine
aminotransferase levels in a blood donor population.
Am J Epidemiol 115, 929-40.
10) Wejstal R, Hansson G, Lindholm A, Norkrans G (1988)
PersistentAlanine Aminotransferaseelevation in healthy
Swedish blood Donors: mainly caused by obesity. Vox
Sang 55,152-6.
11) Robinson D, Whitehead TP (1989) Effect of body mass
and other factors on serum liver enzyme levels in men
attending for well population screening. Ann Clin
Biochem 26, 393-400.
12) Saito Y, Yagyu K, Hattori Y, Ohno K, Okamoto N,
TakahashiA, Ohtsuka S, Ohtsuka T, Maeda K, Okamoto
K, Kato T (1989) A study on fatty liver in health
examination participants. Nippon Eiseigaku Zasshi (Jpn
J Hyg) 44, 953-61 (in Japanese).
13) Shirahama S (1993) Prevalence of chronic hepatitis B
and C virus infection in Kami-Goto-Town, Nagasaki
Prefecture, Japan. Acta Med Nagasaki 38,107-9.
14) Ueda R, Fujishiro K, Mizoue T, Takada K (1994) HCV
antibody positive rate observed in periodec checkup
in Kitakyushu. Sangyoigaku (Jpn J Ind Health) 36, 446-
7 (in Japanese).
Industrial Health 1998, 36, 218-222